ICES Journal of Marine Science: Journal du Conseil

Hidalgo, Manuel; Browman, Howard I

doi: 10.1093/icesjms/fsz067pmid: N/A

Abstract Recent estimates suggest that the mesopelagic zone could contain a total fish biomass of 2-19.5 gigatonnes, roughly equivalent to 100 times the annual catch of all existing fisheries. In addition to the possibility of direct consumption of mesopelagic species, there is interest in their use for fishmeal, as a source of dietary supplements for humans, and to bio-prospect pharmaceuticals. All of this, and the demands for a global food supply that can feed an ever-growing population, has driven interest in the mesopelagic. Thus, accurate quantification of the biomass of mesopelagic resources, their nutritional and genetic composition, their links to other components of the food web, to other oceanic realms and to biological and chemical oceanographic processes and cycles, are the focus of growing research activity. This information is needed to ensure the sustainable management of these resources. In this introduction, we summarize the contributions included in this theme set and provide some “food for thought” on the state-of-the-art in research on the mesopelagic, including identifying the knowledge that must be generated to support its sustainable management (e.g. the effect that extracting significant biomass might have on the pelagic ecosystem and the flow of material and energy through it). Background and motivation for this article theme set Until recently, there was comparatively little research activity on the mesopelagic zone (Figure 1a), and what there was focused on biological oceanographic processes such as carbon flux and on archaea and bacterioplankton (Table 1). Nonetheless, large mesopelagic organisms (mainly fish and squid) have long been proposed as potentially harvestable resources (e.g. FAO, 1997, 2001). Recent estimates suggest that the mesopelagic between 70°N and 70°S could contain a total fish biomass on the order of 9–19.5 gigatonnes (Gt; Irigoien et al., 2014; Proud et al., 2017,, 2019), roughly equivalent to 100 times the annual catch of all existing fisheries. However, other estimates are considerably lower =2.4 to <1.4 Gt (Jennings and Collingridge, 2015; Anderson et al., 2019). In addition to the possibility of direct consumption of mesopelagic species, there is interest in their use for fishmeal, as a source of “Omega-3” oils (as a dietary supplement for humans), and to bio-prospect for other nutraceuticals and pharmaceuticals. All of this, and the demands for a global food supply that can feed an ever-growing population (e.g. Springmann et al., 2018), has driven a renewed interest in the mesopelagic (see IMR et al., 2017; “Blue Growth Strategy,” European Commission, Directorate-General for Maritime Affairs and Fisheries, Director-General, 2018; Figure 1a and b). Thus, accurate quantification of the biomass of mesopelagic resources, their nutritional and genetic composition, their diversity and their links to other components of the food web, to other oceanic realms and to biological and chemical oceanographic processes and cycles, are the focus of growing research activity (Figure 1a; and see Irigoien et al., 2014; St. John et al., 2016). This information is needed to ensure the sustainable management of these resources and the ecosystems to which they belong. Table 1. The ten most cited articles returned by a search of the Web of Science Core Collection for the term “mesopelagic” over the period 1945–2018, conducted on 25 February 2019. Publication . Year of publication . Number of citations . Karner, M. B., DeLong, E. F., and Karl, D. M. Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature, 409: 507–520. 2001 870 DeLong, E. F., Preston, C. M., Mincer, T., Rich, V., Hallam, S. J., Frigaard, N. U., Martínez, A. et al. Community genomics among stratified microbial assemblages in the ocean’s interior. Science, 311: 496–503. 2006 814 Morris, R. M., Rappé, M. S., Connon, S. A., Vergin, K. L., Siebold, W. A., Carlson, C. A. and Giovannoni, S. J. SAR11 clade dominates ocean surface bacterioplankton communities. Nature, 420: 806–810. 2002 627 Jiao, N., Herndl, G. J., Hansell, D. A., Benner, R., Kattner, G., Wilhelm, S. W., Kirchman, D.L et al. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nature Reviews Microbiology, 8: 593–599. 2010 444 Sunagawa, S., Coelho, L. P., Chaffron, S., Kultima, J. R., Labadie, K., Salazar, G., Djahanschiri, B. et al. Structure and function of the global ocean microbiome. Science, 348(6237), 1261359. 2015 380 Schwartzlose, R. A., and Alheit, J. Worldwide large-scale fluctuations of sardine and anchovy populations. African Journal of Marine Science, 21: 289–347. 1999 372 Herndl, G. J., Reinthaler, T., Teira, E., van Aken, H., Veth, C., Pernthaler, A., and Pernthaler, J. Contribution of Archaea to total prokaryotic production in the deep Atlantic Ocean. Applied and Environmental Microbiology, 71: 2303–2309. 2005 369 Furness, R. W., and Camphuysen, K. Seabirds as monitors of the marine environment. ICES Journal of Marine Science, 54: 726–737. 1997 343 Honjo, S. Material fluxes and modes of sedimentation in the mesopelagic and bathypelagic zones. Journal of Marine Research, 38: 53–97. 1980 336 Pauly, D., Trites, A. W., Capuli, E., and Christensen, V. Diet composition and trophic levels of marine mammals. ICES Journal of Marine Science, 55: 467–481. 1998 329 Publication . Year of publication . Number of citations . Karner, M. B., DeLong, E. F., and Karl, D. M. Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature, 409: 507–520. 2001 870 DeLong, E. F., Preston, C. M., Mincer, T., Rich, V., Hallam, S. J., Frigaard, N. U., Martínez, A. et al. Community genomics among stratified microbial assemblages in the ocean’s interior. Science, 311: 496–503. 2006 814 Morris, R. M., Rappé, M. S., Connon, S. A., Vergin, K. L., Siebold, W. A., Carlson, C. A. and Giovannoni, S. J. SAR11 clade dominates ocean surface bacterioplankton communities. Nature, 420: 806–810. 2002 627 Jiao, N., Herndl, G. J., Hansell, D. A., Benner, R., Kattner, G., Wilhelm, S. W., Kirchman, D.L et al. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nature Reviews Microbiology, 8: 593–599. 2010 444 Sunagawa, S., Coelho, L. P., Chaffron, S., Kultima, J. R., Labadie, K., Salazar, G., Djahanschiri, B. et al. Structure and function of the global ocean microbiome. Science, 348(6237), 1261359. 2015 380 Schwartzlose, R. A., and Alheit, J. Worldwide large-scale fluctuations of sardine and anchovy populations. African Journal of Marine Science, 21: 289–347. 1999 372 Herndl, G. J., Reinthaler, T., Teira, E., van Aken, H., Veth, C., Pernthaler, A., and Pernthaler, J. Contribution of Archaea to total prokaryotic production in the deep Atlantic Ocean. Applied and Environmental Microbiology, 71: 2303–2309. 2005 369 Furness, R. W., and Camphuysen, K. Seabirds as monitors of the marine environment. ICES Journal of Marine Science, 54: 726–737. 1997 343 Honjo, S. Material fluxes and modes of sedimentation in the mesopelagic and bathypelagic zones. Journal of Marine Research, 38: 53–97. 1980 336 Pauly, D., Trites, A. W., Capuli, E., and Christensen, V. Diet composition and trophic levels of marine mammals. ICES Journal of Marine Science, 55: 467–481. 1998 329 The total number of publications returned was 2025. Open in new tab Table 1. The ten most cited articles returned by a search of the Web of Science Core Collection for the term “mesopelagic” over the period 1945–2018, conducted on 25 February 2019. Publication . Year of publication . Number of citations . Karner, M. B., DeLong, E. F., and Karl, D. M. Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature, 409: 507–520. 2001 870 DeLong, E. F., Preston, C. M., Mincer, T., Rich, V., Hallam, S. J., Frigaard, N. U., Martínez, A. et al. Community genomics among stratified microbial assemblages in the ocean’s interior. Science, 311: 496–503. 2006 814 Morris, R. M., Rappé, M. S., Connon, S. A., Vergin, K. L., Siebold, W. A., Carlson, C. A. and Giovannoni, S. J. SAR11 clade dominates ocean surface bacterioplankton communities. Nature, 420: 806–810. 2002 627 Jiao, N., Herndl, G. J., Hansell, D. A., Benner, R., Kattner, G., Wilhelm, S. W., Kirchman, D.L et al. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nature Reviews Microbiology, 8: 593–599. 2010 444 Sunagawa, S., Coelho, L. P., Chaffron, S., Kultima, J. R., Labadie, K., Salazar, G., Djahanschiri, B. et al. Structure and function of the global ocean microbiome. Science, 348(6237), 1261359. 2015 380 Schwartzlose, R. A., and Alheit, J. Worldwide large-scale fluctuations of sardine and anchovy populations. African Journal of Marine Science, 21: 289–347. 1999 372 Herndl, G. J., Reinthaler, T., Teira, E., van Aken, H., Veth, C., Pernthaler, A., and Pernthaler, J. Contribution of Archaea to total prokaryotic production in the deep Atlantic Ocean. Applied and Environmental Microbiology, 71: 2303–2309. 2005 369 Furness, R. W., and Camphuysen, K. Seabirds as monitors of the marine environment. ICES Journal of Marine Science, 54: 726–737. 1997 343 Honjo, S. Material fluxes and modes of sedimentation in the mesopelagic and bathypelagic zones. Journal of Marine Research, 38: 53–97. 1980 336 Pauly, D., Trites, A. W., Capuli, E., and Christensen, V. Diet composition and trophic levels of marine mammals. ICES Journal of Marine Science, 55: 467–481. 1998 329 Publication . Year of publication . Number of citations . Karner, M. B., DeLong, E. F., and Karl, D. M. Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature, 409: 507–520. 2001 870 DeLong, E. F., Preston, C. M., Mincer, T., Rich, V., Hallam, S. J., Frigaard, N. U., Martínez, A. et al. Community genomics among stratified microbial assemblages in the ocean’s interior. Science, 311: 496–503. 2006 814 Morris, R. M., Rappé, M. S., Connon, S. A., Vergin, K. L., Siebold, W. A., Carlson, C. A. and Giovannoni, S. J. SAR11 clade dominates ocean surface bacterioplankton communities. Nature, 420: 806–810. 2002 627 Jiao, N., Herndl, G. J., Hansell, D. A., Benner, R., Kattner, G., Wilhelm, S. W., Kirchman, D.L et al. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nature Reviews Microbiology, 8: 593–599. 2010 444 Sunagawa, S., Coelho, L. P., Chaffron, S., Kultima, J. R., Labadie, K., Salazar, G., Djahanschiri, B. et al. Structure and function of the global ocean microbiome. Science, 348(6237), 1261359. 2015 380 Schwartzlose, R. A., and Alheit, J. Worldwide large-scale fluctuations of sardine and anchovy populations. African Journal of Marine Science, 21: 289–347. 1999 372 Herndl, G. J., Reinthaler, T., Teira, E., van Aken, H., Veth, C., Pernthaler, A., and Pernthaler, J. Contribution of Archaea to total prokaryotic production in the deep Atlantic Ocean. Applied and Environmental Microbiology, 71: 2303–2309. 2005 369 Furness, R. W., and Camphuysen, K. Seabirds as monitors of the marine environment. ICES Journal of Marine Science, 54: 726–737. 1997 343 Honjo, S. Material fluxes and modes of sedimentation in the mesopelagic and bathypelagic zones. Journal of Marine Research, 38: 53–97. 1980 336 Pauly, D., Trites, A. W., Capuli, E., and Christensen, V. Diet composition and trophic levels of marine mammals. ICES Journal of Marine Science, 55: 467–481. 1998 329 The total number of publications returned was 2025. Open in new tab Figure 1. Open in new tabDownload slide The number of publications (a) and citations (b) per year returned by a search of the Web of Science Core Collection for the term “mesopelagic” over the period 1945–2018, conducted on 25 February 2019. The total number of publications returned was 2025 and these were cited 54 703 times by 29 434 articles. Figure 1. Open in new tabDownload slide The number of publications (a) and citations (b) per year returned by a search of the Web of Science Core Collection for the term “mesopelagic” over the period 1945–2018, conducted on 25 February 2019. The total number of publications returned was 2025 and these were cited 54 703 times by 29 434 articles. In order to ascertain the present state of knowledge about the mesopelagic, ICES Journal of Marine Science solicited contributions to the article theme set, “Mesopelagic resources—potential and risk.” The intention was to motivate the submission of articles reporting novel advances to: estimate regional and global mesopelagic resources; more completely describe mesopelagic inhabitants, including basic knowledge of their biology and ecology; identify the trophic links between mesopelagic species (and those from other oceanic domains); characterize the functional roles of mesopelagic organisms in the ecosystem, including in the carbon cycle and sequestration of greenhouse gasses and other processes; explore the possible contribution of mesopelagic resources to global food security, human health, and marine bioprospecting; identify the economic and ecological risks of exploiting these resources and discuss the challenges facing the exploitation and sustainable management of mesopelagic resources, including legal responsibilities for regulating these transboundary resources. The contributions to this theme set provide new information on many of these topics. In this introduction, we summarize the contributions included in this theme set and provide some “Food for Thought” on the state-of-the-art in research on the mesopelagic, including identifying the knowledge that must be generated to support its sustainable management. About the articles in this theme set, in the context of what is known/unknown Habitat and distribution Little is known about the composition, diversity, and distribution of mesopelagic communities, mostly owing to sparse data resulting from the challenges of sampling (St. John et al., 2016). Thus, basic questions such as who is down there (biodiversity), what they are up to (trophic ecology), and what ecosystem processes (food web, material and energy fluxes) they support are still poorly known (Glover et al., 2018). Dolan et al. (2019) describe the seasonal variation of heterotrophic protists (tintinnid ciliates, phaecodarian radiolarians, and amphisolenid dinoglagellates) in relation to water column structure. They report contrasting behaviour of the three protist groups compared with expectations, with distinct seasonal patterns among them. This study highlights that the most relevant ecological interactions in the mesopelagic ecosystem occur on fine temporal and spatial scales. Other recent research also supports the importance of looking at small scales to better understand the role of mesopelagic ecosystems in pelagic food webs (Proud et al., 2018), and how regional and mesoscale ecosystems are structured (Proud et al., 2017; Reygondeau et al., 2018). On the opposite extreme in terms of data availability are data-rich monitoring programs that collect larvae of all marine taxa. These monitoring programs are particularly relevant for mesopelagic species because sampling of early life stages provides a more accurate description of mesopelagic taxa compared with adults, which have comparatively lower catchability (Peña, 2019, and references therein). The CalCOFI ichthyoplankton monitoring program in the California upwelling is the longest ichthyoplankton community dataset in the world, with mesopelagic taxa representing the most specious groups. Thus, this dataset provides a unique opportunity to assess the impact of large-scale climate forcing on a mesopelagic community. This is particularly relevant in upwelling ecosystems such as the California current (CC) ecosystem, where mesopelagic and deep sea ecosystems are highly dependent on spatial and temporal variation of midwater oxygen concentration and the extension of the oxygen minimum zone (OMZ). Koslow et al. (2019) enlarge the study area of the CC compared with previous studies by combining information from central, south and Baja California to investigate the evolving response of mesopelagic fishes to declining midwater oxygen concentrations. The study shows a progressive latitudinal effect in the response of mesopelagic species across the three areas, with several warm-water mesopelagic species, apparently adapted to the shallower and intense OMZ off Baja California, increasing despite declining midwater oxygen concentrations. Warm water species are becoming increasingly dominant, initially off Baja California north of the CC. The authors suggest that this response is associated with the warming near-surface, owing to the increased flux of Pacific Equatorial Water into the southern CC during warm phases of the Pacific Decadal Oscillation and the El Niño-Southern Oscillation. This study demonstrates the importance of accounting for synergistic effects between surface and deep water physical conditions to understand the temporal dynamics of mesopelagic species. Trophic links and the vertical transfer of material and energy While the regional and large-scale connections between mesopelagic species and higher trophic levels requires further quantification, their links to lower trophic levels are considerably better known. Six of the thirteen articles in this theme set are on the latter topic. Mei et al. (2019), Olivar et al. (2019), and Richards et al. (2019) use stable isotope ratios (SI; δ13C and δ15N) in muscle tissue, while Contreras et al. (2019) analyse the stomach content of the early life stages to characterize this link. For example, Mei et al. (2019) use SI values of larvae of six mesopelagic fish, and the diet of spawning females, to categorize the spawning strategies of these species as either capital or income breeders. The SI approach also allowed the authors to characterize the dietary niche of overlap of these mesopelagic larvae. Olivar et al. (2019) use transoceanic sampling across the equatorial and tropical Atlantic to assess the trophic position and diet of myctophid fish (the most important component of the mesopelagic fish community that undertakes diel vertical migrations, DVM) across contrasting ecosystems in terms of productivity and oxygen concentration. While species-specific differences were driven by their diets, myctophids inhabiting zones with low oxygen concentration held lower mean trophic level positions. Richards et al. (2019) contrast the diet of mesopelagic fishes (the first dietary descriptions reported for some of them) of different size and DVM behaviour (migrant vs. non-migrants). They show that all species have a similar carbon source, most of them from epipelagic food resources. However, shifts in δ15N signals with increasing body size revealed ontogenetic changes in diet and trophic position. Finally, Contreras et al. (2019) provide novel information on the contrasting trophic ecology of the early life stages of six species of mesopelagic fish, including their diel patterns of feeding and how these change during ontogeny. They also report a general lack of dietary specialization in terms of prey size, mainly composed of copepods, although the diversity of the diet increased during ontogeny. Most prey extracted from predator stomachs are unidentifiable by visual inspection. A more complete characterization of predator diets depends on molecular techniques that are now rapidly being applied to reveal trophic interactions among mesopelagic organisms (Clarke et al., 2019) and to indirectly obtain information on other important species using genetic screening of the stomach content of mesopelagic fish (e.g. Anguilla anguilla in the Sargasso Sea, Jensen et al., 2018). For example, DNA metabarcoding and environmental DNA are supporting rapid increases in knowledge about biodiversity, habitat distribution, and community structure, as well as providing new insights into the evolutionary history of colonization of the mesopelagic realm. A recent gene sequencing study shows that Southern Ocean myctophids are from at least three distant subfamilies, suggesting that colonization has occurred repeatedly, and that spatial divergence of myctophids is rare, likely owing to their enormous abundance and the homogenizing force of ocean currents (Christiansen et al., 2018). Despite the promise in these techniques, they do not as yet support quantitative estimates of biomass or population size. Quantification of carbon fluxes from primary production to mesopelagic fish and other organisms is one of the main challenges in assessing the role of mesopelagic organisms in the biological carbon pump (BCP), as well as their influence in regulating climate in the coming decades (Yool et al., 2013). Anderson et al. (2019) approach this question by investigating carbon transfer via three groups of copepods: detritivores that access sinking particles, vertical migrators, and species that reside in the sub-surface layers. They compared a world ocean model (between 40°N and 40°S) with acoustics-based estimates of mesopelagic fish production. Their study demonstrates the paramount role of migrating organisms in transferring carbon from the surface to the mesopelagic zone. However, they also reveal that their estimates are highly sensitive to the (mostly assumed) trophic pathways within the mesopelagic food web. They explicitly stress that a deeper understanding and parameterization of these linked processes is required to support sustainable management of mesopelagic fish as a harvestable resource. Pakhomov et al. (2019) report a similar exercise (at a local scale) that focuses on the role of pelagic decapods, including large migrators, partial migrators, and non-migrators. They provide estimates of both active and passive carbon transport during the day and night that are not generally included in carbon flux models. In addition, recent studies show that co-occurring mesopelagic fish and crustacean species respond differently to the physical properties and biological factors defining mesopelagic ecoregions (Judkins and Haedrich, 2018), highlighting the contrasting spatial dynamics and dispersal pathways of different groups of mesopelagic organisms and the need to look beyond fish to understand the dynamics of the mesopelagic. The extensive DVMs undertaken by members of the mesopelagic community play a key role in the transfer of material and energy (energy transfer efficiency, ETE) from the sub-surface to demersal and bottom ecosystems (e.g. Proud et al., 2017). This vertical transfer of materials and energy that, at least in the open ocean appears to be higher than expected (Irigoien et al., 2014; Proud et al., 2017), is also a pathway through which the effects of climate change at the surface is transferred to the deep ocean, at a rate much higher than would be the case without the extensive vertical movement of mesopelagic biomass (Smith et al., 2013). Global simulation exercises predict that global warming will increase ETE and result in shallowing of the deep scattering layer (DSL) and a 17% increase in its biomass (Proud et al., 2017). An increase in species richness of mesopelagic zones away from equator is also predicted (Costello and Beyer, 2017). However, Koslow et al. (2019) call for caution when it comes to these future scenarios pending a more complete modelling approach that takes into account additional variables relevant for regional mesoscale processes, such as oxygen concentration. Biomass estimates More global and regional estimates of mesopelagic production are needed (e.g. Irigoien et al., 2014; Proud et al., 2017, 2019), but the various methods currently being used to obtain these estimates have produced wildly different numbers. While estimates based on acoustic surveys range from 14.3 to 19.5 Gt (Irigoien et al., 2014), ten times higher than previously reported (1 Gt, Gjøsæter and Kawaguchi, 1980), estimates based on food-web models yield values around 2.4 Gt (Anderson et al., 2019). Given these hugely different estimates, it is crucial to identify the main uncertainties in each approach-model, quantify them, and present them explicitly. Proud et al. (2019) apply a mesopelagic fish biomass model using acoustic backscatter from the DSL to assess the impact of different sources of uncertainty on estimates of global mesopelagic fish biomass. Their study reveals that there are still considerable sources of uncertainty associated with fish swimbladder volume, length distribution, species morphology, and the proportion of the backscatter that might be from siphonophores vs. fish. Taking these sources of uncertainty into account, they estimate mesopelagic fish biomass could range between ca. 1.8 and 16 Gt. In addition, spatial and temporal changes in these uncertainties could considerably alter the biological reference points that would potentially be used to design harvesting approaches for mesopelagic fish. Although acoustic backscatter is easy to measure, converting it into a species-specific size spectrum and, thereby, into biomass, remains challenging. Most regional estimates of mesopelagic fish are based on acoustic surveys, which are calibrated using samples from mesopelagic trawls. However, catchability of mesopelagic organisms is one of the main limitations in their sampling. Peña (2009) reports a survey-based experiment to assess the influence of vessel lights and noise from the dynamic positioning (DP) system on mesopelagic fish behaviour and vertical distribution. She shows that light triggered a disperse diving of mesopelagics, and this effect was even stronger in response to DP system noise. New technological developments in gear and acoustics, that increase catchability such that it is more representative of what is present in the mesopelagic, are needed. Other articles in this them set focus on the regional scale, making use of information available from monitoring programs. Sassa (2019) present an integrative exercise applying the daily egg production method to estimate spawning stock biomass of an important myctophid in the East China Sea, combining information from fish larval surveys and reproductive parameters available from the literature (i.e. sex ratio, back fecundity, and spawning fraction). These methods could be applied to analogous survey data in other regions in order to generate indirect assessments of myctophid biomass (at a regional scale) when other methods, such as acoustics, are not available or underestimate it. While most approaches to estimating the abundance of mesopelagic organisms assume that biomass is static in space and time, production surely fluctuates on different temporal scales in conjunction with changes in spatial distribution. Fock and Czudaj (2019) compare the size structure of the mesopelagic fish community along a transect from the Equator to the Bay of Biscay during two periods; the 1970s and a survey conducted in 2015. Their study demonstrated that the size structure of 20 out of the 28 species studied had changed (although in different directions), with 8 species showing a greater dominance of small size classes in 2015 compared with earlier, 10 species showing a reduced dominance of older size classes, and 2 showing a clear shift in the modal length of size distributions. Because biomass estimates are dependent on the size structure of mesopelagic fish populations (something that is not easy to accurately determine), better information on the variability of population and community size structure is needed to decrease the uncertainty of these estimates. Improved coordination and integration of food web modelling and DSL-based estimates is needed to better understand the temporal and spatial variation of mesopelagic production, ETE, and their implications on, for example, predation on zooplankton or the role of upper trophic levels, both of which are often omitted by mass balance and end-to-end ecosystem models. Complex models, such as Atlantis (Fulton et al., 2011) and SEAPOYDM (Lehodey et al., 2008, 2015), are starting to include depth integrated processes and associated diel variability, such as the relationship between DSL structure and depth of the euphotic zone. Accurate representation of the BCP in ecological models is important because it feeds directly into climate and earth-system models in which it plays a key role in estimating the transfer of climate forcing from the ocean surface to deep sea ecosystems. Extracting and sustainably managing mesopelagic resources Bioeconomic modelling of possible exploitation scenarios, coupled with alternative management strategies and harvest control rules, are needed to guide the management of mesopelagic resources, particularly for those species for which the present market is undeveloped. Prellezo (2019) presents a case study that assesses the technical, financial, and market viability of mesopelagic resources in the Bay of Biscay. While exploitation is technically possible, it is not a viable alternative to the existing commercial fisheries because of the lower profitability of the landings. However, in the context of the new landing obligation of the EU Common Fisheries Policy, exploitation of mesopelagic resources could have a narrow economic potential because catch quotas of commercial species (and fishing effort) will be limited, leaving the possibility for excess capacity to be used to harvest mesopelagic species. Additional studies such as this are needed, and will have to be updated, if/as the market for mesopelagic species changes. Knowledge gaps and challenges Mesopelagic resources, particularly lanternfish, represent one of the last high biomass groups of fish that is as yet unexploited. If even a small percentage of the estimated biomass of this group is extracted (e.g. 1% of 9 Gt), it would double the present global landings from capture fisheries (=ca. 90 million tonnes, as drawn from Fig. 1 in FAO, 2018). Given the forces alluded to at the beginning of this introduction, it seems likely that it is only a matter of time until this taxonomic group, and other mesopelagic resources, will be exploited. There is an opportunity here to learn from the mistakes (and successes) of the past to sustainably exploit these resources. To achieve that we will need much more information than what is currently available on key aspects of mesopelagic species and systems. Some of these knowledge gaps are mentioned above, others below. Although most of those that follow have been identified earlier (Irigoien et al., 2014; St. John et al., 2016; Proud et al., 2017), they bear repeating. St. John et al. (2016) listed five areas in which more knowledge is needed to inform the sustainable management of the mesopelagic zone: (i) population vital rates, which represent the basics for stock assessments and population dynamic modelling to predict the impact of fishing; (ii) development of stock assessment tools, harvest control rules, etc., adapted to data poor situations; (iii) the interaction between oceanographic scenarios and mesopelagic biomass and biodiversity that enable future projections including those associated with climate change impacts; (iv) food web implications of mesopelagic resource depletion; and (v) the role of mesopelagic species and communities in the sequestration of greenhouse gases. Additionally, there is a need to (vi) develop technology (ships, gear, real-time species identification and biomass estimation, fishing strategies to efficiently capture a diffuse and transboundary resource) to support efficient extraction of mesopelagic species, with minimal bycatch (e.g. Peña, 2019 and references therein); (vii) increase the application of the next generation of genetic tools (e.g. metabarcoding and environmental DNA) to provide a more complete characterization (including quantification) of the biodiversity and ecology of the mesopelagic; (viii) add to our knowledge and understanding of the possible ecosystem consequences of extracting mesopelagic organisms. Although it could be argued that knowledge of how other marine ecosystems react to the removal of large chunks of their biomass can guide us, that is true only to the extent that the food webs of those systems are known, which is often not at all well. Even less is known of the interconnectivity of mesopelagic organisms; (ix) better define how much energy is recycled within the mesopelagic, that is, how important is the “mesopelagic loop?”; (x) given the relative stability of the mesopelagic environment, assess the capacity of mesopelagic organisms to adapt to conditions that will change more rapidly as a result of climate change); and (xi) conduct socioeconomic receptiveness studies and develop market opportunities. There is a general lack of industrial-scale processing technology for these species, and markets for them will have to be developed (IMR et al., 2017; Prellezo, 2019, and references therein). The fact that mesopelagic resources are generally present in regions beyond national jurisdiction presents another challenge in terms of agreeing upon, and implementing, legally binding instruments to govern their exploitation (O’Leary and Roberts, 2018). The sustainable management of the services provided by the mesopelagic ecosystem requires an ecosystem-based framework that balances benefits, risks, and trade-offs (St. John et al., 2016). That is, harvesting this ecosystem can produce more food for human consumption, but the potential consequences associated with the effect of biomass extraction on the (poorly known) role of the mesopelagic in climate regulation, conservation, biodiversity, and ecosystem stability must be carefully considered. Given the present limits on our knowledge of the mesopelagic, and of the effects that large-scale extraction of biomass might have, a precautionary approach has been adopted by some management councils to protect forage fish species such as mesopelagic fish. For example, the Pacific Council, supported by a Comprehensive Ecosystem-Based Amendment (CEBA 1), “prohibits the development of new directed fisheries on forage species that are not currently managed by the Council, or the States, until the Council has had an adequate opportunity to assess the science relating to any proposed fishery and any potential impacts to our existing fisheries and communities.” In contrast, as stated in their “Blue Growth Strategy,” the European Commission is currently open to the exploration and exploitation of new ocean horizons such as the mesopelagic (European Commission, 2018). Global research output on topics such as “ocean acidification” and “marine plastic,” or species such as “Atlantic salmon” and “Atlantic cod,” is 500–900 primary research articles per year (quotation marks identify the search terms used in the Web of Science to obtain these numbers). The present output on “mesopelagic” research topics of about 150 primary research articles per year (Figure 1a) will not produce (on a reasonable timeline) the information needed to meet the challenge of sustainably managing this resource, particularly given the technical and logistic challenges associated with obtaining such information. Although the European Commission (and other agencies) have recently funded projects to investigate some of the questions raised here, considerably more resources will be required to conduct the research needed to support knowledge-based management of mesopelagic resources. Finally, large-scale exploitation of the mesopelagic should not begin until that information is incorporated into management tools. Acknowledgements The authors thank Roland Proud and Mike St. John for their comments on an earlier version of the manuscript. Funding M.H. acknowledges funding from the EU H2020 PANDORA project (Nr. 773713). H.I.B.’s contribution to this article theme set was supported by Project # 83741 (Scientific publishing and editing) from the Institute of Marine Research, Norway. Disclaimer The opinions and positions taken in this article are those of the authors and do not necessarily reflect those of their employers. Conflict of interest statement H.I.B. and M.H. work, respectively, for the Institute of Marine Research, Norway, and the Spanish Institute of Oceanography (IEO, Spain), both of which are government research institutes charged with generating science and providing advice in support of the sustainable use of marine resources. None of the directed funding that he has received to date has been related to the mesopelagic or mesopelagic organisms. H.I.B. is also Editor-in-Chief of this Journal (and M.H. a member of the editorial board), which is owned by the International Council for the Exploration of the Sea (ICES), a global intergovernmental organization that develops science and advice to support the sustainable use of the oceans. The authors declare no conflicts of interest. References Anderson T. R. , Martin A. P., Lampitt R. S., Trueman C. N., Henson S. A., Mayor D. J. 2019 . Quantifying carbon fluxes from primary production to mesopelagic fish using a simple food web model . ICES Journal of Marine Science , 76 : 690 – 701 . Google Scholar Crossref Search ADS WorldCat Christiansen H. , Dettai A., Heindler F. M., Collins M. A., Duhamel G., Hautecoeur M., Steinke D. et al. 2018 . Diversity of mesopelagic fishes in the Southern Ocean—a phylogeographic perspective using DNA barcoding . Frontiers in Ecology and Evolution , 6 : 120 . Google Scholar Crossref Search ADS WorldCat Clarke L. J. , Trebilco R., Walters A., Polanowski A. M., Deagle B. E. 2019 . DNA-based diet analysis of mesopelagic fish from the southern Kerguelen Axis . Deep Sea Research Part II: Topical Studies in Oceanography , doi: 10.1016/j.dsr2.2018.09.001. Google Scholar OpenURL Placeholder Text WorldCat Contreras T. , Olivar M. P., Hulley P. A., Fernández de Puelles M. L. 2019 . Feeding ecology of early life stages of mesopelagic fishes in the equatorial and tropical Atlantic . ICES Journal of Marine Science , 76 : 673 – 689 . Google Scholar Crossref Search ADS WorldCat Costello M. J. , Breyer S. 2017 . Ocean depths: the mesopelagic and implications for global warming . Current Biology , 27 : R36 – R38 . Google Scholar Crossref Search ADS PubMed WorldCat Dolan J. R. , Ciobanu M., Marro S., Coppola L. 2019 . An exploratory study of heterotrophic protists of the mesopelagic Mediterranean Sea . ICES Journal of Marine Science , 76 : 616 – 625 . Google Scholar Crossref Search ADS WorldCat European Commission. 2018 . Blue Bioeconomy: Situation Report and Perspectives. Directorate-General for Maritime Affairs and Fisheries, Director-General, http://www.eumofa.eu/documents/20178/84590/Blue+bioeconomy_Final.pdf. FAO. 1997 . Review of the State of World Fishery Resources: Marine Fisheries. Lanternfishes: A Potential Fishery in the Northern Arabian Sea? Rome: FAO. FAO Fisheries Circular No. 920, FIRM/C.920. FAO. 2001 . Report of the Trilateral Workshop on Lanternfish in the Gulf of Oman, Muscat, Oman, 7–9 May 2001. Muscat: FAO. FAO Fisheries Report No. 665, FIIT/R.665. FAO. 2018 . The State of World Fisheries and Aquaculture 2018 - Meeting the Sustainable Development Goals. Rome, License: CC BY-NC-SA 3.0 IGO. Fock H. O. , Czudaj S. 2019 . Size structure changes of mesopelagic fishes and community biomass size spectra along a transect from the equator to the Bay of Biscay collected in 1966–1979 and 2014–2015 . ICES Journal of Marine Science , 76 : 755 – 770 . Fulton E. A. , Link J. S., Kaplan I. C., Savina‐Rolland M., Johnson P., Ainsworth C., Horne P. et al. 2011 . Lessons in modelling and management of marine ecosystems: the Atlantis experience . Fish and Fisheries , 12 : 171 – 188 . Google Scholar Crossref Search ADS WorldCat Glover A. G. , Wiklunf H., Cheng C., Dahlgren T. G. 2018 . Managing a sustainable deep-sea ‘blue economy’ requires knowledge of what actually lives there . eLife , 7 : e41319. Google Scholar Crossref Search ADS PubMed WorldCat Gjøsæter J. , Kawaguchi K. 1980 . A Review of the World Resources of Mesopelagic Fish. FAO, No. 193. Institute of Marine Research, Nofima, University of Bergen and NIFES. 2017 . Mesopelagic Initiative: Unleashing New Marine Resources for a Growing Human Population. https://www.hi.no/filarkiv/2017/rad-bestander_og_ressurser-mesopelagic_initiative-unleashing_new_marine_resources_for_a_growing_human_population.pdf/nb-no (last accessed 15 November 2018). Irigoien X. , Klevjer T. A., Røstad A., Martinez U., Boyra G., Acuña J. L., Bode A. et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean . Nature Communications , 5 : 3271. Google Scholar Crossref Search ADS PubMed WorldCat Jennings S. , Collingridge K. 2015 . Predicting consumer biomass, size-structure, production, catch potential, responses to fishing and associated uncertainties in the world’s marine ecosystems . PLoS One , 10 : 1 – 28 . Google Scholar OpenURL Placeholder Text WorldCat Jensen M. R. , Knudsen S. W., Munk P., Thomsen P. F., Møller P. R. 2018 . Tracing European eel in the diet of mesopelagic fishes from the Sargasso Sea using DNA from fish stomachs . Marine Biology , 165 : 130 . Google Scholar Crossref Search ADS WorldCat Judkins D. C. , Haedrich R. L. 2018 . The deep scattering layer micronektonic fish faunas of the Atlantic mesopelagic ecoregions with comparison of the corresponding decapod shrimp faunas . Deep Sea Research Part I: Oceanographic Research Papers , 136 : 1 – 3 . Google Scholar Crossref Search ADS WorldCat Koslow J. A. , Davison P., Ferrer E., Jiménez Rosenberg S. P. A., Aceves-Medina G., Watson W. 2019 . The evolving response of mesopelagic fishes to declining midwater oxygen concentrations in the southern and central California current . ICES Journal of Marine Science , 76 : 626 – 638 . Lehodey P. , Senina I., Murtugudde R. 2008 . A spatial ecosystem and populations dynamics model (SEAPODYM) - modelling of tuna and tuna-like populations . Progress in Oceanography , 78 : 304 – 318 . Google Scholar Crossref Search ADS WorldCat Lehodey P. , Conchon A., Senina I., Domokos R., Calmettes B., Jouanno J., Hernandez O. et al. 2015 . Optimization of a micronekton model with acoustic data . ICES Journal of Marine Science , 72 : 1399 – 1412 . Google Scholar Crossref Search ADS WorldCat Mei W. , Umezawa Y., Wan X., Yuan J., Sassa C. 2019 . Feeding habits estimated from weight-related isotope variations of mesopelagic fish larvae in the Kuroshio waters of the northeastern East China Sea . ICES Journal of Marine Science , 76 : 639 – 648 . O’Leary B. C. , Roberts C. M. 2018 . Ecological connectivity across ocean depths: implications for protected area design . Global Ecology and Conservation , 15 : e00431. Google Scholar Crossref Search ADS WorldCat Olivar M. P. , Bode A., López-Pérez C., Hulley P. A., Hernández-León S. 2019 . Trophic position of lanternfishes (Pisces: Myctophidae) of the tropical and equatorial Atlantic estimated using stable isotopes . ICES Journal of Marine Science , 76 : 649 – 661 . Pakhomov E. A. , Podeswa Y., Hunt B. P. V., Kwong L. E. 2019 . Vertical distribution and active carbon transport by pelagic decapods in the North Pacific Subtropical Gyre . ICES Journal of Marine Science , 76 : 702 – 717 . Google Scholar Crossref Search ADS WorldCat Peña M. 2019 . Mesopelagic fish avoidance from the vessel dynamic positioning system . ICES Journal of Marine Science , 76 : 734 – 742 . Prellezo R. 2019 . Exploring the economic viability of a mesopelagic fishery in the Bay of Biscay . ICES Journal of Marine Science , 76 : 771 – 779 . Proud R. , Cox M. J., Brierley A. S. 2017 . Biogeography of the global ocean’s mesopelagic zone . Current Biology , 27 : 113 – 119 . Google Scholar Crossref Search ADS PubMed WorldCat Proud R. , Cox M. J., Le Guen C., Brierley A. S. 2018 . Fine-scale depth structure of pelagic communities throughout the global ocean based on acoustic sound scattering layers . Marine Ecology Progress Series , 598 : 35 – 48 . Google Scholar Crossref Search ADS WorldCat Proud R. , Handegard N. O., Kloser R. J., Cox M. J., Brierley A. S. 2019 . From siphonophores to deep scattering layers: uncertainty ranges for the estimation of global mesopelagic fish biomass . ICES Journal of Marine Science , 76 : 718 – 733 . Reygondeau G. , Guidi L., Beaugrand G., Henson S. A., Koubbi P., MacKenzie B. R., Sutton T. T. et al. 2018 . Global biogeochemical provinces of the mesopelagic zone . Journal of Biogeography , 45 : 500 – 514 . Google Scholar Crossref Search ADS WorldCat Richards T. M. , Gipson E. E., Cook A., Sutton T. T., Wells R. J. D. 2019 . Trophic ecology of meso- and bathypelagic predatory fishes in the Gulf of Mexico . ICES Journal of Marine Science , 76 : 662 – 672 . Sassa C. 2019 . Estimation of the spawning biomass of myctophids based on larval production and reproductive parameters: the case study of Benthosema pterotum in the East China Sea . ICES Journal of Marine Science , 76 : 743 – 754 . Smith K. L. , Ruhl H. A., Kahru M., Huffard C. L., Sherman A. D. 2013 . Deep ocean communities impacted by changing climate over 24 y in the abyssal northeast . Proceedings of the National Academy of Sciences , 110 : 19838 – 19841 . Google Scholar Crossref Search ADS WorldCat Springmann M. , Clark M., Mason-D’Croz D., Wiebe K., Bodirsky B. J., Lassaletta L., de Vries W. et al. 2018 . Options for keeping the food system within environmental limits . Nature , 562 : 519 – 525 . Google Scholar Crossref Search ADS PubMed WorldCat St. John M. A. , Borja A., Chust G., Heath M., Grigorov I., Mariani P., Martin A. P. et al. 2016 . A dark hole in our understanding of marine ecosystems and their services: perspectives from the Mesopelagic Community . Frontiers in Marine Science , 3 : 31 . Google Scholar Crossref Search ADS WorldCat Yool A. , Popova E. E., Coward A. C., Bernie D., Anderson T. R. 2013 . Climate change and ocean acidification impacts on lower trophic levels and the export of organic carbon to the deep ocean . Biogeosciences , 10 : 5831 – 5854 . Google Scholar Crossref Search ADS WorldCat © International Council for the Exploration of the Sea 2019. All rights reserved. For permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) © International Council for the Exploration of the Sea 2019. All rights reserved. For permissions, please email: [email protected]

Dolan, John, R;Ciobanu,, Maria;Marro,, Sophie;Coppola,, Laurent

doi: 10.1093/icesjms/fsx218pmid: N/A

Abstract Is there a mesopelagic protist fauna composed of species different from that of the overlying surface community? Does the mesopelagic community show seasonal changes in abundances and species composition? We addressed these questions by considering three distinct groups in which species identification is relatively unambiguous: tintinnid ciliates, phaeodarian radiolarians, and amphisolenid dinoflagellates. We sampled weekly at 250 m and 30 m depth from January to June a deep-water coastal site characterized by seasonal changes in water column structure; notably, in winter the mixed layer extends down into mesopelagic depths. We found a deep-water community of tintinnid ciliates comprised of forms apparently restricted to deep waters and species also found in the surface layer. This latter group was dominant during the winter mixis period when tintinnid concentrations were highest and subsequently declined with water column stratification. Phaeodarian radiolarians and the amphisolenid dinoflagellates were regularly found in deep samples but were largely absent from surface water samples and showed distinct patterns in the mesopelagic. Phaeodarian radiolarians declined with water column mixing and then increased in concentration with water column stratification whilst amphisolenid dinoflagellates concentrations showed no pattern but species composition varied. We conclude that for all three protists groups there appear to be both distinct mesopelagic forms and seasonal patterns. Introduction Recent reviews have stated is an urgent need to improve our understanding of the biology of the mesopelagic (e.g. St. John et al., 2016; Costello and Breyer, 2017). Primarily, this is because of questions with regard to the standing stocks of deep-water fish (e.g. Irigoien et al., 2014; Sutton, 2013) and as well uncertainties with regard to fluxes of carbon into the bathypelagic and consequently the possible effects of climate change (e.g. Sanders et al., 2016). The food web of the mesopelagic is often depicted as a simple system (compared with the surface layer) in which the inputs are sedimenting particles (aggregates of marine snow, phytoplankton flocs, and zooplankton fecal pellets) exploited and re-packaged by resident or vertically migrating metazoan zooplankton and fish (e.g. Burd et al., 2010; Robinson et al., 2010). Microzooplankton, thought to dominate consumption of primary producers in the surface layer (e.g. Calbet and Landry, 2004), if not entirely absent in depictions of mesopelagic food webs (e.g. Burd et al., 2010), are relegated to a minor role of consumers of free-living prokaryotes (e.g. Giering et al., 2014). This is despite the fact that that microzooplankton in the mesopelagic are known to be a diverse assemblage of both protists and metazoans of distinct ecologies whose aggregate respiration may rival that of the metazoan zooplankton (e.g. Gowing et al., 2003). Microzooplankton in the mesopelagic are thought to be the source of small fecal pellets (minipellets) whose flux may exceed that of metazoan fecal flux (Gowing and Silver, 1985). A surprising variety of mesopelagic consumers prey upon microzooplankton ranging from copepods (Sano et al., 2013), myctophid fishes (Conley and Hopkins, 2004), large tunicates (Hopcroft and Robinson, 1999) to midwater polychaetes (Uttal and Buck, 1996). Close interactions between metazoan consumers and microzooplankton have been proposed as “gardening by metazoans” through fragmenting detrital particles, promoting microbial populations (Mayor et al., 2014). In recent years, a good deal of data on mesopelagic protists has become available through genomic surveys and studies employing flow cytometric analyses (e.g. reviewed in Edgcomb, 2016). These studies suggest that in the mesopelagic protists are diverse, abundant and distinct from the surface layer assemblages. However, studies of temporal variability of different groups of heterotrophic protists in the mesopelagic appear to be quite limited. To the best of our knowledge they concern only the Arabian Sea and the North West Mediterranean Sea. The studies have given contrasting results, likely reflecting differences in the dynamics of the systems examined. Gowing et al. (2003) sampled several stations in the Arabian Sea through four cruises during different monsoonal seasons in 1995. The upper mesopelagic (200–250 m depth) populations of protists, grouped as “dinoflagellates”, “ciliates”, or “sarcodines”, were relatively consistent in concentration, varying in abundance only within a factor of two over time covering distinct monsoonal seasons (Gowing et al., 2003; Supplementary Table S1). Our study site in the N.W. Mediterranean Sea is clearly distinct in terms of some fauna as well dynamics compared with the Arabian Sea. Indeed, the Mediterranean mesopelagic may be distinct from most other systems. With regard to the biological component, the taxonomic composition of fish and metazoan fauna has been described as relatively impoverished in terms of species richness (Sutton et al., 2017). However, mesopelagic fish biomass appears to be comparatively high based on the characteristics of the deep-scattering layer, from which fish biomass has been modelled (Proud et al., 2017). With regard to microbes, the metabolic activity of prokaryotes appears to be higher for mesopelagic Mediterranean populations compared with those found at similar depths in other systems, related possibly to the relatively warm temperature of the deep waters, 13 °C (Luna et al., 2012). Also, the abundances of typical bacterivores, heterotrophic nanoflagellates, appear to be low relative to abundances of prokaryotes compared with other systems perhaps because of high ciliate concentrations (Aristegui et al., 2009). The Mediterrean Sea also differs with regard to water column dynamics. The N.W. Mediterranean is site of winter deep-water formation. In the winter the surface mixed layer extends below 200 m into the mesopelagic and occasionally down to depths of over 2000 m in the Gulf of Lyons (e.g. Herrmann et al., 2008; Houpert et al., 2016). In contrast to clear seasonal changes in water column structure, vertical export flux to the mesopelagic (measured using sediment traps at 200 m) show highly variable fluxes without any clear seasonal pattern or relationship with mixed layer depth (MLD; e.g. Heimbürger et al., 2013). Overall then, it perhaps not surprising that the microbial community of the mesopelagic in the N.W. Mediterranean may be more variable than that of the Arabian Sea. Data are limited to those from only three studies but worth reviewing in some detail. Tanaka and Rassoulzadegan (2002), through bi-monthly sampling, examined changes in the vertical profiles of “bacteria”, “heterotrophic nanoflagellates” and “ciliates” in a deep site in the open N.W. Mediterranean Sea. They found large seasonal changes in upper mesopelagic (200–300 m depth) concentrations of “ciliates” and “heterotrophic nanoflagellates”, ranging well over an order of magnitude between dates whilst concentrations of “bacteria” varied less, within a factor of 3. For all three groups, the highest concentrations correspond with the period of winter water column mixis. Subsequently, Winter et al. (2009a, b) sampling at the same site and depth using bi-monthly sampling found similar seasonal variability in prokaryotic abundances (varying by a factor of 3) and documented seasonal shifts in community composition of prokaryotes. In a more recent study, Weinbauer et al. (2013), through monthly sampling at a coastal deep water site in the NW Mediterranean (the site examined here) over one year, documented seasonal changes in the composition of the mesopelagic prokaryotic community and the abundances of heterotrophic nanoflagellates. They found marked changes in the mesopelagic assemblages. The shifts corresponded with seasonal changes in water column structure; in particular winter water column mixis that preceded marked shifts in the relative abundances of archea and eubacteria as well increases in the abundances of heterotrophic nanoflagellates. However, they reported a pronounced peak in both prokaryotes and heterotrophic flagellate abundance at the beginning of the stratified period. Similar to the findings of Tanaka and Rassoulzadegan, throughout an annual cycle prokaryote abundance varied less, only by a factor of 2, whilst their presumed predators, heterotrophic nanoflagellates, varied by over an order of magnitude (Weinbauer et al., 2013, Figure 4). On the basis of the limited data available protists appear to be a highly dynamic component of the mesopelagic in the N.W. Mediterranean. However, data on species compositions are completely lacking as well as short-term (< month) variability. Here, we build on the study by Weinbauer et al. (2013) using a more intensive sampling at the same deep-water coastal site, sampling at weekly intervals both a surface layer depth and a mesopelagic depth. To examine possible changes in species composition, we focused on distinct groups of protists in which species identifications (based on gross morphology) can be made unambiguously using light microscopy. In this exploratory study, we set out to address two questions: Is there a mesopelagic protist fauna distinct in terms of species composition from that of the overlying surface? Is the mesopelagic protist community dynamic, does it show seasonal changes in species compositions and concentrations? The three groups of protists we investigated are each phylogenetically coherent but are of distinct ecologies. In common are morphologies allowing sampling using fine mesh nets and relatively unambiguous species identifications: tintinnid ciliates, phaeodarian radiolarians, and amphisolenid dinoflagellates. Tintinnid ciliates are grazers on primarily nano-sized prey items (2–20 µm in length) and are characterized by the possession of a (more or less) species-specific lorica or shell into which the ciliate cell may contract. The biology and ecology of tintinnid ciliates is relatively well known (e.g. Dolan et al., 2013). Tintinnids are thought to be largely restricted to the euphotic zone of the oceans although some species are most abundant in the mesopelagic at least in the Adriatic Sea (Kršinić, 1998). Pheodarian radiolarians of the order Phaeogromidae (e.g. Challengarids) are generally considered to be deep-water residents (Anderson, 1983; Kling and Boltovskoy, 1999; Nakamura and Suzuki, 2015) largely feeding on sinking organic aggregates (Gowing and Bentham, 1994; Nothing and Gowing, 1991). Heterotrophic dinoflagellates of the family Amphisoleniaceae group species of the enigmatic genera Amphisolenia and Triplosolenia, species of which lack chloroplasts but host endosymbionts. Amphisolenia can harbour both prokaryotic and eukaryotic phototrophs but nothing is known with regard to their role in the nutrition of the host cell (Gaines and Elbrächter, 1987). Triplosolenia have been included in lists of “shade species” found in deep, poorly lit waters (Taylor and Pollingher, 1987). Triplosolenia have never been found to contain either chloroplasts or food vacuoles (Gaines and Elbrächter, 1987) and like Amphisolenia may contain endosymbiotic cyanobacteria (Saldarriaga and Taylor, 2017). Material and methods Study site Sampling at 250 m and 30 m depth was conducted at “Point C”, a standard sampling site ∼1 km offshore near the entrance of the Bay of Villefranche (43°51′00″N, 07°19′00″E). At the site water column profiles of temperature, salinity conductivity and oxygen of the 300 m depth water column are obtained weekly as part of a French national network of marine sites, SOMLIT (http://somlit.epoc.u-bordeaux1.fr/fr/). Our sampling at Point C was conducted on the same day or within one day of the CTD water column profiling. At this site, as in most of the N.W. Mediterranean Sea, there are large seasonal changes in the depth of the surface mixed layer. During winter (Jan.–March) the mixed layer extends down to the mesopelagic depths (> 200 m) with deep water (> 500 m) formation in areas such as the Gulf of Lions (e.g. D'Ortenzio et al., 2005). The deep-water site, Point C, is located ∼700 m from shallower (80 m depth) main coastal sampling station “Point B”. At this main station, hydrology and biology have been well studied. It is characterized by winter mixing of the water column and annual average water column concentrations of chlorophyll of ∼0.3 µg Chl a l−1 with pronounced maxima in spring and autumn contrasting with low concentrations in the summer (e.g. Bustillos-Guzmán et al., 1995; Mostajir et al., 1995). Sampling and sample processing Sampling was conducted at weekly intervals from 9 January to 1 June 2017. Water samples were obtained using multiple casts of a 30 l Niskin Bottle. The Niskin bottle water was gently emptied into plankton net (20 µm mesh plankton net with a 250 ml collector) placed in a 30-l container. We first sampled at 250 m depth, concentrating material from 120 to 240 l. Water volumes varied as sea conditions did not always permit several casts and volumes sampled were increased during the stratified period when organismal concentrations declined. Elapsed time between bottle casts at 250 m depth was ∼5 min. Net material was preserved with Lugol's fixative (2% final conc.). The net was then thoroughly rinsed and sampling at 30 m depth, material from 90 to 120 l was collected (time between bottle casts was ∼2 min). From the first bottle casts water samples were also taken for counts of heterotrophic prokaryotes and eukaryotes beginning in late February. Here, only data from samples from 250 m are considered. Prokaryotes and heterotrophic protists cell counts using flow cytometer Beginning in late February, samples were taken for flow cytometric analysis. Total cell abundances for heterotrophic prokaryotes and eukaryotes were determined by flow cytometer on 4 ml pre-filtered seawater (water passed through the 20 µm net plankton net) preserved onboard with a mix of glutaraldehyde (0.25% final concentration, EMS)/Pluronic F68 (0.01% final concentration, Sigma–Aldrich) for 15 min at 4 °C, in the dark, then flash frozen in liquid nitrogen and stored at −80 °C until analysis (Marie et al., 2014). All samples were analysed with a FACS Calibur flow cytometer (BD Biosciences, San Jose, CA, USA) equipped with a blue laser emitting at 488 nm at a maximum flow rate of 75 µl min−1. For prokaryotes, 0.5 ml undiluted samples were stained with SYBR Green I (5× final concentration, Invitrogen) and incubated in the dark at room temperature for 10 min. The flow rate varied between 45 µl min−1 (medium) and 75 µl min−1 (high) and acquisition time was 60 s. Their analysis was based on their signature in a side scatter (SSC, related to cell structure) plot versus green fluorescence (FL1) plot (Gasol and Del Giorgio, 2000). One micromitre fluorescent latex beads (Polyscience Inc., Europe) were always used as internal standards. For heterotrophic protists, subsamples of 1 ml from the same thawed sample were stained with SYBR Green I (1× final concentration, Invitrogen) and incubated in the dark, at room temperature for 10 min. A mix of fluorescent beads of different sizes (0.5, 1.0, 3.0, and 10.0 µm, Polyscience Inc., Europe) was used as internal standards. Heterotrophic protists populations were distinguished from large bacteria on an SSC versus FL1 plot as described by Christaki et al. (2011). In our case, the flow rate was established at ∼75 µl min−1 (high) and the data acquisition was achieved for 12 min (see Figure 1 in Supplementary Data Files: cytometric methods). Pure cultures of the heterotrophic flagellate Amastigomonas sp. were used to validate the heterotrophic protists gate on the FL1 versus SSC plot. Undiluted, unstained natural seawater samples were used to better visualize and exclude the autotrophs from the FL3 versus SSC gate. Negative controls represented by 0.2 µm daily filtered seawater plus SYBR Green and the mix of beads were systematically run and never showed any event in the gates of interest. All data were analysed with CellQuest™Pro (BD) software and concentrations expressed as cells ml−1. Figure 1. View largeDownload slide Temporal changes in water column structure: two contrasting periods of “mixed” and “stratified” are indicated based on large differences in the stratification index. Figure 1. View largeDownload slide Temporal changes in water column structure: two contrasting periods of “mixed” and “stratified” are indicated based on large differences in the stratification index. Heterotrophic microplankton cell counts using light microscopy The plankton net material was concentrated through sedimentation in graduated cylinders with a supernate of ∼200 ml removed through gentle siphoning. For each sample, multiple aliquots of (1–3 ml) of the concentrated material were examined using an inverted microscope (Olympus IX71 equipped with DIC optics, a DP71 camera and CellSense image analysis software). Generally, material from at least 100 l from 250 m depth, or 20 l from 30 m depth, was examined for each sample. Species identifications of the three groups considered here were made using a variety of taxonomic works. Those concerning tintinnid ciliates included Abboud-Abi Saab (2008), Balech (1959), Jörgensen (1924), Kofoid and Campbell (1929, 1939), Krsinic (2010). For the phaeogromida phaeodarian radiolarians (i.e. Challengerids, Medusettids, Lirellids) the works consulted were Kling and Boltovskoy (1999), Nakamura and Suzuki (2015), and Borgert (1906, 1911). For the Amphisolenid dinoflagellates (species of Amphisolenia and Triplosolenia) works consulted were Kofoid (1906), Kofoid and Skogsberg (1928), and Taylor (1976). Categorization of species Species were assigned to several (some non-exclusive) categories of occurrence: “rare”, “present”, “common”, “deep water species”, “surface water species”, and “entire water column species” using the following criteria. Occasional or “rare” species were those found on only one or two dates (≤ 10% of sampling dates). These species were not further categorized. Species found on at least three dates were categorized as “present” and further subdivided. Species found on at least three dates in deep water samples were categorized as “deep water species” if fulfilling both the conditions (i) not found on >2 dates in surface samples, and (ii) if found in the surface sample, the concentration did not exceed trace (i.e. >1 cell found). Amongst deep-water species, “common species” were those found on 10 or more dates (≥ 50 of samples). Species found on at least three dates in surface water samples were categorized as surface water species if fulfilling both the conditions (i) not found on >2 dates in deep water samples and if found in a deep water sample, the concentration did not exceed trace (i.e. >1 cell found). Amongst surface water species, “common species” were those found on 10 or more dates (≥ 50 of samples). Species found in both deep-water samples and surface water samples on three or more dates, and in greater than trace concentrations, were categorized as “entire water column species”. We assumed, but did not regularly take samples to prove, that species found at both 30 m and 250 m likely occurred throughout the water column. Complete count data are furnished as a Supplementary Data File. Stratification index and MLD estimations A stratification index was used to characterize the structure of the water column. The index was calculated as the difference in potential density between 10 m and 300 m using the salinity and temperature profiles obtained by the Seabird SBE25 CTD following Behrenfeld et al. (2006), Dave and Lozier (2010), and Lozier et al. (2011). When the difference in potential density is <0.125 kg m3, the upper 300 m can be considered as non-stratified (de Boyer Montégut et al., 2004). The MLD was estimated based on potential density profiles, calculated from pressure, temperature and salinity data. For each potential density profile, the MLD corresponds to the depth where the difference between the potential density at the reference depth (10 dbar) and the measured potential density is higher than the threshold of 0.03 kg m3, which is used in the Mediterranean Sea (D'Ortenzio et al., 2005; de Boyer Montégut et al., 2004). Results Water column structure Seasonal changes in stratification and the depth of the surface mixed layer (Figure 1) were typical for the site. In early January, the mixed layer reached 150 m depth and subsequently deepened to nearly 250 m by early February. Water column stratification began in late March and by mid-April the water column was strongly stratified. The surface mixed layer was restricted to the top 25 m of the water column from early April on and it remained shallow through the end of the sampling period in late June. Based upon the stratification index, the mixed period (index < 0.125) included sampling from 30 January to 9 March (n = 6 dates) and strongly stratified water column (index > 0.6) encompassed sampling from 18 April to 1 June (n = 7 dates). Simple T-tests were used to test for significant differences in organismal concentrations comparing mixed and stratified periods. Heterotrophic prokaryotes (bacteria) and eukaryotes (pico and nanoflagellates) At 250 depth heterotrophic prokaryote (bacteria and archaebacteria) abundance was highest during the mixed period, declining from a peak abundance of 4.6 × 106 cells ml−1 to ∼2 × 105 cells ml−1 during the stratified period, thus varying only by about a factor of 2–3 (Figure 2). Irregular shifts in abundance characterized the heterotrophic eukaryote population. During the mixed period abundance was relatively high ranging from 390 to 420 cells ml−1. During the transitional period between the stratified and mixed periods concentrations were ∼300 cells ml−1. Subsequently, the heterotrophic eukaryote population increased during the beginning of the stratified period to a peak of ∼630 cells ml−1 and subsequently declined to low levels of ∼160 cells ml−1. Overall, the heterotrophic eukaryotes concentrations varied by a factor of 4, considerably more than the prokaryotes. Figure 2. View largeDownload slide Temporal changes in concentrations of heterotrophic prokaryotes and eukaryotes at 250 m during the study period. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 2. View largeDownload slide Temporal changes in concentrations of heterotrophic prokaryotes and eukaryotes at 250 m during the study period. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Tintinnid ciliates A large number of tintinnid species were encountered (see Supplementary Table S1). Out of a total of 74 species, 53 were found in the deep-water samples (250 m), and 58 in the surface samples (30 m). However, some were found only once or twice in either deep water or surface samples (≤ 10% of dates). These numbered 18 out of 58 in the surface samples and 7 out of 53 in the deep-water samples. Leaving aside these rare occurrence species yields then a pool of 46 for the deep-water samples and 40 in the surface samples. The 46 species found in deep waters on at least three dates can be divided into 17 deep-water species, and those also found in the surface samples, 29 species. Of the 34 species in surface samples found on at least three dates, the surface water species numbered 12. Out of the total species pool, the species found on at least three dates were putatively 17 deep-water forms, 22 occurring throughout the water column and 12 surface water species. Relatively few species occurred commonly, defined here as found on at least 50% of the sampling dates. In the deep water, 20 species were common and in surface layer samples only 10 species were common. In the deep-water samples, the 20 commonly occurring species were composed of two sets: 9 deep-water species and 11 species also found in surface samples. This latter group of forms found both in the deep water and surface samples, was 8 of the 10 common species of the surface layer samples. Thus, common deep-water tintinnids were a mixed assemblage of deep-water species and most of the common surface layer forms. The nine common deep-water species, of basically two size-groups, includes five apparently undescribed forms; (see Figure 3). Figure 3. View largeDownload slide Common forms of tintinnid ciliates found in deep waters, absent or nearly absent from surface waters: (a) an undescribed Salpingella species, (b) an undescribed Albatrossiella species, (c) an undescribed Ormosella species, (d) a small undescribed Amphorellopsis species, (e) an undescribed “ringed” Amphorellopsis species, (f) Favella aciculifera, (g) Daturella striata, (h) Xystonellopsis scyphium, (i) Parundella messinensis. Note that there are two basic groups presumably feeding on different sized prey items: forms with small oral opening diameters of 10–20 µm (a–e) and forms with markedly larger oral opening diameters of 40–50 µm (f–g). Figure 3. View largeDownload slide Common forms of tintinnid ciliates found in deep waters, absent or nearly absent from surface waters: (a) an undescribed Salpingella species, (b) an undescribed Albatrossiella species, (c) an undescribed Ormosella species, (d) a small undescribed Amphorellopsis species, (e) an undescribed “ringed” Amphorellopsis species, (f) Favella aciculifera, (g) Daturella striata, (h) Xystonellopsis scyphium, (i) Parundella messinensis. Note that there are two basic groups presumably feeding on different sized prey items: forms with small oral opening diameters of 10–20 µm (a–e) and forms with markedly larger oral opening diameters of 40–50 µm (f–g). Temporal changes in concentrations differed considerably between surface and deep layer populations (Figure 4). The surface population showed much higher variability compared with the deep-water population. Partly this reflected a bloom of salps that corresponded with the near disappearance of tintinnids in the surface water samples in March. Within the deep water population, changes in concentrations largely reflected concentrations of surface and entire water column species in deep waters, which increased markedly with the depth of the mixed layer, compared with shifts in the concentrations of deep water species (Figure 5). Overall concentrations of tintinnids were significantly (p < 0.006) higher (3.3 cells l−1) during the period of mixis compared with the stratified period (1.4 cells l−1). The species richness of deep-water forms was relatively constant with several forms present regardless of cell abundance whilst the numbers of entire water column and surface species showed a distinct peak during the mixed period (Figure 6). Unlike the Phaeogromid radiolarians and Amphisolenid dinoflagellates (see below), there was no clear pattern amongst any deep-water tintinnid species of presence vs. absence during the mixed compared with stratified periods. Figure 4. View largeDownload slide Temporal changes in the concentrations of tintinnids at 250 m and 30 m during the study period. Note the different scales for populations at 250 m and those found at 30 m. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 4. View largeDownload slide Temporal changes in the concentrations of tintinnids at 250 m and 30 m during the study period. Note the different scales for populations at 250 m and those found at 30 m. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 5. View largeDownload slide Temporal changes in the concentrations of tintinnids at 250 m distinguishing two groups—those found only (or nearly only) in the deep water samples, “Deep spp” from other forms, found mainly or exclusively in the surface water samples, “Entire Water Column and Surface ssp”. Note the different scales for the two groups of species. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 5. View largeDownload slide Temporal changes in the concentrations of tintinnids at 250 m distinguishing two groups—those found only (or nearly only) in the deep water samples, “Deep spp” from other forms, found mainly or exclusively in the surface water samples, “Entire Water Column and Surface ssp”. Note the different scales for the two groups of species. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 6. View largeDownload slide Species richness of the deep-water tintinnid assemblage. Temporal changes in the numbers of tintinnid species at 250 m distinguishing two groups—those found only (or nearly only) in the deep water samples, “Deep spp” from other forms, found mainly or exclusively in the surface water samples, “Entire Water Column and Surface ssp”. Note the relatively invariant number of deep water species. Figure 6. View largeDownload slide Species richness of the deep-water tintinnid assemblage. Temporal changes in the numbers of tintinnid species at 250 m distinguishing two groups—those found only (or nearly only) in the deep water samples, “Deep spp” from other forms, found mainly or exclusively in the surface water samples, “Entire Water Column and Surface ssp”. Note the relatively invariant number of deep water species. Phaeogromid radiolarians A total of nine phaeogromid species were found in the samples from 250 m, and none was encountered in the samples from 30 m (see Supplementary Table S1 and Supplementary Data File for details). In the samples from 250 m, seven species were found on at least three dates: and four occurred commonly (Figure 7). The lowest concentrations were recorded from the beginning of the mixed period and increased over time to peak at the start of the stratified period (Figure 8). Consequently, the average concentration in the mixed period (0.05 cells l−1) compared with the average during the stratified period (0.14 cells l−1) was significantly (p < 0.015) lower. The higher concentrations of the stratified period reflected mainly the concentrations of species which were not encountered during the mixed period or present in trace concentrations, but consistently found in the stratified period: Challengeranium diodon, Challengeron willemoseii, Euphysetta lucani and Euphysetta pusilla. Species richness was highest during the stratified period (Figure 8). Figure 7. View largeDownload slide Common forms of phaeogromid radiolarians found in deep waters: (a) Medusetta parthenopaea, (b) Challengeranium diodon, (c) Challengeria xiphodon, (d) Challengeron willemoesii. Figure 7. View largeDownload slide Common forms of phaeogromid radiolarians found in deep waters: (a) Medusetta parthenopaea, (b) Challengeranium diodon, (c) Challengeria xiphodon, (d) Challengeron willemoesii. Figure 8. View largeDownload slide Temporal changes in the concentrations and species richness (numbers of species) of phaeogromid radiolarians at 250 m. Note that changes in concentrations and species richness are similar. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 8. View largeDownload slide Temporal changes in the concentrations and species richness (numbers of species) of phaeogromid radiolarians at 250 m. Note that changes in concentrations and species richness are similar. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Amphisolenid dinoflagellates Only four amphisolenid species were encountered, three species of Amphisolenia (A. bidentata, A. extensa, A. globulosa) and Triplosolenia bicornis. They were occasionally encountered in surface water samples (4 of 20 dates) in trace concentrations but with exception of the late January sampling, consistently present at 250 m. Two species were found commonly, Amphisolenia globulosa and T. bicornis (Figure 9). There was no significant difference between concentrations in the mixed period vs. the stratified period, both averaged 0.06 cells l−1. However, there was a marked shift in the composition of the assemblage from the mixed period with relatively abundant Amphisolenia and an absence of Triplosolenia to the stratified period dominated by T. bicornis (Figure 10). Figure 9. View largeDownload slide Common species of amphisolenid dinoflagellates found in deep waters: (a) Amphisolenia globulosa, and (b) Triposolenia bicornis. Figure 9. View largeDownload slide Common species of amphisolenid dinoflagellates found in deep waters: (a) Amphisolenia globulosa, and (b) Triposolenia bicornis. Figure 10. View largeDownload slide Temporal changes in the concentrations of species of Amphisolenia and the concentration of Triposolenia bicornis at 250 m. Note the distinct temporal trends of the two genera. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Figure 10. View largeDownload slide Temporal changes in the concentrations of species of Amphisolenia and the concentration of Triposolenia bicornis at 250 m. Note the distinct temporal trends of the two genera. For water column characteristics distinguishing “mixed” from “stratified” periods indicated see Figure 1. Discussion We found clear, consistent and large differences between the surface layer and mesopelagic layer populations both in terms of abundances and species compositions. The differences are unlikely to be due to sampling variability. A previous a study of sampling variability, with regard to tintinnids conducted at a nearby site, using sample sizes considerably smaller than those employed in the present study (10 l) found the major differences amongst replicate samplings to be the number and identity of rare species (Dolan and Stoeck, 2011). The seasonal changes we documented in water column structure and mesopelagic abundances of heterotrophic prokaryotes and eukaryotes largely conform to previous studies in terms of magnitudes. Both Tanaka and Rassoulzadegan (2002) and Winter et al. (2009a, b) investigated populations at the open water Dyfamed site in the NW Mediterranean whilst Weinbauer et al. (2013) reported on samples from our study site. We found prokaryote abundances to range between 1.7 and 4.6 × 105 cells ml−1 and heterotrophic eukaryotes concentrations ranging from 157 to 629 cells ml−1 (Figure 2). Tanaka and Rassoulzadegan found prokaryote abundances at 300 m to vary between 1 and 4 × 105 cells ml−1 from January to May whilst heterotrophic nanoflagellate concentrations varied between 40 and 200 cells ml−1. Winter et al. (2009b) reported mesopelagic prokaryote abundances of 1–2.1 × 105 cells ml−1 between February and June. For the period between March and June, Weinbauer et al. (2013) reported prokaryote abundance of 2.5–3.2 × 105 cells ml−1 and the heterotrophic nanoflagellate concentration from 120 to 250 cells ml−1. Despite largely identical seasonal changes in water column structure, the periods of peak abundance varied amongst the previous studies and compared with our findings. For example Tanaka and Rassoulzadegan (2002) and Winter et al. (2009b), both reporting on samples from the open water Dyfamed site, reported peak prokaryote abundances in February and May respectively. Weinbauer et al. (2013), sampling at the same site as the present report, found peak prokaryote and heterotrophic nanoflagellate abundance in May. We found peak prokaryote and eukaryote abundances in March and May, respectively (Figure 2). However, these apparent differences may be due to the relatively coarse sampling of bi-monthly to monthly previously employed. To the best of our knowledge, our study is the first to address weekly temporal changes amongst mesopelagic protists and document changes in the abundances of individual species. All three groups we examined contained species absent from surface water samples. All three groups showed marked changes in either species compositions, or concentrations, or both over the study period that covered contrasting water column conditions. In the assemblage of tintinnid ciliates nine species were found commonly in the deep-water samples. The deep-sea tintinnids were a minority of the tintinnid assemblage during the period of water column mixis when surface water species were found at depth but dominated the assemblage in the stratified period. Peak abundance occurred at the beginning of the mixed period. The phaeogromid radiolarians, completely absent from the surface samples, showed increased abundances and changes in species richness comparing the mixed and stratified periods. They showed peak abundance at the beginning of the stratified period. Amongst the amphisolenid dinoflagellates peak abundances were recorded in the middle of the mixed period. Species of Amphisolenia were found during the mixed period and T. bicornata, absent during the mixed period, dominated numbers during the stratified period. The protist species we found that appear to be mainly mesopelagic residents includes both forms suspected to be deep-water species and those known to inhabit deep-water layer. For example, amongst the Amphisolenid dinoflagellates, T. bicornis was originally described from plankton net tow material gathered from 250 -to 50 depth, “rarely from surface hauls” (Kofoid, 1906) and Jörgensen (1924) described it as almost exclusively from deep water in the Mediterranean and missing from winter samples, similar to our findings. Amphisolenia species, whilst known from surface water samples (e.g. Gómez et al., 2011), are most abundant in sub-surface waters (100–200) but occur in down to 400 m depth (Tarangkoon et al., 2010). With regard to phaeogromid radiolarians, the species we denoted as common in the mesopelagic are also known from the mesopelagic of the Adriatic Sea (Kršinić and Kršinić, 2012) and the western and central Pacific Ocean (Yamashita et al., 2002). Similar to our findings in the NW Mediterranean, in the Adriatic the radiolarians are very rarely found in the surface layer (Krsinic and Krsinic, 2010). On the other hand, the deep-water tintinnid fauna we encountered may be unique to our study site. The 17 tintinnid species we found to be common or present at 250 m (thus absent or rare in surface samples) differed not only in number (eight species) but also identity of those described as deep water forms in the Adriatic Sea (Kršinić, 1998). In common with the Adriatic list of Kršinić (1998), we found Parundella lohmanni, P. messinensis, Xystonellopsis scyphium, and Favella aciculifera. However, of the five other species reported as Adriatic deep-water forms we found one in both surface and deep water samples (Ormosella trachleum), two are common in the Bay of Villefranche, Amphorides teragona and Cyttarocylis ampulla (Dolan, 2017), and one is not known to date from the study area (Xystonellopsis cymatica). However, it should be noted that the studies in the Adriatic employed a different sampling method, vertical tows of a 53 µm mesh net thus sampling large volumes of water (several m3) but with a net size unlikely to catch the small forms we commonly found (Figure 3). Concentrations of all three groups varied by over an order of magnitude but quite asynchronously. Tintinnid ciliates, phaeogromid radiolarians, and amphisolenid dinoflagellates, are of overall similar sizes and were found in similar concentrations. They would appear to represent roughly similar food items for consumers of microplankton so their distinct abundance patterns argues against strong top-down control by grazers since this would presumably have yielded similar abundance patterns for the three groups. By default the distinct abundance patterns support a hypothesis of variability in group-specific resources and/or variability in group-specific mortality. Whilst we know little about sources of mortality, the three groups in principle exploit very different food resources. Tintinnid ciliates graze on suspended food items with optimal prey size related to the lorica opening diameter. The average optimal prey size is ∼0.25 lorica oral diameter (Dolan, 2010). The tintinnids common in the mesopelagic were of two groups in terms of lorica oral diameter (Figure 3) those with diameters of ∼10–20 µm and forms with diameters of ∼40–50 µm. The mesopelagic tintinnid ciliates presumably relied on prey of ∼4 and 11 µm diameter respectively, thus nano-plankton-size prey such as heterotrophic nanoflagellates. The phaeogromid radiolarians are thought to feed on sinking organic aggregates (Gowing, 1986). A study of the food vacuole content of several species including species recorded in our study (Challengeria xiphodon, C. willemoesii, E. pusilla), found the major food items to be bacteria and eukaryotic algae presumably from ingesting suspended organic aggregates (Gowing and Bentham, 1994). Simple correlation analysis revealed no significant relationship with prokaryotic or eukaryotic heterotrophs enumerated using flow cytometry with tintinnid or radiolarian abundances. However, it should be noted that we have a relatively small number of data points for the flow cytometry counts (n = 13). With regard to amphosolenid dinoflagellate nutrition, virtually nothing is known. They are known to harbour endosymbiotic phototrophs, some with both prokaryotic and eukaryotic endosymbionts; presumably the dinoflagellate host cell profit from the presence of endosymbionts but their role in the nutrition of the host cell has not been examined (Daugbjerg et al., 2013). However, based on the Sechi disk depths, which varied between 8 and 21 m (data not shown), the euphotic zone reached a maximum depth of ∼40 m arguing against a role for endosymbiont photosynthesis in the populations at 250 m depth. The possibility that amphisolenid dinoflagellates below the photic zone digest their symbionts for nutrition cannot be excluded. Overall, our study admittedly raises many more questions than the two we set out to answer. For example, the factors controlling the abundance of the mesopelagic protists considered here clearly remain to be revealed and the possibility that the morpho-species considered may or may not be genetically distinct merits attention. Conclusions We examined short-term temporal changes in the species composition and abundance of three distinct groups of mesopelagic protists in the N.W. Mediterranean Sea over the period corresponding to large changes in water column structure. All three groups examined contained species absent from surface water samples. All three groups showed marked changes in either species compositions, or concentrations, or both over the study period. We established for all three protists groups (tintinnid ciliates, phaeogromid radiolarians, and amphisolenid dinoflagellates) that there appear to be specific mesopelagic forms and the three groups display distinct seasonal patterns of abundance and species composition. We conclude that the mesopelagic protist assemblage is clearly diverse and dynamic at least with regard to groups in which species can be identified using light microscopy. Acknowledgements We thank the skipper, Jean-Yves Carval, and crew members Pierre Cohen and Julien Garde, of the vessel “Sagitta 3” for ship operations and aid in sampling. Water column profile data used to calculate MLD and the stratification index for Point C were kindly supplied by the SOMLIT coastal observation program. Sabine Agatha and Sylvia Angerer kindly commented on some species identifications. However, the responsibility for all species identification, for better or for worse, remains with the authors. Funding This work was funded by the French Agence National de la Recherche through the AncesStram project coordinated by David Moreira, Université Paris-Sud. References Abboud-Abi Saab M. 2008 . Tintinnids of the Lebanese Coastal Waters (Eastern Mediterranean). CNRS-Lebanon/UNEP/MAP/RAC/SPA, Lebanon, 192 pp. Anderson O. R. 1983 . Radiolaria . Springer-Verlag , New York. 355 pp. Aristegui J. , Gasol J. M. , Duarte C. M. , Herndld G. J. 2009 . Microbial oceanography of the dark ocean's pelagic realm . Limnology and Oceanography , 54 : 1501 – 1529 . Google Scholar Crossref Search ADS Balech E. 1959 . Tintinnoinea del Mediterraneo . Trabajos Del Instituto Español De Oceanografia , 28 : 1 – 88 . Behrenfeld M. J. , O’Malley R. T. , Siegel D. A. , McClain C. R. , Sarmiento J. L. , Feldman G. C. , Milligan A. J. , et al. 2006 . Climate-driven trends in contemporary ocean productivity . Nature , 444 : 752 – 755 . Google Scholar Crossref Search ADS PubMed Borgert A. 1906 . Die tripyleen radiolarian der plankton-expedition . Medusettidae. Ergebnisee Plankton-Expedition Der Humboldt-Stiftung , 3 : 133 – 193 . Borgert A. 1911 . Die tripyleen radiolarian der plankton-expedition . Challengeridae. Ergebnisee Plankton-Expedition Der Humboldt-Stiftung , 3 : 148 – 536 . Burd A. B. , Hansell D. A. , Steinberg D. K. , Anderson T. R. , Arístegui J. , Baltar F. , Beaupré S. R. , et al. 2010 . Assessing the apparent imbalance between geochemical and biochemical indicators of meso-and bathypelagic biological activity: what the@ $♯! is wrong with present calculations of carbon budgets? Deep Sea Research Part II: Topical Studies in Oceanography , 57 : 1557 – 1571 . Google Scholar Crossref Search ADS Bustillos-Guzmán J. , Claustre H. , Marty J. C. 1995 . Specific phytoplankton signatures and their relationship to hydrographic conditions in the coastal northwestern Mediterranean Sea . Marine Ecology Progress Series , 124 : 247 – 258 . Google Scholar Crossref Search ADS Calbet A. , Landry M. R. 2004 . Phytoplankton growth, microzooplankton grazing, and carbon cycling in marine systems . Limnology and Oceanography , 49 : 51 – 57 . Google Scholar Crossref Search ADS Christaki U. , Courties C. , Massana R. , Catala P. , Lebaron P. , Gasol J. M. , Zubkov M. V. 2011 . Optimized routine flow cytometric enumeration of heterotrophic flagellates using SYBR Green I . Limnology and Oceanography Methods , 9 : 329 – 339 . Google Scholar Crossref Search ADS Conley W. J. , Hopkins T. L. 2004 . Feeding ecology of lanternfish (Pisces: Myctophidae) larvae: prey preferences as a reflection of morphology . Bulletin of Marine Science , 75 : 361 – 379 . Costello M. J. , Breyer S. 2017 . Ocean depths: the mesopelagic and implications for global warming . Current Biology , 27 : R36 – R38 . Google Scholar Crossref Search ADS PubMed Daugbjerg N. , Jensen M. H. , Hansen P. J. 2013 . Using nuclear-encoded lsu and ssu rDNA sequences to identify the eukaryotic endosymbiont in Amphisolenia bidentata (Dinophyceae) . Protist , 164 : 411 – 422 . Google Scholar Crossref Search ADS PubMed Dave A. C. , Lozier M. S. 2010 . Local stratification control of marine productivity in the subtropical North Pacific . Journal of Geophysical Research: Oceans , 115 , doi: 10.1029/2010JC006507. de Boyer Montégut C. , Madec G. , Fischer A. S. , Lazar A. , Iudicone D. 2004 . Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology . Journal of Geophysical Research: Oceans , 109 , doi: 10.1029/2004JC002378. Dolan J. R. 2010 . Morphology and ecology in tintinnid ciliates of the marine plankton: correlates of lorica dimensions . Acta Protozoologica , 49 : 235 – 344 . Dolan J. R. 2017 . Historical trends in the species inventory of tintinnids (ciliates of the microzooplankton) in the Bay of Villefranche (NW Mediterranean Sea): Shifting Baselines . European Journal of Protistology , 57 : 16 – 25 . Google Scholar Crossref Search ADS PubMed Dolan J. R. , Stoeck T. 2011 . Repeated sampling reveals differential variability in measures of species richness and community composition in planktonic protists . Environmental Microbiology Reports , 3 : 661 – 666 . Google Scholar Crossref Search ADS PubMed Dolan J. R. , Agatha S. , Coats D. W. , Montagnes D. J. S. , Stocker D. K. (Eds.) 2013 . Biology and Ecology of Tintinnid Ciliates: Models for Marine Plankton , Blackwell , Oxford . D'Ortenzio F. , Iudicone D. , de Boyer Montegut C. , Testor P. , Antoine D. , Marullo S. , Santoleri R. , et al. 2005 . Seasonal variability of the mixed layer depth in the Mediterranean Sea as derived from in situ profiles . Geophysical Research Letters , 32 , doi: 10.1029/2005GL022463. Edgcomb V. P. 2016 . Marine protist associations and environmental impacts across trophic levels in the twilight zone and below . Current Opinion in Microbiology , 31 : 169 – 175 . Google Scholar Crossref Search ADS PubMed Gasol J. M. , Del Giorgio P. A. 2000 . Using flow cytometry for counting natural planktonic bacteria and understanding the structure of planktonic bacterial communities . Scientia Marina , 64 : 197 – 224 . Google Scholar Crossref Search ADS Gaines G. , Elbrächter M. 1987 . Heterotrophic nutrition. In The Biology of Dinoflagellates, Ed. by F. J. R. Taylor. Blackwell, Oxford, pp. 224 – 68 . Giering S. L. C. , Sanders R. , Lampitt R. S. , Anderson T. R. , Tamburini C. , Boutrif M. , Zubkov M. V. , et al. 2014 . Reconciliation of the carbon budget in the ocean's twilight zone . Nature , 507 : 480. Google Scholar Crossref Search ADS PubMed Gómez F. , López-García P. , Moreira D. 2011 . Molecular phylogeny of dinophysoid dinoflagellates: the systematic position of Oxyphysis oxytoxoides and the Dinophysis hastata group (Dinophysales, Dinophyceae) . Journal of Phycology , 47 : 393 – 406 . Google Scholar Crossref Search ADS PubMed Gowing M. M. 1986 . Trophic biology of phaeodarian radiolarians and flux of living radiolarians in the upper 2000 m of the North Pacific central gyre . Deep Sea Research Part A , 33 : 655 – 674 . Google Scholar Crossref Search ADS Gowing M. M. , Bentham W. N. 1994 . Feeding ecology of phaeodarian radiolarians at the VERTEX North Pacific Time series site . Journal of Plankton Research , 16 : 707 – 719 . Google Scholar Crossref Search ADS Gowing M. M. , Silver M. W. 1985 . Minipellets: a new and abundant size class of marine fecal pellets . Journal of Marine Research , 43 : 395 – 418 . Google Scholar Crossref Search ADS Gowing M. M. , Garrison D. L. , Wishner K. F. , Gelfman C. 2003 . Mesopelagic microplankton of the Arabian Sea . Deep Sea Research Part I , 50 : 1205 – 1234 . Google Scholar Crossref Search ADS Heimbürger L. E. , Lavigne H. , Migon C. , D’Ortenzio F. , Estournel C. , Coppola L. , Miquel J. C. 2013 . Temporal variability of vertical export flux at the DYFAMED time-series station (Northwestern Mediterranean Sea) . Progress in Oceanography , 119 : 59 – 67 . Google Scholar Crossref Search ADS Herrmann M. , Somot S. , Sevault F. , Estournel C. , Déqué M. 2008 . Modeling the deep convection in the northwestern Mediterranean Sea using an eddy-permitting and an eddy-resolving model: case study of winter 1986–1987 . Journal of Geophysical Research: Oceans , 113 , doi: 10.1029/2006JC003991. Hopcroft R. R. , Robison B. H. 1999 . A new mesopelagic larvacean, Mesochordaeus erythrocephalus, sp. nov., from Monterey Bay, with a description of its filtering house . Journal of Plankton Research , 21 : 1923 – 1937 . Google Scholar Crossref Search ADS Houpert L. , Durrieu de Madron X. , Testor P. , Bosse A. , D'Ortenzio F. , Bouin M. N. , Dausse D. , et al. 2016 . Observations of open-ocean deep convection in the northwestern Mediterranean Sea: seasonal and interannual variability of mixing and deep water masses for the 2007–2013 period . Journal of Geophysical Research, Oceans , 121 : 8139 – 8171 . Google Scholar Crossref Search ADS Irigoien, X., Klevjer, T. A., Røstad, A., Martinez, U., Boyra, G., Acuña, J. L. et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean. Nature communications , 5 , 3271 . Jörgensen E. 1924 . Mediterranean Tintinnids. Report on the Danish Oceanographical Expeditions 1908–10 to the Mediterranean and Adjacent Seas. 2 J.3. (Biology): 1 – 110 . Kling S. A. , Boltovskoy D. 1999 . Radiolaria phaeodaria. In South Altantic Zooplankton , pp. 231 – 264 . Ed. by Boltovskoy D. . Backhuys Publishers , Leiden . Kofoid C. A. 1906 . Dinoflagellata of the San Diego region II. On Triposolenia, a new genus of the Dinophysidae . University of California Publications in Zoology , 3 : 93 – 116 . Kofoid C. A. , Campbell A. S. 1929 . A conspectus of the marine and freshwater Ciliata belonging to the suborder Tintinnoinea, with descriptions of new species principally from the Agassiz expedition to the eastern tropical pacific 1904–1905 . University of California Publications in Zoology , 34 : 403 . Kofoid C. A. , Campbell A. S. 1939 . The Tintinnoinea of the Eastern Tropical Pacific . Bulletin of the Museum of Comparative Zoology at Harvard College , 84 : 1 – 473 . Kofoid C. A. , Skogsberg T. 1928 . The Dinoflagellata: The Dinophysoidae. No. 35 Reports on the Scientific Results of the Expedition to the Eastern Tropical Pacific in charge of Alexander Agassiz, by the U. S. Fish Commission Steamer ‘Albatross’, from October 1904, to March 1905, Lieut. Commander L. M. Garrett, U.S.N., commanding . Bulletin of the Museum of Comparative Zoölogy at Harvard College , 51 : 1 – 766 . Kršinić F. 1998 . Vertical distribution of protozoan and microcopepod communities in the South Adriatic Pit . Journal of Plankton Research , 20 : 1033 – 1060 . Google Scholar Crossref Search ADS Krsinic F. 2010 . Tintinnids (Tintinnida, Choreotrichia, Ciliata) in the Adriatic Sea, Mediterranean. Part I. Taxonomy, Part II. Ecology . Institute of Oceanography and Fisheries , Split, Croatia . Kršinić F. , Kršinić A. 2012 . Radiolarians in the Adriatic Sea plankton (Eastern Mediterranean) . Acta Adriatica , 53 : 189 – 211 . Lozier M. S. , Dave A. C. , Palter J. B. , Gerber L. M. , Barber R. T. 2011 . On the relationship between stratification and primary productivity in the North Atlantic . Geophysical Research Letters , 38 , doi: 10.1029/2011GL049414. Luna G. M. , Bianchelli S. , Decembrini F. , De Domenico E. , Danovaro R. , Dell'Anno A. 2012 . The dark portion of the Mediterranean Sea is a bioreactor of organic matter cycling . Global Biogeochemical Cycles , 26 : GB2017. Google Scholar Crossref Search ADS Marie D. , Rigaut-Jalabert F. , Vaulot D. 2014 . An improved protocol for flow cytometry analysis of phytoplankton cultures and natural samples . Cytometry Part A , 85 : 962 – 968 . Google Scholar Crossref Search ADS Mayor D. J. , Sanders R. , Giering S. L. , Anderson T. R. 2014 . Microbial gardening in the ocean's twilight zone: detritivorous metazoans benefit from fragmenting, rather than ingesting, sinking detritus . BioEssays , 36 : 1132 – 1137 . Google Scholar Crossref Search ADS PubMed Mostajir B. , Dolan J. R. , Rassoulzadegan F. 1995 . Seasonal variations of pico- and nano-detrital particles (DAPI Yellow Particles, DYP) in the Ligurian Sea (NW Mediterranean) . Aquatic Microbial Ecology , 9 : 259 – 266 . Google Scholar Crossref Search ADS Nakamura Y. , Suzuki N. 2015 . Phaeodaria: diverse marine cercozoans. In Marine Protists Diversity and Dynamics , pp. 179 – 222 . Ed. by Ohtsuka S. , Suzuki T. , Horiguchi T. , Suzuki N. , Not F. . Springer , Tokyo . Nothing E. M. , Gowing M. M. 1991 . Late winter abundance and distribution of phaeodarian radiolarians, other large protozooplankton and copepod nauplii in the Weddell Sea, Antarctica . Marine Biology , 111 : 473 – 484 . Google Scholar Crossref Search ADS Proud R. , Cox M. J. , Brierley A. S. 2017 . Biogeography of the global ocean’s mesopelagic zone . Current Biology , 27 : 113 – 119 . Google Scholar Crossref Search ADS PubMed Uttal L. , Buck K. R. 1996 . Dietary study of the midwater polychaete Poeobius meseres in Monterey Bay, California . Marine Biology , 125 : 333 – 343 . Google Scholar Crossref Search ADS Saldarriaga J. F. , Taylor F. J. R. 2017 . Dinoflagellata. In Handbook of the Protists , 2nd edn. , 1, pp. 625 – 678 . Ed. by Archibald J. M. , Simpson A. G. B. , Slamovits C. H. . Springer, New York. Sano M. , Maki K. , Nishibe Y. , Nagata T. , Nishida S. 2013 . Feeding habits of mesopelagic copepods in Sagami Bay: insights from integrative analysis . Progress in Oceanography , 110 : 11 – 26 . Google Scholar Crossref Search ADS Sanders R. J. , Henson S. A. , Martin A. P. , Anderson T. R. , Bernardello R. , Enderlein P. , Fielding S. , et al. 2016 . Controls over ocean mesopelagic interior carbon storage (COMICS): fieldwork, synthesis, and modeling efforts . Frontiers in Marine Science , 3 : 136. Google Scholar Crossref Search ADS Robinson C. , Steinberg D. K. , Anderson T. R. , Arístegui J. , Carlson C. A. , Frost J. R. , Ghiglione J-F. , et al. 2010 . Mesopelagic zone ecology and biogeochemistry – a synthesis . Deep Sea Research Part II , 57 : 1504 – 1518 . Google Scholar Crossref Search ADS Sutton T. T. 2013 . Vertical ecology of the pelagic ocean: classical patterns and new perspectives . Journal of Fish Biology , 83 : 1508 – 1527 . Google Scholar Crossref Search ADS PubMed Sutton T. T. , Clark M. R. , Dunn D. C. , Halpin P. N. , Rogers A. D. , Guinotte J. , Bograd , et al. 2017 . A global biogeographic classification of the mesopelagic zone . Deep Sea Research Part I , 126 : 85 – 102 . Google Scholar Crossref Search ADS St. John M. A. , Borja A. , Chust G. , Heath M. , Grigorov I. , Mariani P. , Martin A. P , et al. 2016 . A dark hole in our understanding of marine ecosystems and their services: Perspectives from the mesopelagic community . Frontiers in Marine Science , 3 : 31 . Google Scholar Crossref Search ADS Tanaka T. , Rassoulzadegan F. 2002 . Full-depth profile (0–2000 m) of bacteria, heterotrophic nanoflagellates and ciliates in the NW Mediterranean Sea: vertical partitioning of microbial trophic structures . Deep Sea Research Part II , 49 : 2093 – 2107 . Google Scholar Crossref Search ADS Tarangkoon W. , Hansen G. , Hansen P. J. 2010 . Spatial distribution of symbiont-bearing dinoflagellates in the Indian Ocean in relation to oceanographic regimes . Aquatic Microbial Ecology , 58 : 197 – 213 . Google Scholar Crossref Search ADS Taylor F. J. R. 1976 . Dinoflagellates from the International Indian Ocean Expedition. A report of material collected by the R.V. “Anton Brun” 1963–1964 . Bibliotheca Botanica , 132 : 1 – 234 . Taylor F. J. R. , Pollingher P. 1987 . Ecology of dinoflagellates. In The Biology of Dinoflagellates , pp. 399 – 529 . Ed. by Taylor F. J. R. . Blackwell , Oxford . Weinbauer M. G. , Liu J. , Motegi C. , Maier C. , Pedrotti M. L. , Dai M. , Gattuso J. P. 2013 . Seasonal variability of microbial respiration and bacterial and archaeal community composition in the upper twilight zone . Aquatic Microbial Ecology , 71 : 99 – 115 . Google Scholar Crossref Search ADS Winter C. , Kerros M. E. , Weinbauer M. G. 2009a . Seasonal changes of bacterial and archaeal communities in the dark ocean: evidence from the Mediterranean Sea . Limnology and Oceanography , 54 : 160 – 170 . Google Scholar Crossref Search ADS Winter C. , Kerros M. E. , Weinbauer M. G. 2009b . Seasonal and depth-related dynamics of prokaryotes and viruses in surface and deep waters of the northwestern Mediterranean Sea . Deep Sea Research Part I , 56 : 1972 – 1982 . Google Scholar Crossref Search ADS Yamashita H. , Takahashi K. , Fujitani N. 2002 . Zonal and vertical distribution of radiolarians in the western and central Equatorial Pacific in January 1999 . Deep Sea Research Part II: Topical Studies in Oceanography , 49 : 2823 – 2862 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2017. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Koslow, J, Anthony;Davison,, Pete;Ferrer,, Erica;Jiménez Rosenberg, S Patricia, A;Aceves-Medina,, Gerardo;Watson,, William

doi: 10.1093/icesjms/fsy154pmid: N/A

Abstract Declining oxygen concentrations in the deep ocean, particularly in areas with pronounced oxygen minimum zones (OMZs), are a growing global concern related to global climate change. Its potential impacts on marine life remain poorly understood. A previous study suggested that the abundance of a diverse suite of mesopelagic fishes off southern California was closely linked to trends in midwater oxygen concentration. This study expands the spatial and temporal scale of that analysis to examine how mesopelagic fishes are responding to declining oxygen levels in the California Current (CC) off central, southern, and Baja California. Several warm-water mesopelagic species, apparently adapted to the shallower, more intense OMZ off Baja California, are shown to be increasing despite declining midwater oxygen concentrations and becoming increasingly dominant, initially off Baja California and subsequently in the CC region to the north. Their increased abundance is associated with warming near-surface ocean temperature, the warm phase of the Pacific Decadal oscillation and Multivariate El Niño-Southern Oscillation Index, and the increased flux of Pacific Equatorial Water into the southern CC. Introduction Oxygen concentrations are declining globally in the world’s oceans (Helm et al., 2011; Ito et al., 2017; Schmidtko et al., 2017), consistent with the predictions of global climate models: oxygen is less soluble and ventilation is reduced in a warmer, more stratified ocean (Matear and Hirst, 2003; Shaffer et al., 2009). However, while models predict a decline of 1–7% in global ocean oxygen content by 2100 (Keeling et al., 2010; Schmidtko et al., 2017), declines on the order of 20–25% have been observed in recent decades at midwater depths in an extensive region characterized by oxygen minimum zones (OMZs) around the rim of the North Pacific Ocean (Bograd et al., 2008; Stramma et al., 2008; McClatchie et al., 2010; Crawford and Peña, 2016). These amplified declines in oxygen are believed to be related to the Pacific Decadal Oscillation (PDO) and associated basin-scale wind systems and temperature-related amplification of microbial metabolism (Deutsch et al., 2011, 2014). This amplification provides an opportunity to examine the potential long-term ecological impacts of climate change-induced deoxygenation, as well as its effects on decadal time scales in particularly dynamic regions. Ecosystems in these regions are likely to be most sensitive to varying oxygen conditions, since oxygen concentrations in the vicinity of OMZs are already at critical levels that affect organismal behaviour and physiology (Vaquer-Sunyer and Duarte, 2008). The long-term ecological consequences of declining midwater oxygen concentrations and expanding OMZs remain poorly understood. There are few time series for mesopelagic ecosystems (Robison, 2009; Koslow and Couture, 2015). Furthermore, although OMZs are common in the world ocean, their depth range and extent vary substantially. Data from the World Ocean Atlas reveal that even within the California Current (CC), the upper bound of the OMZ, where oxygen concentrations are ≤0.5 ml l−1, shoals from ∼600 m depth off central and southern California (32–35°N latitude) to about 200 m off southern Baja California (23° N latitude) (Garcia et al., 2013). Off southern California, the deep scattering layer (DSL) generally resides in daytime within the hypoxic boundary layer (HBL) (0.5 ml l−1 ≤ [O2] <1.5 ml l−1), which extends for several hundred metres above the OMZ, with the depth of the DSL varying with the depth of the OMZ and HBL (Koslow et al., 2011; Bianchi et al., 2013, Netburn and Koslow, 2015). Since 1984, oxygen concentrations off southern California declined by 21% at 300 m depth, and the OMZ shoaled on average 41 m (Bograd et al., 2008), leading to a mean increase in a factor of 2.5 in the light level available to visually orienting predators within the HBL (Koslow et al., 2011). Given that diel vertical migration serves the mesopelagic micro-nekton primarily as a means to reduce mortality to visually orienting predators, the DSL cannot shoal indefinitely in response to a shoaling OMZ. In regions with a notably shallow OMZ (depth <200 m), such as off southern Baja California or within the Eastern Tropical Pacific (ETP) or the Humboldt Current off Peru, the dominant mesopelagic fishes such as Triphoturus mexicanus and Vinciguerria lucetia typically reside within the OMZ itself during the day and vertically migrate at night into near-surface waters to feed and re-oxygenate (Robison, 1972; Cornejo and Koppelmann, 2006). A suite of physical and physiological adaptations, including metabolic suppression, are required for these taxa to reside within suboxic waters for extended periods (Childress and Seibel, 1998; Seibel, 2011). We hypothesize that continued declines in midwater oxygen concentrations and the concomitant shoaling of the OMZ and HBL will lead eventually to changes in the mesopelagic fauna, such that it descends into the OMZ proper during daytime to avoid predation or it is replaced by taxa adapted to do so. The waters off southern California serve as an ecotone or transition zone for coastal fishes and intertidal and nearshore invertebrates, with Pt. Conception often considered a biogeographic boundary between the cool-water fauna of the Oregonian Province (coextensive with the northern and central CC) and the warmer-water fauna of the San Diego Province within the southern CC (Horn et al., 2006; Blanchette et al., 2008) (Figure 1). The Ensenada Front, situated approximately offshore of the US–Mexico border, may serve as a further oceanographic and faunal boundary, separating the warm-water fauna of subtropical/tropical affinity from the cooler-water fauna to the north (Haury et al., 1993; Lara-Lopez et al., 2012). The distribution of mesopelagic fishes is generally more extensive than that of coastal faunas, with the range of cool-water taxa in the CC often extending into sub-Arctic waters or across the Transition Zone to the western Pacific, while the range of warm-water taxa often extends into low-latitude subtropical/tropical waters. These faunas meet in the waters off southern California (Horn et al., 2006; Moser and Watson, 2006). Figure 1. View largeDownload slide (a) Schematic of the California Current (CC) (adapted from Checkley and Barth, 2009), showing its customary divisions into the Central and Southern CC at around Pt Conception. The southern CC is further sub-divided into the southern California Bight characterized by a gyral circulation, and the CC off Baja California, south of the quasi-permanent Ensenada Front. (b) The CalCOFI sampling stations off central and southern California. The onshore–offshore line demarcates the stations used for southern California region south of Pt Conception and the central California region north of it, while the offshore line demarcates offshore stations that were sampled less consistently and therefore not included in the study. (c) The combined CalCOFI/IMECOCAL sampling stations off Baja California. The box around stations in Vizcaino Bay, Baja California contains the most consistently sampled stations used in the study. Figure 1. View largeDownload slide (a) Schematic of the California Current (CC) (adapted from Checkley and Barth, 2009), showing its customary divisions into the Central and Southern CC at around Pt Conception. The southern CC is further sub-divided into the southern California Bight characterized by a gyral circulation, and the CC off Baja California, south of the quasi-permanent Ensenada Front. (b) The CalCOFI sampling stations off central and southern California. The onshore–offshore line demarcates the stations used for southern California region south of Pt Conception and the central California region north of it, while the offshore line demarcates offshore stations that were sampled less consistently and therefore not included in the study. (c) The combined CalCOFI/IMECOCAL sampling stations off Baja California. The box around stations in Vizcaino Bay, Baja California contains the most consistently sampled stations used in the study. Mesopelagic taxa represent the most speciose group in the CalCOFI ichthyoplankton data set for southern California, comprising 38% of the taxa, and they are the second most abundant after the coastal pelagic species, with 20% of the total numbers (Moser and Watson, 2006). The dominant pattern in this data set from 1951 to 2008 was a coherent pattern of change in mesopelagics, correlated with the annual mean midwater oxygen concentration over this period. The overall abundance of mesopelagic fishes in the CalCOFI time series declined 63% between periods of high and low midwater oxygen concentration (Koslow et al., 2011). This massive decline has potentially profound implications for ocean food webs, given the widespread occurrence of OMZs in relation to eastern boundary currents and other productive ocean regions and the ongoing and predicted future declines in mid-depth oxygen concentrations (Helly and Levin, 2004; Breitburg et al., 2018). Among fishes, mesopelagics comprise the dominant zooplankton consumers in most of the world’s oceans, with an estimated biomass on the order of 109 tons, ∼100-fold greater than global annual marine fish landings (Irigoien et al., 2014). Our objective in this study is to extend our analysis to a wider region, both north of Pt. Conception off central California and off Baja California, south of the US–Mexico border. The decline in midwater oxygen has been a regional phenomenon, reported from the subarctic Northeast Pacific (Crawford and Peña, 2016) to the ETP (Stramma et al., 2008), as well as off southern California (Bograd et al., 2008). In extending our analysis to the waters north and south of southern California, our initial hypothesis was that the strong correlation between mesopelagic fish abundance and midwater oxygen concentration would be found regionally, consistent with the coherent ecosystem responses observed over much of the CC to major ENSO events (Chelton et al., 1982) or PDO regime shifts on decadal time scales (Hare and Mantua, 2000). We propose to test for regional coherence both by comparing multivariate analyses carried out separately in the different regions and by carrying out combined analyses on the most common taxa across adjacent regions. If the dominant temporal patterns are coherent, we expect that taxa will load consistently across regions on the dominant modes. If they are out of phase, taxa from the two regions will load highly but with opposite signs on the dominant modes of variance. And if patterns from two regions are unrelated, it is expected that taxa from mostly one region will load highly on one mode, and taxa from the second will load highly on another. We also extend our previous time series off southern California from 2008 to 2015 in this paper, which enables us to test the stability of previously observed relationships between the abundance of the mesopelagic community and ocean drivers. Methods Data In this paper, we use larval fish time series from the California Cooperative Oceanic Fisheries Investigations (CalCOFI) and Investigaciones Mexicanas de la Corriente California (IMECOCAL) ichthyoplankton survey programs as proxies for the spawning stock biomass of fishes off central, southern, and Baja California. Larval fishes are predominantly sampled in these surveys during the early preflexion stage when net-sampler avoidance is minimal and post-spawning losses to mortality are still relatively low, so larval abundance is highly congruent with adult spawning stock biomass, as shown for a range of taxa in the CC for which there are stock assessments or other data on adult stock biomass: northern anchovy and Pacific sardine (Koslow et al., 2011), rockfishes (Sebastes) (Moser et al., 2000), various rocky nearshore taxa (Moser et al., 2001), and California halibut (Moser and Watson, 1990). From 1951 through 1966 the CalCOFI program sampled the region from central California to Baja California monthly with a systematic sampling grid (Figure 1). From 1967 to 1984, surveys over the annual cycle were carried out only every third year. From 1984 onward, surveys were carried out on a quarterly basis but restricted primarily to southern California, the region from Pt. Conception to the US–Mexico border. In 1998, Mexico resumed full sampling of the former CalCOFI stations in Mexican waters using CalCOFI protocols under the IMECOCAL program, so our time series for Baja California combines the CalCOFI (1951–1981) and IMECOCAL (1998–2011) sampling periods. Analysis of the ichthyoplankton time series off southern California and comparison to the time series off Baja California were based on the annual mean abundance of larval fish taxa averaged over the six most consistently sampled transects from the US–Mexico border to just north of Pt. Conception (Figure 1). The use of annual means enables us to include all taxa within the region, regardless of spawning season. The annual means, expressed in numbers per square metre, were calculated as the mean of the seasonal mean abundances, which in turn were calculated as the mean abundance for all cruises within a season. Cruise means were calculated as the arithmetic mean abundance for all stations within a cruise. Since 1984, CalCOFI cruises have been conducted seasonally, so the annual means were simply the mean abundance over the four annual cruises. However, the estimation of annual means from seasonal means avoids potential bias in those years with more than one cruise within a season. Years were included in the analyses only if they were sampled over three or four seasons. Taxa off southern California spawn primarily in the winter and spring, and these seasons were consistently sampled. Prior to carrying out further analyses, the annual mean larval abundances x were log10-transformed: log10(x + 1). In 2003 sampling resumed on transects north of Pt. Conception to San Francisco in spring to better cover the spawning distribution of Pacific sardine; these northern CalCOFI transects were also covered in the spring-time of 1991 and 1998. To enable an unbiased comparison to the data set from the core-CalCOFI survey region, comparative analyses for the central and southern California regions were based on the mean abundance of taxa in the spring months (March–May). The spring-only time series also provided several additional years of data (through 2015), because analysis of the spring ichthyoplankton samples was prioritized for stock assessment purposes. The spring time series for central and southern California were based on the six transects north and five transects south of Pt. Conception, respectively, the widely accepted biogeographic boundary for the region (Horn et al., 2006) (Figure 1). Oceanographic sampling technology has advanced considerably since 1951, leading to an expanded range of ocean measurements on CalCOFI and IMECOCAL cruises (Ohman and Venrick, 2003; McClatchie, 2014). However, the core measurements have remained comparable, with physical, chemical, and biogeochemical sampling carried out at each station to 500 m depth based on a mix of continuous CTD and discrete water sample measurements: temperature, salinity, oxygen concentration, chlorophyll a, nutrients, among others. An oblique plankton tow is carried out at each station, with a flowmeter attached to the net to measure sampling volume so larval abundance per unit volume and area may be estimated. Our analyses were carried out based on abundances per m2. Zooplankton displacement volume (DV) was measured from each sample, all ichthyoplankton were removed, identified to the lowest possible taxon, and enumerated. The reader is referred to Kramer et al (1972) and Smith and Richardson (1977) for detailed descriptions of the ichthyoplankton protocols and McClatchie (2014) for a comprehensive modern review. There were two key changes in the ichthyoplankton sampling protocol over time. Initially sampling was with a 1-m diameter ring net towed with a bridle in front of the mouth from the surface to 140 m depth [200 metres wire out (mwo)]. In 1969, the sampling depth was increased to 210 m (300 mwo). This led us to eliminate several mesopelagic taxa from the analysis that are distributed predominantly in the deeper stratum (140–210 m) (Ahlstrom, 1959; Moser and Smith, 1993), whose abundance in the time series underwent an obvious step-change in 1969. The second key change occurred in 1977, when the 1-m diameter ring net with tow bridle was superseded by a 71-cm diameter bongo net without a tow bridle. The tow bridle enhances avoidance of the net (Thompson et al., 2017), but the effect is fairly minor since it predominantly affects post-flexion larvae and appears consistent across taxa and therefore does not affect our multivariate analysis. A fuller treatment of these changes and their potential impacts on our study is provided in the Supplementary. The biogeographic distributions of mesopelagic taxa were classified based on Moser (1996), as well as the authors’ expert knowledge. Taxa largely restricted to the CC were classified as CC endemics. Taxa with distributions extending into tropical–subtropical waters south of the CC were classified as having warm-water affinities, while those with distributions extending into the Transition Zone and sub-Arctic of the North Pacific were classified as having cold-water affinities. A few taxa with broad, virtually cosmopolitan distributions were classified as “broad.” Both local and regional ocean variables were used to examine potential relationships between patterns of larval fish abundance and potential physical forcing mechanisms. The primary local ocean variables were mean oxygen concentration at 200–400 m depth and temperature at 10 m depth, which were obtained on the same ocean cruises and stations as the larval fish samples and averaged in the same way, first by cruise, then by season and year. Midwater oxygen concentration off southern California was used in analyses with the Baja California time series because the midwater oxygen time series off Baja California is missing values for 1998–2001. The midwater oxygen time series off southern and Baja California were significantly correlated (r = 0.39, n = 25, p < 0.05). The following large-scale ocean variables were incorporated in the analysis: the multivariate El Niño-Southern Oscillation (ENSO) Index (MEI) (Wolter and Timlin, 1998) as maintained by NOAA: https://www.esrl.noaa.gov/psd/enso/mei/; the Pacific Decadal Oscillation (PDO), which is the first principal component (PC) of the SST fields of the Pacific north of 20° N latitude (Mantua et al., 1997) obtained from: http://research.jisao.washington.edu/pdo/PDO.latest.txt (Supplementary Figure S1); and the North Pacific Gyre Oscillation (NPGO), which is PC 2 of the North Pacific sea-surface height (SSH) fields (PC1 of the North Pacific SST and SSH fields are high correlated) (Di Lorenzo et al., 2008), was obtained from http://www.oces.us/npgo/npgo.php (Supplementary Figure S1). Statistical analysis PC analysis was used to reduce the dimensionality of the ichthyoplankton data sets and to extract the dominant patterns in the time series. Our PC analyses were based on the correlation matrix for the log-transformed annual mean abundances of all taxa using the Statistical Package for the Social Sciences (SPSS). PCA based on the correlation matrix normalizes the data to deviations from the mean in standard deviation units and thus weights taxa equally; analysis based on the variance–covariance matrix would be highly weighted by a relatively few dominant taxa (Legendre and Legendre, 2012). The loading of each variable on a PC is equivalent to the correlation between the time series for that variable and the PC. The loadings thus enable evaluation of the contribution of each variable to a PC and of the coherence of taxa with the PC time series. In presenting the results of a PCA, the loadings of all taxa that nominally loaded significantly on the PC are shown, based on the loading-correlation and degrees of freedom (df). Taxa were only included in the analysis if they were present in at least half the years of the time series, such that 57 taxa were included in the Baja California analysis, 75 from southern California (the core CalCOFI sampling area), and 28 from central California (CalCOFI stations from Pt. Conception to San Francisco) (Figure 1). Off Baja California, only stations in the Vizcaino Bay area were consistently sampled over the course of the time series, so the analysis was restricted to those stations (Figure 1). The PC1Baja time series extends from 1951 to 2011 but is missing data from 1984 to 1997 and 2009. The dominant multivariate pattern, PC 1, from all three areas was dominated by a broad suite of mesopelagic fishes, which serves as the focus of our study. Comparison of the dominant multivariate patterns of larval fish abundance with ocean forcing was based primarily on simple graphical and statistical analyses: plots of the time series, scatterplots of one against another, and correlation analyses. We considered our analyses largely exploratory, so significance levels were not corrected for possible autocorrelation. Where there appeared to be a potential regime shift or step-change in time series, this was tested with sequential t-test analysis of regime shifts (STARS) (Rodionov, 2004). Analysis of covariance was used to test for changes in relationships between variables between different time periods. However, our paper does not focus on apparent regime shifts because our ichthyoplankton time series are not well-suited to such analysis: CalCOFI sampling was only triennial during the period of the 1977 PDO regime shift (Mantua et al., 1997) and sampling off Baja California only resumed in 1998, which precludes examining the time series for the apparent 1998 regime shift reported elsewhere in the CC (Peterson and Schwing, 2003). Results The rank order of the 15 most abundant taxa in the ichthyoplankton time series indicates considerable congruence in the community structure of the fishes of the CC from San Francisco to Baja California, with a gradual transition from communities dominated by cool-water to warm-water distributions in proceeding from higher to lower latitudes. Most taxa were among the top 15 in at least two regions: four shared between Baja California and southern California and three between southern and central California, with five taxa (33%) among the top 15 in all three regions (Table 1). Taxa shared among all three regions were all dominant CC endemics: the northern anchovy (Engraulis mordax), Pacific hake (Merluccius productus), Pacific jack mackerel (Trachurus symmetricus), rockfishes (Sebastes spp.), and California smoothtongue (Leuroglossus stilbius). Taxa only ranked among the top 15 off Baja California and southern California are generally noted for their warm-water affinities: Panama lightfish (Vinciguerria lucetia), Mexican lampfish (Triphoturus mexicanus), and Pacific sardine (Sardinops sagax). Similarly, the three mesopelagic fishes that were among the top 15 taxa in only southern and central CalCOFI are noted for their predominantly cool-water distributions: the northern lampfish (Stenobrachius leucopsarus), eared blacksmelt (Lipolagus ochotensis), and California flashlightfish (Protomyctophum crockeri). Table 1. The 15 most abundant taxa by sampling region listed in rank order. Baja California Southern California Central California Engraulis mordaxa Engraulis mordaxa Stenobrachius leucopsarusc Triphoturus mexicanusb Merluccius productusa Merluccius productus Vinciguerria lucetiab Leuroglossus stilbiusa Engraulis mordaxa Citharichthys spp.d Sebastes spp.a Sebastes spp.a Sardinops sagaxb Vinciguerria lucetiab Lipolagus ochotensisc Diogenichthys spp Stenobrachius leucopsarusc Leuroglossus stilbiusa Sebastes spp.a Sardinops sagaxb Nannobrachium spp.d Bathylagoides wesethib Trachurus symmetricusc Tarletonbeania crenularis Synodus spp Sebastes jordani Trachurus symmetricusc Leuroglossus stilbiusa Lipolagus ochotensisc Protomyctophum crockeria Merluccius productusa Sciaenidae Diaphus spp. Cyclothone spp. Triphoturus mexicanusb Melamphaes spp. Nannobrachium spp.d Bathylagoides wesethib Citharichthys spp.d Scomber japonicus Protomyctophum crockeria Icichthys lockingtoni Protomyctophum crockeria Ceratoscopelus townsendi Symbolophorus californiensis Baja California Southern California Central California Engraulis mordaxa Engraulis mordaxa Stenobrachius leucopsarusc Triphoturus mexicanusb Merluccius productusa Merluccius productus Vinciguerria lucetiab Leuroglossus stilbiusa Engraulis mordaxa Citharichthys spp.d Sebastes spp.a Sebastes spp.a Sardinops sagaxb Vinciguerria lucetiab Lipolagus ochotensisc Diogenichthys spp Stenobrachius leucopsarusc Leuroglossus stilbiusa Sebastes spp.a Sardinops sagaxb Nannobrachium spp.d Bathylagoides wesethib Trachurus symmetricusc Tarletonbeania crenularis Synodus spp Sebastes jordani Trachurus symmetricusc Leuroglossus stilbiusa Lipolagus ochotensisc Protomyctophum crockeria Merluccius productusa Sciaenidae Diaphus spp. Cyclothone spp. Triphoturus mexicanusb Melamphaes spp. Nannobrachium spp.d Bathylagoides wesethib Citharichthys spp.d Scomber japonicus Protomyctophum crockeria Icichthys lockingtoni Protomyctophum crockeria Ceratoscopelus townsendi Symbolophorus californiensis a Taxa were among the 15 most abundant in all three regions. b Taxa among the top 15 in both Baja and southern California. c Taxa among the top 15 in southern California and central California. d Taxa among the top 15 off Baja and central California but not southern California. Table 1. The 15 most abundant taxa by sampling region listed in rank order. Baja California Southern California Central California Engraulis mordaxa Engraulis mordaxa Stenobrachius leucopsarusc Triphoturus mexicanusb Merluccius productusa Merluccius productus Vinciguerria lucetiab Leuroglossus stilbiusa Engraulis mordaxa Citharichthys spp.d Sebastes spp.a Sebastes spp.a Sardinops sagaxb Vinciguerria lucetiab Lipolagus ochotensisc Diogenichthys spp Stenobrachius leucopsarusc Leuroglossus stilbiusa Sebastes spp.a Sardinops sagaxb Nannobrachium spp.d Bathylagoides wesethib Trachurus symmetricusc Tarletonbeania crenularis Synodus spp Sebastes jordani Trachurus symmetricusc Leuroglossus stilbiusa Lipolagus ochotensisc Protomyctophum crockeria Merluccius productusa Sciaenidae Diaphus spp. Cyclothone spp. Triphoturus mexicanusb Melamphaes spp. Nannobrachium spp.d Bathylagoides wesethib Citharichthys spp.d Scomber japonicus Protomyctophum crockeria Icichthys lockingtoni Protomyctophum crockeria Ceratoscopelus townsendi Symbolophorus californiensis Baja California Southern California Central California Engraulis mordaxa Engraulis mordaxa Stenobrachius leucopsarusc Triphoturus mexicanusb Merluccius productusa Merluccius productus Vinciguerria lucetiab Leuroglossus stilbiusa Engraulis mordaxa Citharichthys spp.d Sebastes spp.a Sebastes spp.a Sardinops sagaxb Vinciguerria lucetiab Lipolagus ochotensisc Diogenichthys spp Stenobrachius leucopsarusc Leuroglossus stilbiusa Sebastes spp.a Sardinops sagaxb Nannobrachium spp.d Bathylagoides wesethib Trachurus symmetricusc Tarletonbeania crenularis Synodus spp Sebastes jordani Trachurus symmetricusc Leuroglossus stilbiusa Lipolagus ochotensisc Protomyctophum crockeria Merluccius productusa Sciaenidae Diaphus spp. Cyclothone spp. Triphoturus mexicanusb Melamphaes spp. Nannobrachium spp.d Bathylagoides wesethib Citharichthys spp.d Scomber japonicus Protomyctophum crockeria Icichthys lockingtoni Protomyctophum crockeria Ceratoscopelus townsendi Symbolophorus californiensis a Taxa were among the 15 most abundant in all three regions. b Taxa among the top 15 in both Baja and southern California. c Taxa among the top 15 in southern California and central California. d Taxa among the top 15 off Baja and central California but not southern California. Southern/Baja California comparison We tested initially for the robustness of the PCA to changes in sampling period. The addition of data since 2008 did not significantly alter the PC 1 time series for the core CalCOFI (southern California) region, with a correlation of 0.98 between PCs 1 derived from time series 1951–2008 and 1951–2011, the most recent data available for the full annual cycle off southern California. PC 1 for the Baja California region (PC1Baja) explained 29.7% of the variance of the data set and is the only PC considered here. (In contrast, PC2Baja explained 9.9% of the variance.) Its time series was significantly correlated with PC 1 for southern California (PC1So Calif) (r = 0.60, df = 32, p < 0.001), indicating coherence in the dominant ecological trends in the two regions. However, the correlation was based on the relationship during the CalCOFI sampling period to 1981, when the correlation was considerably higher (r = 0.78, df = 20, p < 0.001). Because of the 17-year gap in sampling off Baja California, formal breakpoint analysis is not possible, but it is apparent that the relationship breaks down after 1998 (r = 0.17, ns): PC1Baja increased while PC1So Calif declined (Figure 2, Supplementary Figure S2). Figure 2. View largeDownload slide Scatterplots of PC 1 for Baja California with (a) PC1 for southern California and (b) central California larval fish time series. The time series were significantly correlated prior to 1998, but post-1998, PC 1 (Baja) notably increased while PC 1 for southern and central California declined. Figure 2. View largeDownload slide Scatterplots of PC 1 for Baja California with (a) PC1 for southern California and (b) central California larval fish time series. The time series were significantly correlated prior to 1998, but post-1998, PC 1 (Baja) notably increased while PC 1 for southern and central California declined. The positive correlation of PC1Baja with PC1So Calif was based on a preponderance of mesopelagic taxa loading positively on PC1Baja, including all 15 taxa with the highest loadings (Supplementary Table S1), similar to the positive loading of predominantly mesopelagic taxa on PC1So Calif. Virtually all of the mesopelagic taxa loading positively have warm-water affinities. Three of the six taxa that loaded significantly negatively on PC1Baja were CC endemic and cool-water affinity taxa: S. leucopsarus, M. productus, and E. mordax (Supplementary Table S1). The apparent coherence in overall trends across the southern CC among mesopelagic fishes represented by PC 1 was supported by the results of a combined PCA for southern and Baja California based on the abundance of the 15 most abundant taxa in each region. PC 1 for this analysis explained 31.5% of the variance of the total data set. Fourteen of the 15 variables loading significantly positively on PC 1 were mesopelagics from both regions, eight from Baja California and six from off southern California (Supplementary Table S2), indicating both the high degree of congruence between the two regions and the similarity of this combined PC to PC1Baja and PC1So Calif: the correlations with these two PCs were 0.93 and 0.77, respectively. A number of California Current endemics and cool-water affinity taxa loaded significantly negatively, six from each data set, including S. leucopsarus, M. productus, and E. mordax from both regions (Supplementary Table S2). The combined Baja–southern California PC 1 was significantly correlated with SST (southern California) (r = 0.72, df = 30, p < 0.001) PDO (r = 0.52, p < 0.01), and NPGO (r = 0.37, p < 0.05). The combined PC 1, like PC1Baja, was not significantly correlated with deepwater oxygen concentration over the entire time series (r = 0.23, ns) but was significantly correlated for the time series prior to 1998 (r = 0.68, p < 0.001). Despite the strong positive correlation of PC1So Calif with both PC1Baja (noted above) and midwater oxygen concentration (r = 0.74, p < 0.001), PC1Baja was not correlated with midwater oxygen concentration (r = 0.06) (Table 2). A scatterplot of PC1Baja and midwater oxygen concentration shows that they were positively correlated for the first period of the Baja California time series (1951–1981) (r = 0.48, p < 0.05, df = 19), but that the relationship broke down in the second part of the time series from 1998 to 2011 (Figure 3). Table 2. Correlations of PC 1 from Baja (PC1Baja) and southern California (PC1So Calif) time series with oxygen concentrations at 200–400 m depth (O2Deep) and mean sea temperature at 10 m depth (T10) in the southern California (core CalCOFI survey region), the multivariate ENSO index (MEI), Pacific Decadal Oscillation (PDO), and North Pacific Gyre Oscillation (NPGO). O2Deep MEI PDO NPGO T10 PC1So Calif 0.74*** 0.50*** 0.58*** −0.21† 0.53*** PC1Baja 0.06† 0.27† 0.39* 0.38* 0.68*** O2Deep MEI PDO NPGO T10 PC1So Calif 0.74*** 0.50*** 0.58*** −0.21† 0.53*** PC1Baja 0.06† 0.27† 0.39* 0.38* 0.68*** * p < 0.05; ** p < 0.01; *** p < 0.001; ?p < 0.10; †p > 0.10. N = 31–34 for correlations with PC1Baja; N = 45–49 for correlations with PC1So Calif. Table 2. Correlations of PC 1 from Baja (PC1Baja) and southern California (PC1So Calif) time series with oxygen concentrations at 200–400 m depth (O2Deep) and mean sea temperature at 10 m depth (T10) in the southern California (core CalCOFI survey region), the multivariate ENSO index (MEI), Pacific Decadal Oscillation (PDO), and North Pacific Gyre Oscillation (NPGO). O2Deep MEI PDO NPGO T10 PC1So Calif 0.74*** 0.50*** 0.58*** −0.21† 0.53*** PC1Baja 0.06† 0.27† 0.39* 0.38* 0.68*** O2Deep MEI PDO NPGO T10 PC1So Calif 0.74*** 0.50*** 0.58*** −0.21† 0.53*** PC1Baja 0.06† 0.27† 0.39* 0.38* 0.68*** * p < 0.05; ** p < 0.01; *** p < 0.001; ?p < 0.10; †p > 0.10. N = 31–34 for correlations with PC1Baja; N = 45–49 for correlations with PC1So Calif. Figure 3. View largeDownload slide Scatterplots of PC1Baja in relation to (a) mean annual oxygen concentration (in ml l−1) at 200–400 m depth; (b) the PDO; and sea temperature at 10 m depth. The data points are labelled by year. There is a significant correlation of PC1Baja with midwater oxygen during the first part of the time series (1951–1978) (r = 0.61, p < 0.01) but not after the time series resumed in 1998. The correlation of PC1Baja with the PDO is significant over the entire time series (r = 0.39, p < 0.05), but the plot indicates two distinct relationships: the first part of the time series (1951–1981) and the second (1998–2011). There is a significant correlation overall with sea temperature at 10 m depth with no indication of a change in the relationship. Figure 3. View largeDownload slide Scatterplots of PC1Baja in relation to (a) mean annual oxygen concentration (in ml l−1) at 200–400 m depth; (b) the PDO; and sea temperature at 10 m depth. The data points are labelled by year. There is a significant correlation of PC1Baja with midwater oxygen during the first part of the time series (1951–1978) (r = 0.61, p < 0.01) but not after the time series resumed in 1998. The correlation of PC1Baja with the PDO is significant over the entire time series (r = 0.39, p < 0.05), but the plot indicates two distinct relationships: the first part of the time series (1951–1981) and the second (1998–2011). There is a significant correlation overall with sea temperature at 10 m depth with no indication of a change in the relationship. Although PC1Baja was significantly correlated with the PDO over the entire time series, a scatterplot of the two variables suggests a regime shift or break in the relationship during the second part of the time series, post-1998 (Figure 3). The relationship between PC1Baja and the PDO is approximately parallel in the two periods but offset. The relationship with the PDO is statistically significant during the post-1998 period (r = 0.65, p < 0.05) and marginally significant (r = 0.42, p = 0.054) for the period 1951–1981. The analyses carried out thus far indicate a distinct break in the relationship between the mesopelagic fish community off southern and Baja California and between mesopelagic fishes off Baja California and midwater oxygen concentration after 1997. However, a plot of the difference between PC1Baja and PC1So Calif in relation to midwater oxygen concentration indicates that there has been an approximately linear relationship over time, with mesopelagic fishes off Baja California increasingly outperforming those off southern California as oxygen concentration declined (r = −0.61, p < 0.001) (Figure 4). This suggests that the mesopelagic assemblage off Baja California is better adapted to reduced oxygen concentrations and a shoaling HBL. Figure 4. View largeDownload slide The difference in time series values of PC1Baja and PC1CalCOFI in relation to annual mean oxygen concentration at 200–400 m depth. Figure 4. View largeDownload slide The difference in time series values of PC1Baja and PC1CalCOFI in relation to annual mean oxygen concentration at 200–400 m depth. PC1Baja was highly correlated with annual mean sea temperature at 10 m depth (T10) (r = 0.68, p < 0.001) (Table 2), and the relationship does not appear to shift over time (Figure 3). T10 is positively correlated with deepwater oxygen concentration (r = 0.52, df = 50, p < 0.001), as it is with the PDO (r = 0.40, p < 0.01) and MEI (r = 0.35, p < 0.05). These inter-correlations complicate attempts to ascertain causal relationships, but for the Baja California mesopelagic community, the relationship with temperature seems strongest and most consistent. Comparison of the rank order of abundance of the Baja California fish community during years of low (1960–1963, 2009–2011) and high (1952–1959, 1966–1978) midwater oxygen concentrations indicated a significant shift toward increasing dominance of warm-water affinity mesopelagic fish taxa. Of the 67 taxa in our analysis, 25 were mesopelagic, of which 17 were classified as warm-water affinity (predominantly subtropical–tropical in distribution), three as cool-water affinity (distributions primarily extending to high latitudes), two as a CC endemic, and three as broadly distributed (no clear warm- or cool-water preference). The community is thus comprised predominantly of tropical–subtropical taxa, which inhabit the shallow OMZ observed from Baja California to the ETP and further south to the Humboldt Current at low latitudes. Of the 17 warm-water affinity mesopelagic taxa, 12 increased in relative dominance (i.e. moved to a lower rank), three declined in relative abundance, and two were unchanged in their rank order of abundance in the data set, indicating that the warm-water affinity mesopelagic taxa became significantly more dominant in the fish assemblage off Baja California in the low- vs. high-oxygen years (p < 0.05, binomial distribution) (Table 3). Table 3. The rank order of abundance of warm-water affinity (tropical–subtropical) mesopelagic fish taxa during the period of relatively low midwater oxygen concentration (1960–1963, 2009–2011) and high oxygen concentrations (1952–1959, 1960–1978) off Baja California. Taxon Rank order, low [O2] Rank order, high [O2] Change Vinciguerria lucetia 2 4 + Triphoturus mexicanus 3 3 0 Diogenichthys spp. 7 8 + Bathylagoides wesethi 8 11 + Cyclothone spp. 12 12 0 Stomias atriventer 18 22 + Ceratoscopelus townsendi 19 29 + Symbolophorus californiensis 20 40 + Gonichthys tenuiculus 30 37 + Hygophum spp. 36 28 − Myctophum nitidulum 38 55 + Nansenia crassa 42 36 − Lampadena urophaos 43 52 + Diaphus spp. 54 61 + Loweina rara 62 64 + Scopelogadus mizolepis bispinosus 63 65 + Idiacanthus antrostomus 67 66 − Taxon Rank order, low [O2] Rank order, high [O2] Change Vinciguerria lucetia 2 4 + Triphoturus mexicanus 3 3 0 Diogenichthys spp. 7 8 + Bathylagoides wesethi 8 11 + Cyclothone spp. 12 12 0 Stomias atriventer 18 22 + Ceratoscopelus townsendi 19 29 + Symbolophorus californiensis 20 40 + Gonichthys tenuiculus 30 37 + Hygophum spp. 36 28 − Myctophum nitidulum 38 55 + Nansenia crassa 42 36 − Lampadena urophaos 43 52 + Diaphus spp. 54 61 + Loweina rara 62 64 + Scopelogadus mizolepis bispinosus 63 65 + Idiacanthus antrostomus 67 66 − The sign of the change refers to a relatively higher (+) or lower (−) relative abundance during periods of low oxygen concentration. Table 3. The rank order of abundance of warm-water affinity (tropical–subtropical) mesopelagic fish taxa during the period of relatively low midwater oxygen concentration (1960–1963, 2009–2011) and high oxygen concentrations (1952–1959, 1960–1978) off Baja California. Taxon Rank order, low [O2] Rank order, high [O2] Change Vinciguerria lucetia 2 4 + Triphoturus mexicanus 3 3 0 Diogenichthys spp. 7 8 + Bathylagoides wesethi 8 11 + Cyclothone spp. 12 12 0 Stomias atriventer 18 22 + Ceratoscopelus townsendi 19 29 + Symbolophorus californiensis 20 40 + Gonichthys tenuiculus 30 37 + Hygophum spp. 36 28 − Myctophum nitidulum 38 55 + Nansenia crassa 42 36 − Lampadena urophaos 43 52 + Diaphus spp. 54 61 + Loweina rara 62 64 + Scopelogadus mizolepis bispinosus 63 65 + Idiacanthus antrostomus 67 66 − Taxon Rank order, low [O2] Rank order, high [O2] Change Vinciguerria lucetia 2 4 + Triphoturus mexicanus 3 3 0 Diogenichthys spp. 7 8 + Bathylagoides wesethi 8 11 + Cyclothone spp. 12 12 0 Stomias atriventer 18 22 + Ceratoscopelus townsendi 19 29 + Symbolophorus californiensis 20 40 + Gonichthys tenuiculus 30 37 + Hygophum spp. 36 28 − Myctophum nitidulum 38 55 + Nansenia crassa 42 36 − Lampadena urophaos 43 52 + Diaphus spp. 54 61 + Loweina rara 62 64 + Scopelogadus mizolepis bispinosus 63 65 + Idiacanthus antrostomus 67 66 − The sign of the change refers to a relatively higher (+) or lower (−) relative abundance during periods of low oxygen concentration. To examine whether similar shifts in community composition or in the relationship with midwater oxygen concentration may be occurring off southern or central California, we analysed time series based on ichthyoplankton abundance in the spring only, because more recent data are only available from this season. However, this is the dominant season for spawning for many taxa, and there was a strong correlation between PC 1 based on annual and spring means: r = 0.83, p < 0.001. PC1So Calif-spr explained 18.0% of the variance in the time series, which contained 54 years of data from 1951 to 2015 (Supplementary Table S3). PC1So Calif-spr was highly correlated with annual mean oxygen concentration at 200–400 m depth for the period 1951–2003 (r = 0.60, df = 39, p < 0.001), but the relationship began to shift in 2004 and weaken although it remained significant (r = 0.43, df = 46, p < 0.01). After 2011, there was a marked shift, with anomalously high values for PC1So Calif-spr relative to midwater oxygen concentrations, which remained relatively low during this period (Figure 5), and the correlation with deepwater oxygen over the entire time series was no longer significant (r = 0.17, df = 50, ns). Analysis of covariance indicated a significant shift in the relationship with deepwater oxygen in both 2004 and 2011 (p < 0.01). The correlation of PC1So Calif-spr with annual mean sea temperature at 10 m depth from the CalCOFI cruises, which had been significant until 2010 (r = 0.34, df = 46, p < 0.05) also broke down thereafter (r = 0.23, df = 51, ns). Figure 5. View largeDownload slide (a) Scatterplot of annual mean oxygen concentration at 200–400 m depth in the standard CalCOFI sampling area off southern California with PC 1 for that area based on spring data only. The relationship, which was highly correlated for the period 1951–2003, can be seen to begin to shift in 2004. After 2011, there is a marked shift and the correlation is no longer significant. (b) Time series of the PDO and PC 1 for southern California. Figure 5. View largeDownload slide (a) Scatterplot of annual mean oxygen concentration at 200–400 m depth in the standard CalCOFI sampling area off southern California with PC 1 for that area based on spring data only. The relationship, which was highly correlated for the period 1951–2003, can be seen to begin to shift in 2004. After 2011, there is a marked shift and the correlation is no longer significant. (b) Time series of the PDO and PC 1 for southern California. Mesopelagic fishes with tropical/subtropical distributions became increasingly dominant off southern California in the most recent period (2011–2015) when the relationship between PC1So Calif-spr and midwater oxygen concentration significantly shifted (Figure 5). Comparing the period 2011–2015 with the period prior to 2004, overall larval abundance declined by almost 50% due to the decline in cool-water dominant taxa (Koslow et al., 2015) and mesopelagic fish abundance remained approximately unchanged, but the abundance of warm-water mesopelagic taxa increased 73% (Supplementary Table S4). The rank-order of abundance of the two most abundant warm-water affinity mesopelagic fishes, V. lucetia and T. mexicanus, which were the second and third most abundant larval fishes in the Baja California data set, rose from 10 to 19 to six and seven, respectively, off southern California in the most recent period. The tropical–subtropical fish taxa that dominate the mesopelagic fish community in the region from Baja California to the northern Humboldt Current are thus becoming increasingly dominant off southern California as the OMZ there has shoaled. Although the correlation of PC1So Calif-spr with deepwater oxygen concentration breaks down in the most recent portion of the time series, PC1So Calif-spr remained significantly correlated with the PDO and to a lesser extent, the MEI, over the time series from 1951 to 2015 (r = 0.48 and r = 0.38, respectively, df = 51, p < 0.01). Time series of PC1So Calif-spr and the PDO suggests that the warm-water affinity mesopelagic fishes that dominate PC1So Calif-spr follow both the interannual variability in the PDO and the decadal-scale trends: the period from 1951 to the end of the 1970s when both the PDO and PC1So Calif-spr were predominantly negative followed by a period of relatively high values for both indices until the end of the 1990s, followed by another period of low values until the very end of the time series (Figure 5). It is notable that 2015, the year of the “warm blob” off the west coast of North America, marked a significant El Niño and a peak in the PDO and PC1So Calif-spr. Central California, comparison to Baja California and southern California PC 1 based on the PCA for the central California ichthyoplankton (CalCOFI transects north of Pt Conception to San Francisco) (PC1Cent Calif) explained 22.2% of the variance of the data set. Eight of the 16 taxa that loaded significantly on PC 1 were mesopelagic fishes, and all of these that were identified to species and whose biogeographic affinities could be determined had cool-water affinities (Supplementary Table S5), consistent with the increased dominance of mesopelagic fishes with predominantly high-latitude distributions north of Pt. Conception. PC1Cent Calif was significantly correlated with PC1So Calif (Figure 6) and with the MEI, PDO, SST and deepwater oxygen concentration (Table 4). PC1Cent Calif was significantly correlated with PC1Baja during the initial sampling period, 1951–1981 (r = 0.72, df = 19, p < 0.001), but the relationship broke down after 1998, leading to a non-significant correlation overall (Figure 2). Examining the difference between PCs 1 from southern and central California, there was a significant step-change in the relationship in the early 2000s, based on STARS analysis (Rodionov, 2004) (Figure 6). PC1So Calif was still significantly correlated with PC1Cent Calif but was at a relatively higher level in the latter period. Although the relationship of PC1Cent Calif with midwater oxygen concentration remained significant over the entire time series (1951–2015), there is evidence of a shift in the relationship in recent years (Figure 7), with a significantly different relationship after 2003 and again after 2010 (ANCOVA, p < 0.01). However, the relationship of PC1Cent Calif with the PDO remained consistent throughout the time series. Table 4. Correlations between PC 1 for the CalCOFI region north of Pt. Conception to San Francisco (PC1Cent Calif) with PC 1 for the core CalCOFI region (PC1So Calif), PC 1 for the Baja California (PC1Baja), the multivariate ENSO index (MEI), the Pacific Decadal Oscillation (PDO), North Pacific Gyre Oscillation (NPGO), near-surface (10 m) sea temperature (SST), and mean oxygen concentration at 200–400 m depth in the core CalCOFI region (Deep O2). PC1So Calif PC1 Baja MEI PDO NPGO SST Deep O2 PC1Cent Calif 0.56*** 0.33? 0.47** 0.40* −0.31? 0.38* 0.55** (36) (28) (36) (36) (36) (36) (27) PC1So Calif PC1 Baja MEI PDO NPGO SST Deep O2 PC1Cent Calif 0.56*** 0.33? 0.47** 0.40* −0.31? 0.38* 0.55** (36) (28) (36) (36) (36) (36) (27) Significance levels based on N-2 df, shown in parentheses: *** p < 0.001; ** p < 0.01; * p < 0.05; ? 0.10 > p > 0.05. Table 4. Correlations between PC 1 for the CalCOFI region north of Pt. Conception to San Francisco (PC1Cent Calif) with PC 1 for the core CalCOFI region (PC1So Calif), PC 1 for the Baja California (PC1Baja), the multivariate ENSO index (MEI), the Pacific Decadal Oscillation (PDO), North Pacific Gyre Oscillation (NPGO), near-surface (10 m) sea temperature (SST), and mean oxygen concentration at 200–400 m depth in the core CalCOFI region (Deep O2). PC1So Calif PC1 Baja MEI PDO NPGO SST Deep O2 PC1Cent Calif 0.56*** 0.33? 0.47** 0.40* −0.31? 0.38* 0.55** (36) (28) (36) (36) (36) (36) (27) PC1So Calif PC1 Baja MEI PDO NPGO SST Deep O2 PC1Cent Calif 0.56*** 0.33? 0.47** 0.40* −0.31? 0.38* 0.55** (36) (28) (36) (36) (36) (36) (27) Significance levels based on N-2 df, shown in parentheses: *** p < 0.001; ** p < 0.01; * p < 0.05; ? 0.10 > p > 0.05. Figure 6. View largeDownload slide (a) The relationship between PC 1 off southern California and central California based on spring-time CalCOFI cruises. The relationships between the PCs are significant for both periods (1951–2004 and 2005–2015) but offset. (b) Time series of the difference between PC 1 off southern California and PC 1 off central California. There is a step change, with PC 1 off southern California relatively higher than PC1 off central California after 2005 than from 1951 to 2004, based on STARS analysis. Figure 6. View largeDownload slide (a) The relationship between PC 1 off southern California and central California based on spring-time CalCOFI cruises. The relationships between the PCs are significant for both periods (1951–2004 and 2005–2015) but offset. (b) Time series of the difference between PC 1 off southern California and PC 1 off central California. There is a step change, with PC 1 off southern California relatively higher than PC1 off central California after 2005 than from 1951 to 2004, based on STARS analysis. Figure 7. View largeDownload slide The relationship between PC1 for central California ichthyoplankton (mean springtime abundance) with (a) mean annual oxygen concentration at 200–400 m depth and (b) the PDO. The relationship with deepwater oxygen concentration shifts after 2004. Figure 7. View largeDownload slide The relationship between PC1 for central California ichthyoplankton (mean springtime abundance) with (a) mean annual oxygen concentration at 200–400 m depth and (b) the PDO. The relationship with deepwater oxygen concentration shifts after 2004. PC 1 from a PCA carried out on 18 of the most abundant taxa common to both the central and southern California data sets explained 24.2% of the variance of the combined data set. Of 14 taxa with loadings > 0.5, 10 were mesopelagic fishes, and seven were from the southern California and seven from the central California time series, indicating close coherence between the two areas (Supplementary Table S6). Although the combined PC 1 time series was not significantly correlated with deepwater oxygen concentration due to a breakdown in the relationship after 2010, similar to the breakdown in the relationship off southern California, there was a consistent correlation with the MEI (r = −0.41, df = 36, p < 0.05) and PDO (r = 0.35, df = 36, p < 0.05). It is interesting that this group of predominantly cool-water affinity mesopelagic fishes appears to respond coherently and positively to the warm phase of the PDO and ENSO cycles, consistent with the relationship displayed by warm-water affinity mesopelagic taxa off Baja California. Discussion Large marine ecosystems often respond coherently to environmental forcing across time scales ranging from the interannual (e.g. ENSO: Chelton et al., 1982) to the decadal [e.g. the PDO: Mantua et al., 1997 or North Atlantic Oscillation (Alheit and Bakun, 2010; Beaugrand et al., 2015)]. Most examples are from fisheries, for which the longest time series are available, with case studies as well from zooplankton time series (Beaugrand, 2003; Brinton and Townsend, 2003). However, despite the magnitude of these patterns, the mechanisms underlying them often remain unclear, an indication of how poorly marine ecosystems and their dynamics are understood. This presents a pressing issue for oceanographers, given that the global climate system appears set to change on a scale unprecedented in human experience. Larval fish time series indicate that a broad suite of mesopelagic fishes in the Southern California Bight underwent a cycle of expansion and decline since the middle of the last century, with a 63% decline between periods of high and low abundance. This pattern, observed over the period 1951–2008, was high correlated with changes in midwater oxygen concentration in the CC (Koslow et al., 2011), which had declined since the 1980s by 20–25% (Bograd et al., 2008). Similar changes in deepwater oxygen concentration have been reported around the basin of the North Pacific from the waters off Japan and the sub-Arctic (Ono et al., 2001; Emerson et al., 2004), along the extent of the CC (Crawford and Peña, 2016) and down to the ETP (Stramma et al., 2008). This raised the question whether the changes in the mesopelagic fish community observed off southern California would be observed more widely across the CC or even beyond. Our study indicates that the response of mesopelagic fishes to ocean forcing within the CC from central California to Baja California is not a simple one. During the half-cycle of midwater oxygen conditions from ∼1951 to 1981, when low oxygen concentrations trended upward, there was a coherent increasing trend of mesopelagic fish abundance from Baja California to central California across a broad suite of both warm- and cool-water affinity taxa, highly correlated with the trend in oxygen conditions. However, the pattern across the southern CC diverged at the end of the 1990s: although midwater oxygen concentrations across the region trended downward, the mesopelagic community increased off Baja California while it declined off southern and central California. After 2010, the positive correlation with midwater oxygen off southern California appeared to have broken down as well. Underlying this divergence was the varying response of warm- and cool-water affinity mesopelagic faunas to changing ocean conditions. The dominant trend (PC 1) of mesopelagic fishes off Baja California remained positively correlated with trends in ocean temperature and the warm phases of the PDO and ENSO cycle, with a dramatic increase in the abundance of warm-water affinity mesopelagic fishes such as V. lucetia and T. mexicanus. These taxa, particularly V. lucetia, dominate throughout the subtropical–tropical waters of the eastern Pacific, extending through the ETP to the Humboldt Current. The OMZ is most strongly developed throughout this region, with hypoxic waters often extending to within a hundred to several hundred metres of the surface (Cornejo and Koppelmann, 2006). These shallow-OMZ specialists have evolved various adaptations to these conditions that enable them to vertically migrate on a diel basis into hypoxic and suboxic waters (Childress and Seibel, 1998; Seibel, 2011). The southern CC is at or near the northernmost extent of their distribution. As midwater oxygen concentrations declined in the southern CC, these taxa dramatically increased in abundance, first off Baja California and more recently off southern California as well, indicating a shift in their distribution to higher latitudes with warming. Off central California, where a cool-water affinity mesopelagic fauna dominates, the relationship between mesopelagic fish abundance and oxygen concentration continues to hold, although there are signs of a possible transition there as well. This trend of increasing abundance of warm-water affinity mesopelagic fishes is consistent with the recent increased flux of Pacific Equatorial Water into the waters off southern California (McClatchie et al., 2018). This divergent response of mesopelagic fishes to environmental forcing is consistent with the region between Pt. Conception and the US–Mexico border being an ecotone, where the cool-water affinity fauna of the Oregonian Province to the north mixes with the warm-water affinity fauna of the San Diegan Province to the south (Horn and Allen, 1978). As Hubbs (1974) noted, fish distributions are fluid: taxa with tropical–subtropical affinities shift back and forth across Pt. Conception in relation to El Niño–La Niña events, as well as on geological time scales. Thus, although the CC is often referred to as a large marine ecosystem (Sherman and Duda, 1999) or an oceanographic province (Longhurst, 1998), there are distinct mesopelagic faunas with different biogeographic distributions within this ecosystem. As Brinton and Townsend (2003) showed for the euphausiids of the CC, taxa with warm- or cool-water affinities may respond oppositely to oceanographic drivers, such as the ENSO cycle or the PDO. What are the implications of our observations in the CC for the future abundance of mesopelagic fishes both within the CC and more globally? Global climate models generally predict that deep ocean oxygen levels will decline as a consequence of global warming, which will enhance stratification and hence inhibit ventilation of the deep ocean (Keeling et al., 2010; Schmidtko et al., 2017). In regions with an OMZ, where these effects will be most pronounced, our study indicates these trends will likely lead to increased dominance of mesopelagic taxa particularly adapted to low oxygen conditions and a shallow oxycline. Thus, there are likely to be winners and losers, rather than a single trend across the mesopelagic fish fauna. However, it would be valuable to examine time series for mesopelagic fishes in other oceanographic systems affected by OMZs in the Pacific Ocean and beyond before generalizing. The warm-water affinity mesopelagic fishes still do not extend significantly north of Pt Conception, and it remains unclear how the cool-water mesopelagic fauna may adapt (or not) to a shoaling OMZ. Although there is an ocean of ignorance with respect to the mesopelagic fauna and the factors regulating their ecology, there is also considerable interest in predicting how global climate change will impact various faunas, and mesopelagic fishes are no exception. Proud et al. (2017) recently predicted that mesopelagic fish abundance will increase globally over the coming century, based on the positive relationship they observed in data on acoustic backscatter from the DSL and temperature at the depth of the DSL obtained from oceanographic cruises within 14 of Longhurst’s (1998) 32 ocean provinces. Many issues can no doubt be raised in relation to such bold attempts to generalize globally from a limited data set. For example, oxygen was not included in their model, despite the published impacts on mesopelagic fish abundance of declining midwater oxygen levels (Koslow et al., 2011). However, oxygen may not have been included because there were no data in their model from any of the key regions of the global ocean influenced by OMZs. Nor were there data from any of the ocean’s eastern boundary currents. Furthermore, as the authors note, their predicted increase in mesopelagic fish biomass runs counter to the decline in primary productivity predicted for the global ocean as a consequence of increasing temperature. These issues lead us to conclude that it is premature to predict even the future direction of change in global mesopelagic fish biomass based on a comparative analysis of acoustic backscatter in relation primarily to a single variable, temperature, across a disparate and unrepresentative data set. Our caution in extrapolating too far into a future dominated by secular warming and away from the CC is based in large part on the divergence observed in this study: the correlation of mesopelagic fish abundance with midwater oxygen concentration that once seemed coherent across the region’s cool- and warm-water affinity faunas (Koslow et al., 2011) now appears to have diverged between the southern and central sectors of the CC. This so-called “breakdown” in a statistical relationship seems to reinforce the oft-repeated lesson in fisheries science that statistical relationships of fish abundance with environmental correlates cannot be relied upon to continue indefinitely. This may result from the relationships varying over time as species and communities adapt to long-term environmental change, as well as from the relationships having been fundamentally spurious and flawed. In either case, this serves as a cautionary tale for attempts to predict the impacts of climate change on particular species or communities based on simple linear statistical relationships. The impacts of secular climate change are unlikely to be manifested as simply an extension of the warm phase of the ENSO or PDO cycles, and it is unlikely that correlations based on past experience will serve as a reliable guide for future ecosystem trajectories. As the dominant fish consumers of zooplankton production in the global ocean (Irigoien et al., 2014; Koslow and Davison, 2015) and key players in the global carbon cycle (Davison et al., 2013), it is critical to assess the impact on this fauna of human-induced climate change. The strong patterns observed in our study suggest that such understanding is attainable. However, to do so, it is critical to understand the mechanisms underlying trends in mesopelagic fish abundance. Our observation of common trends in abundance indicates that common responses to factors, such as food availability, vulnerability to predators within the DSL, and/or physiological responses to temperature and limited oxygen availability, influence abundance across a range of mesopelagic fish taxa. Targeted field and laboratory studies of these factors, combined with extending mesopelagic fish time series across a broader range of ocean systems and development of mechanistic ecosystem models, will provide the way forward. Funding We gratefully acknowledge funding from the NOAA Fisheries and the Environment (FATE) program that supported P Davison and J Koslow and funded workshops that furthered our collaboration. NSF Research Experience for Undergraduates program supported E. Ferrer as a summer intern. Instituto Politécnico Nacional, IMECOCAL Program as well as SNI, EDI and COFAA programs supported G. Aceves and S. P. A. Jimenez. References Ahlstrom E. H. 1959 . Vertical distribution of pelagic fish eggs and larvae off California and Baja California . Fishery Bulletin , 60 : 107 – 146 . Alheit J. , Bakun A. 2010 . Population synchronies within and between ocean basins: apparent teleconnections and implications as to physical–biological linkage mechanisms . Journal of Marine Systems , 79 : 267 – 285 . Google Scholar Crossref Search ADS Beaugrand G. 2003 . Long-term changes in copepod abundance and diversity in the north-east Atlantic in relation to fluctuations in the hydroclimatic environment . Fisheries Oceanography , 12 : 270 – 283 . Google Scholar Crossref Search ADS Beaugrand G. , Conversi A. , Chiba S. , Edwards M. , Fonda-Umani S. , Greene C. , Mantua N. et al. 2015 . Synchronous marine pelagic regime shifts in the Northern Hemisphere . Philosophical Transactions of the Royal Society B , 370 : 20130272. Google Scholar Crossref Search ADS Bianchi D. , Galbraith E. D. , Carozza D. A. , Mislan K. A. S. , Stock C. A. 2013 . Intensification of open-ocean oxygen depletion by vertically migrating animals . Nature Geoscience , 6 : 545 – 548 . Google Scholar Crossref Search ADS Blanchette C. A. , Melissa Miner C. , Raimondi P. T. , Lohse D. , Heady K. E. , Broitman B. R. 2008 . Biogeographical patterns of rocky intertidal communities along the Pacific coast of North America . Journal of Biogeography , 35 : 1593 – 1607 . Google Scholar Crossref Search ADS Bograd S. J. , Castro C. G. , Di Lorenzo E. , Palacios D. M. , Bailey H. , Gilly W. , Chavez F. P. 2008 . Oxygen declines and the shoaling of the hypoxic boundary in the California Current . Geophysical Research Letters , 35 : L12607 . Google Scholar Crossref Search ADS Breitburg D. , Levin L. A. , Oschlies A. , Grégoire M. , Chavez F. P. , Conley D. J. , Garçon V. et al. 2018 . Declining oxygen in the global ocean and coastal waters . Science , 359 : eaam7240. Google Scholar Crossref Search ADS PubMed Brinton E. , Townsend A. 2003 . Decadal variability in abundances of the dominant euphausiid species in southern sectors of the California Current . Deep Sea Research Part II , 50 : 2449 – 2472 . Google Scholar Crossref Search ADS Checkley D. M. , Barth J. A. 2009 . Patterns and process in the California Current . Progress in Oceanography , 83 : 49 – 64 . Google Scholar Crossref Search ADS Chelton D. B. , Bernal P. A. , McGowan J. A. 1982 . Large-scale interannual physical and biological interaction in the California Current . Journal of Marine Research , 40 : 95 – 112 . Childress J. J. , Seibel B. A. 1998 . Life at stable low oxygen: adaptations of animals to oceanic oxygen minimum layers . Journal of Experimental Biology , 201 : 1223 – 1232 . Google Scholar PubMed Cornejo R. , Koppelmann R. 2006 . Distribution patterns of mesopelagic fishes with special reference to Vinciguerria lucetia Garman 1899 (Phosichthyidae: Pisces) in the Humboldt Current Region off Peru . Marine Biology , 149 : 1519 – 1537 . Google Scholar Crossref Search ADS Crawford W. R. , Peña M. A. 2016 . Decadal trends in oxygen concentration in subsurface waters of the Northeast Pacific Ocean . Atmosphere-Ocean , 54 : 171 – 192 . Google Scholar Crossref Search ADS Davison P. C. , Checkley D. M. Jr , Koslow J. A. , Barlow J. 2013 . Carbon export mediated by mesopelagic fishes in the northeast Pacific Ocean . Progress in Oceanography , 116 : 14 – 30 . Google Scholar Crossref Search ADS Deutsch C. , Brix H. , Ito T. , Frenzel H. , Thompson L. 2011 . Climate-forced variability of ocean hypoxia . Science , 333 : 336 – 339 . Google Scholar Crossref Search ADS PubMed Deutsch C. , Berelson W. , Thunell R. , Weber T. , Tems C. , McManus J. , Crusius J. et al. 2014 . Centennial changes in North Pacific anoxia linked to tropical trade winds . Science , 345 : 665 – 668 . Google Scholar Crossref Search ADS PubMed Di Lorenzo E. , Schneider N. , Cobb K. M. , Chhak K. , Franks P. J. S. , Miller A. J. , McWilliams J. C. et al. 2008 . North Pacific Gyre Oscillation links ocean climate and ecosystem change . Geophysical Research Letters , 35 : L08607 . Google Scholar Crossref Search ADS Emerson S. , Watanabe Y. W. , Ono T. , Mecking S. 2004 . Temporal trends in apparent oxygen utilization in the upper pycnocline of the North Pacific: 1980–2000 . Journal of Oceanography , 60 : 139 – 147 . Google Scholar Crossref Search ADS Garcia H. E. , Locarnini R. A. , Boyer T. P. , Antonov J. I. , Mishonov A. V. , Baranova O. K. , Zweng M. M. et al. 2013 . World Ocean Atlas Vol. 3: dissolved oxygen, apparent oxygen utilization, and oxygen saturation . In NOAA Atlas NESDIS 75 : 27 pp. Hare S. R. , Mantua N. J. 2000 . Empirical evidence for North Pacific regime shifts in 1977 and 1989 . Progress in Oceanography , 47 : 103 – 146 . Google Scholar Crossref Search ADS Haury L. R. , Venrick E. L. , Fey C. L. , McGowan J. A. , Niiler P. P. 1993 . The Ensenada Front: July 1985 . CalCOFI Reports , 34 : 69 – 88 . Helly J. J. , Levin L. A. 2004 . Global distribution of naturally occurring marine hypoxia on continental margins . Deep Sea Research Part I , 51 : 1159 – 1168 . Google Scholar Crossref Search ADS Helm K. P. , Bindoff N. L. , Church J. A. 2011 . Observed decreases in oxygen content of the global ocean . Geophysical Research Letters , 38 : L23602 . Google Scholar Crossref Search ADS Horn M. H. , Allen L. G. 1978 . A distributional analysis of California coastal marine fishes . Journal of Biogeography , 5 : 23 – 42 . Google Scholar Crossref Search ADS Horn M. H. , Allen L. G. , Lea R. N. 2006 . Biogeography. In The Ecology of Marine Fishes. California and Adjacent Waters , pp. 3 – 25 . Ed. by Allen L. G. , Pondella D. J. , Horn M. H. . University of California Press , Berkeley . Hubbs C. L. 1974 . Review: marine Zoogeography by John C. Briggs . Copeia , 1974 : 1002 – 1005 . Google Scholar Crossref Search ADS Irigoien X. , Klevjer T. A. , Røstad A. , Martinez U. , Boyra G. , Acuña J. L. , Bode A. et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean . Nature Communications , 5 , doi: 10.1038/ncomms4271. Ito T. , Minobe S. , Long M. C. , Deutsch C. 2017 . Upper ocean O2 trends: 1958–2015 . Geophysical Research Letters , 44 : 2017GL073613 . Keeling R. F. , Kortzinger A. , Gruber N. 2010 . Ocean deoxygenation in a warming world . Annual Review Marine Science , 2 : 199 – 229 . Google Scholar Crossref Search ADS Koslow J. A. , Goericke R. , Lara-Lopez A. , Watson W. 2011 . Impact of declining intermediate-water oxygen on deepwater fishes in the California Current . Marine Ecology Progress Series , 436 : 207 – 218 . Google Scholar Crossref Search ADS Koslow J. A. , Couture J. 2015 . Pacific Ocean observation programs: gaps in ecological time series . Marine Policy , 51 : 408 – 414 . Google Scholar Crossref Search ADS Koslow J. A. , Miller E. F. , McGowan J. A. 2015 . Dramatic declines in coastal and oceanic fish communities off California . Marine Ecology Progress Series , 538 : 221 – 227 . Google Scholar Crossref Search ADS Koslow J. A. , Davison P. 2015 . Productivity and biomass of fishes in the California Current large marine ecosystem: comparison of fishery-dependent and independent time series . Environmental Development , 17 : 23 – 32 . Google Scholar Crossref Search ADS Kramer D. , Kalin M. J. , Stevens E. G. , Thrailkill J. R. , Zweifel J. R. 1972 . Collecting and Processing Data on Fish Eggs and Larvae in the California Current Region, 37. 38 pp. Lara-Lopez A. , Davison P. , Koslow J. A. 2012 . Abundance and community composition of micronekton across a frontal system off Southern California . Journal of Plankton Research , 34 : 828 – 848 . Google Scholar Crossref Search ADS Legendre P. , Legendre L. 2012 . Numerical Ecology . Elsevier , 853 pp. Longhurst A. R. 1998 . The Ecological Geography of the Sea . Academic Press , London . Mantua N. J. , Hare S. R. , Zhang Y. , Wallace J. M. , Francis R. C. 1997 . A Pacific interdecadal climate oscillation with impacts on salmon production . Bulletin of the American Meteorological Society , 78 : 1069 – 1079 . Google Scholar Crossref Search ADS Matear R. J. , Hirst A. C. 2003 . Long-term changes in dissolved oxygen concentrations in the ocean caused by protracted global warming . Global Biogeochemical Cycles , 17 : 1125 . McClatchie S. , Goericke R. , Cosgrove R. , Auad G. , Vetter R. D. 2010 . Oxygen in the Southern California Bight: multidecadal trends and implications for demersal fisheries . Geophysical Research Letters , 37 : L19602 . Google Scholar Crossref Search ADS McClatchie S. 2014 . Regional Fisheries Oceanography of the California Current System: The CalCOFI Program . Springer , Dordrecht . 235 pp. McClatchie S. , Gao J. , Drenkard E. J. , Thompson A. R. , Watson W. , Ciannelli L. , Bograd S. J. et al. 2018 . Inter-annual and secular variability of larvae of mesopelagic and forage fishes in the southern California Current System . Journal of Geophysical Research: Oceans , doi: 10.1029/2018JC014011. Moser H. G. , Watson W. 1990 . Distribution and abundance of early life history stages of the California halibut, Paralichthys californicus and comparisons with the fantail sole, Xystreurys liolepis. California Department of Fish and Game's Fish Bulletin, 174 : 31 – 84 . Moser H. G. , Smith P. E. 1993 . Larval fish assemblages of the California Current region and their horizontal and vertical distributions across a front . Bulletin of Marine Science , 53 : 645 – 691 . Moser H. G. 1996 . The Early Stages of Fishes in the California Current Region . Allen Press , Lawrence, KS . 1505 pp. Moser H. G. , Charter R. L. , Watson W. , Ambrose D. A. , Butler J. L. , Charter S. R. , Sandknop E. M. 2000 . Abundance and distribution of rockfish (Sebastes) larvae in the Southern California Bight in relation to environmental conditions and fishery exploitation . CalCOFI Reports , 41 : 32 – 47 . Moser H. G. , Charter R. L. , Watson W. , Ambrose D. A. , Hill K. T. , Smith P. E. , Butler J. L. et al. 2001 . The CalCOFI ichthyoplankton time series: potential contributions to the management of rocky-shore fishes . CalCOFI Reports , 42 : 112 – 128 . Moser H. G. , Watson W. 2006 . Ichthyoplankton. In Ecology of California Marine Fishes , pp. 269 – 319 . Ed. by Allen L. G. , Pondella D. J. , Horn M. H. . University of California Press , Berkeley . Netburn A. N. , Koslow J. A. 2015 . Dissolved oxygen as a constraint on daytime deep scattering layer depth in the southern California current ecosystem . Deep Sea Research Part I: Oceanographic Research Papers , 104 : 149 – 158 . Google Scholar Crossref Search ADS Ohman M. D. , Venrick E. L. 2003 . CalCOFI in a changing ocean . Oceanography , 16 : 76 – 85 . Google Scholar Crossref Search ADS Ono T. , Midorikawa T. , Watanabe Y. W. , Tadokoro K. , Saino T. 2001 . Temporal increases of phosphate and apparent oxygen utilization in the subsurface waters of western subarctic Pacific from 1968 to 1998 . Geophysical Research Letters , 28 : 3285 – 3288 . Google Scholar Crossref Search ADS Peterson W. T. , Schwing F. B. 2003 . A new climate regime in northeast Pacific ecosystems . Geophysical Research Letters , 30 . Proud R. , Cox M. J. , Brierley A. S. 2017 . Biogeography of the global ocean's mesopelagic zone . Current Biology , 27 : 113 – 119 . Google Scholar Crossref Search ADS PubMed Robison B. H. 1972 . Distribution of the midwater fishes of the Gulf of California . Copeia , 1972 : 448 – 461 . Google Scholar Crossref Search ADS Robison B. 2009 . Conservation of deep pelagic biodiversity . Conservation Biology , 23 : 847 – 858 . Google Scholar Crossref Search ADS PubMed Rodionov S. N. 2004 . A sequential algorithm for testing climate regime shifts . Geophysical Research Letters , 31 : L09204 . Google Scholar Crossref Search ADS Schmidtko S. , Stramma L. , Visbeck M. 2017 . Decline in global oceanic oxygen content during the past five decades . Nature , 542 : 335 – 339 . Google Scholar Crossref Search ADS PubMed Seibel B. A. 2011 . Critical oxygen levels and metabolic suppression in oceanic oxygen minimum zones . Journal of Experimental Biology , 214 : 326 – 336 . Google Scholar Crossref Search ADS PubMed Shaffer G. , Olsen S. M. , Pedersen O. P. 2009 . Long-term ocean oxygen depletion in response to carbon dioxide emissions from fossil fuels . Nature Geoscience , 2 : 105 – 109 . Google Scholar Crossref Search ADS Sherman K. , Duda A. M. 1999 . Large marine ecosystems: an emerging paradigm for fishery sustainability . Fisheries , 24 : 15 – 26 . Google Scholar Crossref Search ADS Smith P. E. , Richardson S. L. 1977 . Standard techniques for pelagic fish egg and larva surveys . FAO Fisheries Technology Paper , 175 : 1 – 100 . Stramma L. , Johnson G. C. , Sprintall J. , Mohrholz V. 2008 . Expanding oxygen-minimum zones in the tropical oceans . Science , 320 : 655 – 658 . Google Scholar Crossref Search ADS PubMed Thompson A. R. , McClatchie S. , Weber E. D. , Watson W. , Lennert-Cody C. E. 2017 . Correcting for bias in CalCOFI ichthyoplankton abundance estimates associated with the 1977 transition from ring net to bongo sampling . CalCOFI Reports , 58 : 113 – 123 . Vaquer-Sunyer R. , Duarte C. M. 2008 . Thresholds of hypoxia for marine biodiversity . Proceedings National Academy of Science , 105 : 15452 – 15457 . Google Scholar Crossref Search ADS Wolter K. , Timlin M. S. 1998 . Measuring the strength of ENSO events—how does 1997/98 rank? Weather , 53 : 315 – 324 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2018. All rights reserved. For permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Mei,, Weiping;Umezawa,, Yu;Wan,, Xin;Yuan,, Jinghan;Sassa,, Chiyuki

doi: 10.1093/icesjms/fsy016pmid: N/A

Abstract Bulk carbon and nitrogen stable isotope (SI) ratios (δ13C and δ15N) were analysed to investigate the feeding habits of six taxa of mesopelagic fish larvae inhabiting the Kuroshio waters of the northeastern East China Sea. Large variation in tissue SI during early larval periods suggested maternal effects from parent fishes, and non-selective feeding on a variety of plankton species due to poor swimming ability. The similarity between SI ratios measured in larval tissues and those estimated for eggs of an “income breeder” in the spawning area support an “income breeder” strategy in Diaphus slender type and Vinciguerria nimbaria, while Lipolagus ochotensis seemed to show “capital breeder”-like characteristics. SI ratios of the fish larvae studied became relatively constant at species-specific body dry-weights (0.5–1.0 mg), probably due to the commencement of selective feeding, meaning SI ratios during late larval periods could be used for trophic position analysis. There was great overlap (44.6–76.5%) in trophic niche among the larval fishes within the same taxonomic family of Myctophidae. Even if principal diet components cannot be identified with gut contents analyses, diet information from other fish species occupying a similar isotopic niche can thus improve our understanding of the diets of larval fishes. Introduction Mesopelagic fishes, with a biomass of at least 10 billion tons, dominate the world’s total fish biomass (Irigoien et al., 2014). These fishes are an important food source for commercially important fish species such as yellowfin tuna (Potier et al., 2007), and, since most species show active diurnal vertical migration (Watanabe et al., 1999; Luo et al., 2000; Yatsu et al., 2005), also have important implications for biogeochemical cycling in the ocean. For instance, mesopelagic fishes provide trophic connectivity and transport organic carbon between the surface and the mesopelagic layers (Kaartvedt et al., 2012; Irigoien et al., 2014). The Kuroshio region in the northeastern East China Sea (ECS) is regarded as an important spawning and nursery ground, especially in winter and early spring, for commercial fishes such as Japanese sardine (Sardinops melanostictus) and chub mackerel (Scomber japonicus) (Sugisaki et al., 2010; Chen et al., 2014), whereas various larvae of mesopelagic fishes such as Japanese lanternfish (Notoscopelus japonicus) and Eared blacksmelt (Lipolagus ochotensis) also frequently occur and may compete for food resources and marine habitat (Sassa and Hirota, 2013). When larval fishes replace their diets with various exogenous sources after yolk absorption they still have a weak swimming ability due to undeveloped body structures such as fins and muscles, and thus are at enhanced risk of predation. Hence, growth and nutrient condition are commonly considered to be the main factors determining larval fish survival rates (Anderson, 1988; Bailey and Houde, 1989). Therefore, clarifying the feeding habits of these mesopelagic fish species in their larval or juvenile stage is important for understanding growth and survival rates, which contribute greatly to the abundance of fishery resources. Several studies have shown that six taxa of mesopelagic fishes, Diaphus slender type, Myctophum asperum, and N. japonicus (Myctophidae), L. ochotensis (Microstomatidae), Sigmops gracilis (Gonostomatidae), and Vinciguerria nimbaria (Phosichthyidae) are dominant species in the northeastern ECS during winter, with the species composition, spatial and vertical distributions, and reproductive seasonality of these mesopelagic fishes larvae having been described in detail (Sassa et al., 2004; Watanabe et al., 2010; Sassa and Hirota, 2013; Sassa and Konishi, 2015). Specifically, the depth preferences of these fish larvae show large variation: Diaphus slender type, N. japonicus, M. asperum, and V. nimbaria occurred mainly in the 25- to 80-m depth layers, while L. ochotensis and S. gracilis were centered in relatively deeper layers at 30- to 100- and 55- to 100-m depth, respectively (Boehlert et al., 1992; Watanabe et al., 2010). Feeding habits of these mesopelagic fishes have been investigated mainly based on stomach contents. For example, V. nimbaria in its adult stage mainly preys upon Oncaeidae and Corycaeidae in the equatorial zone (Champalbert et al., 2008). In the larval stage, M. asperum feeds mainly on ostracods and polychaetes, while Diaphus spp. are reported to feed on appendicularian houses, copepod nauplii, calanoid copepodites and Oithona spp. depending on species and body size (Sassa and Kawaguchi, 2004, 2005). However, feeding habits of most of these mesopelagic fishes in the larval stage remain unclear. Stable isotopes (SIs) are effectively used for elucidating the time-integrated structures and dynamics of food webs, based on their enrichments between each trophic level (e.g. Minagawa and Wada, 1984; Post, 2002; McCutchan et al., 2003). For example, stable carbon and nitrogen isotope ratios (δ13C and δ15N) has been used to demonstrate spatial variations in the diet of Lanternfish (Myctophidae, 42- to 122-mm standard body length (SL)) (Flynn and Kloser, 2012), as well as differences in trophic positions within the overlapping distribution of 18 dominant mesopelagic fish species (13- to 193-mm SL) in the western Mediterranean (Valls et al., 2014). SIs in fish tissues have also been studied with other biotic factors (e.g. body length, body weight, and body-size ratio) to clarify the shift in trophic level, diet, habitat, and migration between juvenile and adult stages in epipelagic fishes (Deudero et al., 2004; Wells and Rooker, 2009; Laiz-Carrión et al., 2015) and mesopelagic fishes (Cherel et al., 2010; Choy et al., 2012). Shifts in tissue SI with body size in the larval stage have also been studied for some epipelagic fishes, such as Sardina pilchardus (Laiz-Carrión et al., 2011), Engraulis encrasicolus (Quintanilla et al., 2015) and Thunnus thynnus (García et al., 2017), but the presence of such shifts are still largely unknown for mesopelagic fish larvae. When SI variations are examined for fishes before the flexion stage, there is the potential for a maternal effect on larval tissue SI, i.e. the SI in fish larvae at this time reflect two factors: the isotopes of their diets, and the isotopes of parent fishes (Uriarte et al., 2016). McBride et al. (2015) reported that δ13C and δ15N differences among the fish species during the early period of larval stage may be linked to energy acquisition and allocation to egg production. Some fish species (e.g. Atlantic salmon and Winter flounder) spawn and feed in separate areas, during different seasons by storing energy and utilizing it later for reproduction (termed as “capital breeder”), whereas some other fish species (e.g. Zebrafish and Bay anchovy) spawn using energy acquired locally, allocating energy directly to reproduction (termed an “income breeder”) (McBride et al., 2015). Based on diet-switching feeding experiments, Tanaka et al. (2016) successfully demonstrated that Japanese anchovy (Engraulis japonicus) were “income breeders,” because the δ13C and δ15N ratios in eggs closely follow the isotope ratios of the food of adult fish (hereafter, termed as “δadultfood”) which the fishes had incorporated just before spawning (i.e. Δδ13C = δ13Cegg—δ13Cadultfood = 0.1–1.6‰, Δδ15N = δ15Negg—δ15Nadultfood = 0.9–2.0‰). Even in the field, therefore, the difference or similarity between the SI ratios measured in early larval fishes and the SI ratios estimated for “income breeder” eggs from that spawning ground represents a potentially useful tool for identifying the breeding type of targeted fishes. Recently, a quantitative analytical approaches in SIBER (Stable Isotope Bayesian Ellipses in R) was introduced by Jackson et al. (2011), in which SI data sets are used for comparing isotopic niches among and within communities. In this study, we examined δ13C and δ15N of six taxa of larvae of mesopelagic fishes (i.e. M. asperum, N. japonicus and Diaphus slender type, L. ochotensis, S. gracilis, and V. nimbaria) which are dominant in the micronektonic fish communities of the Kuroshio region of the East China Sea during the late winter, and aimed to assess: (i) biotic and environmental factors affecting the weight-specific isotopic shifts in their larval stages; and (ii) the species-specific feeding habits based on the isotopic niche overlaps indicated by the SIBER approach. Material and methods Study area and sample collection Larval fish samples were collected at 112 stations in the shelf-break region of the northeastern ECS (Figure 1) during cruises of the TV Wakatori-Maru (Tottori Prefecture, Japan) from 01 to 24 February 2009 and 28 January to 21 February 2010. A paired bongo net (Posgay and Marak, 1980) with 70-cm mouth diameter, 335-μm mesh, flowmeters, and a depth metre were used for quantitative sampling. A double-oblique tow was conducted at each station from the surface down to 100-m depth or 10 m above the bottom at shallow stations. Because the targeted species are generally distributed in the epipelagic layer without diel vertical migration to mesopelagic layer in their larval stages (Moser and Smith, 1993; Sassa et al., 2002, 2004), sampling was performed regardless of day or night conditions. In order to consider the maternal effect on isotope ratios in the larval fishes, adult fish samples of Diaphus slender type, L. ochotensis and M. asperum were collected using midwater otter trawls (9-mm mesh) during a cruise of the RV Kaiyo-Maru No.7 (Nippon Kaiyo Co. Ltd., Japan) between 18 February and 12 March 2008. Adult fish samples of S. gracilis and V. nimbaria were collected using Matsuda-Oozeki-Hu trawl tows (MOHT, 1.59-mm mesh net) during a cruise of the RV Yoko-Maru (Japan Fisheries Research and Education Agency) from 18 to 28 February 2015. Adult N. japonicus were not caught during the surveys. The position of the Kuroshio axis during each cruise was determined based on the location of 16.5°C isotherm at 200-m depth (Kawai, 1972). The sampling area in the study was thus divided into two parts: the area east of the Kuroshio axis and the area west of the Kuroshio axis (hereafter “AEKA” and “AWKA,” respectively; Sassa et al., 2004) to describe the spatial differences in SIs. Figure 1. View largeDownload slide Adult fish sampling stations (a) for Diaphus slender type, L. ochotensis, and M. asperum in February and March 2008 and S. gracilis and V. nimbaria in February 2015, and larval fish sampling stations (black dots) in (b) February 2009 and (c) February 2010 in the Kuroshio waters of the northeastern East China Sea, on a map of sea water temperature at 200-m depth layer. ECS, East China Sea; KBCNT, Kuroshio Branch Current North of Taiwan; TSWC, Tsushima Warm Current; YSWC, Yellow Sea Warm Current; CRDW, Changjiang River Diluted Water. The black solid lines in (b) and (c) represent the position of the Kuroshio axis as defined by the 16.5°C isotherm at 200-m depth (Kawai, 1972). The sampling areas were defined as AEKA, the area east of the Kuroshio axis including the axis, and AWKA, the area west of the Kuroshio axis (Sassa et al., 2004). Figure 1. View largeDownload slide Adult fish sampling stations (a) for Diaphus slender type, L. ochotensis, and M. asperum in February and March 2008 and S. gracilis and V. nimbaria in February 2015, and larval fish sampling stations (black dots) in (b) February 2009 and (c) February 2010 in the Kuroshio waters of the northeastern East China Sea, on a map of sea water temperature at 200-m depth layer. ECS, East China Sea; KBCNT, Kuroshio Branch Current North of Taiwan; TSWC, Tsushima Warm Current; YSWC, Yellow Sea Warm Current; CRDW, Changjiang River Diluted Water. The black solid lines in (b) and (c) represent the position of the Kuroshio axis as defined by the 16.5°C isotherm at 200-m depth (Kawai, 1972). The sampling areas were defined as AEKA, the area east of the Kuroshio axis including the axis, and AWKA, the area west of the Kuroshio axis (Sassa et al., 2004). All larval fish specimens were first fixed in 5% buffered formalin seawater for 6 h and then transferred to 70% isopropyl alcohol after rinsing formalin with freshwater on board. After all samples were identified in laboratory, they were transferred to 90% ethanol for preservation. All adult fish specimens were first fixed in 10% buffered formalin seawater on board, and transferred to 10% formalin for preservation after identification in laboratory. The six mesopelagic fishes belonging to four families [i.e. Diaphus slender type, M. asperum, and N. japonicus (Myctophidae), L. ochotensis (Microstomatidae), S. gracilis (Gonostomatidae), and V. nimbaria (Phosichthyidae)] were sorted and identified according to Okiyama (2014). Standard body lengths of the larval and adult fish samples were measured to the nearest 0.01 mm with a digital calliper. Samples preserved with ethanol were rinsed with distilled water and vacuum-freeze dried (DRT140FB, ADVANTEC) overnight after removing gut contents. The body weights (hereafter dry-weight) were measured to the nearest 0.1 µg with ultra-microbalances (Mettler Toledo: XP2U, Switzerland). A total of 19 428 larval fish specimens (i.e. n = 10 048 in 2009 and n = 9380 in 2010, respectively) were sampled for the six targeted taxa. Of these, 1– 2 larval fish individuals for each taxon at each station (in total, n = 354 in 2009 and n = 279 in 2010, respectively) were subjected to stable isotopes analysis (SIA). In addition, 8–10 adult fish individuals for each taxon were used for SIA. The SL of the larval fish samples used for SIA during 2009 and 2010 ranged from 3.1 mm in M. asperum to 17.8 mm in V. nimbaria. The dry-weight ranged from 0.04 mg in V. nimbaria to 2.99 mg in S. gracilis. The SL measured for all fishes showed strong nonlinear correlations with dry-weight (Supplementary Table S1). Sample preparation and analysis Lipids were removed using a chloroform-ethanol solution (2:1, V/V) for 12 h to remove the potential effect of lipids on δ13C (Rau et al., 1992). Samples were then dried on a hotplate (60°C, 4 h) to remove any remaining solvent. Whole fish samples (for small larvae fish with 0.02–0.8 mg dry weight) or 0.6–0.8 mg subsamples of the dorsal muscle (for large fish including adult fish) were weighted into 5 × 8 mm silver capsules, and dried at 60°C after acidification with a few drops of 1.0 N HCl to remove inorganic carbon (e.g. CaCO3). SIA was conducted using a continuous-flow elemental analyzer/isotope-ratio mass spectrometer (Flash EA, Delta V Advantage, Thermo Fisher Scientific) with narrow-diameter customized combustion and reduction columns δ or ultra-sensitive analysis of small samples (Ogawa et al., 2010). SI ratios of 13C/12C and 15N/14N were expressed in δ notation defined as follows: δ δ13C,δ15N (‰)=(Rsample/Rstandard−1)×1000 where the term R denotes the ratios of 13C/12C or 15N/14N, and Vienna Pee Dee Belemnite and atmospheric nitrogen were used as standards for carbon and nitrogen isotopes, respectively. Quality assurance of SI ratios was tested by running one known standard (L-Alanine SS09, SI Science Co., Ltd., Japan) for each five unknown (fish) samples. Based on replicate measurement of this standard, analytical precision was generally better than ±0.2‰ SD for both δ13C in sample sizes > 18 µg-C and δ15N in sample sizes > 7 µg-N. The δ13C and δ15N of fish samples are presented without any correction for the effect of organic solvent preservation, except where fish SI ratios are compared with other organic matter [e.g. particulate organic matter (POM)] in which case corrected values are used (see “Discussion” and Supplementary Material sections). Statistical analysis Normality (Shapiro-Wilk test) and homogeneity of variance (Levene’s test) of data were verified before statistical analysis. Spearman’s rank correlation was conducted to test the relationship between isotopes in fish tissues and other variables (i.e. sea surface temperature and salinity in their sampling locations, body length, body weight and CN ratio of the targeted fishes). Isotopic distributions were compared between the sampling locations (i.e. AWKA and AEKA) using a Mann-Whitney U test. Isotopes shifts between the two years were compared for the early larval fishes and late larval fishes using Mann-Whitney U test (see “Results” section for the definitions of “early larval” and “late larval” fishes). Median SI ratios of the late larval fishes were compared among the six taxa using the Kruskal test followed by Dunn post-hoc test. SIBER analyses in the R package SIAR were conducted to compare isotopic niche overlap among the six taxa of late larval fishes, including: (i) the standard ellipse area (SEA) for core isotopic niche width (a proxy for trophic niche width), SEAc for SEA after small sample size correction; and (ii) isotopic niche overlap area and overlap percentage, based on comparison of isotopic niche width using a Bayesian modelling approach (Jackson et al., 2011; Layman et al., 2012). Statistical analyses were conducted using R 3.3.0 (www.R-project.org). Results Isotopic compositions and its shift with body weight The δ13C of the larval fishes sampled in 2009 varied from −21.6‰ in L. ochotensis to −18.8‰ in V. nimbaria, while the δ15N varied from 4.2‰ in M. asperum to 10.1‰ in L. ochotensis. The ranges in δ13C and δ15N were similar in 2010 (−21.5 to −18.6 and 4.2–9.6‰, respectively). The median SL, dry-weight, δ13C and δ15N of larval fishes were significantly different among the six taxa in 2009 and 2010 (Table 1). Table 1. Isotopic compositions (δ13C and δ15N) of six taxa of mesopelagic fish larvae in the ECS. Years Fish species n SL (mm) Dry-W (mg) δ13C (‰) δ15N (‰) Median Range Median Range Median Range Median Range 2009 M. asperum 61 4.5c 3.1–8.8 0.103c 0.048–1.504 −20.3bc −21.4 to –19.3 6.0d 4.2–7.0 Diaphus slender 61 5.4b 3.7–10.0 0.164ab 0.054–1.896 –20.4c –21–2 to –19.6 6.1cd 4.7–7.6 N. japonicus 64 4.5c 3.1–6.7 0.144bc 0.054–0.602 –20.2b –21.0 to –19.1 6.2c 4.8–7.5 L. ochotensis 30 6.8a 6.0–14.0 0.244a 0.089–1.348 –20.9d –21.6 to –20.2 7.7a 5.7–10.1 V. nimbaria 64 7.0a 4.3–17.8 0.156a 0.036–2.720 –19.8a –20.5 to –18.8 6.1c 4.8–7.8 S. gracilis 74 6.6a 3.8–17.6 0.210a 0.066–2.990 –19.8a –20.3 to –19.1 6.6b 5.7–7.5 2010 M. asperum 51 5.1c 3.1–8.6 0.223c 0.070–1.410 –20.2c –21.0 to –18.7 5.6de 4.2–7.5 Diaphus slender 38 5.5c 4.2–8.5 0.151d 0.083–1.136 –20.2c –21.2 to –19.4 6.0cd 5.2–6.8 N. japonicus 54 5.1c 4.1–8.5 0.231c 0.068–0.968 –20.4c –21.3 to –18.6 5.5e 4.3–8.0 L. ochotensis 30 8.3b 6.2–13.0 0.218c 0.105–0.894 –20.8d –21.5 to –19.5 7.3a 4.9–9.6 V. nimbaria 41 10.3a 7.0–17.4 0.500a 0.181–2.730 –19.5a –20.5 to –18.7 6.0bc 5.0–7.3 S. gracilis 65 8.6b 6.2–14.2 0.343b 0.151–1.500 –19.8b –20.8 to –19.1 6.1b 4.8–7.7 Years Fish species n SL (mm) Dry-W (mg) δ13C (‰) δ15N (‰) Median Range Median Range Median Range Median Range 2009 M. asperum 61 4.5c 3.1–8.8 0.103c 0.048–1.504 −20.3bc −21.4 to –19.3 6.0d 4.2–7.0 Diaphus slender 61 5.4b 3.7–10.0 0.164ab 0.054–1.896 –20.4c –21–2 to –19.6 6.1cd 4.7–7.6 N. japonicus 64 4.5c 3.1–6.7 0.144bc 0.054–0.602 –20.2b –21.0 to –19.1 6.2c 4.8–7.5 L. ochotensis 30 6.8a 6.0–14.0 0.244a 0.089–1.348 –20.9d –21.6 to –20.2 7.7a 5.7–10.1 V. nimbaria 64 7.0a 4.3–17.8 0.156a 0.036–2.720 –19.8a –20.5 to –18.8 6.1c 4.8–7.8 S. gracilis 74 6.6a 3.8–17.6 0.210a 0.066–2.990 –19.8a –20.3 to –19.1 6.6b 5.7–7.5 2010 M. asperum 51 5.1c 3.1–8.6 0.223c 0.070–1.410 –20.2c –21.0 to –18.7 5.6de 4.2–7.5 Diaphus slender 38 5.5c 4.2–8.5 0.151d 0.083–1.136 –20.2c –21.2 to –19.4 6.0cd 5.2–6.8 N. japonicus 54 5.1c 4.1–8.5 0.231c 0.068–0.968 –20.4c –21.3 to –18.6 5.5e 4.3–8.0 L. ochotensis 30 8.3b 6.2–13.0 0.218c 0.105–0.894 –20.8d –21.5 to –19.5 7.3a 4.9–9.6 V. nimbaria 41 10.3a 7.0–17.4 0.500a 0.181–2.730 –19.5a –20.5 to –18.7 6.0bc 5.0–7.3 S. gracilis 65 8.6b 6.2–14.2 0.343b 0.151–1.500 –19.8b –20.8 to –19.1 6.1b 4.8–7.7 Superscripts “a, b, c, d, e” indicate significant differences at α = 0.05 level. SL, standard body length; dry-W, dry weight after removing gut contents and lipids; n, number of samples. Table 1. Isotopic compositions (δ13C and δ15N) of six taxa of mesopelagic fish larvae in the ECS. Years Fish species n SL (mm) Dry-W (mg) δ13C (‰) δ15N (‰) Median Range Median Range Median Range Median Range 2009 M. asperum 61 4.5c 3.1–8.8 0.103c 0.048–1.504 −20.3bc −21.4 to –19.3 6.0d 4.2–7.0 Diaphus slender 61 5.4b 3.7–10.0 0.164ab 0.054–1.896 –20.4c –21–2 to –19.6 6.1cd 4.7–7.6 N. japonicus 64 4.5c 3.1–6.7 0.144bc 0.054–0.602 –20.2b –21.0 to –19.1 6.2c 4.8–7.5 L. ochotensis 30 6.8a 6.0–14.0 0.244a 0.089–1.348 –20.9d –21.6 to –20.2 7.7a 5.7–10.1 V. nimbaria 64 7.0a 4.3–17.8 0.156a 0.036–2.720 –19.8a –20.5 to –18.8 6.1c 4.8–7.8 S. gracilis 74 6.6a 3.8–17.6 0.210a 0.066–2.990 –19.8a –20.3 to –19.1 6.6b 5.7–7.5 2010 M. asperum 51 5.1c 3.1–8.6 0.223c 0.070–1.410 –20.2c –21.0 to –18.7 5.6de 4.2–7.5 Diaphus slender 38 5.5c 4.2–8.5 0.151d 0.083–1.136 –20.2c –21.2 to –19.4 6.0cd 5.2–6.8 N. japonicus 54 5.1c 4.1–8.5 0.231c 0.068–0.968 –20.4c –21.3 to –18.6 5.5e 4.3–8.0 L. ochotensis 30 8.3b 6.2–13.0 0.218c 0.105–0.894 –20.8d –21.5 to –19.5 7.3a 4.9–9.6 V. nimbaria 41 10.3a 7.0–17.4 0.500a 0.181–2.730 –19.5a –20.5 to –18.7 6.0bc 5.0–7.3 S. gracilis 65 8.6b 6.2–14.2 0.343b 0.151–1.500 –19.8b –20.8 to –19.1 6.1b 4.8–7.7 Years Fish species n SL (mm) Dry-W (mg) δ13C (‰) δ15N (‰) Median Range Median Range Median Range Median Range 2009 M. asperum 61 4.5c 3.1–8.8 0.103c 0.048–1.504 −20.3bc −21.4 to –19.3 6.0d 4.2–7.0 Diaphus slender 61 5.4b 3.7–10.0 0.164ab 0.054–1.896 –20.4c –21–2 to –19.6 6.1cd 4.7–7.6 N. japonicus 64 4.5c 3.1–6.7 0.144bc 0.054–0.602 –20.2b –21.0 to –19.1 6.2c 4.8–7.5 L. ochotensis 30 6.8a 6.0–14.0 0.244a 0.089–1.348 –20.9d –21.6 to –20.2 7.7a 5.7–10.1 V. nimbaria 64 7.0a 4.3–17.8 0.156a 0.036–2.720 –19.8a –20.5 to –18.8 6.1c 4.8–7.8 S. gracilis 74 6.6a 3.8–17.6 0.210a 0.066–2.990 –19.8a –20.3 to –19.1 6.6b 5.7–7.5 2010 M. asperum 51 5.1c 3.1–8.6 0.223c 0.070–1.410 –20.2c –21.0 to –18.7 5.6de 4.2–7.5 Diaphus slender 38 5.5c 4.2–8.5 0.151d 0.083–1.136 –20.2c –21.2 to –19.4 6.0cd 5.2–6.8 N. japonicus 54 5.1c 4.1–8.5 0.231c 0.068–0.968 –20.4c –21.3 to –18.6 5.5e 4.3–8.0 L. ochotensis 30 8.3b 6.2–13.0 0.218c 0.105–0.894 –20.8d –21.5 to –19.5 7.3a 4.9–9.6 V. nimbaria 41 10.3a 7.0–17.4 0.500a 0.181–2.730 –19.5a –20.5 to –18.7 6.0bc 5.0–7.3 S. gracilis 65 8.6b 6.2–14.2 0.343b 0.151–1.500 –19.8b –20.8 to –19.1 6.1b 4.8–7.7 Superscripts “a, b, c, d, e” indicate significant differences at α = 0.05 level. SL, standard body length; dry-W, dry weight after removing gut contents and lipids; n, number of samples. The δ13C and δ15N of larval fishes showed large variation (Table 1), but SI ratios became relatively constant as body weight increased, especially in terms of δ15N (Figure 2). When these constant values were defined as being SI ratios with SDs < 0.4‰, the dry-weight above which SI ratios become constant (hereafter W1), was estimated as follows for each species: ca. 0.8 mg (SL: 8.2 mm) for Diaphus slender type; ca. 0.7 mg (6.8 mm) for M. asperum; ca. 0.5 mg for L. ochotensis (10.2 mm), N. japonicus (6.6 mm) and S. gracilis (9.6 mm); ca. 1.0 mg (12.7 mm) for V. nimbaria. In this study, the fish larvae of “post-larval stage” were categorized into two growth periods based on W1 (Supplementary Figure S1): “early larval period” with dry-weight less than W1 and “late larval period” with dry-weight larger than W1. Figure 2. View largeDownload slide Relationships between body weight (dry-weight) and tissue isotopes (δ13C and δ15N) in six taxa of larval fishes in February 2009 and 2010. The scatter plots refer to larval fishes, while the boxplots on the left side of the y-axis denote the ranges (minimum, first quartile, median, third quartile and maximum) of isotopes ratios of adult fishes (no data for adult N. japonicus). Dotted lines indicate the body weight in which tissue SI become nearly constant (see text). The numeric values in blue colour under x-axis indicate the back-calculated SL (mm). And the back-calculated SL (mm) was calculated based on the average relationship between SL and dry-weight (mg) for each species, because the SL is variable depending on the morphology and fertility of each individual while the body weight simply increases with growth. Figure 2. View largeDownload slide Relationships between body weight (dry-weight) and tissue isotopes (δ13C and δ15N) in six taxa of larval fishes in February 2009 and 2010. The scatter plots refer to larval fishes, while the boxplots on the left side of the y-axis denote the ranges (minimum, first quartile, median, third quartile and maximum) of isotopes ratios of adult fishes (no data for adult N. japonicus). Dotted lines indicate the body weight in which tissue SI become nearly constant (see text). The numeric values in blue colour under x-axis indicate the back-calculated SL (mm). And the back-calculated SL (mm) was calculated based on the average relationship between SL and dry-weight (mg) for each species, because the SL is variable depending on the morphology and fertility of each individual while the body weight simply increases with growth. During the early larval period, δ13C showed a dispersed distribution and varied between −21.6 and −18.6‰ (Figure 2a–f). Likewise, δ15N varied between 4.2 and 8.0‰ for most of the species, with only L. ochotensis having higher values up to 10.1‰ (Figure 2g–l). During the late larval period, three taxa (M. asperum, Diaphus slender type and N. japonicus) had similarly lower median δ13C (ca. −20.5‰) and δ15N (ca. 5.7‰), while two species (S. gracilis and V. nimbaria) had similarly higher median δ13C (ca. −19.6‰) and δ15N (ca. 6.2‰) (Supplementary Table S2). Although the median δ13C of L. ochotensis was varied from −21.2 to −20.8‰, the median δ15N (ca. 6.9‰) was relatively higher than the other taxa (Supplementary Table S2). For the adult fishes, δ13C was similar across fish species [D. fulgens (Diaphus slender type) = −21.1 ± 0.5‰ (mean ± SD, n = 8); L. ochotensis = −20.2 ± 0.3‰ (n = 8); M. asperum = −21.0 ± 0.5‰ (n = 8); S. gracilis = −20.1 ± 0.6‰ (n = 10) and V. nimbaria = −20.5 ± 0.3‰ (n = 8)] (Figure 2a–f). The δ15N ratios in adult fishes were highest for S. gracilis [12.1 ± 0.8‰ (mean ± SD, n = 10)], followed by D. fulgens, M. asperum, and L. ochotensis [δ15N = 9.6 ± 0.7, 9.4 ± 0.6, and 9.3 ± 1.0‰ in 2008 (n = 8 for each fish species)], while they were lowest for V. nimbaria [7.9 ± 0.3‰ (n = 10)] (Figure 2g–l). Spatial distributions of δ13C and δ15N Although the species-specific diet should be considered using late larval period samples, during which time isotope ratios become relatively constant, the number of such late stage fish samples (all fishes, n ≤ 5 in AWKA, except S. gracilis in 2010, n = 13 in AWKA) was too small for statistical analysis comparison of δ13C and δ15N between AEKA and AWKA. Therefore, the spatial distributions of δ13C and δ15N between AWKA and AEKA were compared based on datasets including both early and late larval periods (Supplementary Table S3). During 2009, δ13C in L. ochotensis, N. japonicus, S. gracilis, and V. nimbaria was significantly higher in AWKA (t test, all p < 0.006), while only δ15N in N. japonicus was significantly higher in AEKA (t test, t62 = −2.212, p = 0.0307). During 2010, δ13C in M. asperum, N. japonicus, and V. nimbaria was significantly higher in AWKA (t test, all p < 0.012), while δ15N in M. asperum, S. gracilis, and V. nimbaria was also significantly higher in AWKA (t test, all p < 0.017). Isotopic comparisons of the larval fishes between 2009 and 2010 The δ13C of early larval period fishes showed no significant difference between 2009 and 2010 (Mann-Whitney U test, all p > 0.05), while the δ15N of early larval period fishes showed significant difference for some fish species (Mann-Whitney U test, p < 0.05 for N. japonicus, S. gracilis, and V. nimbaria) (data not shown). For late larval period fishes, on the other hand, δ15N showed no significant difference (Mann-Whitney U test, all p > 0.21), whereas δ13C only showed significant differences for two fish species (Mann-Whitney U test, p < 0.02 for M. asperum and V. nimbaria) (Supplementary Table S2). Isotopic niches overlap among six species fishes in the late larval period Since there were no significant differences between February 2009 and 2010 for δ15N of all species and δ13C of most species during the late larval period, the isotopes ratios representing only during the later larval period from the 2 different years were compiled together and used for isotopic niche analysis (Figure 3). The δ13C of S. gracilis and V. nimbaria were significantly higher than the other four species (Kruskal test and Dunn post-hoc test, all p < 0.05). The standard ellipses area (SEAc, ‰2) as a proxy of trophic niches were 0.40, 1.10, 0.44, 0.60, 0.51, and 0.63 for Diaphus slender type, L. ochotensis, M. asperum, N. japonicus, S. gracilis, and V. nimbaria, respectively. Figure 3. View largeDownload slide Plots for the core isotopic niche spaces of six taxa of fishes during the late larval period. Figure 3. View largeDownload slide Plots for the core isotopic niche spaces of six taxa of fishes during the late larval period. Furthermore, we quantified the isotopic niche overlap among the six taxa of fishes during the late larval period using SIBER (Table 2). The fishes’ isotopic niches of the taxonomic family Myctophidae (i.e. M. asperum, N. japonicus, and Diaphus slender type) overlapped from 44.6% to 76.5%. Microstomatidae fish species, L. ochotensis, overlapped only 0–4.7% with the other five taxa of fishes. Fishes from different taxonomic families, S. gracilis (Gonostomatidae) and V. nimbaria (Phosichthyidae), which belong to the same taxonomic order of Stomiiformes, also showed high overlap (33.6–41.2%) with each other. Moreover, all six taxa showed little trophic niche overlap (0–13.1%) among the different taxonomic families or orders. Table 2. Overlap percentage of SEAc among six taxa of mesopelagic fishes during late larval period (dry-weight > W1). Dsa Maa Nja Lob Sgc Vnc Dsa 57.3% 67.1% 13.1% 0.0% 0.0% Maa 52.4% 76.5% 0.0% 0.0% 0.0% Nja 44.6% 55.6% 2.1% 0.0% 0.0% Lob 4.7% 0.0% 1.2% 0.0% 0.0% Sgc 0.0% 0.0% 0.0% 0.0% 41.2% Vnc 0.0% 0.0% 0.0% 0.0% 33.6% Dsa Maa Nja Lob Sgc Vnc Dsa 57.3% 67.1% 13.1% 0.0% 0.0% Maa 52.4% 76.5% 0.0% 0.0% 0.0% Nja 44.6% 55.6% 2.1% 0.0% 0.0% Lob 4.7% 0.0% 1.2% 0.0% 0.0% Sgc 0.0% 0.0% 0.0% 0.0% 41.2% Vnc 0.0% 0.0% 0.0% 0.0% 33.6% The table should be read horizontally, as each number in the cell refers to the percentage of overlap of the area of the group indicated in each row (e.g. 52.4% is the percentage of the SEAc of Ma that is overlapped with the Ds, while 57.3% is the percentage of the SEAc of the Ds overlapped with the Ma). SEAc, core isotopic niche width (SEA, standard ellipse area) after small sample size correction. Superscripts “a, b, c” refer to the same family Myctophidae fishes (Ds, Diaphus slender type; Ma, M. asperum; Nj, N. japonicus), family Microstomatidae fish (Lo, L. ochotensis) and the same order Stomiiformes fishes (Sg, S. gracilis; Vn, V. nimbaria). Table 2. Overlap percentage of SEAc among six taxa of mesopelagic fishes during late larval period (dry-weight > W1). Dsa Maa Nja Lob Sgc Vnc Dsa 57.3% 67.1% 13.1% 0.0% 0.0% Maa 52.4% 76.5% 0.0% 0.0% 0.0% Nja 44.6% 55.6% 2.1% 0.0% 0.0% Lob 4.7% 0.0% 1.2% 0.0% 0.0% Sgc 0.0% 0.0% 0.0% 0.0% 41.2% Vnc 0.0% 0.0% 0.0% 0.0% 33.6% Dsa Maa Nja Lob Sgc Vnc Dsa 57.3% 67.1% 13.1% 0.0% 0.0% Maa 52.4% 76.5% 0.0% 0.0% 0.0% Nja 44.6% 55.6% 2.1% 0.0% 0.0% Lob 4.7% 0.0% 1.2% 0.0% 0.0% Sgc 0.0% 0.0% 0.0% 0.0% 41.2% Vnc 0.0% 0.0% 0.0% 0.0% 33.6% The table should be read horizontally, as each number in the cell refers to the percentage of overlap of the area of the group indicated in each row (e.g. 52.4% is the percentage of the SEAc of Ma that is overlapped with the Ds, while 57.3% is the percentage of the SEAc of the Ds overlapped with the Ma). SEAc, core isotopic niche width (SEA, standard ellipse area) after small sample size correction. Superscripts “a, b, c” refer to the same family Myctophidae fishes (Ds, Diaphus slender type; Ma, M. asperum; Nj, N. japonicus), family Microstomatidae fish (Lo, L. ochotensis) and the same order Stomiiformes fishes (Sg, S. gracilis; Vn, V. nimbaria). Discussion Correlations between isotopes and other variables Fish tissue δ13C and δ15N is often correlated with body size, reflecting a shift in the trophic level of food sources as they grow (Deudero et al., 2004). Furthermore, these isotopes potentially change among habitats because the isotopes in phytoplankton, which is the basis of many open ocean food webs, also shift depending on several environmental and physiological conditions [e.g. the availability of terrestrial dissolved inorganic nitrogen (DIN) and dissolved inorganic carbon (DIC), and growth rate]. Therefore, environmental (salinity and temperature) and biotic (dry-weight and SL) parameters were compared with both isotopes (δ13C and δ15N) for all fish species in the larval stage (Table 3) to understand factors causing the variation observed in tissue isotopes. Table 3. Relationships (Spearman r values) between isotopes of fishes in the larval stage and other variables. Years Isotopes Factors Fish species M. asperum Diaphus slender N. japonicus L. ochotensis V. nimbaria S. gracilis 2009 δ13C SST –0.05 –0.03 –0.25* –0.18 –0.03 –0.12 SL –0.42*** –0.48*** –0.54*** –0.43*** 0.06 0.24* dry-W –0.57*** –0.58*** –0.74*** –0.57*** 0.16 0.19 Salinity –0.23 –0.21 –0.45*** –0.23 –0.52*** –0.31** δ15N SST 0.19 0.19 0.28* 0.23 0.03 0.2 SL –0.17 –0.37*** –0.29* –0.72*** –0.23 –0.12 dry-W –0.23 –0.45*** –0.52*** –0.89*** –0.4*** –0.21 Salinity 0.38*** 0.11 0.12 0.01 0.34** 0.08 2010 δ13C SST –0.24 0.11 –0.43*** 0.19 –0.23 0 SL –0.09 –0.38* –0.04 0.23 0.48*** 0.23 dry-W –0.03 –0.32* –0.18 0.2 0.42*** 0.27* Salinity –0.51*** –0.15 –0.42*** –0.09 –0.51*** –0.1 δ15N SST –0.36** –0.2 –0.17 0.08 –0.38** –0.22 SL 0.05 –0.15 –0.12 –0.64*** 0.55*** 0.34** dry-W 0 0.07 –0.15 –0.64*** 0.5*** 0.42*** Salinity –0.28* –0.35* –0.17 0.19 –0.21 –0.23 Years Isotopes Factors Fish species M. asperum Diaphus slender N. japonicus L. ochotensis V. nimbaria S. gracilis 2009 δ13C SST –0.05 –0.03 –0.25* –0.18 –0.03 –0.12 SL –0.42*** –0.48*** –0.54*** –0.43*** 0.06 0.24* dry-W –0.57*** –0.58*** –0.74*** –0.57*** 0.16 0.19 Salinity –0.23 –0.21 –0.45*** –0.23 –0.52*** –0.31** δ15N SST 0.19 0.19 0.28* 0.23 0.03 0.2 SL –0.17 –0.37*** –0.29* –0.72*** –0.23 –0.12 dry-W –0.23 –0.45*** –0.52*** –0.89*** –0.4*** –0.21 Salinity 0.38*** 0.11 0.12 0.01 0.34** 0.08 2010 δ13C SST –0.24 0.11 –0.43*** 0.19 –0.23 0 SL –0.09 –0.38* –0.04 0.23 0.48*** 0.23 dry-W –0.03 –0.32* –0.18 0.2 0.42*** 0.27* Salinity –0.51*** –0.15 –0.42*** –0.09 –0.51*** –0.1 δ15N SST –0.36** –0.2 –0.17 0.08 –0.38** –0.22 SL 0.05 –0.15 –0.12 –0.64*** 0.55*** 0.34** dry-W 0 0.07 –0.15 –0.64*** 0.5*** 0.42*** Salinity –0.28* –0.35* –0.17 0.19 –0.21 –0.23 SST, sea surface temperature; SL, standard body length; dry-W, dry-weight after removing gut contents and lipids. Asterisks indicate significant correlations between two variables. * p < 0.05; ** p < 0.01; *** p < 0.001. Table 3. Relationships (Spearman r values) between isotopes of fishes in the larval stage and other variables. Years Isotopes Factors Fish species M. asperum Diaphus slender N. japonicus L. ochotensis V. nimbaria S. gracilis 2009 δ13C SST –0.05 –0.03 –0.25* –0.18 –0.03 –0.12 SL –0.42*** –0.48*** –0.54*** –0.43*** 0.06 0.24* dry-W –0.57*** –0.58*** –0.74*** –0.57*** 0.16 0.19 Salinity –0.23 –0.21 –0.45*** –0.23 –0.52*** –0.31** δ15N SST 0.19 0.19 0.28* 0.23 0.03 0.2 SL –0.17 –0.37*** –0.29* –0.72*** –0.23 –0.12 dry-W –0.23 –0.45*** –0.52*** –0.89*** –0.4*** –0.21 Salinity 0.38*** 0.11 0.12 0.01 0.34** 0.08 2010 δ13C SST –0.24 0.11 –0.43*** 0.19 –0.23 0 SL –0.09 –0.38* –0.04 0.23 0.48*** 0.23 dry-W –0.03 –0.32* –0.18 0.2 0.42*** 0.27* Salinity –0.51*** –0.15 –0.42*** –0.09 –0.51*** –0.1 δ15N SST –0.36** –0.2 –0.17 0.08 –0.38** –0.22 SL 0.05 –0.15 –0.12 –0.64*** 0.55*** 0.34** dry-W 0 0.07 –0.15 –0.64*** 0.5*** 0.42*** Salinity –0.28* –0.35* –0.17 0.19 –0.21 –0.23 Years Isotopes Factors Fish species M. asperum Diaphus slender N. japonicus L. ochotensis V. nimbaria S. gracilis 2009 δ13C SST –0.05 –0.03 –0.25* –0.18 –0.03 –0.12 SL –0.42*** –0.48*** –0.54*** –0.43*** 0.06 0.24* dry-W –0.57*** –0.58*** –0.74*** –0.57*** 0.16 0.19 Salinity –0.23 –0.21 –0.45*** –0.23 –0.52*** –0.31** δ15N SST 0.19 0.19 0.28* 0.23 0.03 0.2 SL –0.17 –0.37*** –0.29* –0.72*** –0.23 –0.12 dry-W –0.23 –0.45*** –0.52*** –0.89*** –0.4*** –0.21 Salinity 0.38*** 0.11 0.12 0.01 0.34** 0.08 2010 δ13C SST –0.24 0.11 –0.43*** 0.19 –0.23 0 SL –0.09 –0.38* –0.04 0.23 0.48*** 0.23 dry-W –0.03 –0.32* –0.18 0.2 0.42*** 0.27* Salinity –0.51*** –0.15 –0.42*** –0.09 –0.51*** –0.1 δ15N SST –0.36** –0.2 –0.17 0.08 –0.38** –0.22 SL 0.05 –0.15 –0.12 –0.64*** 0.55*** 0.34** dry-W 0 0.07 –0.15 –0.64*** 0.5*** 0.42*** Salinity –0.28* –0.35* –0.17 0.19 –0.21 –0.23 SST, sea surface temperature; SL, standard body length; dry-W, dry-weight after removing gut contents and lipids. Asterisks indicate significant correlations between two variables. * p < 0.05; ** p < 0.01; *** p < 0.001. Decreases in δ13C in POM are observed in western Kyushu when primary productivity decreases due to nutrient depletion associated with salinity increases (i.e. negative relationship; Ozaki, 2016), partly because (i) the δ13C of phytoplankton-derived organic matter often decreases during bacterial decomposition (Lehmann et al., 2002) and (ii) relative isotopic discrimination increases at lower growth rates (Rau et al., 1996). This isotopic change in POM can generally be reflected in the SI ratios of large zooplankton and other consumers at higher trophic levels. Therefore, the negative correlation between δ13C in larval tissues and salinity may indicate that larval fishes inhabiting the AWKA and on the continental shelf where relatively lower salinities occur have higher δ13C (Supplementary Table S3) due to feeding on plankton supported by phytoplankton with higher δ13C. On the other hand, SI ratios of most larval fishes showed a negative correlation with both SL and body weight in our study. A similar result was observed for bluefin tuna larvae in the Eastern and Western Gulf of Mexico, although the reason was not clearly identified (Laiz-Carrión et al., 2015). Because both SL and body weight did not show any correlation with salinity in our dataset, the negative correlation between SL and δ13C in the larval tissues is not likely to indicate larval migration to AEKA (i.e. high salinity zone) during larval growth. Potential reasons are further considered from the viewpoint of maternal effects in the following section. Fish species- and weight-specific δ15N ratios in the larval stage The isotope ratios of fishes during the early larval period showed large within species variation depending on the fish species, but trends reached a relative constant in all the species when weights were larger than W1, especially for δ15N. Here we considered the potential factors causing these species-, and weight-specific variations in δ13C and δ15N in the larval stage from the viewpoint of (i) the effect of nutrition from parent fish, and (ii) diet shifts during morphological development. Maternal effect Uriarte et al. (2016) reported that the yolk sac of T. thynnus were fully consumed at ca. 3.0-mm NL (notochord length), while its effect on SI ratios continued to ca. 4.0-mm NL. In other words, there was little or no isotopic discrimination between fish egg and yolk-sac tissue before ca. 4.0-mm NL. In the case of M. asperum, its yolk sac was reported to be fully consumed before ca. 3.0-mm NL (Sassa and Kawaguchi, 2004). Because the SL range of M. asperum in this study was 3.1–8.8 mm (Supplementary Table S1), SI ratios of fish larvae may continue to reflect the SI ratios of egg for several days (e.g. before ca. 4.0-mm NL). Therefore, the SI ratios of the smallest larvae of several fish species were used as a proxy of SI ratios of fish eggs, to assess the type of breeder strategy (see “Introduction” section and Supplementary Figure S2). In the spawning area of this study (mainly AEKA), the estimated δ15N of eggs from “income breeder” fishes (δegg) were ca. 5.0–6.1‰, based on the estimated SI ratios of potential food for parent fishes (i.e. large zooplankton in AEKA) and laboratory-determined isotopic enrichments between adult “income breeder” food and their eggs (see Figure 4 and Supplementary Material). However, δ15N of the smallest larvae (δlarvae) of L. ochotensis (8.6 ± 0.7‰) was significantly higher than the estimated SI ratios of eggs from “income breeder” fishes in AEKA (Mann-Whitney U test, p < 0.001). This clear discrepancy suggested that at least L. ochotensis in our sampling years did not show “income breeder”-like characteristics (Figure 4). L. ochotensis is known to migrate to the high productivity subarctic Oyashio region before returning to the subtropical Kuroshio region before the spawning season (Sassa et al., 2004). If “capital breeder”-like characteristics contributed significantly to the spawning strategy of L. ochotensis, the energy stored in parents could be used for reproduction. This hypothesis is likely to explain the rapid shift in SI ratios of the L. ochotensis larvae from values close to those of adult fishes to values reflecting the food sources for their larvae in AEKA. Figure 4. View largeDownload slide SIs measured in the targeted mesopelagic fish larvae, and those estimated for eggs of an “income breeder,” for estimation of spawning strategy. For the targeted three species of mesopelagic fishes, SI ratios of only the smallest fish individuals with body weights <0.2 mg were used as the beginning of the early larval fishes (i.e. a proxy of fish egg), and compared with the estimated SI for eggs of an “income breeder” in the spawning AEKA area. Mesopelagic fish data were corrected based on the effect of organic solvent preservation. The borders of all boxes indicate the variations (mean ± 1 SD) of δ13Ccorrected and δ15Ncorrected. POM SI ratios in the AEKA area were cited from Cheung et al. (2017) and H. Saito unpublished data (Supplementary Table S4). SI ratios for eggs of an “income breeder” were estimated based on POM SI ratios and the reported enrichment factors between phytoplankton and zooplankton, and between food for parent fish and their eggs. See more detailed explanations in Supplementary Material and Supplementary Figure S2. Figure 4. View largeDownload slide SIs measured in the targeted mesopelagic fish larvae, and those estimated for eggs of an “income breeder,” for estimation of spawning strategy. For the targeted three species of mesopelagic fishes, SI ratios of only the smallest fish individuals with body weights <0.2 mg were used as the beginning of the early larval fishes (i.e. a proxy of fish egg), and compared with the estimated SI for eggs of an “income breeder” in the spawning AEKA area. Mesopelagic fish data were corrected based on the effect of organic solvent preservation. The borders of all boxes indicate the variations (mean ± 1 SD) of δ13Ccorrected and δ15Ncorrected. POM SI ratios in the AEKA area were cited from Cheung et al. (2017) and H. Saito unpublished data (Supplementary Table S4). SI ratios for eggs of an “income breeder” were estimated based on POM SI ratios and the reported enrichment factors between phytoplankton and zooplankton, and between food for parent fish and their eggs. See more detailed explanations in Supplementary Material and Supplementary Figure S2. On the other hand, Sassa et al. (2016) found that all three species of Diaphus stubby type fishes (Diaphus garmani, D. chrysorhynchus, and D. watasei) were primarily “income breeders” based on multiple physiological characteristics (i.e. asynchronous oocyte development, multiple batch spawning, and active feeding during the spawning season). Therefore, it is expected that Diaphus slender type should mainly behave as an “income breeder.” In the case of Diaphus slender type in this study, the SI differences between the early larval period fishes and the potential food of parent fish in AEKA (i.e. Δδ13CDs = δ13Clarvae – δ13Cadultfood = 0.0–0.5‰, Δδ15NDs = δ15Nlarvae – δ15Nadultfood = 0.0–1.5‰) were equivalent to the differences in SI between eggs and foods of parent fishes just before the spawning period for the “income breeder” (i.e. Δδ = δegg – δadultfood) reported in Tanaka et al. (2016). Therefore, the isotopes signatures of Diaphus slender type during the early larval period were consistent with the biological observations of Sassa et al. (2016) suggesting that Diaphus spp. are “income breeders.” Similarly, V. nimbaria primarily belongs to the “income breeders” based on asynchronous oocyte development and multiple batch spawning (Stequert et al., 2003). This characteristic also seems to be supported by the overlap between the δ13C and δ15N ranges of early larval V. nimbaria and the estimated isotopic ranges in eggs of “income breeders” that feed and spawn in AEKA. Here, we tested only three taxa of fishes as a first approximation, focussing on those whose larvae had distinctly different SI ratios from the estimated SI ratios in eggs of an income breeder, or which had other evidence of being income breeders based on physiological characteristics. In other species, many uncertainties in the SI signatures used for the above estimations (i.e. lack of actual zooplankton and parent SI data, potential SI variation in space and time, and species-specific SI enrichment factors, etc.) might obscure the breeder type-depended isotopic characteristics. However, the good accordance between evidence from SI ratios and physiological characteristics suggests that SI ratios are a potentially effective tool for understanding reproductive strategies in the field, especially under conditions where energy acquisition and allocation to egg production is variable and depends on food availability during the spawning period (McBride et al., 2015). Further studies, including accumulating SI and physiological evidence through the intensive field surveys and rearing experiments to check species-specific isotopic enrichment between food and eggs, could improve our understanding of the spawning strategy of mesopelagic fishes. Effect of morphological development The δ13C and δ15N in POM showed large variation (Supplementary Table S4), because POM consists of a variety of components (e.g. diazotrophs, phytoplankton, small zooplankton) which have different nitrogen sources [e.g. terrestrial and upwelling-derived DIN and dissolved organic nitrogen (DON)], have differing isotopic fractionation associated with DIN and DIC uptake (e.g. Rau et al., 1996), and are at different stages of decomposition. Reflecting these variations, the SI ratios in zooplankton at each subsequent trophic step also differ between zooplankton groups, species and development stage (Koppelmann et al., 2009). Therefore, large variation in SI ratios during early larval periods may be explained by non-selective feeding on a variety of organisms with different SI ratios due to poor swimming ability, in addition to maternal effect from adult fishes with a broad range of diets and thus tissue SI. When some morphological characteristics (e.g. caudal and anal fins) are completely developed, it is plausible that tissue SI become relatively constant because the fishes can then start to selectively prey on certain diet species with specific and relatively constant δ13C and δ15N. Larvae of Diaphus theta and M. asperum in the transition region of the western North Pacific have been reported to be daytime visual feeders (Sassa and Kawaguchi, 2004, 2005), thus visual targeting of specific small zooplankton is possible in the larval stage. Actually, the caudal fin of N. japonicus is completely developed when its’ size become > 7 mm (Okiyama, 2014), which is consistent with the SL at the timing of W1 of N. japonicus larvae. Moreover, gut contents analyses (GCAs) in the transition region of the western North Pacific Ocean indicated that larval D. theta, which have morphological characteristics of the slender type in this study in the larval stage, started to change their diet from copepod nauplii to calanoid copepodites at body lengths > 8.0 mm (Sassa and Kawaguchi, 2005), which is consistent with the SL in the timing of W1 of Diaphus slender type larvae. This suggests that the timing of W1 based on SI values may be effective for determining the timing of feeding changes of fishes in the larval stage. Food sources and isotopic niches overlap among six species fish larvae As discussed earlier, tissue isotopes of fishes during the early larval period were related not only to food source isotopes but also maternal effects. Therefore, only isotope data obtained during the late larval period, during which selective feeding is expected to start, was considered suitable for trophic niche analysis (SIBER) of the larvae of targeted fishes. The isotopic niche for late larval fishes within the same taxonomic family (family Myctophidae: Diaphus slender type, M. asperum and N. japonicus) was highly overlapped. This suggests that there is competition for diets and/or habitats for late larval fishes belonging to the same taxonomic family. However, even though S. gracilis (family Gonostomatidae) and V. nimbaria (family Phosichthyidae) belong to different taxonomic families, their larvae showed relatively high isotopic niche overlap with each other. Although the main habitat depths of these larvae show small differences between S. gracilis (55–100 m; Watanabe et al., 2010) and V. nimbaria (25–75 m; Boehlert et al., 1992), the vertical mixing of the Kuroshio waters (Watanabe et al., 2010), sinking POM, and active vertical migration of zooplankton (Steinberg et al., 2002; Bianchi et al., 2013) may cause similar isotopic niches (food source) between these fishes. GCAs of these larval fishes in the northwestern ECS were also conducted by C. Sassa (unpublished data). Most gut contents (80–95%) in Diaphus slender type, N. japonicus and L. ochotensis were digested materials and not identified. However, the high isotopic niche overlap among the species in the family Myctophidae and the GCA of M. asperum indicated that Copepoda and other Crustaceans may be the common diets of larval Myctophidae fishes in the northwestern ECS. On the other hand, the gut of larval V. nimbaria is straight, and it was almost impossible to find any remaining food. The diet of larval S. gracilis was almost completely identified as Calanoid copepodite and other Crustaceans. Therefore, some of the diets of the late larval V. nimbaria were also expected to be Calanoid copepodite and other Crustacean species similar to S. gracilis, because their isotopic niches overlapped. Isotopic niche of larval L. ochotensis (Microstomatidae) overlapped only 0–4.7% with the other five taxa of larval fishes. The potential contributions from other minor diet components, such as appendicularian houses, may be identified with molecular-based analysis using the digested materials in the future (e.g. Hirai et al., 2017). Conclusion Tissue isotopes of mesopelagic fishes during the early larval period showed large variation, which probably reflected maternal transmission from parent fishes, and non-selective feeding on a variety of plankton species due to poor swimming ability. The isotopes of late larval fish from six mesopelagic fish taxa could be appropriately used for trophic position analysis because the maternal effect was shown to disappear. Tissue isotopes became almost constant at the following body size: ca. 0.8 mg (back-calculated SL: 8.2 mm) for Diaphus slender type; ca. 0.7 mg (6.8 mm) for M. asperum; ca. 0.5 mg for L. ochotensis (10.2 mm), N. japonicus (6.6 mm) and S. gracilis (9.6 mm); ca. 1.0 mg (12.7 mm) for V. nimbaria. Combined with physiological observations, the similarity or difference between measured tissue SI of the early larval fishes and the estimated SI ratios of eggs for “income breeder” in the spawning area supported the idea that Diaphus slender type and V. nimbaria are mainly “income breeders,” while L. ochotensis didn’t show “income breeder”-like characteristics in this study. The isotopic niche overlap ranged from 44.6–76.5% among species in the same taxonomic family of Myctophidae larval fishes, followed by 33.6–41.2% overlap between S. gracilis and V. nimbaria. Consequently, there may be intense competition for diets or habitats in larval fishes within the same taxonomic family. Even if principal diets are not identified with GCAs due to digestion or evacuation, diet information from other fish species having similar isotopic niches can improve our understanding of the diets of larval fishes. Acknowledgements We gratefully thank Dr Satoshi Kitajima of Seikai National Fisheries Research Institute, Japan Fisheries Research and Education Agency for statistical analysis and valuable discussions. We thank Cuixia Fei (Nagoya Univ.) for the assistances with sample pre-treatment process. We are also grateful to the captains, officers, and crews of TV Wakatori-Maru, RV Kaiyo-Maru No. 7, and Yoko-Maru for their assistances in the field sampling. This Study was financially supported by the Study of Kuroshio Ecosystem Dynamics for Sustainable Fisheries (SKED) funded by the MEXT (Ministry of Education, Culture, Sports, Science and Technology), Japan. References Anderson J. T. 1988 . A Review of Size Dependent Survival During Pre-Recruit Stages of Fishes in Relation to Recruitment . Journal of Northwest Atlantic Fishery Science , 8 : 55 – 66 . Google Scholar Crossref Search ADS Bailey K. M. , Houde E. D. 1989 . Predation on eggs and larvae of marine fishes and the recruitment problem . Advances in Marine Biology , 25 : 1 – 83 . Google Scholar Crossref Search ADS Bianchi D. , Stock C. , Galbraith E. D. , Sarmiento J. L. 2013 . Diel vertical migration: ecological controls and impacts on the biological pump in a one-dimensional ocean model . Global Biogeochemical Cycles , 27 : 478 – 491 . Google Scholar Crossref Search ADS Boehlert G. W. , Watson W. , Sun L. C. 1992 . Horizontal and vertical distributions of larval fishes around an isolated oceanic island in the tropical Pacific . Deep Sea Research , 39 : 459 – 466 . Google Scholar Crossref Search ADS Champalbert G. A. , Kouamé B. , Pagano M. , Marchal E. 2008 . Feeding behavior of adult Vinciguerria nimbaria (Phosichthyidae), in the tropical Atlantic (0-4°N, 15°W) . Marine Biology , 156 : 79 – 95 . Google Scholar Crossref Search ADS Chen W. Y. , Lee M. A. , Lan K. W. , Gong G. C. 2014 . Distributions and assemblages of larval fish in the East China Sea during the northeasterly and southwesterly monsoon seasons of 2008 . Biogeosciences , 11 : 547 – 561 . Google Scholar Crossref Search ADS Cherel Y. , Fontaine C. , Richard P. , Labatc J. P. 2010 . Isotopic niches and trophic levels of myctophid fishes and their predators in the Southern Ocean . Limnology and Oceanography , 55 : 324 – 332 . Google Scholar Crossref Search ADS Cheung S. , Suzuki K. , Saito H. , Umezawa Y. , Xia X. , Liu H. 2017 . Highly heterogeneous diazotroph communities in the Kuroshio Current and the Tokara Strait, Japan . PLoS ONE , 12 : e0186875 . Google Scholar Crossref Search ADS PubMed Choy C. A. , Davison P. C. , Drazen J. C. , Flynn A. , Gier E. J. , Hoffman J. C. , McClain-Counts J. P. et al. 2012 . Global trophic position comparison of two dominant mesopelagic fish families (Myctophidae, Stomiidae) using amino acid nitrogen isotopic analyses . PLoS ONE , 7 : e50133 . Google Scholar Crossref Search ADS PubMed Deudero S. , Pinnegar J. K. , Polunin N. V. C. , Morey G. , Morales-Nin B. 2004 . Spatial variation and ontogenic shifts in the isotopic composition of Mediterranean littoral fishes . Marine Biology , 145 : 971 – 981 . Google Scholar Crossref Search ADS Flynn A. J. , Kloser R. J. 2012 . Cross-basin heterogeneity in lanternfish (family Myctophidae) assemblages and isotopic niches (δ13C and δ15N) in the southern Tasman Sea abyssal basin . Deep Sea Research Part I , 69 : 113 – 127 . Google Scholar Crossref Search ADS García A. , Laiz-Carrión R. , Uriarte A. , Quintanilla J. M. , Morote E. , Rodríguez J. M. , Alemany F. 2017 . Differentiated stable isotopes signatures between pre- and post-flexion larvae of Atlantic bluefin tuna (Thunnus thynnus) and of its associated tuna species of the Balearic Sea (NW Mediterranean) . Deep Sea Research Part II , 140 : 18 – 24 . Google Scholar Crossref Search ADS Hirai J. , Hidaka K. , Nagai S. , Ichikawa T. 2017 . Molecular-based diet analysis of the early post-larvae of Japanese sardine Sardinops melanostictus and Pacific round herring Etrumeus teres . Marine Ecology Progress Series , 564 : 99 – 113 . Google Scholar Crossref Search ADS Irigoien X. , Klevjer T. A. , Røstad A. , Martinez U. , Boyra G. , Acuña J. L. , Bode A. , et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean . Nature Communications , 5 : 1 – 17 . Google Scholar Crossref Search ADS Jackson A. L. , Inger R. , Parnell A. C. , Bearhop S. 2011 . Comparing isotopic niche widths among and within communities: sIBER - Stable Isotope Bayesian Ellipses in R . Journal of Animal Ecology , 80 : 595 – 602 . Google Scholar Crossref Search ADS PubMed Kaartvedt S. , Staby A. , Aksnes D. L. 2012 . Efficient trawl avoidance by mesopelagic fishes causes large underestimation of their biomass . Marine Ecology Progress Series , 456 : 1 – 6 . Google Scholar Crossref Search ADS Kawai H. 1972 . Hydrography of the Kuroshio Extension. In Kuroshio - Its Physical Aspects , pp. 235 – 354 . Ed. by Stommel H. , Yoshida K. . University of Tokyo Press , Tokyo (in Japanese). Koppelmann R. , Böttger-Schnack R. , Möbius J. , Weikert H. 2009 . Trophic relationships of zooplankton in the eastern Mediterranean based on stable isotope measurements . Journal of Plankton Research , 31 : 669 – 686 . Google Scholar Crossref Search ADS Laiz-Carrión R. , Quintanilla J. M. , Mercado J. M. , García A. 2011 . Combined study of daily growth variability and nitrogen-carbon isotopic signature analysis of schooling Sardina pilchardus larvae . Journal of Fish Biology , 79 : 896 – 914 . Google Scholar Crossref Search ADS PubMed Laiz-Carrión R. , Gerard T. , Uriarte A. , Malca E. , Quintanilla J. M. , Muhling B. A. , Alemany F. , et al. 2015 . Trophic Ecology of Atlantic Bluefin Tuna (Thunnus thynnus) Larvae from the Gulf of Mexico and NW Mediterranean Spawning Grounds: a Comparative Stable Isotope Study . PLoS One , 10 : e0133406 . Google Scholar Crossref Search ADS PubMed Layman C. A. , Araujo M. S. , Boucek R. , Hammerschlag-Peyer C. M. , Harrison E. , Jud Z. R. , Matich P. , et al. 2012 . Applying stable isotopes to examine food-web structure: an overview of analytical tools . Biological Reviews , 87 : 545 – 562 . Google Scholar Crossref Search ADS PubMed Lehmann M. F. , Bernasconi S. M. , Barbieri A. , McKenzie J. A. 2002 . Preservation of organic matter and alteration of its carbon and nitrogen isotope composition during simulated and in situ early sedimentary diagenesis . Geochimica et Cosmochimica Acta , 66 : 3573 – 3584 . Google Scholar Crossref Search ADS Luo J. , Ortner P. B. , Forcucci D. , Cummings S. R. 2000 . Diel vertical migration of zooplankton and mesopelagic fish in the Arabian Sea . Deep Sea Research Part II , 47 : 1451 – 1473 . Google Scholar Crossref Search ADS McBride R. S. , Somarakis S. , Fitzhugh G. R. , Albert A. , Yaragina N. A. , Wuenschel M. J. , Alonso-Fernández A. , et al. 2015 . Energy acquisition and allocation to egg production in relation to fish reproductive strategies . Fish and Fisheries , 16 : 23 – 57 . Google Scholar Crossref Search ADS McCutchan J. H. , Lewis W. M. , Kendall C. , McGrath C. C. 2003 . Variation in trophic shift for stable isotope ratios of carbon, nitrogen, and sulfur . Oikos , 102 : 378 – 390 . Google Scholar Crossref Search ADS Minagawa M. , Wada E. 1984 . Stepwise enrichment of 15N along food chains: further evidence and the relation between δ15N and animal age . Geochimica et Cosmochimica Acta , 48 : 1135 – 1140 . Google Scholar Crossref Search ADS Moser H. G. , Smith P. E. 1993 . Larval fish assemblages of the California Current region and their horizontal and vertical distributions across a front . Bulletin of Marine Science , 53 : 645 – 691 . Ogawa N. O. , Nagata T. , Kitazato H. , Ohkouchi N. 2010 . Ultra-sensitive elemental analyzer/isotope ratio mass spectrometer for stable nitrogen and carbon isotope analyses. In Earth, Life, and Isotopes , pp. 339 – 353 . Ed. by Ohkouchi N. , Tayasu I. , Koba K. . Kyoto University Press , Kyoto . Okiyama M. 2014 . An Atlas of the Early Stage Fishes in Japan . Tokai University Press , Tokyo (in Japanese). Ozaki T. 2016 . Seasonal dynamics of nutrients, dissolved and particulate organic matter, and contribution to remineralized nutrients at Ariake Bay. Master thesis, Nagasaki University, Nagasaki, Japan (in Japanese). Posgay J. A. , Marak R. R. 1980 . The MARMAP Bongo Zooplankton Samplers . Journal of Northwestern Atlantic Fisheries of Science , 1 : 91 – 99 . Google Scholar Crossref Search ADS Post D. M. 2002 . Using stable isotopes to estimate trophic position: models, methods, and assumptions . Ecology , 83 : 703 – 718 . Google Scholar Crossref Search ADS Potier M. , Marsac F. , Cherel Y. , Lucas V. , Sabatié R. , Maury O. , Ménard F. 2007 . Forage fauna in the diet of three large pelagic fishes (lancetfish, swordfish and yellowfin tuna) in the western equatorial Indian Ocean . Fisheries Research , 83 : 60 – 72 . Google Scholar Crossref Search ADS Quintanilla J. , Laiz-Carrión R. , Uriarte A. , García A. 2015 . Influence of trophic pathways on daily growth patterns of western Mediterranean anchovy Engraulis encrasicolus larvae . Marine Ecology Progress Series , 531 : 263 – 275 . Google Scholar Crossref Search ADS Rau G. H. , Ainley D. G. , Bengtson J. L. , Torres J. J. , Hopkins T. L. 1992 . 15N/14N and 13C/12C in Weddell Sea birds, seals, and fish: implications for diet and trophic structure . Marine Ecology Progress Series , 84 : 1 – 8 . Google Scholar Crossref Search ADS Rau G. H. , Riebesell U. , Wolf-Gladrow D. 1996 . A model of photosynthetic 13C fractionation by marine phytoplankton based on diffusive molecular CO2 uptake . Marine Ecology Progress Series , 133 : 275 – 285 . Google Scholar Crossref Search ADS Sassa C. , Kawaguchi K. 2004 . Larval feeding habits of Diaphus garmani and Myctophum asperum (Pisces: myctophidae) in the transition region of the western North Pacific . Marine Ecology Progress Series , 278 : 279 – 290 . Google Scholar Crossref Search ADS Sassa C. , Kawaguchi K. , Mori K. 2004 . Late winter larval mesopelagic fish assemblage in the Kuroshio waters of the western North Pacific . Fisheries Oceanography , 13 : 121 – 133 . Google Scholar Crossref Search ADS Sassa C. , Kawaguchi K. 2005 . Larval feeding habits of Diaphus theta, Protomyctophum thompsoni, and Tarletonbeania taylori (Pisces: myctophidae) in the transition region of the western North Pacific . Marine Ecology Progress Series , 298 : 261 – 276 . Google Scholar Crossref Search ADS Sassa C. , Hirota Y. 2013 . Seasonal occurrence of mesopelagic fish larvae on the onshore side of the Kuroshio off southern Japan . Deep Sea Research Part I , 81 : 49 – 61 . Google Scholar Crossref Search ADS Sassa C. , Konishi Y. 2015 . Late winter larval fish assemblage in the southern East China Sea, with emphasis on spatial relations between mesopelagic and commercial pelagic fish larvae . Continental Shelf Research , 108 : 97 – 111 . Google Scholar Crossref Search ADS Sassa C. , Moser J. G. , Kawaguchi K. 2002 . Horizontal and vertical distribution patterns of larval myctophid fishes in the Kuroshio region . Fisheries Oceanography , 11 : 1 – 10 . Google Scholar Crossref Search ADS Sassa C. , Tanaka H. , Ohshimo S. 2016 . Comparative reproductive biology of three dominant myctophids of the genus Diaphus on the slope region of the East China Sea . Deep Sea Research Part I , 115 : 145 – 158 . Google Scholar Crossref Search ADS Steinberg D. K. , Goldthwait S. A. , Hansell D. A. 2002 . Zooplankton vertical migration and the active transport of dissolved organic and inorganic nitrogen in the Sargasso Sea . Deep Sea Research Part I , 49 : 1445 – 1461 . Google Scholar Crossref Search ADS Stequert B. , Menard F. , Marchal E. 2003 . Reproductive biology of Vinciguerria nimbaria in the equatorial waters of the eastern Atlantic Ocean . Journal of Fish Biology , 62 : 1116 – 1136 . Google Scholar Crossref Search ADS Sugisaki H. , Nonaka M. , Ishizaki S. , Hidaka K. , Kameda T. , Hirota Y. , Oozeki Y. , et al. 2010 . Marine Ecosystems of the North Pacific Ocean, 2003-2008. Marine Ecosystems of the North Pacific Ocean, PICES Special Publication, 4: 393. Tanaka H. , Yoneda M. , Kitano H. , Kawamura K. , Imanaga Y. , Matsuyama M. , Okamura K. , et al. 2016 . Stable isotope evidence for income resource allocation to egg production in the Japanese anchovy Engraulis japonicus . Marine Biology , 163 : 28 – 33 . Google Scholar Crossref Search ADS Uriarte A. , García A. , Ortega A. , De la Gándara F. , Quintanilla J. , Laiz-Carrión R. , Laiz-Carrión R. 2016 . Isotopic discrimination factors and nitrogen turnover rates in reared Atlantic bluefin tuna larvae (Thunnus thynnus): effects of maternal transmission . Scientia Marina , 80 : 447 – 456 . Google Scholar Crossref Search ADS Valls M. , Olivar M. P. , Fernández de Puelles M. L. , Molí B. , Bernal A. , Sweeting C. J. 2014 . Trophic structure of mesopelagic fishes in the western Mediterranean based on stable isotopes of carbon and nitrogen . Journal of Marine Systems , 138 : 160 – 170 . Google Scholar Crossref Search ADS Watanabe H. , Moku M. , Kawaguchi K. , Ishimaru K. , Ohno A. 1999 . Diel vertical migration of myctophid fishes (Family Myctophidae) in the transitional waters of the western North Pacific . Fisheries Oceanography , 8 : 115 – 127 . Google Scholar Crossref Search ADS Watanabe H. , Sassa C. , Ishida M. 2010 . Late winter vertical distribution of mesopelagic fish larvae in the Kuroshio Current region of the western North Pacific . Bulletin of the Japanese Society of Fisheries of Oceanography , 74 : 153 – 158 . Wells R. J. D. , Rooker J. R. 2009 . Feeding ecology of pelagic fish larvae and juveniles in slope waters of the Gulf of Mexico . Journal of Fish Biology , 75 : 1719 – 1732 . Google Scholar Crossref Search ADS PubMed Yatsu A. , Sassa C. , Moku M. , Kinoshita T. 2005 . Night-time vertical distribution and abundance of small epipelagic and mesopelagic fishes in the upper 100 m layer of the Kuroshio-Oyashio Transition Zone in Spring . Fisheries Science , 71 : 1280 – 1286 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2018. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Olivar, M, Pilar;Bode,, Antonio;López-Pérez,, Cristina;Hulley, P, Alexander;Hernández-León,, Santiago

doi: 10.1093/icesjms/fsx243pmid: N/A

Abstract Lanternfishes (Myctophidae) constitute the most important component of the daily vertically migrating mesopelagic fish community. This research addresses the estimation of the trophic position and diet of myctophids using stable isotope analyses. Fishes were collected across the central Atlantic, from a very productive zone influenced by the Mauritanian upwelling to the western oligotrophic equatorial waters. The survey also encompassed a zone of low oxygen concentration in the mesopelagic layers. Determinations of δ13C and δ15N values were made on the 20 most frequent and abundant myctophids, from small-sized species (e.g. Notolychnus valdivae) to larger ones (e.g. Myctophum punctatum). Isotope analyses on the seston and several plankton groups were also performed to assess the influence of zonal differences in trophic position (TP) calculations, and to use as food sources in diet estimations. Myctophids displayed a narrow range of trophic positions, being greater than 2 and less than 4, except for N. valdiviae (TP = 1.7). Comparisons of diets estimated through an isotopic mixing model differentiated the smallest species, with a strong seston signature (Diogenichthys atlanticus and N. valdiviae), from the Diaphus species of medium sizes, (D. brachycephalus, D. holti, and D. rafinesquii), which feed on prey of higher TP values. Introduction Mesopelagic fishes are, both numerically and in terms of biomass, the greatest component of teleosteans of open oceans. Although the large biomasses of mesopelagic fishes reported are due to different families [lanternfishes (Myctophidae), bristlemouths (Gonostomatidae), hatchetfishes (Sternoptychidae), lightfishes (Phosycthydae), and dragonfishes (Stomiidae)], the family Myctophidae presents the highest species diversity both in oceanic and slope regions (Gjøsaeter and Kawaguchi, 1980; Hulley, 1981; Gibbs and Krueger, 1987). Lanternfishes are small fishes, with adult stages commonly ranging from 2 to 10 cm and with individual wet weights generally <4 mg. Their main behavioural trait is a capacity for performing daily vertical migration through the water column. They move to mesopelagic layers during the day in search of predator refuge (Mehner and Kasprzak, 2011; Sutton, 2013) and ascend to the near surface layers at night for feeding following the nightly ascension of zooplankton (Merrett and Roe, 1974; Gartner et al. 1987). Therefore, these fishes play a substantial role in the active carbon transport through the water column, carrying the ingested zooplankton from the upper layers to deeper zones where they respire, excrete, and even serve as prey for fishes (both demersal and pelagic) (Bosley et al., 2004; Pakhomov et al., 2006; Revill et al., 2009; Valls et al., 2011; Battaglia et al., 2013; Olafsdottir et al., 2016), mammals and sea birds (Hedd and Montevecchi, 2006; Cherel et al., 2008; Navarro et al., 2009). Mesopelagic fishes alone may contribute 8–40% of deep carbon flux (Davison et al., 2013; Hudson et al., 2014; Trueman et al., 2014), making this group of fishes an important unit in open ocean foodwebs and linking the lower trophic levels to upper predators. Trophic ecology of myctophid communities has been investigated through studies of gut content analyses (e.g. Clarke, 1980; Hopkins and Gartner, 1992, Hopkins et al., 1996; Bernal et al., 2015; McClain-Counts et al., 2017) which identified these groups of fishes as mainly zooplankton feeders. More recent studies based on bulk stable carbon (δ13C) and nitrogen (δ15N) signatures have determined the trophic levels of several species in the Tasman Sea (Flynn and Kloser, 2012), Mediterranean Sea (Fanelli et al., 2011; Valls et al., 2014), and Gulf of Mexico (McClain-Counts et al., 2017). Scattered information on isotopic values of myctophids collected in general foodweb structure studies or in diet analysis of top predators are also frequently reported (Nilsen et al., 2008; Cherel et al., 2010; Fanelli et al., 2011, 2014; Colaço et al., 2013; Quintana-Rizzo et al., 2015). Stable isotope values of carbon (δ13C) and nitrogen (δ15N) are commonly used to determine carbon (C) sources and trophic levels, based on the similar C isotopic composition of consumers and prey and the enrichment of nitrogen (N) isotopes with each trophic level (Vander Zanden and Rasmussen, 1999). The isotopic composition of a consumer’s tissues is considered to be a result of its diet integrated through time and space, as light isotopes are preferentially mobilized in metabolic reactions (Fry, 2006). Therefore, stable isotopes can provide a measure of the trophic niche dimensions (Layman et al., 2012) and have been used to estimate the contributions of various prey sources to the diet of an organism (e.g. Carassou et al., 2008; Valls et al., 2014; McClain-Counts et al., 2017). Estimations of the trophic position (TP) of a given consumer can be made from the δ15N values of the consumer tissues and a reference baseline at the base of the foodweb (Post, 2002). Recent studies showed that isotopic enrichment varies along the foodweb (Caut et al., 2009; Hussey et al., 2014), and several methods have been proposed to deal with this variability when estimating TPs and consumer diets. The classical approach is to consider average enrichment factors between prey and consumer tissues for both carbon and nitrogen isotopes (e.g. Cherel et al., 2010, Valls et al., 2014). However, meta-analysis studies showed that the isotopic enrichment, particularly in the case of N, decreased in upper trophic levels and suggested corrections based on δ15N values of the reference baseline (Hussey et al., 2014) or the isotopic composition of the diet (Caut et al., 2009). Research on mesopelagic fishes from the equatorial and tropical Atlantic has been mainly devoted to their taxonomic composition and diversity as well as their main patterns of vertical migration (Backus et al., 1970; Badcock and Merrett, 1976; Hulley, 1981; Hulley and Paxton 2016a, b; Olivar et al., 2017). Knowledge of myctophids diets from this region is more scarce (Merrett and Roe, 1974; Kinzer and Schulz, 1985), the closest and more complete investigations being those by Hopkins et al. (1996) and McClain-Counts et al. (2017) in the Gulf of Mexico. Additional dietary data on myctophids are available for more northern locations in temperate regions: Northeast Atlantic (Gjøsaeter, 1973; Pusch et al., 2004; Pepin, 2013; Hudson et al., 2014) and western Mediterranean Sea (Scotto di Carlo et al., 1982; Bernal et al., 2013, 2015; Battaglia et al., 2014, 2016). This study aims to investigate the TP of myctophids from the equatorial and tropical Atlantic using C and N stable isotope analysis (SIA). The investigation covered specimens from the oligotrophic western sector (Morel et al., 2010) to near the North African coast, which is one of the most productive regions of the world due to the Mauritanian upwelling (Mittelstaedt, 1983). Additionally, a permanent oxygen minimum zone (OMZ) at 300–600 m characterizes the eastern tropical North Atlantic (Karstensen et al., 2016). We selected the 20 most abundant species of this family ranging from small-sized species such as N. validivae to larger species such as Myctophum punctatum. Possible geographical differences arising from different environmental zones across the Atlantic and in relation to differences in the selected baseline were investigated. Fish isotopic signatures were also analysed in combination with those from several groups of plankton in order to find out how these species differ in their diet preferences. Material and methods Sample collection Samples were collected during a cruise conducted across the equatorial and tropical Atlantic from off the Brazilian coast to south of the Canary Islands. At each station, a CTD equipped with Niskin bottles was deployed to examine the oceanographic features (temperature, salinity, fluorescence, and oxygen) and collect water samples for chlorophyll concentrations and seston analyses. Zooplankton samples were collected in the epipelagic layer (0–200 m) from eight stations (Figure 1) representing the available prey for vertically migrating myctophids. Samples were collected with a MOCNESS-1 net (1 m2 of mouth area and 200-µm mesh size) during both daylight and night periods. Tows were oblique at a ship speed of 1.5–2.5 knots and winch retrieval of 0.3 m s−1. Random subsamples were stored frozen until the sorting of specimens in the laboratory for SIA. Specimens were identified to the lowest taxonomical level possible (species), but were later grouped at higher levels to provide enough samples for the isotopic determinations. Calanoid copepods from the mesozooplankton were used as the reference baseline for TP estimations as they are primarily herbivores (Bode et al., 2015). For myctophid diet estimations, the isotopic composition of copepods was averaged by herbivores or omnivores and carnivores based on the information provided in the literature (Bode et al., 2015; Benedetti et al., 2016) and other specimens grouped as crustacean macrozooplankton (Euphausiacea, Amphipoda, and Decapoda) or Chaetognatha. Additional reference material was provided by seston samples obtained by filtration of up to 4 l of seawater collected at three depths in the epipelagic layer (surface, deep chlorophyll maximum, and 200 m), with Niskin bottles attached to a rosette equipped with fluorometer. Samples were filtered onto glass fibre filters (Whatman GF/F, nominal pore size 0.7 µm) that were dried and processed as plankton samples for isotopic analysis. Isotopic values for seston at each station were weighted-average by considering the depth distribution of C and N (e.g. Fernández et al., 2014). Figure 1. View largeDownload slide Location of sampling stations across the equatorial and tropical Atlantic during April 2015. Dots indicate CTD profiles and circles the stations where plankton and myctophids samples for isotopic analysis were collected. Stations were grouped in four broad zones. Figure 1. View largeDownload slide Location of sampling stations across the equatorial and tropical Atlantic during April 2015. Dots indicate CTD profiles and circles the stations where plankton and myctophids samples for isotopic analysis were collected. Stations were grouped in four broad zones. Fish samples for SIA were obtained from these stations using a 35-m2 pelagic net, the “Mesopelagos” (Olivar et al., 2017). The 20 most abundant myctophid species occurring in the epipelagic layers at night were selected for the present investigation. Fishes were sorted, identified on board, and kept frozen (–20 °C) until later SIA. In the laboratory, the standard length (SL) of each fish was measured before freeze drying. Stable isotope analyses Stable isotope determinations were made on oven dried plankton (50 °C, 24 h) and freeze-dried fish (18 h). In the case of fish samples, portions of the dorsal musculature were selected, except for very small specimens that were analysed whole, the gut and gonads having been previously eliminated. All samples were ground to a fine and homogeneous powder with mortar and pestle, and subsamples were weighed (± 2 µg) and packed into tin capsules which were fed into an elemental analyser (Carlo Erba CHNSO 1108) coupled to an isotope-ratio mass spectrometer (Finnigan Matt Delta Plus). Values of natural abundance of stable isotopes were reported as δ13C or δ15N (%) relative to Vienna Pee Dee Belemnite or atmospheric nitrogen, respectively (Coplen, 2011). International Atomic Energy Agency USGS40 and L-alanine isotope standards were analysed with the samples along with internal acetanilide and sample standards (cyanobacteria culture of known isotope composition used as an internal control). Precision (±SE) of replicate determinations of standards and samples for both isotopes was <0.1 and <0.3% (n = 3), respectively. The analytical offset between certified and measured values was <0.1%. All isotopic determinations were made in the Servicio de Análisis Instrumental of the Universidade da Coruña (Spain). Samples were not treated to remove lipids or carbonates before analysis to prevent loss of material. For the estimation of diets, all δ13C ratios were normalized for lipid content using the equation for aquatic animals provided by Post et al. (2007): Δδ13C=–3.32 + 0.99 · C:N (1) where Δδ13C is the change in δ13C caused by lipids and C: N is the carbon-to-nitrogen ratio (by mass) of the sample. Seston values were not normalized on account that it is mostly composed of phytoplankton and protozoa and that it has been reported that there is not a relationship between C: N and Δδ13C in aquatic plants (Post et al., 2007). Estimation of TP Estimations of the TP of myctophids were obtained using three alternative models to compare the effect of using different trophic enrichment assumptions. First, the classical additive model (TPA) was employed (Post, 2002): TPA=[(δ15Nmyct–δ15Nbase)/TEF]+2 (2) where δ15Nmyc and δ15Nbase are the N isotopic signatures of the myctophids and the reference baseline (calanoid copepods), TEF (trophic enrichment factor) is the average enrichment between trophic levels (3.4%, e.g. Stowasser et al., 2012), and 2 is the trophic level of the reference baseline. Second, an additive model was used, but TEF was variable (TEFv) and estimated from the δ15N value of the diet using the equation derived from the meta-analysis in Caut et al. (2009) for fish muscle, assuming that copepods were part of the diet of myctophids: TEFv=–0.281 δ15Nbase+5.879 (3) TPC=[(δ15Nmyct–δ15Nbase)/TEFv]+2 (4) Finally, estimates were calculated using a decreasing isotopic enrichment with the increase in trophic level, using the scaled model (TPS) of Hussey et al. (2014): TPS =2 +[log (δ15Nlim–δ15Nbase)– log (δ15Nlim–δ15Nmyct)]/k (5) where δ15Nlim is the saturating isotope limit as TP increases and k the rate at which δ15Nmyct approaches δ15Nlim. Values for δ15Nlim = 21.926 and k = 0.137 and were provided by the meta-analysis of Hussey et al. (2014). As for the other estimations, it is assumed that the reference zooplankton is a primary consumer with a TP of 2. Significant differences among species, stations, and sampling zones were tested with ANOVA and the a posteriori Dunnett-C test. Diet estimation from stable isotope mixing model The contribution of several potential preys to the diet of the different species was estimated by applying the Mix SIAR Bayesian mixing model (SIAR v4.1.3, Stable Isotope Analysis in R) of Stock and Semmens (2013). Potential prey sources included seston and the plankton groups detailed earlier (herbivorous and omnivorous–carnivorous copepods, crustacean macrozooplankton, and chaetognaths) from samples taken at the same stations where each of the studied myctophid species were caught. Because fishes have also been reported as prey in several diet studies (Clarke, 1980; Hopkins and Gartner, 1992; Watanabe and Kawaguchi, 2003; Bernal et al., 2015), two trials including small mesopelagic fishes as potential prey were also performed. First, the smallest myctophid (Notolychnus valdiviae) (maximum attainable size 25 mm) was used. A second analysis was performed including data of Cyclothone alba, a small fish of the family Gonostomatidae, selected as reference on the possible contribution of the genus Cyclothone [the most abundant mesopelagic fishes (Nelson, 2006)] in the diet of myctophids. Because no TEFs were available for myctophids, we used those calculated by Sweeting et al. (2007a, b) for fish muscle (mean 0.97, s.d. 1.08 for δ13C and mean 3.15, SD 1.28 for δ15N) and previously applied in mesopelagic fishes isotopic studies of mesopelagic fishes (Valls et al., 2014) as well as in other marine fishes (Sweeting et al., 2007a, b). Results Environmental features A general overview of the main oceanographic features during the survey has been presented in Olivar et al. (2017). In summary, stratification and relatively high sea surface temperatures characterized most of the equatorial and tropical Atlantic, except for the three stations closer to the African coast. East North Atlantic Central Waters were present in the epipelagic layers of the three stations closer to the African coast (10–12), while South Atlantic Central Waters occurred from near Brazil to Station 6 (Figure 1). In terms of production, the region also showed an increasing gradient from the oligotrophic waters of the western sector (Morel et al., 2010) to the eastern sector, closer to the African coast, which received enrichment due to the effect of the Northwest African upwelling. In accordance with this, the surface concentration of chlorophyll a (Chl a) showed an increase northward across the region, with the lowest values in the western equatorial stations, where the presence of the deep chlorophyll maximum (DCM) was detected from 30 to 70 m (Stations 4–9). Conspicuously high Chl a values from surface to 50 m were observed off Cape Blanc (Station 11) (Figure 2). The last sampled station (12), despite its proximity to the upwelling region, presented a lower surface Chl a, but a wide and relatively deep DCM (Figure 2). Another important oceanographic feature was the OMZ between 200 and 600 m in the vicinity of the Cape Verde Islands stations (7–10) (Olivar et al., 2017). Taking into account these features and the results of a PCA analysis of oceanographic parameters (Olivar et al., 2017), comparisons were performed grouping the data into four zones (Zone 0: Station 12; Zone 1: Station 11; Zone 2: Stations 7–10; and Zone 3: Stations 4–6). Figure 2. View largeDownload slide Vertical profiles of Chl a concentration in the first 200 m of the water column in Stations 4–6 (equatorial), 7–9 (Cape Verde), and 10–12 (near the African coast). Figure 2. View largeDownload slide Vertical profiles of Chl a concentration in the first 200 m of the water column in Stations 4–6 (equatorial), 7–9 (Cape Verde), and 10–12 (near the African coast). TP of myctophid species There were almost no significant differences in the stable isotope composition and C: N values of the seston and plankton groups among zones, but copepods and crustacean macrozooplankton showed high δ15N values in zone 2 (Table 1). Such variability suggests changes in the isotopic baseline that must be taken into account when estimating TPs. Table 1. Mean and SD values of the isotopic composition of seston and plankton prey groups in the zones marked in Figure 1. Zone Group δ15N δ13C C: N Mean SD C-test Mean SD C-test Mean SD C-test n 0 Seston 6.51 –22.03 7.77 1 Copepoda H 4.90 0.47 a –21.95 0.13 a 5.76 0.66 a 2 Copepoda OC 6.20 0.40 a –22.12 0.84 a 6.81 1.07 a 8 Macrozooplankton 6.83 0.84 b –21.43 1.78 a 5.39 1.21 a 2 Chaetognaths 8.56 0.65 a –20.84 1.14 a 5.53 0.71 a 2 1 Seston 4.89 –20.56 7.62 1 Copepoda H 4.35 0.95 a –20.82 0.69 a 10.38 5.04 b 5 Copepoda OC 4.94 0.79 a –21.47 0.79 a 9.77 5.04 b 9 Macrozooplankton 4.72 1.14 a –20.57 0.87 a 7.29 2.36 a 8 Chaetognaths 6.55 0.55 a –19.94 0.06 a 5.99 0.02 a 2 2 Seston 6.64 0.81 a –23.23 0.71 a 8.43 0.61 a 4 Copepoda H 7.44 1.07 b –21.92 0.83 a 5.97 1.16 a 29 Copepoda OC 7.45 1.38 b –22.10 0.74 a 6.34 1.57 a 45 Macrozooplankton 7.48 1.04 b –21.53 0.62 a 5.64 0.61 a 27 Chaetognaths 8.80 1.11 a –21.12 0.50 a 5.74 0.59 a 18 3 Seston 4.93 0.93 a –23.73 0.01 a 8.10 0.41 a 2 Copepoda H 5.47 1.30 a –21.77 0.88 a 5.19 0.34 a 9 Copepoda OC 5.97 2.13 a –22.14 0.93 a 6.91 2.31 a 24 Macrozooplankton 6.05 1.28 a, b –21.64 1.28 a 5.78 1.68 a 16 Chaetognaths 7.46 1.48 a –21.49 1.03 a 5.69 1.21 a 7 Zone Group δ15N δ13C C: N Mean SD C-test Mean SD C-test Mean SD C-test n 0 Seston 6.51 –22.03 7.77 1 Copepoda H 4.90 0.47 a –21.95 0.13 a 5.76 0.66 a 2 Copepoda OC 6.20 0.40 a –22.12 0.84 a 6.81 1.07 a 8 Macrozooplankton 6.83 0.84 b –21.43 1.78 a 5.39 1.21 a 2 Chaetognaths 8.56 0.65 a –20.84 1.14 a 5.53 0.71 a 2 1 Seston 4.89 –20.56 7.62 1 Copepoda H 4.35 0.95 a –20.82 0.69 a 10.38 5.04 b 5 Copepoda OC 4.94 0.79 a –21.47 0.79 a 9.77 5.04 b 9 Macrozooplankton 4.72 1.14 a –20.57 0.87 a 7.29 2.36 a 8 Chaetognaths 6.55 0.55 a –19.94 0.06 a 5.99 0.02 a 2 2 Seston 6.64 0.81 a –23.23 0.71 a 8.43 0.61 a 4 Copepoda H 7.44 1.07 b –21.92 0.83 a 5.97 1.16 a 29 Copepoda OC 7.45 1.38 b –22.10 0.74 a 6.34 1.57 a 45 Macrozooplankton 7.48 1.04 b –21.53 0.62 a 5.64 0.61 a 27 Chaetognaths 8.80 1.11 a –21.12 0.50 a 5.74 0.59 a 18 3 Seston 4.93 0.93 a –23.73 0.01 a 8.10 0.41 a 2 Copepoda H 5.47 1.30 a –21.77 0.88 a 5.19 0.34 a 9 Copepoda OC 5.97 2.13 a –22.14 0.93 a 6.91 2.31 a 24 Macrozooplankton 6.05 1.28 a, b –21.64 1.28 a 5.78 1.68 a 16 Chaetognaths 7.46 1.48 a –21.49 1.03 a 5.69 1.21 a 7 Significant means are indicated by different letters (ANOVA and Dunett C tests, p < 0.05). n: number of samples, Copepoda H: herbivorous copepods, Copepoda OC: omnivorous and carnivorous copepods, macrozooplankton: crustacean macrozooplankton (Decapoda, Euphausiacea and Amphipoda). δ13C and C: N values were not corrected for lipids. Table 1. Mean and SD values of the isotopic composition of seston and plankton prey groups in the zones marked in Figure 1. Zone Group δ15N δ13C C: N Mean SD C-test Mean SD C-test Mean SD C-test n 0 Seston 6.51 –22.03 7.77 1 Copepoda H 4.90 0.47 a –21.95 0.13 a 5.76 0.66 a 2 Copepoda OC 6.20 0.40 a –22.12 0.84 a 6.81 1.07 a 8 Macrozooplankton 6.83 0.84 b –21.43 1.78 a 5.39 1.21 a 2 Chaetognaths 8.56 0.65 a –20.84 1.14 a 5.53 0.71 a 2 1 Seston 4.89 –20.56 7.62 1 Copepoda H 4.35 0.95 a –20.82 0.69 a 10.38 5.04 b 5 Copepoda OC 4.94 0.79 a –21.47 0.79 a 9.77 5.04 b 9 Macrozooplankton 4.72 1.14 a –20.57 0.87 a 7.29 2.36 a 8 Chaetognaths 6.55 0.55 a –19.94 0.06 a 5.99 0.02 a 2 2 Seston 6.64 0.81 a –23.23 0.71 a 8.43 0.61 a 4 Copepoda H 7.44 1.07 b –21.92 0.83 a 5.97 1.16 a 29 Copepoda OC 7.45 1.38 b –22.10 0.74 a 6.34 1.57 a 45 Macrozooplankton 7.48 1.04 b –21.53 0.62 a 5.64 0.61 a 27 Chaetognaths 8.80 1.11 a –21.12 0.50 a 5.74 0.59 a 18 3 Seston 4.93 0.93 a –23.73 0.01 a 8.10 0.41 a 2 Copepoda H 5.47 1.30 a –21.77 0.88 a 5.19 0.34 a 9 Copepoda OC 5.97 2.13 a –22.14 0.93 a 6.91 2.31 a 24 Macrozooplankton 6.05 1.28 a, b –21.64 1.28 a 5.78 1.68 a 16 Chaetognaths 7.46 1.48 a –21.49 1.03 a 5.69 1.21 a 7 Zone Group δ15N δ13C C: N Mean SD C-test Mean SD C-test Mean SD C-test n 0 Seston 6.51 –22.03 7.77 1 Copepoda H 4.90 0.47 a –21.95 0.13 a 5.76 0.66 a 2 Copepoda OC 6.20 0.40 a –22.12 0.84 a 6.81 1.07 a 8 Macrozooplankton 6.83 0.84 b –21.43 1.78 a 5.39 1.21 a 2 Chaetognaths 8.56 0.65 a –20.84 1.14 a 5.53 0.71 a 2 1 Seston 4.89 –20.56 7.62 1 Copepoda H 4.35 0.95 a –20.82 0.69 a 10.38 5.04 b 5 Copepoda OC 4.94 0.79 a –21.47 0.79 a 9.77 5.04 b 9 Macrozooplankton 4.72 1.14 a –20.57 0.87 a 7.29 2.36 a 8 Chaetognaths 6.55 0.55 a –19.94 0.06 a 5.99 0.02 a 2 2 Seston 6.64 0.81 a –23.23 0.71 a 8.43 0.61 a 4 Copepoda H 7.44 1.07 b –21.92 0.83 a 5.97 1.16 a 29 Copepoda OC 7.45 1.38 b –22.10 0.74 a 6.34 1.57 a 45 Macrozooplankton 7.48 1.04 b –21.53 0.62 a 5.64 0.61 a 27 Chaetognaths 8.80 1.11 a –21.12 0.50 a 5.74 0.59 a 18 3 Seston 4.93 0.93 a –23.73 0.01 a 8.10 0.41 a 2 Copepoda H 5.47 1.30 a –21.77 0.88 a 5.19 0.34 a 9 Copepoda OC 5.97 2.13 a –22.14 0.93 a 6.91 2.31 a 24 Macrozooplankton 6.05 1.28 a, b –21.64 1.28 a 5.78 1.68 a 16 Chaetognaths 7.46 1.48 a –21.49 1.03 a 5.69 1.21 a 7 Significant means are indicated by different letters (ANOVA and Dunett C tests, p < 0.05). n: number of samples, Copepoda H: herbivorous copepods, Copepoda OC: omnivorous and carnivorous copepods, macrozooplankton: crustacean macrozooplankton (Decapoda, Euphausiacea and Amphipoda). δ13C and C: N values were not corrected for lipids. Values for myctophids (Table 2) also varied between species, stations, and sampling zones (ANOVA, p < 0.05, Supplementary Table S1). The species with the lowest average δ15N values were Hygophum macrochir, Diogenichthys atlanticus, Ceratoscopelus warmingii, and N. valdiviae, while the maximum values were recorded for three species of Diaphus (D. mollis, D. rafinesquii, and D. dumerilii). However, there was a large overlap between mean values for the different species (Figure 3). Table 2. Mean and SD values of δ15N, δ13C, C: N molar ratio, and TP estimates for myctophid species in each station. Species SL (mm) n Station δ15N δ13C C: N TP estimates Mean SD Mean SD Mean SD Mean SD TPA TPC TPS Error B. glaciale (Reinhardt, 1837) 34.3 1.2 3 11 8.95 0.27 –20.50 0.29 7.82 1.01 3.30 2.92 2.88 0.35 B. suborbitale (Gilbert, 1913) 28.7 4.0 3 7 8.36 0.54 –20.26 0.12 4.46 0.09 2.76 2.57 2.52 0.30 24.0 1.0 3 12 8.41 0.07 –20.34 0.54 4.86 0.37 2.84 2.63 2.58 0.23 C. warmingii (Lütken, 1892) 58.7 5.0 3 6 7.43 0.45 –20.94 0.49 5.47 0.33 3.11 2.74 2.70 0.54 Diaphus brachycephalus (Tåning, 1928) 48.7 0.6 3 6 9.77 0.46 –19.49 0.19 4.46 0.17 3.85 3.23 3.26 0.54 33.0 1.7 3 11 8.16 0.15 –19.02 0.13 4.32 0.11 3.05 2.74 2.69 0.31 Diaphus dumerilii (Bleeker, 1856) 53.0 4.0 3 4 9.50 1.65 –20.29 0.82 4.94 0.59 2.61 2.52 2.46 0.95 Diaphus holti (Tåning, 1918) 44.3 5.5 3 11 9.13 0.74 –18.25 0.65 4.38 0.33 3.35 2.95 2.93 0.50 D. mollis (Tåning, 1928) 53.3 1.2 3 9 9.99 1.08 –19.72 0.71 4.24 0.33 2.46 2.42 2.37 0.65 D. rafinesquii (Cocco, 1838) 57.3 0.6 2 11 9.35 0.60 –19.12 0.20 4.67 0.02 3.42 3.00 2.98 0.45 58.0 1 12 10.14 –19.33 5.23 3.39 3.04 3.01 0.21 D. vanhoeffeni (Brauer, 1906) 35.3 0.6 3 7 8.81 0.42 –20.54 0.42 4.73 0.29 2.90 2.68 2.63 0.27 D. atlanticus (Tåning, 1928) 19.0 1.3 3 7 7.05 0.34 –20.74 0.57 4.24 0.28 2.35 2.26 2.23 0.24 H. macrochir (Günther, 1864) 41.0 1.0 3 11 6.53 0.12 –18.63 0.50 4.49 0.12 2.54 2.38 2.34 0.30 L. alatus (Goode and Bean, 1896) 48.3 0.6 3 7 8.44 0.28 –19.14 0.19 4.00 0.01 2.78 2.59 2.54 0.22 49.7 1.5 3 11 8.54 0.38 –20.13 0.17 5.88 0.43 3.17 2.82 2.78 0.38 L. nobilis (Tåning, 1928) 47.7 3.1 3 4 9.31 0.32 –19.12 0.16 3.99 0.03 2.55 2.47 2.41 0.54 L. pusillus (Johnson, 1890) 34.5 0.7 2 11 8.90 0.94 –20.07 0.45 5.64 0.38 3.28 2.90 2.87 0.56 34.0 1 12 8.99 –20.85 5.80 3.03 2.77 2.72 0.21 L. guentheri (Goode and Bean, 1896) 52.0 8.2 3 4 9.47 0.14 –19.09 0.00 4.05 0.02 2.60 2.51 2.45 0.48 L. dofleini (Zugmayer, 1911) 28.5 2.1 2 11 8.91 0.70 –20.29 0.23 5.40 0.48 3.28 2.91 2.87 0.48 28.0 1 12 9.94 –20.15 4.80 3.32 2.99 2.96 0.21 M. nitidulum (Garman, 1899) 73.0 1.4 2 7 8.14 0.10 –18.53 0.09 4.01 0.19 2.69 2.52 2.47 0.17 69.0 1 8 8.27 –18.97 3.95 2.32 2.27 2.23 0.15 59.5 4.9 2 10 9.37 0.01 –19.38 0.65 4.08 0.17 2.61 2.51 2.46 0.21 56.0 1 12 8.53 –19.31 4.11 2.88 2.66 2.61 0.21 M. punctatum (Rafinesque, 1810) 66.0 2.6 3 11 8.68 0.13 –18.09 1.65 4.52 1.03 3.21 2.85 2.81 0.31 N. valdiviae (Brauer, 1904) 20.6 1.0 7 9 7.46 0.57 –22.51 0.24 7.52 0.67 1.67 1.70 1.76 0.49 N. resplendens (Richardson, 1845) 35.5 0.7 2 7 8.23 0.14 –19.99 0.03 4.15 0.06 2.72 2.54 2.49 0.18 36.0 1 8 10.55 –19.72 4.26 3.03 2.86 2.81 0.15 41.0 1.7 3 12 8.69 0.19 –19.85 0.91 4.15 0.15 2.93 2.70 2.65 0.27 C. alba (Brauer, 1906) 25.7 0.6 3 8 9.16 0.32 –19.98 0.45 4.17 0.10 2.27 2.25 2.21 0.43 Species SL (mm) n Station δ15N δ13C C: N TP estimates Mean SD Mean SD Mean SD Mean SD TPA TPC TPS Error B. glaciale (Reinhardt, 1837) 34.3 1.2 3 11 8.95 0.27 –20.50 0.29 7.82 1.01 3.30 2.92 2.88 0.35 B. suborbitale (Gilbert, 1913) 28.7 4.0 3 7 8.36 0.54 –20.26 0.12 4.46 0.09 2.76 2.57 2.52 0.30 24.0 1.0 3 12 8.41 0.07 –20.34 0.54 4.86 0.37 2.84 2.63 2.58 0.23 C. warmingii (Lütken, 1892) 58.7 5.0 3 6 7.43 0.45 –20.94 0.49 5.47 0.33 3.11 2.74 2.70 0.54 Diaphus brachycephalus (Tåning, 1928) 48.7 0.6 3 6 9.77 0.46 –19.49 0.19 4.46 0.17 3.85 3.23 3.26 0.54 33.0 1.7 3 11 8.16 0.15 –19.02 0.13 4.32 0.11 3.05 2.74 2.69 0.31 Diaphus dumerilii (Bleeker, 1856) 53.0 4.0 3 4 9.50 1.65 –20.29 0.82 4.94 0.59 2.61 2.52 2.46 0.95 Diaphus holti (Tåning, 1918) 44.3 5.5 3 11 9.13 0.74 –18.25 0.65 4.38 0.33 3.35 2.95 2.93 0.50 D. mollis (Tåning, 1928) 53.3 1.2 3 9 9.99 1.08 –19.72 0.71 4.24 0.33 2.46 2.42 2.37 0.65 D. rafinesquii (Cocco, 1838) 57.3 0.6 2 11 9.35 0.60 –19.12 0.20 4.67 0.02 3.42 3.00 2.98 0.45 58.0 1 12 10.14 –19.33 5.23 3.39 3.04 3.01 0.21 D. vanhoeffeni (Brauer, 1906) 35.3 0.6 3 7 8.81 0.42 –20.54 0.42 4.73 0.29 2.90 2.68 2.63 0.27 D. atlanticus (Tåning, 1928) 19.0 1.3 3 7 7.05 0.34 –20.74 0.57 4.24 0.28 2.35 2.26 2.23 0.24 H. macrochir (Günther, 1864) 41.0 1.0 3 11 6.53 0.12 –18.63 0.50 4.49 0.12 2.54 2.38 2.34 0.30 L. alatus (Goode and Bean, 1896) 48.3 0.6 3 7 8.44 0.28 –19.14 0.19 4.00 0.01 2.78 2.59 2.54 0.22 49.7 1.5 3 11 8.54 0.38 –20.13 0.17 5.88 0.43 3.17 2.82 2.78 0.38 L. nobilis (Tåning, 1928) 47.7 3.1 3 4 9.31 0.32 –19.12 0.16 3.99 0.03 2.55 2.47 2.41 0.54 L. pusillus (Johnson, 1890) 34.5 0.7 2 11 8.90 0.94 –20.07 0.45 5.64 0.38 3.28 2.90 2.87 0.56 34.0 1 12 8.99 –20.85 5.80 3.03 2.77 2.72 0.21 L. guentheri (Goode and Bean, 1896) 52.0 8.2 3 4 9.47 0.14 –19.09 0.00 4.05 0.02 2.60 2.51 2.45 0.48 L. dofleini (Zugmayer, 1911) 28.5 2.1 2 11 8.91 0.70 –20.29 0.23 5.40 0.48 3.28 2.91 2.87 0.48 28.0 1 12 9.94 –20.15 4.80 3.32 2.99 2.96 0.21 M. nitidulum (Garman, 1899) 73.0 1.4 2 7 8.14 0.10 –18.53 0.09 4.01 0.19 2.69 2.52 2.47 0.17 69.0 1 8 8.27 –18.97 3.95 2.32 2.27 2.23 0.15 59.5 4.9 2 10 9.37 0.01 –19.38 0.65 4.08 0.17 2.61 2.51 2.46 0.21 56.0 1 12 8.53 –19.31 4.11 2.88 2.66 2.61 0.21 M. punctatum (Rafinesque, 1810) 66.0 2.6 3 11 8.68 0.13 –18.09 1.65 4.52 1.03 3.21 2.85 2.81 0.31 N. valdiviae (Brauer, 1904) 20.6 1.0 7 9 7.46 0.57 –22.51 0.24 7.52 0.67 1.67 1.70 1.76 0.49 N. resplendens (Richardson, 1845) 35.5 0.7 2 7 8.23 0.14 –19.99 0.03 4.15 0.06 2.72 2.54 2.49 0.18 36.0 1 8 10.55 –19.72 4.26 3.03 2.86 2.81 0.15 41.0 1.7 3 12 8.69 0.19 –19.85 0.91 4.15 0.15 2.93 2.70 2.65 0.27 C. alba (Brauer, 1906) 25.7 0.6 3 8 9.16 0.32 –19.98 0.45 4.17 0.10 2.27 2.25 2.21 0.43 The gonostomatid fish C. alba is also added for comparison. The error of the estimates was computed by additive propagation of errors in δ15N of myctophids and the herbivorous copepods used as the reference baseline. n: number of samples. δ13C and C:N values were not corrected for lipids. Table 2. Mean and SD values of δ15N, δ13C, C: N molar ratio, and TP estimates for myctophid species in each station. Species SL (mm) n Station δ15N δ13C C: N TP estimates Mean SD Mean SD Mean SD Mean SD TPA TPC TPS Error B. glaciale (Reinhardt, 1837) 34.3 1.2 3 11 8.95 0.27 –20.50 0.29 7.82 1.01 3.30 2.92 2.88 0.35 B. suborbitale (Gilbert, 1913) 28.7 4.0 3 7 8.36 0.54 –20.26 0.12 4.46 0.09 2.76 2.57 2.52 0.30 24.0 1.0 3 12 8.41 0.07 –20.34 0.54 4.86 0.37 2.84 2.63 2.58 0.23 C. warmingii (Lütken, 1892) 58.7 5.0 3 6 7.43 0.45 –20.94 0.49 5.47 0.33 3.11 2.74 2.70 0.54 Diaphus brachycephalus (Tåning, 1928) 48.7 0.6 3 6 9.77 0.46 –19.49 0.19 4.46 0.17 3.85 3.23 3.26 0.54 33.0 1.7 3 11 8.16 0.15 –19.02 0.13 4.32 0.11 3.05 2.74 2.69 0.31 Diaphus dumerilii (Bleeker, 1856) 53.0 4.0 3 4 9.50 1.65 –20.29 0.82 4.94 0.59 2.61 2.52 2.46 0.95 Diaphus holti (Tåning, 1918) 44.3 5.5 3 11 9.13 0.74 –18.25 0.65 4.38 0.33 3.35 2.95 2.93 0.50 D. mollis (Tåning, 1928) 53.3 1.2 3 9 9.99 1.08 –19.72 0.71 4.24 0.33 2.46 2.42 2.37 0.65 D. rafinesquii (Cocco, 1838) 57.3 0.6 2 11 9.35 0.60 –19.12 0.20 4.67 0.02 3.42 3.00 2.98 0.45 58.0 1 12 10.14 –19.33 5.23 3.39 3.04 3.01 0.21 D. vanhoeffeni (Brauer, 1906) 35.3 0.6 3 7 8.81 0.42 –20.54 0.42 4.73 0.29 2.90 2.68 2.63 0.27 D. atlanticus (Tåning, 1928) 19.0 1.3 3 7 7.05 0.34 –20.74 0.57 4.24 0.28 2.35 2.26 2.23 0.24 H. macrochir (Günther, 1864) 41.0 1.0 3 11 6.53 0.12 –18.63 0.50 4.49 0.12 2.54 2.38 2.34 0.30 L. alatus (Goode and Bean, 1896) 48.3 0.6 3 7 8.44 0.28 –19.14 0.19 4.00 0.01 2.78 2.59 2.54 0.22 49.7 1.5 3 11 8.54 0.38 –20.13 0.17 5.88 0.43 3.17 2.82 2.78 0.38 L. nobilis (Tåning, 1928) 47.7 3.1 3 4 9.31 0.32 –19.12 0.16 3.99 0.03 2.55 2.47 2.41 0.54 L. pusillus (Johnson, 1890) 34.5 0.7 2 11 8.90 0.94 –20.07 0.45 5.64 0.38 3.28 2.90 2.87 0.56 34.0 1 12 8.99 –20.85 5.80 3.03 2.77 2.72 0.21 L. guentheri (Goode and Bean, 1896) 52.0 8.2 3 4 9.47 0.14 –19.09 0.00 4.05 0.02 2.60 2.51 2.45 0.48 L. dofleini (Zugmayer, 1911) 28.5 2.1 2 11 8.91 0.70 –20.29 0.23 5.40 0.48 3.28 2.91 2.87 0.48 28.0 1 12 9.94 –20.15 4.80 3.32 2.99 2.96 0.21 M. nitidulum (Garman, 1899) 73.0 1.4 2 7 8.14 0.10 –18.53 0.09 4.01 0.19 2.69 2.52 2.47 0.17 69.0 1 8 8.27 –18.97 3.95 2.32 2.27 2.23 0.15 59.5 4.9 2 10 9.37 0.01 –19.38 0.65 4.08 0.17 2.61 2.51 2.46 0.21 56.0 1 12 8.53 –19.31 4.11 2.88 2.66 2.61 0.21 M. punctatum (Rafinesque, 1810) 66.0 2.6 3 11 8.68 0.13 –18.09 1.65 4.52 1.03 3.21 2.85 2.81 0.31 N. valdiviae (Brauer, 1904) 20.6 1.0 7 9 7.46 0.57 –22.51 0.24 7.52 0.67 1.67 1.70 1.76 0.49 N. resplendens (Richardson, 1845) 35.5 0.7 2 7 8.23 0.14 –19.99 0.03 4.15 0.06 2.72 2.54 2.49 0.18 36.0 1 8 10.55 –19.72 4.26 3.03 2.86 2.81 0.15 41.0 1.7 3 12 8.69 0.19 –19.85 0.91 4.15 0.15 2.93 2.70 2.65 0.27 C. alba (Brauer, 1906) 25.7 0.6 3 8 9.16 0.32 –19.98 0.45 4.17 0.10 2.27 2.25 2.21 0.43 Species SL (mm) n Station δ15N δ13C C: N TP estimates Mean SD Mean SD Mean SD Mean SD TPA TPC TPS Error B. glaciale (Reinhardt, 1837) 34.3 1.2 3 11 8.95 0.27 –20.50 0.29 7.82 1.01 3.30 2.92 2.88 0.35 B. suborbitale (Gilbert, 1913) 28.7 4.0 3 7 8.36 0.54 –20.26 0.12 4.46 0.09 2.76 2.57 2.52 0.30 24.0 1.0 3 12 8.41 0.07 –20.34 0.54 4.86 0.37 2.84 2.63 2.58 0.23 C. warmingii (Lütken, 1892) 58.7 5.0 3 6 7.43 0.45 –20.94 0.49 5.47 0.33 3.11 2.74 2.70 0.54 Diaphus brachycephalus (Tåning, 1928) 48.7 0.6 3 6 9.77 0.46 –19.49 0.19 4.46 0.17 3.85 3.23 3.26 0.54 33.0 1.7 3 11 8.16 0.15 –19.02 0.13 4.32 0.11 3.05 2.74 2.69 0.31 Diaphus dumerilii (Bleeker, 1856) 53.0 4.0 3 4 9.50 1.65 –20.29 0.82 4.94 0.59 2.61 2.52 2.46 0.95 Diaphus holti (Tåning, 1918) 44.3 5.5 3 11 9.13 0.74 –18.25 0.65 4.38 0.33 3.35 2.95 2.93 0.50 D. mollis (Tåning, 1928) 53.3 1.2 3 9 9.99 1.08 –19.72 0.71 4.24 0.33 2.46 2.42 2.37 0.65 D. rafinesquii (Cocco, 1838) 57.3 0.6 2 11 9.35 0.60 –19.12 0.20 4.67 0.02 3.42 3.00 2.98 0.45 58.0 1 12 10.14 –19.33 5.23 3.39 3.04 3.01 0.21 D. vanhoeffeni (Brauer, 1906) 35.3 0.6 3 7 8.81 0.42 –20.54 0.42 4.73 0.29 2.90 2.68 2.63 0.27 D. atlanticus (Tåning, 1928) 19.0 1.3 3 7 7.05 0.34 –20.74 0.57 4.24 0.28 2.35 2.26 2.23 0.24 H. macrochir (Günther, 1864) 41.0 1.0 3 11 6.53 0.12 –18.63 0.50 4.49 0.12 2.54 2.38 2.34 0.30 L. alatus (Goode and Bean, 1896) 48.3 0.6 3 7 8.44 0.28 –19.14 0.19 4.00 0.01 2.78 2.59 2.54 0.22 49.7 1.5 3 11 8.54 0.38 –20.13 0.17 5.88 0.43 3.17 2.82 2.78 0.38 L. nobilis (Tåning, 1928) 47.7 3.1 3 4 9.31 0.32 –19.12 0.16 3.99 0.03 2.55 2.47 2.41 0.54 L. pusillus (Johnson, 1890) 34.5 0.7 2 11 8.90 0.94 –20.07 0.45 5.64 0.38 3.28 2.90 2.87 0.56 34.0 1 12 8.99 –20.85 5.80 3.03 2.77 2.72 0.21 L. guentheri (Goode and Bean, 1896) 52.0 8.2 3 4 9.47 0.14 –19.09 0.00 4.05 0.02 2.60 2.51 2.45 0.48 L. dofleini (Zugmayer, 1911) 28.5 2.1 2 11 8.91 0.70 –20.29 0.23 5.40 0.48 3.28 2.91 2.87 0.48 28.0 1 12 9.94 –20.15 4.80 3.32 2.99 2.96 0.21 M. nitidulum (Garman, 1899) 73.0 1.4 2 7 8.14 0.10 –18.53 0.09 4.01 0.19 2.69 2.52 2.47 0.17 69.0 1 8 8.27 –18.97 3.95 2.32 2.27 2.23 0.15 59.5 4.9 2 10 9.37 0.01 –19.38 0.65 4.08 0.17 2.61 2.51 2.46 0.21 56.0 1 12 8.53 –19.31 4.11 2.88 2.66 2.61 0.21 M. punctatum (Rafinesque, 1810) 66.0 2.6 3 11 8.68 0.13 –18.09 1.65 4.52 1.03 3.21 2.85 2.81 0.31 N. valdiviae (Brauer, 1904) 20.6 1.0 7 9 7.46 0.57 –22.51 0.24 7.52 0.67 1.67 1.70 1.76 0.49 N. resplendens (Richardson, 1845) 35.5 0.7 2 7 8.23 0.14 –19.99 0.03 4.15 0.06 2.72 2.54 2.49 0.18 36.0 1 8 10.55 –19.72 4.26 3.03 2.86 2.81 0.15 41.0 1.7 3 12 8.69 0.19 –19.85 0.91 4.15 0.15 2.93 2.70 2.65 0.27 C. alba (Brauer, 1906) 25.7 0.6 3 8 9.16 0.32 –19.98 0.45 4.17 0.10 2.27 2.25 2.21 0.43 The gonostomatid fish C. alba is also added for comparison. The error of the estimates was computed by additive propagation of errors in δ15N of myctophids and the herbivorous copepods used as the reference baseline. n: number of samples. δ13C and C:N values were not corrected for lipids. Figure 3. View largeDownload slide Mean ± SE δ15N for myctophids sorted by descending values. Different letters indicate significantly different means (ANOVA and Dunnett-C a posteriori test, p < 0.05). Figure 3. View largeDownload slide Mean ± SE δ15N for myctophids sorted by descending values. Different letters indicate significantly different means (ANOVA and Dunnett-C a posteriori test, p < 0.05). TPs estimated by the three methods were significantly correlated (Pearson r, p < 0.001), but TPS produced lower values than the other methods, equivalent to ca. 1.4 or 1 TPs less than TPA or TPC, respectively (Figure 4). However, all estimates indicated that the myctophids displayed a narrow range of 1.5–2 TPs (± 0.34 SE). TPs of individual species were between 2 and 4, except for the small-sized fish N. valdiviae, which presented the lowest TP, 1.7 (Table 2). Figure 4. View largeDownload slide Comparison of TP estimates of myctophid species using three different models. TPA, additive model with constant isotopic enrichment; TPC, additive model with variable isotopic enrichment; TPS, scaled enrichment model (see “Material and methods” section for details of the models). Model II regression lines for TPA (continuous line) or TPC (dashed line) are displayed. r: correlation coefficient. Figure 4. View largeDownload slide Comparison of TP estimates of myctophid species using three different models. TPA, additive model with constant isotopic enrichment; TPC, additive model with variable isotopic enrichment; TPS, scaled enrichment model (see “Material and methods” section for details of the models). Model II regression lines for TPA (continuous line) or TPC (dashed line) are displayed. r: correlation coefficient. Mean values of δ15N for the herbivorous copepods selected as the reference baseline showed zonal patterns different from those of myctophids. Although the reference baseline had the lowest mean in Zone 1 (influenced by Northwest African upwelling waters) (Figure 5a), there were no significant differences between the mean δ15N value of the myctophids (Figure 5b). This resulted in the mean TPS for myctophids in Zone 1 being significantly higher than those found in the other zones, with the lowest TPS observed in Zone 2 (Figure 5c). Figure 5. View largeDownload slide Mean (±SE) δ15N values for herbivorous copepods (a) and myctophids (b), and TP estimates (c, TPS) by the zones indicated in Figure 1. Different letters indicate significantly different means (ANOVA and Dunnett-C a posteriori tests, p < 0.05). Figure 5. View largeDownload slide Mean (±SE) δ15N values for herbivorous copepods (a) and myctophids (b), and TP estimates (c, TPS) by the zones indicated in Figure 1. Different letters indicate significantly different means (ANOVA and Dunnett-C a posteriori tests, p < 0.05). Interspecies differences in diet estimations All the myctophids analysed had C: N ratios >4, denoting high lipid content; therefore, δ13C values were lipid normalized following Post et al. (2007) before applying the stable isotope mixing model. N. valdiviae and Benthosema glaciale were the species with the highest lipid content. The isospace plot of δ13C and δ15N for those myctophids analysed and the feasible contribution of potential preys showed that the majority of the species fell inside the area defined by the values of the prey sources. Fishes collected at Station 11 were those that fitted best inside the main prey area (Figure 6), while the three species which had the lowest N values appeared in the periphery of the plots. The two tests, first using a small myctophid (N. valdiviae) and then a small gonostomatid (C. alba) as potential fish prey, showed similar patterns. The two models identify the same few species for which piscivory represents more than 20% of the diet. However, fish contributions were always lower when using the gonostomatid than the small myctophid (Supplementary Tables S2 and S3). Figure 6. View largeDownload slide Isospace of δ13C and δ15N for myctophids included in the MixSIAR mixing model and feasible contribution of potential prey (solid black dots) to their diets. Bars denote 95% CIs. Species abbreviations, first letter of genus, and species (see Table 2 for full names). Figure 6. View largeDownload slide Isospace of δ13C and δ15N for myctophids included in the MixSIAR mixing model and feasible contribution of potential prey (solid black dots) to their diets. Bars denote 95% CIs. Species abbreviations, first letter of genus, and species (see Table 2 for full names). The most outstanding result of the model (Figure 7 and Supplementary Table S2) was the high isotopic signature of seston in the diet of D. atlanticus (76%) and N. valdiviae (50%) and to a lesser extent in other species such as Lobianchia dofleini, Benthosema suborbitale, and small Diaphus brachycephalus (Station 11), Diaphus vanhoeffeni, Lampanyctus alatus (Station 11), and Lampanyctus pusillus (35–43%). Copepod contributions ranged from 12 to 65% and emerged as the most abundant prey in more than half of the species studied, with the highest contribution observed in the diet of B. suborbitale and Notoscopelus resplendens (Station 12) (>60%), followed by H. macrochir, C. warmingii, Lampanyctus nobilis, L. alatus, Lepidophanes guentheri, D. dumerilii, Myctophum nitidulum, and D. mollis (30–47%), and with a higher contribution of herbivorous copepods in B. suborbitale, H. macrochir, and N. resplendens. Crustacean macrozooplankton contributions ranged just from 5 to 24%, being >20% only in H. macrochir, L. nobilis, L. guentheri, and D. dumerilii. The inputs of chaetognaths ranged from 3 to 22%, appearing as a relative important prey source just in large D. brachycephalus (49 mm from Station 6), Diaphus holti, and M. punctatum (18–22%). The model also reported fish as an important potential prey in large D. brachycephalus (40%) and D. rafinesquii (36%), followed by L. dofleini, D. holti, L. pusillus, and L. guentheri (26–28%). Results for the species H. macrochir deserve some comments because this is a common and abundant species across the central Atlantic for which no previous diet information is available. Our model showed a diet dominated by herbivorous copepods (38%), crustacean macrozooplankton (24%), and seston (21%). Figure 7. View largeDownload slide Mean relative contribution of seston, copepods (H, herbivorous; OC, omnivorous–carnivorous), crustacean macrozooplankton, chaetognaths, and a small myctophid (N. valdiviae) as potential prey sources to the diet of the most abundant myctophids across the equatorial and tropical Atlantic. Mean SL of the specimens included in the analysis and station number indicated after species names. Figure 7. View largeDownload slide Mean relative contribution of seston, copepods (H, herbivorous; OC, omnivorous–carnivorous), crustacean macrozooplankton, chaetognaths, and a small myctophid (N. valdiviae) as potential prey sources to the diet of the most abundant myctophids across the equatorial and tropical Atlantic. Mean SL of the specimens included in the analysis and station number indicated after species names. Discussion There are numerous studies examining the feeding regimes of myctophids from different oceans and periods. However, there are still a large number of species and regions for which information is scarce or unavailable. Most previous investigations were made using microscopical examination of gut contents, which gives information of what has been recently eaten (Merrett and Roe, 1974; Hopkins and Gartner, 1992; Hopkins et al., 1996; Bernal et al. 2015; Battaglia et al., 2016). This is a time-consuming process requiring high skills in identifying prey items, a job often complicated by digestive processes. All of these result in a lack of data on where and what a particular fish is eating. The expansion of indirect studies such as those based on isotope analyses provides information on longer time-scales of assimilated material. A recent investigation on mesopelagic fishes from the Gulf of Mexico combined gut content and isotope analyses to identify the trophic structure of mesopelagic fishes (McClain-Counts et al., 2017) and showed not only the usefulness of this last method, but also the strength of combining the two methods. Reference studies on gut content analysis characterized myctophids as mainly opportunist zooplanktivorous fishes, selecting for particular prey among the available zooplankton (Merrett and Roe, 1974; Clarke, 1980; Hopkins and Gartner, 1992; Gartner et al., 1997). Larval myctophids consume eggs and larval stages of copepods (Sabatés and Saiz, 2000; Contreras et al., 2015), while adult copepods, euphausiids, amphipods, and ostracods are among the most cited and frequent prey for adult myctophids, with an increase in prey size with increasing length (Hopkins and Gartner, 1992; Pusch et al., 2004; Suntsov and Brodeur, 2008; Shreeve et al., 2009; Bernal et al., 2013; Hudson et al., 2014; Saunders et al., 2015). Some myctophid species were reported to also consume relatively large amounts of salps and gelatinous organisms (Suntsov and Brodeur, 2008; Shreeve et al., 2009; Hudson et al., 2014; Saunders et al., 2015; McClain-Counts et al., 2017), as well as other fishes (Clarke, 1980; Kinzer, 1982; Hopkins and Gartner, 1992; Pusch et al., 2004; Bernal et al., 2015). Studies using stable isotopes and fatty acid markers also conclude that the main prey items are calanoid copepods and euphausiids (Petursdottir et al., 2008; Valls et al., 2014). Therefore, myctophids occupy a key position in the foodweb for the intermediate transfer of nutrients from plankton to upper trophic levels, and this was confirmed by studies of their isotopic composition (Table 3). Table 3. Upper and lower values of TP of myctophid species from the literature compared with those found in this study. Number of species TEF TP Zone References Lower Upper 2 3.80 3.3 4.0 N Atlantic (subpolar) Petursdottir et al. (2008) 3 2.75 3.0 4.0 W Mediterranean Fanelli et al. (2011) 13 3.15 3.3 4.0 W Mediterranean Valls et al. (2014) 12 3.20 2.3 3.9 S Pacific (Tasman Sea) Flynn and Kloser (2012) 14 3.20 3.3 4.2 Southern Ocean Cherel et al. (2010) 8 3.40 3.7 4.4 Southern Ocean Stowasser et al. (2012) 5 3.40 3.1 3.6 N Atlantic (temperate) Colaço et al. (2013) 10 2.54 2.8 4.2 W Mediterranean Fanelli et al. (2014) 16 3.40 2.9 3.5 S Pacific (subtropical) Hunt et al. (2015) 9 2.00 1.0 4.2 W Atlantic (Gulf of Mexico) McClain-Counts et al. (2017) 20 3.40 1.7 3.9 Equatorial and subtropical Atlantic This study 20 Scaled 1.8 3.3 Equatorial and subtropical Atlantic This study Number of species TEF TP Zone References Lower Upper 2 3.80 3.3 4.0 N Atlantic (subpolar) Petursdottir et al. (2008) 3 2.75 3.0 4.0 W Mediterranean Fanelli et al. (2011) 13 3.15 3.3 4.0 W Mediterranean Valls et al. (2014) 12 3.20 2.3 3.9 S Pacific (Tasman Sea) Flynn and Kloser (2012) 14 3.20 3.3 4.2 Southern Ocean Cherel et al. (2010) 8 3.40 3.7 4.4 Southern Ocean Stowasser et al. (2012) 5 3.40 3.1 3.6 N Atlantic (temperate) Colaço et al. (2013) 10 2.54 2.8 4.2 W Mediterranean Fanelli et al. (2014) 16 3.40 2.9 3.5 S Pacific (subtropical) Hunt et al. (2015) 9 2.00 1.0 4.2 W Atlantic (Gulf of Mexico) McClain-Counts et al. (2017) 20 3.40 1.7 3.9 Equatorial and subtropical Atlantic This study 20 Scaled 1.8 3.3 Equatorial and subtropical Atlantic This study TEF, trophic enrichment factor. All factors from the literature were estimated using the additive model of constant TEF while for comparative purposes both additive (TFA) and scaled (TFS) values are provided in this study. Table 3. Upper and lower values of TP of myctophid species from the literature compared with those found in this study. Number of species TEF TP Zone References Lower Upper 2 3.80 3.3 4.0 N Atlantic (subpolar) Petursdottir et al. (2008) 3 2.75 3.0 4.0 W Mediterranean Fanelli et al. (2011) 13 3.15 3.3 4.0 W Mediterranean Valls et al. (2014) 12 3.20 2.3 3.9 S Pacific (Tasman Sea) Flynn and Kloser (2012) 14 3.20 3.3 4.2 Southern Ocean Cherel et al. (2010) 8 3.40 3.7 4.4 Southern Ocean Stowasser et al. (2012) 5 3.40 3.1 3.6 N Atlantic (temperate) Colaço et al. (2013) 10 2.54 2.8 4.2 W Mediterranean Fanelli et al. (2014) 16 3.40 2.9 3.5 S Pacific (subtropical) Hunt et al. (2015) 9 2.00 1.0 4.2 W Atlantic (Gulf of Mexico) McClain-Counts et al. (2017) 20 3.40 1.7 3.9 Equatorial and subtropical Atlantic This study 20 Scaled 1.8 3.3 Equatorial and subtropical Atlantic This study Number of species TEF TP Zone References Lower Upper 2 3.80 3.3 4.0 N Atlantic (subpolar) Petursdottir et al. (2008) 3 2.75 3.0 4.0 W Mediterranean Fanelli et al. (2011) 13 3.15 3.3 4.0 W Mediterranean Valls et al. (2014) 12 3.20 2.3 3.9 S Pacific (Tasman Sea) Flynn and Kloser (2012) 14 3.20 3.3 4.2 Southern Ocean Cherel et al. (2010) 8 3.40 3.7 4.4 Southern Ocean Stowasser et al. (2012) 5 3.40 3.1 3.6 N Atlantic (temperate) Colaço et al. (2013) 10 2.54 2.8 4.2 W Mediterranean Fanelli et al. (2014) 16 3.40 2.9 3.5 S Pacific (subtropical) Hunt et al. (2015) 9 2.00 1.0 4.2 W Atlantic (Gulf of Mexico) McClain-Counts et al. (2017) 20 3.40 1.7 3.9 Equatorial and subtropical Atlantic This study 20 Scaled 1.8 3.3 Equatorial and subtropical Atlantic This study TEF, trophic enrichment factor. All factors from the literature were estimated using the additive model of constant TEF while for comparative purposes both additive (TFA) and scaled (TFS) values are provided in this study. Trophic position All the myctophids from this study are daily vertical migrators and were collected in the epipelagic layer at night (Olivar et al., 2017), where they were feeding, as has been observed for these same species in other investigations (Clarke, 1980; Kinzer and Schulz, 1985; Hopkins and Gartner, 1992; Bernal et al., 2015) and confirmed by their full stomachs (stomach weight represented from 10 to 20% of total body weight). Therefore, the influence of increasing δ15N values associated with feeding depth, as observed in other mesopelagic species (McClain-Counts et al., 2017), can be ruled out. However, a number of other aspects, such as developmental stage, may affect isotopic composition. In this study, all the analysed specimens were adults except for D. atlanticus (19 mm, immature), L. nobilis (48 mm, immature), and the small size class of N. resplendens (36 mm, immature). This may explain the generally higher δ15N values in the present study when compared with those of the same species in the Gulf of Mexico by McClain-Counts et al. (2017), which were reported to be mostly juveniles. In addition, changes in the source of nitrogen for the primary producers needs to be taken into account when estimating tropic position, since it results in local changes in the baseline that need to be properly taken into account when estimating TP (e.g. Choy et al., 2012). In most studies, TP values of individual species varied between ca. two (fully herbivore) and four (secondary carnivore). Even when there are uncertainties caused by the selection of the appropriate baseline and TEF, the estimates are very similar in all oceans. In our study, the use of a scaled TEF produced lower TP values than those using constant TEF, but maintained the essential difference among species. The scaling approach was useful in studies of large foodwebs (including upper trophic levels, e.g. sharks) that resulted in being underestimated with constant TEF (e.g. Hussey et al., 2014). However, for studies of the lower part of the foodweb, the scaling of the TEF by the baseline δ15 N value tends to produce lower TP values than the use of a constant increase, as shown in this study with myctophids and other studies with planktivores (Bode et al., 2017). In any case, the distribution of TPs among myctophid species in this and other studies reflects their opportunism in consuming different prey. For example, investigations on B. glaciale of similar sizes to those studied here (30–42 mm) and from different regions showed similar TPA, 3.3 in the subpolar North Atlantic (Petursdottir et al., 2008), between 3.5 and 3.8 in the western Mediterranean (Fanelli et al., 2014; Valls et al., 2014), and 3.3 in our study (Table 2, TPA). Mean values of TPA reported for other species in the western Mediterranean, such as M. punctatum, L. dofleini, D. holti, and L. pusillus (Fanelli et al., 2011, 2014; Valls et al., 2014), also differed by <0.3 TPs of the equivalent estimates of the same species in our study. This suggests that these species play a similar trophic role in very different environments. The relatively high values of δ15 N in seston (between 5 and 6%) and plankton (between 6 and 9% found in all zones in this study support the idea of the low influence of diazotrophic nitrogen in the plankton foodweb (e.g. Fernández et al., 2014). Differences in δ15N values among zones for herbivorous copepods can be related to changes in the species composition of zooplankton and also to their phytoplankton prey. Large isotope differences have been reported among different species of the same zooplankton groups (e.g. for chaetognaths; Bohata and Koppelmann, 2013). The differences in water masses, productivity, and oxygen concentrations between zones may also account for the observed differences in δ15N due to changes in the source of nitrogen for primary production (e.g. Aberle et al., 2010). In particular, stations in Zone 2 (with low phytoplankton biomass in surface layers and with the OMZ in the mesopelagic layers) were characterized by higher δ15N and lower δ13C values than at Station 11 in Zone 1 (affected by the highly productive waters of the Northwest African upwelling and characterized by high phytoplankton biomass). The equatorial waters of Zone 3 had very low phytoplankton biomass and a very deep and weak chlorophyll maximum (Olivar et al., 2017), but the isotopic signatures of plankton were quite similar to those of Zones 0 and 2, where the foodweb would most likely be based on the diffusive flux of inorganic nitrogen across the pycnocline (e.g. Fernández-Castro et al., 2015). Comparative feeding preferences The reliability of diet determination from isotopic mixing models has some uncertainties that are related particularly to TEF and to the sources included (Fry, 2006; Caut et al., 2009). Because no experimental enrichment data are available for myctophids, we used the data from Sweeting et al. (2007a, b) previously applied to the muscle of marine fishes (Sweeting et al., 2007a, b; Valls et al., 2014). The lack of a particular prey group implies that its contribution cannot be computed and it is assigned then into the groups actually introduced. For example, the lack of data on ostracods or gelatinous plankton affects the actual diet estimations. Although aware of these shortcomings, the present data are not intended to present the diet spectrum for each species, but to present a comparative framework. In our model, seston emerges as an important component in the diet of small size species such as D. atlanticus (19 mm) and N. valdiviae (20 mm) (27 and 25 mm of maximum attainable sizes, respectively, according to Hulley and Paxton, 2016b), while published data on gut content analysis indicated a diet almost exclusively (80–90%) based on copepods (Clarke, 1980; Kinzer and Schulz, 1985; Hopkins and Gartner, 1992; McClain-Counts et al., 2017). Our results suggest that in the zone where these species were caught (Zone 2, south of Cape Verde Islands, where OMZ occupies the mesopelagic layers), prey availability may be different from that in other regions, and they probably fed on prey of lower TPs than previously observed. According to our estimations, among the studied myctopids, the top predators (with higher proportions of fishes and chaetognaths) are large D. brachycephalus (49 mm), D. rafinesquii (57 mm), and D. holti (44 mm), although neither fishes nor chaetognaths were cited among prey items in previously reported gut contents analysis of D. rafinesquii (Podrazhanskaya, 1993). Copepods arose as the main prey for several species of myctophids with different maximum sizes and body morphology, coinciding with most previous investigations on myctophid diets (Clark, 1980; Kinzer and Schulz, 1985; Alwin and Gjosaeter, 1988; Hopkins and Gartner, 1992; Hopkins et al., 1996; McClain-Counts et al., 2017). It is interesting to note how the same species, but from different stations, showed different proportions of prey. In some cases, this may be attributed to size, such as in D. brachycephalus and N. resplendens, where small specimens showed a higher seston signature. However, B. suborbitale of similar size group from Stations 7 and 12 showed a higher seston signature for the first station, which, as for D. atlanticus and N. valdiviae, could be explained by a different copepod availability in the two zones, reflecting the opportunistic feeding of the species. The two fishes selected as potential prey for myctohids were chosen because their small size (ca. 25 mm), but N. valdiviae co-occurs with the rest of myctophids in the upper layers at night, while C. alba is a non-migrator species, as are the other bristlemouths (Olivar et al., 2017). The relative contribution of N. valdiviae in the estimated diets was always higher than that of C. alba, which could be related to this lack of co-occurrence. Nevertheless, the general agreement in the species denoting piscivory, no matter which species is used in the model, indicated that the method as used here is useful to differentiate coarse groups in the diet (in this case, fish prey), but not conclusive to resolve to species level. In summary, stable δ13C and δ15N isotope analysis determined TP and established wide diet comparisons among myctophids from the central Atlantic. There are important overlaps in TP among species, but significant differences were detected between species in the upper and lower TPs, with very small-sized myctophids in the lower range. However, size is not the only determinant for these differences because the largest species in our study, M. punctatum, M. nitidulum, and C. warmingii, occupy intermediate TPs. In spite of the variability in absolute δ15N values from the species of this study and other regions, the estimated species TP varies in a relatively small range, thus placing myctophids between ca. two (fully herbivore) and four (secondary carnivore). The similarity between TP of species in this study and in other regions suggests that they play a similar role in very different environments. The use of C and N in diet estimations through isotopic mixing models identifies copepods as the main prey for most of the species and also allows for the differentiation of a group of small species, i.e. D. atlanticus, N. valdiviae, N. resplendens, and B. suborbitale, feeding on small prey items (seston and copepods) from three relatively large species of genus Diaphus, i.e. D. brachycephahus, D. rafinesquii, and D. holti preying on fish and chaetognaths. Nevertheless, this cannot be related just to size because these were not the largest species of the study, and other factors related to feeding behaviour, e.g. mouth morphology, or visual adaptations, may also have an important role in feeding behaviour. Acknowledgements The authors are most grateful to the captain, crew, and technicians who worked with them on board the RV “Hesperides”. Special mention is due to M. Teresa Álvarez-Ossorio and M. Luz Fernández de Puelles for their help with the plankton samples, and to Ángel F. Lamas for the preparation of isotope samples. This research has been funded by the Spanish Ministerio de Economía y Competitividad through projects CTM2012-39587-C04 and CTM2016-78853-R. References Aberle N. , Hansen T. , Boettger-Schnack R. , Burmeister A. , Post A. F. , Sommer U. 2010 . Differential routing of ‘new” nitrogen toward higher trophic levels within the marine food web of the Gulf of Aqaba, Northern Red Sea . Marine Biology , 157 : 157 – 169 . Google Scholar Crossref Search ADS Alwin A. , Gjosaeter J. 1988 . Feeding behaviour of Diaphus dumerilii in NW Africa with notes on its relation to other myctophids in the area . Flodevigen Rapportse , 1 : 55 – 69 . Backus R. H. , Craddock J. E. , Haedrich R. L. , Shores D. L. 1970 . The distribution of mesopelagic fishes in the equatorial and western North Atlantic Ocean . Journal of Marine Research , 28 : 179 – 201 . Badcock J. , Merrett N. R. 1976 . Midwater fishes in the eastern North Atlantic–I. Vertical distribution and associated biology in 30°N, 23°W, with developmental notes on certain myctophids . Progress in Oceanography , 7 : 3 – 58 . Google Scholar Crossref Search ADS Battaglia P. , Andaloro F. , Consoli P. , Esposito V. , Malara D. , Musolino S. , Pedà C. et al. 2013 . Feeding habits of the Atlantic bluefin tuna, Thunnus thynnus (L. 1758), in the central Mediterranean Sea (Strait of Messina) . Helgoland Marine Research , 67 : 97 – 107 . Google Scholar Crossref Search ADS Battaglia P. , Esposito V. , Malara D. , Falautano M. , Castriota L. , Andaloro F. 2014 . Diet of the spothead lanternfish Diaphus metopoclampus (Cocco, 1829) (Pisces: Myctophidae) in the central Mediterranean Sea . Italian Journal of Zoology , 81 : 530 – 543 . Google Scholar Crossref Search ADS Battaglia P. , Andaloro F. , Esposito V. , Granata A. , Guglielmo L. , Guglielmo R. , Musolino S. , Romeo T. , Zagami G. 2016 . Diet and trophic ecology of the lanternfish Electrona risso (Cocco 1829) in the Strait of Messina (central Mediterranean Sea) and potential resource utilization from the Deep Scattering Layer (DSL) . Journal of Marine Systems , 159 : 100 – 108 . Google Scholar Crossref Search ADS Benedetti F. , Gasparini S. , Ayata S. D. 2016 . Identifying copepod functional groups from species functional traits . Journal of Plankton Research , 38 : 159 – 166 . Google Scholar Crossref Search ADS PubMed Bernal A. , Olivar M. P. , Fernández de Puelles M. L. 2013 . Feeding patterns of Lampanyctus pusillus (Pisces, Myctophidae) throughout its ontogenetic development . Marine Biology , 160 : 81 – 95 . Google Scholar Crossref Search ADS Bernal A. , Olivar M. P. , Maynou F. , Fernández de Puelles M. L. 2015 . Diet and feeding strategies of mesopelagic fishes in the western Mediterranean . Progress in Oceanography , 135 : 1 – 17 . Google Scholar Crossref Search ADS Bode M. , Hagen W. , Schukat A. , Teuber L. , Fonseca-Batista D. , Dehairs F. , Auel H. 2015 . Feeding strategies of tropical and subtropical calanoid copepods throughout the eastern Atlantic Ocean “Latitudinal and bathymetric aspects . Progress in Oceanography , 138 : 268 – 282 . Google Scholar Crossref Search ADS Bode A. , Carrera P. , González-Nuevo G. , Nogueira E. , Riveiro I. , Santos M. B. 2017 . A trophic index for sardine (Sardina pilchardus) and its relationship to population abundance in the southern Bay of Biscay and adjacent waters of the NE Atlantic . Progress in Oceanography , doi: 10.1016/j.pocean.2017.08.005. Bohata K. , Koppelmann R. , Browman H. 2013 . Chaetognatha of the Namibian upwelling region: taxonomy, distribution and trophic position . Plos One , 8 : e53839. Google Scholar Crossref Search ADS PubMed Bosley K. L. , Lavelle J. W. , Brodeur R. D. , Wakefield W. W. , Emmett R. L. , Baker E. T. , Rehmke K. M. 2004 . Biological and physical processes in and around Astoria submarine Canyon, Oregon, USA . Journal of Marine Systems , 50 : 21 – 37 . Google Scholar Crossref Search ADS Carassou L. , Kulbicki M. , Nicola T. J. R. , Polunin N. V. C. 2008 . Assessment of fish trophic status and relationships by stable isotope data in the coral reef lagoon of New Caledonia, southwest Pacific . Aquatic Living Resources , 21 : 1 – 12 . Google Scholar Crossref Search ADS Caut S. , Angulo E. , Courchamp F. 2009 . Variation in discrimination factors (Δ15N and Δ13C): the effect of diet isotopic values and applications for diet reconstruction . Journal of Applied Ecology , 46 : 443 – 453 . Google Scholar Crossref Search ADS Cherel Y. , Ducatez S. , Fontaine C. , Richard P. , Guinet C. 2008 . Stable isotopes reveal the trophic position and mesopelagic fish diet of female southern elephant seals breeding on the Kerguelen Islands . Marine Ecology Progress Series , 370 : 239 – 247 . Google Scholar Crossref Search ADS Cherel Y. , Fontaine C. , Richard P. , Labat J. P. 2010 . Isotopic niches and trophic levels of myctophid fishes and their predators in the Southern Ocean . Limnology and Oceanography , 55 : 324 – 332 . Google Scholar Crossref Search ADS Choy C. A. , Davison P. C. , Drazen J. C. , Flynn A. , Gier E. J. , Hoffman J. C. , McClain-Counts J. P. et al. 2012 . Global trophic position comparison of two dominant mesopelagic fish families (Myctophidae, Stomiidae) using amino acid nitrogen isotopic analyses . PLoS ONE , 7 : e50133. Google Scholar Crossref Search ADS PubMed Clarke T. A. 1980 . Diets of fouteen species of vertically migrating mesopelagic fishes in Hawaiian waters . Fishery Bulletin, US , 78 : 619 – 640 . Colaço A. , Giacomello E. , Porteiro F. , Menezes G. M. 2013 . Trophodynamic studies on the Condor seamount . Deep Sea Research II , 98 : 178 – 189 . Google Scholar Crossref Search ADS Contreras T. , Olivar M. P. , Bernal A. , Sabatés A. 2015 . Comparative feeding patterns of early stages of mesopelagic fishes with vertical habitat partitioning . Marine Biology , 162 : 2265 – 2277 . Google Scholar Crossref Search ADS Coplen T. B. 2011 . Guidelines and recommended terms for expression of stable-isotope-ratio and gas-ratio measurement results . Rapid Communications in Mass Spectrometry , 25 : 2538 – 2560 . Google Scholar Crossref Search ADS PubMed Davison P. C. , Checkley D. M. Jr. , Koslow J. A. , Barlow J. 2013 . Carbon export mediated by mesopelagic fishes in the northeast Pacific Ocean . Progress in Oceanography , 116 : 14 – 30 . Google Scholar Crossref Search ADS Fanelli E. , Cartes J. , Papiol V. 2011 . Food web structure of deep-sea macrozooplankton and micronekton off the Catalan slope: Insight from stable isotopes . Journal of Marine Systems , 87 : 79 – 89 . Google Scholar Crossref Search ADS Fanelli E. , Papiol V. , Cartes J. E. , Rodriguez-Romeu O. 2014 . Trophic ecology of Lampanyctus crocodilus on north-west Mediterranean Sea slopes in relation to reproductive cycle and environmental variables . Journal of Fish Biology , 84 : 1654 – 1688 . Google Scholar Crossref Search ADS PubMed Fernández A. , Marañón E. , Bode A. 2014 . Large-scale meridional and zonal variability in the nitrogen isotopic composition of plankton in the Atlantic Ocean . Journal of Plankton Research , 36 : 1060 – 1073 . Google Scholar Crossref Search ADS Fernández-Castro B. , Mouriño-Carballido B. , Marañón E. , Chouciño P. , Gago J. , Ramírez T. et al. 2015 . Importance of salt fingering for new nitrogen supply in the oligotrophic ocean . Nature Communications , 6 : 8002. Google Scholar Crossref Search ADS PubMed Flynn A. J. , Kloser R. J. 2012 . Cross-basin heterogeneity in lanternfish (family Myctophidae) assemblages and isotopic niches (δ13C and δ15N) in the southern Tasman Sea abyssal basin . Deep-Sea Research Part I-Oceanographic Research Papers , 69 : 113 – 127 . Google Scholar Crossref Search ADS Fry B. 2006 . Stable Isotope Ecology . Springer Science+Business Media, LLC , New York . 308 pp. Gartner J. V. , Crabtree R. E. , Kenneth J. S. 1997 . Feeding at depth. In Deep-Sea Fishes , pp. 115 – 182 . Ed. by Randall D. J. , Farrell A. P. . Academic Press , San Diego . Gartner J. V. , Hopkins T. L. , Baird R. C. , Milliken D. M. 1987 . The lanternfishes (Pisces, Myctophidae) of the eastern Gulf of Mexico . Fishery Bulletin, US , 85 : 81 – 98 . Gibbs R. Jr. , Krueger W. H. (Eds). 1987 . Biology of midwater fishes of the Bermuda Ocean Acre. Smithsonian Contributions to Zoology, 452: 1 – 187 . Gjøsaeter J. 1973 . Food of myctophid fish, Benthosema glaciale (Reinhardt), from Western Norway . Sarsia , 52 : 53 – 58 . Google Scholar Crossref Search ADS Gjøsaeter J. , Kawaguchi K. 1980 . A review of the world resources of mesopelagic fish . FAO Fisheries Technical Paper , 13 : 1 – 151 . Hedd A. , Montevecchi W. A. 2006 . Diet and trophic position of Leach’s storm-petrel Oceanodroma leucorhoa during breeding and moult, inferred from stable isotope analysis of feathers . Marine Ecology Progress Series , 322 : 291 – 301 . Google Scholar Crossref Search ADS Hopkins T. L. , Gartner J. V. 1992 . Resource-partitioning and predation impact of a low-latitude myctophid community . Marine Biology , 114 : 185 – 197 . Google Scholar Crossref Search ADS Hopkins T. L. , Sutton T. , Lancraft T. M. 1996 . The trophic structure and predation impact of a low latitude midwater fish assemblage . Progress in Oceanography , 38 : 205 – 239 . Google Scholar Crossref Search ADS Hudson J. M. , Steinberg D. K. , Sutton T. T. , Graves J. E. , Latour R. J. 2014 . Myctophid feeding ecology and carbon transport along the northern Mid-Atlantic Ridge . Deep-Sea Research Part I: Oceanographic Research Papers , 93 : 104 – 116 . Google Scholar Crossref Search ADS Hulley P. A. 1981 . Results of the research cruise of FRV ‘‘Walter Herwig to South America. Family Myctophidae (Osteichthyes, Myctophiformes) . Archives Fischereiwissenschaft , 31 : 1 – 300 . Hulley P. A. , Paxton J. R. 2016a . Neoscopelidae. In Bony Fishes, Part 1 (Elopiformes-Scorpaeniformes), the Living Marine Resources of the Eastern Central Atlantic , Vol 1 , pp. 1855 – 1859 . Ed. by Carpenter K. , De Angelis N. . FAO , Rome . Hulley P. A. , Paxton J. R. 2016b . Myctophidae. In Bony Fishes, Part 1 (Elopiformes-Scorpaeniformes), the Living Marine Resources of the Eastern Central Atlantic , Vol 3 , pp. 1860 – 1928 . Ed. by Carpenter K. , De Angelis N. . FAO , Rome . Hunt B. P. V. , Allain V. , Menkes C. , Lorrain A. , Graham B. , Rodier M. , Pagano M. , Carlotti F. 2015 . A coupled stable isotope-size spectrum approach to understanding pelagic food-web dynamics: A case study from the southwest sub-tropical Pacific . Deep Sea Research Part II: Topical Studies in Oceanography , 113 : 208 – 224 . Google Scholar Crossref Search ADS Hussey N. E. , Macneil M. A. , McMeans B. C. , Olin J. A. , Dudley S. F. J. , Cliff G. , Wintner S. P. , Fennessy S. T. , Fisk A. T. 2014 . Rescaling the trophic structure of marine food webs . Ecology Letters , 17 : 239 – 250 . Google Scholar Crossref Search ADS PubMed Karstensen J. , Schütte F. , Pietri A. , Krahmann G. , Fiedler B. , Grundle D. , Hauss H. et al. 2016 . Upwelling and isolation in oxygen-depleted anticyclonic modewater eddies and implications for nitrate cycling . Biogeosciences Discussions , 14 : 2167 – 2181 . Google Scholar Crossref Search ADS Kinzer J. 1982 . The food of four myctophid fish species off Northwest Africa . Rapports et Procès-Verbaux des Réunions - Conseil Permanent International pour ĺExploration de la Mer , 180 : 378 – 383 . Kinzer J. , Schulz K. 1985 . Vertical distribution and feeding patterns of midwater fish in the central equatorial Atlantic . Marine Biology , 85 : 313 – 322 . Google Scholar Crossref Search ADS Layman C. A. , Araujo M. S. , Boucek R. , Hammerschlag-Peyer C. M. , Harrison E. , Jud Z. R. , Matich P. et al. 2012 . Applying stable isotopes to examine food-web structure: an overview of analytical tools . Biological Reviews , 87 : 545 – 562 . Google Scholar Crossref Search ADS PubMed McClain-Counts J. P. , Demopoulos A. W. J. , Ross S. W. 2017 . Trophic structure of mesopelagic fishes in the Gulf of Mexico revealed by gut content and stable isotope analyses . Marine Ecology , 38 : e12449. Google Scholar Crossref Search ADS Mehner T. , Kasprzak P. 2011 . Partial diel vertical migrations in pelagic fish . Journal of Animal Ecology , 80 : 761 – 770 . Google Scholar Crossref Search ADS PubMed Merrett N. R. , Roe H. S. 1974 . Patterns and selectivity in the feeding of certain mesopelagic fishes . Marine Biology , 28 : 115 – 126 . Google Scholar Crossref Search ADS Mittelstaedt E. 1983 . The upwelling area off Northwest Africa: a description of phenomena related to coastal upwelling . Progress in Oceanography , 12 : 307 – 331 . Google Scholar Crossref Search ADS Morel A. , Claustre H. , Gentili B. 2010 . The most oligotrophic subtropical zones of the global ocean: similarities and differences in terms of chlorphyll and yellow substance . Biogeosciences , 7 : 3139 – 3151 . Google Scholar Crossref Search ADS Navarro J. , Louzao M. , Igual J. M. , Oro D. , Delgado A. , Arcos J. M. , Genovart M. et al. 2009 . Seasonal changes in the diet of a critically endangered seabird and the importance of trawling discards . Marine Biology , 156 : 2571 – 2578 . Google Scholar Crossref Search ADS Nelson J. S. 2006 . Fishes of the World , 4th edn . John Wiley & Sons, Inc ., Hoboken, NJ . 624 pp. Nilsen M. , Pedersen T. , Nilssen E. M. , Fredriksen S. 2008 . Trophic studies in a high-latitude fjord ecosystem - a comparison of stable isotope analyses (δ13C and δ15N) and trophic-level estimates from a mass-balance model . Canadian Journal of Fisheries and Aquatic Sciences , 65 : 2791 – 2806 . Google Scholar Crossref Search ADS Olafsdottir D. , MacKenzie B. R. , Chosson-P V. , Ingimundardottir T. 2016 . Dietary evidence of mesopelagic and pelagic foraging by Atlantic bluefin una (Thunnus thynnus L.) during autumn migrations to the Iceland Basin . Frontiers in Marine Science , 3 : 108 . Google Scholar Crossref Search ADS Olivar M. P. , Hulley P. A. , Castellón A. , Emelianov M. , López C. , Tuset V. M. , Contreras T. et al. 2017 . Mesopelagic fishes across the tropical and equatorial Atlantic: Biogeographical and vertical patterns . Progress in Oceanography , 151 : 116 – 137 . Google Scholar Crossref Search ADS Pakhomov E. A. , Bushula T. , Kaehler S. , Watkins B. P. , Leslie R. W. 2006 . Structure and distribution of the slope fish community in the vicinity of the sub-Antarctic Prince Edward Archipelago . Journal of Fish Biology , 68 : 1834 – 1866 . Google Scholar Crossref Search ADS Pepin P. 2013 . Distribution and feeding of Benthosema glaciale in the western Labrador Sea: Fish-zooplankton interaction and the consequence to calanoid copepod populations . Deep-Sea Research Part I: Oceanographic Research Papers , 75 : 119 – 134 . Google Scholar Crossref Search ADS Petursdottir H. , Gislason A. , Falk-Petersen S. , Hop H. , Svavarsson J. 2008 . Trophic interactions of the pelagic ecosystem over the Reykjanes Ridge as evaluated by fatty acid and stable isotope analyses . Deep Sea Research Part II: Topical Studies in Oceanography , 55 : 83 – 93 . Google Scholar Crossref Search ADS Podrazhanskaya S. G. 1993 . Feeding habits of mesopelagic species of fish and estimation of plankton graze in the northwest Atlantic . NAFO Scientific Council Studies , 19 : 79 – 85 . Post D. M. 2002 . Using stable isotopes to estimate trophic position: Models, methods, and assumptions . Ecology , 83 : 703 – 718 . Google Scholar Crossref Search ADS Post D. M. , Layman C. A. , Arrington D. A. , Takimoto G. , Quattrochi J. , Montaña C. G. 2007 . Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable isotope analyses . Oecologia , 152 : 179 – 189 . Google Scholar Crossref Search ADS PubMed Pusch C. , Hulley P. A. , Kock K. H. 2004 . Community structure and feeding ecology of mesopelagic fishes in the slope waters of King George Island (South Shetland Islands, Antarctica) . Deep-Sea Research , 51 : 1685 – 1708 . Google Scholar Crossref Search ADS Quintana-Rizzo E. , Torres J. J. , Ross S. W. , Romero I. , Watson K. , Goddard E. , Hollander D. 2015 . Delta C-13 and delta N-15 in deep-living fishes and shrimps after the Deepwater Horizon oil spill, Gulf of Mexico . Marine Pollution Bulletin , 94 : 241 – 250 . Google Scholar Crossref Search ADS PubMed Revill A. T. , Young J. W. , Lansdell M. 2009 . Stable isotopic evidence for trophic groupings and bio-regionalization of predators and their prey in oceanic waters off eastern Australia . Marine Biology , 156 : 1241 – 1253 . Google Scholar Crossref Search ADS Sabatés A. , Saiz E. 2000 . Intra-and interspecific variability in prey size and niche breadth of myctophiform larvae . Marine Ecology Progress Series , 201 : 261 – 271 . Google Scholar Crossref Search ADS Saunders R. A. , Collins M. A. , Ward P. , Stowasser G. , Hill S. L. , Shreeve R. , Tarling G. A. 2015 . Predatory impact of the myctophid fish community on zooplankton in the Scotia Sea (Southern Ocean) . Marine Ecology Progress Series , 541 : 45 – 64 . Google Scholar Crossref Search ADS Scotto di Carlo B. , Costanzo G. , Fresi E. , Guglielmo L. , lanora A. 1982 . Feeding ecology and stranding mechanisms in two lanternfishes, Hygophum benoiti and Myctophum punctatum . Marine Ecology Progress Series , 9 : 13 – 24 . Google Scholar Crossref Search ADS Shreeve R. , Collins M. , Tarling G. , Main C. , Ward P. , Johnston N. 2009 . Feeding ecology of myctophid fishes in the northern Scotia Sea . Marine Ecology Progress Series , 386 : 221 – 236 . Google Scholar Crossref Search ADS Stock B. C. , Semmens B. X. 2013 . MixSIAR GUI User Manual. Version 3.1. https://github.com/brianstock/MixSIAR, doi:10.5281/zenodo.56159. Stowasser G. , Atkinson A. , McGill R. A. R. , Phillips R. A. , Collins M. A. , Pond D. W. 2012 . Food web dynamics in the Scotia Sea in summer: A stable isotope study . Deep Sea Research II , 59-60 : 208 – 221 . Google Scholar Crossref Search ADS Suntsov A. V. , Brodeur R. D. 2008 . Trophic ecology of three dominant myctophid species in the northern California Current region . Marine Ecology Progress Series , 373 : 81 – 96 . Google Scholar Crossref Search ADS Sutton T. T. 2013 . Vertical ecology of the pelagic ocean: classical patterns and new perspectives . Journal of Fish Biology , 83 : 1508 – 1527 . Google Scholar Crossref Search ADS PubMed Sweeting C. J. , Barry J. , Barnes C. , Polunin N. V. C. , Jennings S. 2007a . Effects of body size and environment on diet-tissue δ15N fractionation in fishes . Journal of Experimental Marine Biology and Ecology , 340 : 1 – 10 . Google Scholar Crossref Search ADS Sweeting C. J. , Barry J. T. , Polunin N. V. C. , Jennings S. 2007b . Effects of body size and environment on diet-tissue delta C-13 fractionation in fishes . Journal of Experimental Marine Biology and Ecology , 352 : 165 – 176 . Google Scholar Crossref Search ADS Trueman C. N. , Johnston G. , ÓHea B. , MacKenzie K. M. 2014 . Trophic interactions of fish communities at midwater depths enhance long-term carbon storage and benthic production on continental slopes . Proceedings of the Royal Society B-Biological Sciences , 281 : 1787. Google Scholar Crossref Search ADS Valls M. , Olivar M. P. , Fernández de Puelles M. L. , Molí B. , Bernal A. , Sweeting C. J. 2014 . Trophic structure of mesopelagic fishes in the western Mediterranean based on stable isotopes of carbon and nitrogen . Journal of Marine Systems , 138 : 160 – 170 . Google Scholar Crossref Search ADS Valls M. , Quetglas A. , Ordines F. , Moranta J. 2011 . Feeding ecology of demersal elasmobranchs from the shelf and slope off the Balearic Sea (western Mediterranean) . Scientia Marina , 75 : 633 – 639 . Google Scholar Crossref Search ADS Vander Zanden M. J. , Rasmussen J. B. 1999 . Primary consumer delta C-13 and delta N-15 and the trophic position of aquatic consumers . Ecology , 80 : 1395 – 1404 . Google Scholar Crossref Search ADS Watanabe H. , Kawaguchi K. 2003 . Decadal change in the diets of the surface migratory myctophid fish Myctophum nitidulum in the Kuroshio region of the western North Pacific: Predation on sardine larvae by myctophids . Fisheries Science , 69 : 716 – 721 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2018. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Richards, Travis, M;Gipson, Emily, E;Cook,, April;Sutton, Tracey, T;Wells, R J, David

doi: 10.1093/icesjms/fsy074pmid: N/A

Abstract The trophic ecology of eight circumglobal meso- and bathypelagic fishes (Anoplogaster cornuta, Chauliodus sloani, Coccorella atlantica, Gigantura chuni, G. indica, Omosudis lowii, Photostomias guernei, and Stomias affinis) with contrasting vertical migration habits (vertical migrators vs. non-migrators) were examined using stable isotope analysis (SIA). Mean δ13C values of these predators were similar among species, ranging from –18.17 to –18.99 ‰, suggesting that all species are supported by a similar carbon source. This finding was supported by mixing-model analysis; all of these deep-living predators received the majority (>73%) of their carbon from epipelagic food resources. Mean δ15N values of the predators ranged from 9.18 to 11.13 ‰, resulting in trophic position estimates between the third and fourth trophic level, although significant shifts in δ15N with increasing body size suggest that some of these species undergo ontogenetic shifts in trophic position. Bayesian standard ellipses, used to estimate isotopic niche areas, differed in size among species, with those occupying the highest relative trophic positions possessing the largest isotopic niches. These results, which provide the first trophic descriptions using dietary tracers for several of these species, offer insight into the trophic structure of deep-sea ecosystems and will help inform the construction of ecosystem-based models. Introduction The deep-pelagic zone (waters deeper than 200 m to just above the seabed) represents the largest cumulative habitat on earth and is home to a diverse array of specialized fauna adapted to its abiotic and biotic conditions (Angel, 1997; Robison, 2004, 2009). The deep sea and its inhabitants provide an array of ecosystem services that are important to humans, including carbon sequestration, nutrient regeneration, fisheries production, and waste absorption (Danovaro et al., 2008; Mengerink et al., 2014; Thurber et al., 2014). Despite its enormous volume and the economic and ecological importance of its fauna, deep-pelagic ecosystems remain chronically understudied (Webb et al., 2010) and face an increasing number of stressors including climate change, ocean acidification, overfishing, and natural resource extraction (Morato et al., 2006; Ramirez-Llodra et al., 2011; Mengerink et al., 2014). As threats to the diversity and stability of marine ecosystems increase and expand into deeper oceanic environments, there has been increasing concern regarding the status of deep-sea communities and a renewed interest in describing and understanding deep-sea ecosystem structure. Central to our understanding of ecosystem and community structure is a thorough knowledge of foodwebs (Polis and Strong, 1996; McCann, 2000). In addition to providing important information regarding ecosystem functioning, the study of foodwebs provides understanding of how animal communities are structured and sheds light on the mechanisms underlying species coexistence and persistence. While our knowledge of deep-pelagic foodwebs has advanced considerably over the past few decades (Robison, 2009; Sutton, 2013), fundamental information in many regions, including species-specific feeding relationships, trophic position estimates, and delineations of energy pathways connecting disparate trophic levels and communities, is lacking (Mengerink et al., 2014; Drazen and Sutton, 2017). Fishes are a dominant component of deep-pelagic ecosystems worldwide and are among the main taxa that undertake diel vertical migrations (DVM). While the standardized abundance (no. per unit volume) of meso- and bathypelagic fishes is relatively low (Angel and Baker, 1982), their global distributions have resulted in high cumulative biomass estimated at 7–10 billion tonnes (Gjøsaeter and Kawaguchi, 1980; Irigoien et al., 2014). Due to their sheer numbers and vertical migration behaviour, which can exceed 1000 m in vertical extent, it is increasingly being recognized that fishes play key ecological and biogeochemical roles in open-ocean ecosystems (Wilson et al., 2009; Drazen and Sutton, 2017). As highly abundant mid-level consumers, deep-pelagic fishes help regulate zooplankton populations (Hopkins and Gartner, 1992; Pakhomov et al., 1996). Deep-pelagic fishes also serve as trophic links between zooplankton and higher-order consumers such as epipelagic fishes (Moteki et al., 2001; Choy et al., 2013), marine mammals (Pauly et al., 1998), and seabirds (Raclot et al., 1998; Cherel et al., 2008). DVM of fishes have been shown to connect the epi-, meso-, and bathypelagic habitats with each other and with deep-benthic habitats (Porteiro and Sutton, 2007; Trueman et al., 2014). Stable isotope analysis (SIA) has been widely used to delineate foodweb structure and provides an integrated view of an organism’s diet over time-scales relevant to tissue turnover rates rather than digestion rates (Peterson and Fry, 1987; Post, 2002). Carbon isotopes undergo relatively small amounts of fractionation during trophic transfers and are useful for determining the relative contributions of carbon sources to the production of consumers (Peterson and Fry, 1987). Stable isotopes of nitrogen undergo comparatively large levels of fractionation (∼3–5 ‰) during trophic transfer, resulting in predictable differences in the isotopic signatures of consumers and their prey (Post, 2002; Hussey et al., 2014). The relatively predictable level of enrichment of 15 N during trophic transfer allows for the determination of trophic levels and can be used to identify trophic relationships within assemblages of organisms (Peterson and Fry, 1987; Post, 2002). To date, much of the research describing the trophic ecology of deep-pelagic fishes has focused on zooplanktivorous groups (myctophids, sternoptychids, gonostomatids), while less attention has been paid to micronektonivores (stomiids, alepisauroids) that occupy higher trophic levels. The numerical importance of micronektonivores (Hopkins et al., 1996; Sutton and Hopkins, 1996a), their propensity to prey heavily on zooplanktivorous fishes (Clarke, 1982; Hopkins et al., 1996; Sutton and Hopkins, 1996b), and documented importance as prey for higher trophic level consumers (Moteki et al., 2001; Choy et al., 2013) provides the rationale for further describing their trophic dynamics. Here, we describe the trophic ecology of eight putative high-trophic-level fishes: Anoplogaster cornuta, Chauliodus sloani, Coccorella atlantica, Gigantra chuni, G. indica, Omosudis lowii, Photostomias guernei, and Stomias affinis. These species are meso- and bathypelagic fishes with circumglobal distributions, some of which have been documented as numerically important components of deep-pelagic assemblages (Sutton and Hopkins, 1996a; Moore et al., 2003; Sutton et al., 2008). Specific goals of this study are to provide estimates of trophic position, describe the isotopic niche areas and the extent of niche overlap among species, detail ontogenetic shifts in trophic position, and quantify the relative carbon contributions of particulate organic matter (POM) from the epi-, meso-, and bathypelagic zones to each of these species. Material and methods Sample collection and study site Fishes were collected from the northern Gulf of Mexico (GOM) during three oceanographic cruises conducted during 2011 in spring (22 March–11 April), summer (23 June–13 July), and fall (8–27 September). All cruises were part of the Offshore Nekton Sampling and Analysis Program (ONSAP) that was implemented following the Deepwater Horizon oil spill in support of NOAA’s Natural Resource Damage Assessment (NRDA). ONSAP stations are the same as stations currently used by the long-term Southeast Area Monitoring and Assessment Program (SEAMAP) and are situated every half degree of longitude and latitude in the northern GOM (Figure 1). Specimens were collected using large midwater trawls fitted with large-mesh panels (∼80 cm) near the mouth that gradually tapered to smaller mesh (∼6 cm) sizes before the codend. Trawls were fished obliquely from the surface to depths of either 700 or 1400 m. Once the trawls were retrieved, animals were sorted, enumerated, and visually identified to species. Samples for SIA were selected haphazardly in an effort to maximize spatial and temporal coverage. All specimens for SIA were frozen whole at –20°C until processed at Texas A&M University at Galveston. Figure 1. View largeDownload slide Map of ONSAP sampling grid, locations of POM samples, and locations where fishes were collected for SIA (specimen number denoted by circle diameter) in the GOM. Figure 1. View largeDownload slide Map of ONSAP sampling grid, locations of POM samples, and locations where fishes were collected for SIA (specimen number denoted by circle diameter) in the GOM. Stable isotope analysis SIA was conducted on 212 specimens, with sample sizes of each species ranging from 19 to 37 individuals (Table 1). White muscle tissue for SIA was dissected from the dorsal musculature of fishes and visually examined under a dissecting microscope for the presence of bones, which were subsequently removed. Cleaned samples were rinsed with deionized water, frozen, and lyophilized for 48 h. Freeze-dried samples were homogenized using a mortar and pestle, weighed, wrapped in tin capsules, and shipped to the Stable Isotope Facility at the University of California Davis for analysis. Analysis of muscle tissue δ13C and δ15N was carried out using an elemental analyser (PDZ Europa ANCA-GSL) coupled with an isotope ratio mass spectrometer (PDZ Europa 20-20). The long-term standard deviation of the facility at UC Davis is 0.2 ‰ for δ13C and 0.3 ‰ for δ15N. Stable isotope data are expressed relative to international standards of Vienna PeeDee belemnite and atmospheric N2 for carbon and nitrogen, respectively. The C: N of fishes in this study were low (species mean C: N range 3.31–3.86; 92% of individuals C: N < 4.0) compared with C: N from similar species collected in the Atlantic and Southern oceans (C: N 3.3–12.5; Hoffman and Sutton, 2010), suggesting that lipids did not significantly confound the interpretation of δ13C data. Therefore, all statistical analyses were performed on uncorrected δ13C values. Table 1. Species-specific sample descriptions and bulk δ13C and δ15N isotope data (mean ± SD). Species n Spring Summer Fall Standard length range (mm) Mean standard length (mm) ± SD δ13C (‰) ± SD δ15N (‰) ± SD C: Nbulk ± SD A. cornuta1 23 2 12 9 84–148 114.35 ± 19.09 –18.93 ± 0.67 11.14 ± 0.96 3.66 ± 0.44 C. sloani2 30 10 20 0 143–237 191.43 ± 23.32 –18.68 ± 0.43 9.51 ± 0.42 3.42 ± 0.23 C. atlantica2 19 0 19 0 44–125 89.53 ± 26.53 –18.50 ± 0.47 9.96 ± 0.83 3.67 ± 0.34 G. chuni1 24 6 9 9 34–186 134.57 ± 40.11 –18.25 ± 0.44 11.13 ± 1.08 3.31 ± 0.15 G. indica1 21 8 6 7 75–192 141.95 ± 28.83 –18.25 ± 0.94 10.70 ± 0.64 3.43 ± 0.29 O. lowii1 32 12 10 10 36–261 120.66 ± 61.60 –19.04 ± 0.32 9.79 ± 0.60 3.40 ± 0.08 P. guernei2 37 14 13 10 56–127 98.08 ± 14.03 –18.61 ± 0.40 9.18 ± 0.63 3.44 ± 0.13 S. affinis2 26 5 16 5 55–205 127.31 ± 38.66 –19.38 ± 0.84 9.98 ± 0.89 3.86 ± 0.63 Species n Spring Summer Fall Standard length range (mm) Mean standard length (mm) ± SD δ13C (‰) ± SD δ15N (‰) ± SD C: Nbulk ± SD A. cornuta1 23 2 12 9 84–148 114.35 ± 19.09 –18.93 ± 0.67 11.14 ± 0.96 3.66 ± 0.44 C. sloani2 30 10 20 0 143–237 191.43 ± 23.32 –18.68 ± 0.43 9.51 ± 0.42 3.42 ± 0.23 C. atlantica2 19 0 19 0 44–125 89.53 ± 26.53 –18.50 ± 0.47 9.96 ± 0.83 3.67 ± 0.34 G. chuni1 24 6 9 9 34–186 134.57 ± 40.11 –18.25 ± 0.44 11.13 ± 1.08 3.31 ± 0.15 G. indica1 21 8 6 7 75–192 141.95 ± 28.83 –18.25 ± 0.94 10.70 ± 0.64 3.43 ± 0.29 O. lowii1 32 12 10 10 36–261 120.66 ± 61.60 –19.04 ± 0.32 9.79 ± 0.60 3.40 ± 0.08 P. guernei2 37 14 13 10 56–127 98.08 ± 14.03 –18.61 ± 0.40 9.18 ± 0.63 3.44 ± 0.13 S. affinis2 26 5 16 5 55–205 127.31 ± 38.66 –19.38 ± 0.84 9.98 ± 0.89 3.86 ± 0.63 1Denotes no DVM, 2denotes asynchronous DVM (not all individuals of population migrate vertically each day). References for vertical migration patterns: A. cornuta (Clarke and Wagner, 1976), C. sloani (Sutton and Hopkins, 1996a), C. atlantica (McEachran and Fechhelm, 1998), G. chuni (McEachran and Fechhelm, 1998), G. indica (Sutton et al., 2010), O. lowii (McEachran and Fechhelm, 1998; Sutton et al., 2010), P. guernei (Sutton and Hopkins, 1996a), S. affinis (Sutton and Hopkins, 1996a). Table 1. Species-specific sample descriptions and bulk δ13C and δ15N isotope data (mean ± SD). Species n Spring Summer Fall Standard length range (mm) Mean standard length (mm) ± SD δ13C (‰) ± SD δ15N (‰) ± SD C: Nbulk ± SD A. cornuta1 23 2 12 9 84–148 114.35 ± 19.09 –18.93 ± 0.67 11.14 ± 0.96 3.66 ± 0.44 C. sloani2 30 10 20 0 143–237 191.43 ± 23.32 –18.68 ± 0.43 9.51 ± 0.42 3.42 ± 0.23 C. atlantica2 19 0 19 0 44–125 89.53 ± 26.53 –18.50 ± 0.47 9.96 ± 0.83 3.67 ± 0.34 G. chuni1 24 6 9 9 34–186 134.57 ± 40.11 –18.25 ± 0.44 11.13 ± 1.08 3.31 ± 0.15 G. indica1 21 8 6 7 75–192 141.95 ± 28.83 –18.25 ± 0.94 10.70 ± 0.64 3.43 ± 0.29 O. lowii1 32 12 10 10 36–261 120.66 ± 61.60 –19.04 ± 0.32 9.79 ± 0.60 3.40 ± 0.08 P. guernei2 37 14 13 10 56–127 98.08 ± 14.03 –18.61 ± 0.40 9.18 ± 0.63 3.44 ± 0.13 S. affinis2 26 5 16 5 55–205 127.31 ± 38.66 –19.38 ± 0.84 9.98 ± 0.89 3.86 ± 0.63 Species n Spring Summer Fall Standard length range (mm) Mean standard length (mm) ± SD δ13C (‰) ± SD δ15N (‰) ± SD C: Nbulk ± SD A. cornuta1 23 2 12 9 84–148 114.35 ± 19.09 –18.93 ± 0.67 11.14 ± 0.96 3.66 ± 0.44 C. sloani2 30 10 20 0 143–237 191.43 ± 23.32 –18.68 ± 0.43 9.51 ± 0.42 3.42 ± 0.23 C. atlantica2 19 0 19 0 44–125 89.53 ± 26.53 –18.50 ± 0.47 9.96 ± 0.83 3.67 ± 0.34 G. chuni1 24 6 9 9 34–186 134.57 ± 40.11 –18.25 ± 0.44 11.13 ± 1.08 3.31 ± 0.15 G. indica1 21 8 6 7 75–192 141.95 ± 28.83 –18.25 ± 0.94 10.70 ± 0.64 3.43 ± 0.29 O. lowii1 32 12 10 10 36–261 120.66 ± 61.60 –19.04 ± 0.32 9.79 ± 0.60 3.40 ± 0.08 P. guernei2 37 14 13 10 56–127 98.08 ± 14.03 –18.61 ± 0.40 9.18 ± 0.63 3.44 ± 0.13 S. affinis2 26 5 16 5 55–205 127.31 ± 38.66 –19.38 ± 0.84 9.98 ± 0.89 3.86 ± 0.63 1Denotes no DVM, 2denotes asynchronous DVM (not all individuals of population migrate vertically each day). References for vertical migration patterns: A. cornuta (Clarke and Wagner, 1976), C. sloani (Sutton and Hopkins, 1996a), C. atlantica (McEachran and Fechhelm, 1998), G. chuni (McEachran and Fechhelm, 1998), G. indica (Sutton et al., 2010), O. lowii (McEachran and Fechhelm, 1998; Sutton et al., 2010), P. guernei (Sutton and Hopkins, 1996a), S. affinis (Sutton and Hopkins, 1996a). The stable isotope data of POM used in this study are derived from the published dataset of Fernández-Carrera et al. (2016). For detailed descriptions of methodologies and sample locations, see Fernández-Carrera et al. (2016), but a brief description of methodologies follows. POM samples were collected during summer 2011 (2–21 July) in the northern GOM. In addition to samples collected in pelagic waters, the complete published dataset included samples taken from waters over the continental shelf and from waters in close proximity to the Mississippi River. In order to maximize the spatial overlap between POM samples and the collection locations of fishes, only POM data collected within close proximity to ONSAP sampling stations in waters ≥1000 m deep (Figure 1) were utilized. POM samples were collected throughout the water column using remotely fired 10-l Niskin bottles. Samples were then filtered across 47-mm glass fibre filters at low pressure and dried at 60°C for 24 h prior to isotope analysis (Fernández-Carrera et al., 2016). In order to determine if the isotopic signatures of POM samples changed with depth, we used collection depth to designate POM samples as epipelagic (0–200 m), mesopelagic (200–1000 m), or bathypelagic (>1000 m) so that statistical comparisons could be made. Data analysis Multivariate analysis of variance (MANOVA) was used to test for differences in δ13C and δ15N among species and POM depth zones. Species and season were included as factors in the linear model and tested for the presence of an interaction. If significant differences were found, univariate tests for both δ13C and δ15N were performed using analysis of variance among fish species and POM depth zones. A posteriori differences among means were detected using Tukey’s honestly significant difference (HSD) test. Using equation 4 from Post et al. (2007), trophic position was calculated for each species: TrLi= [(δ15Ni– δ15Nbase)/Δ15N] + λ (1) where δ15Ni is the mean species δ15N, δ15Nbase is the mean δ15N of the primary producer or primary consumer being used to set the isotopic baseline, Δ15N is the trophic discrimination factor, and λ represents the trophic level of the organism being used to set the baseline. Because primary consumer data were not available for the time-period of this study, trophic position estimates made using mean δ15N values of POM collected from the epipelagic zone were compared with estimates calculated from published δ15N values of a group of primary consumers (euphausiids) collected in the pelagic northern GOM during 2007 (McClain-Counts et al., 2017). In order to explore the relationship between fish size and δ13C and δ15N, least-squares linear regression analysis was conducted for each species. Spatial variation in δ13C and δ15N of both fishes and POM was investigated using least-squares linear regression between isotopic values and longitude and latitude (0.5° intervals). Because every species was not collected at every sampling location, isotope data were pooled across species within each line of longitude and latitude (more than one site along each 0.5° of longitude). Additionally, because not all species were collected across a range of latitudes and longitudes within each season, the effect of season on the spatial relationships of the isotope data was not explored. All statistical analyses were performed in R (R Development Core Team, 2016) v. 3.3.2. The trophic breadth of each species and trophic similarity among species were assessed by calculating standard ellipse areas (SEA) using the R package SIBER (Jackson et al., 2011) and following methods outlined by Jackson et al. (2011). Bayesian standard ellipses encompass ∼40% of the isotope data for each species are less affected by increases in sample size or statistical outliers than convex hull analysis, and represent the core isotopic niche area of a species (Jackson et al., 2011). Size-corrected SEAs (SEAc) were calculated for each species, which adjusts for underestimation of ellipse area at small sample sizes and allows for comparison of ellipse sizes to other studies (Jackson et al., 2011). Overlap of size-corrected ellipses was used as a proxy for trophic similarity and was examined by calculating the extent of overlap between each pairwise combination of species. The percentage of overlap between species pairs was calculated by dividing the area of overlap (‰2) by the total combined ellipse area (‰2) of the two species being compared. Isotopic niche overlap was considered significant when overlap between two species was >50%. Differences in size-corrected ellipse area, a proxy for trophic breadth that assumes species with larger SEAc feed more broadly within the foodweb than those with smaller SEAc were compared among species and considered to be significantly different when 95% of posterior draws were smaller in one species compared with the other. The Bayesian mixing model, MixSIAR (Stock and Semmens, 2015), was used to estimate the relative contribution of epi- (0–200 m), meso- (200–1000 m), and bathypelagic (>1000 m) POM to each species. Bayesian mixing models provide the most accurate estimations of source or prey contributions when tissue and species-specific discrimination factors are used (Caut et al., 2008), but discrimination factors for meso- and bathypelagic fishes are currently unknown. We chose to run mixing models using discrimination factors of 3.15 ‰ ± 1.28 ‰ and 0.97 ‰ ± 1.08 ‰ for δ15N and δ13C, respectively (Sweeting et al., 2007a,b), which have been previously used to study the trophic structure of meso- and bathypelagic fishes (Valls et al., 2014). Mixing models in MixSIAR estimate probability density functions using Markov chain Monte Carlo methods, and each model was run with identical parameters (number of chains = 3; chain length = 100 000; burn in = 50 000; thin = 50). Model convergence was determined using Gelman-Rubin and Geweke diagnostic tests (Stock and Semmens, 2015). Results Stable isotopes Individual consumer δ13C values ranged from –21.49 to –16.63 ‰, while mean δ13C values were similar among species, with a difference of 1.13 ‰ separating the most depleted (S. affinis: –19.38 ‰ ± 0.83) and most enriched species (G. chuni: –18.25 ‰ ± 0.44 and G. indica: –18.25 ‰ ± 0.94) (Table 1; Figure 2). Individual δ15N values varied between 7.10 and 13.07 ‰, with 1.96 ‰ separating the mean δ15N values of the most enriched (A. cornuta: 11.14 ‰ ± 0.96) and depleted species (P. guernei: 9.18 ‰ ± 0.63) (Table 1; Figure 2). Species-specific differences in δ13C and δ15N were significant (F14, 382 = 17.24, p < 0.001); however, no significant seasonal effects were found (F14 382 = 1.29, p = 0.27), and no significant interaction effect among species and season was detected (F22 382 = 1.05, p = 0.40). Significant differences in δ13C values among species (one-way ANOVA; F7204 = 11.62, p < 0.001) were driven by G. chuni and G. indica, which were enriched in 13 C compared with more 13 C-depleted species such as O. lowii and S. affinis (Figure 2). Significant differences in δ15N among species (one-way ANOVA; F7204 = 25.55, p < 0.001) were primarily driven by A. cornuta, G. chuni, and G. indica, which were enriched in 15 N compared with C. sloani and P. guernei (Figure 2). Results of all pairwise comparisons for δ13C and δ15N values among species are listed in Supplementary Table S1. Figure 2. View largeDownload slide Isotope bi-plot of δ13C and δ15N values from POM (squares) and fishes (circles). Data points represent means and error bars represent ± 1 SD. Figure 2. View largeDownload slide Isotope bi-plot of δ13C and δ15N values from POM (squares) and fishes (circles). Data points represent means and error bars represent ± 1 SD. The δ13C values of fishes were significantly correlated with latitude (r = 0.08, p < 0.01) and longitude (r = 0.04, p < 0.01), while δ15N values were not (latitude p = 0.46; longitude p = 0.19). Due to limited spatial coverage within each species, spatial trends were tested by pooling all fish species together. Because spatial variation could not be tested within each species and due to the low correlation coefficients observed among fish δ13C values and latitude and longitude, isotope data for each species were pooled across lines of latitude and longitude during subsequent analysis. A total of 154 samples of POM collected from depths ranging from 1 to 2500 m were utilized (Fernández-Carrera et al. 2016). POM exhibited a wide range of δ13C (–27.17 to –16.41) and δ15N values (–3.58 to 11.69), with POM samples generally becoming more 15 N enriched with increasing depth (Figure 2). Significant differences in POM δ13C and δ15N among vertical depth zones (MANOVA: F4302 = 14.54, p < 0.001) were observed. Significant differences in δ15N were found among depth zones (ANOVA: F2151 = 34.41, p < 0.001), with epipelagic POM more 15 N depleted than POM collected from mesopelagic and bathypelagic depths (p < 0.001). The δ13C values of POM did not significantly differ across depth zones (F2151 = 0.42, p = 0.66). Latitudinal and longitudinal variation in POM δ13C and δ15N was minimal, with the only significant correlation occurring between epipelagic POM δ13C and longitude, although correlation coefficients were low (r = 0.04, p = 0.043). All other pairwise combinations between δ13C and δ15N and latitude or longitude within the epi-, meso-, and bathypelagic depth zones were non-significant (Supplementary Table S2). Trophic position estimates The use of primary producers or primary consumers to set the isotopic baseline had no effect on the relative trophic positions among consumers, but resulted in slight differences (0.32 TL) in calculated trophic levels. When primary producer (POM) data were used to set the baseline, consumer TPs ranged from 2.8 (P. guernei) to 3.4 (A. cornuta, G. chuni), while all species fell within the third and fourth trophic levels when primary consumers were used to set the baseline (P. guernei = 3.1, A. cornuta and G. chuni = 3.7) (Figure 3; Supplementary Table S3). Figure 3. View largeDownload slide Trophic level estimates calculated using δ15N data of each species. Letters represent significant differences in TL among species, with like letters being similar and non-like letters significantly different. Dashed lines represent the δ 15N threshold values of TL 3 and TL 4 when using primary consumers (euphausiids) to set the isotopic baseline; dotted lines represent the δ 15N threshold values of TL 3 and TL 4 when using primary producers (POM) to establish isotopic baseline. For species-specific TP estimates (± SD), see Supplementary Table S3. Figure 3. View largeDownload slide Trophic level estimates calculated using δ15N data of each species. Letters represent significant differences in TL among species, with like letters being similar and non-like letters significantly different. Dashed lines represent the δ 15N threshold values of TL 3 and TL 4 when using primary consumers (euphausiids) to set the isotopic baseline; dotted lines represent the δ 15N threshold values of TL 3 and TL 4 when using primary producers (POM) to establish isotopic baseline. For species-specific TP estimates (± SD), see Supplementary Table S3. Of the species examined, A. cornuta (r = 0.63, p < 0.001), C. atlantica (r = 0.74, p < 0.001), G. chuni (r = 0.41, p < 0.001), C. sloani (r = 0.22, p < 0.001), P. guernei (r = 0.25, p < 0.001), and S. affinis (r = 0.53, p < 0.001) exhibited significant positive relationships between δ15N and SL (Figure 4). Relationships between δ13C and SL were more variable than those observed with δ15N (Figure 4). Two species, G. chuni (r = 0.33, p < 0.01) and O. lowii (r = 0.44, p < 0.001) displayed significant positive relationships between δ13C and SL (Figure 4). Figure 4. View largeDownload slide Results of least-squares regression analysis between standard length (mm) and δ15N and δ13C values: (a) A. cornuta, (b) C. sloani, (c) C. atlantica, (d) G. chuni, (e) G. indica, (f) O. lowii, (g) P. guernei, and (h) S. affinis. Figure 4. View largeDownload slide Results of least-squares regression analysis between standard length (mm) and δ15N and δ13C values: (a) A. cornuta, (b) C. sloani, (c) C. atlantica, (d) G. chuni, (e) G. indica, (f) O. lowii, (g) P. guernei, and (h) S. affinis. Isotopic niche breadth, calculated using SEAc, was largest for the piscivorous S. affinis (SEAc= 2.27), G. indica (SEAc = 1.98), A. cornuta (SEAc = 1.96), and G. chuni (SEAc = 1.53), which collectively occupied the highest trophic positions within the guild of predators examined. C. atlantica (SEAc = 0.1.19) occupied an intermediate relative trophic position and intermediate-sized trophic niche. P. guernei (SEAc = 0.71) and O. lowii (SEAc = 0.62), which feed primarily on crustaceans and cephalopods, respectively, occupied lower relative trophic positions and were characterized by relatively small isotopic niches (Figure 5, Table 2). Interestingly, the smallest isotopic niche also belonged to a known piscivore, C. sloani (SEAc = 0.56), although the small calculated isotopic niche area could have been due to a limited sampled size range (Figure 4). In the 20 instances where overlap in SEAc occurred, the percentage of shared isotopic niche space ranged from 1% (A. cornuta and C. atlantica; G. indica and O. lowii) to 27% (between G. chuni and G. indica) (Table 3). Directional overlap, or the percentage of one species’ ellipse covering the ellipse of another species, varied widely from 2 to 100%. Differences in directional overlap was greatest between C. sloani and C. atlantica (81 vs. 38%), C. sloani and S. affinis (63 vs. 15%), and O. lowii and S. affinis (100 vs. 27%) (Table 3, Figure 5). Table 2. Metrics for estimating isotopic niche size in eight meso- and bathypelagic predators. Species TA SEA SEAc CD A. cornuta 6.81 1.87 1.96 0.97 C. sloani 2.68 0.53 0.56 0.46 C. atlantica 3.02 1.13 1.19 0.82 G. chuni 5.82 1.46 1.53 0.83 G. indica 5.17 1.88 1.98 0.98 O. lowii 2.33 0.60 0.62 0.58 P. guernei 2.58 0.69 0.71 0.66 S. affinis 7.01 2.18 2.27 1.04 Species TA SEA SEAc CD A. cornuta 6.81 1.87 1.96 0.97 C. sloani 2.68 0.53 0.56 0.46 C. atlantica 3.02 1.13 1.19 0.82 G. chuni 5.82 1.46 1.53 0.83 G. indica 5.17 1.88 1.98 0.98 O. lowii 2.33 0.60 0.62 0.58 P. guernei 2.58 0.69 0.71 0.66 S. affinis 7.01 2.18 2.27 1.04 TA, total area (expressed in ‰) encompassed by all data points of each species; SEA, standardized ellipse area for each species; SEAc, size-corrected standardized ellipse area; CD, centroid distance calculated by taking average distance of each data point from the centroid for each species. Table 2. Metrics for estimating isotopic niche size in eight meso- and bathypelagic predators. Species TA SEA SEAc CD A. cornuta 6.81 1.87 1.96 0.97 C. sloani 2.68 0.53 0.56 0.46 C. atlantica 3.02 1.13 1.19 0.82 G. chuni 5.82 1.46 1.53 0.83 G. indica 5.17 1.88 1.98 0.98 O. lowii 2.33 0.60 0.62 0.58 P. guernei 2.58 0.69 0.71 0.66 S. affinis 7.01 2.18 2.27 1.04 Species TA SEA SEAc CD A. cornuta 6.81 1.87 1.96 0.97 C. sloani 2.68 0.53 0.56 0.46 C. atlantica 3.02 1.13 1.19 0.82 G. chuni 5.82 1.46 1.53 0.83 G. indica 5.17 1.88 1.98 0.98 O. lowii 2.33 0.60 0.62 0.58 P. guernei 2.58 0.69 0.71 0.66 S. affinis 7.01 2.18 2.27 1.04 TA, total area (expressed in ‰) encompassed by all data points of each species; SEA, standardized ellipse area for each species; SEAc, size-corrected standardized ellipse area; CD, centroid distance calculated by taking average distance of each data point from the centroid for each species. Table 3. Isotopic niche overlap measured in percentage of shared space (‰) between each pairwise combination of species. A. cornuta C. sloani C. atlantica G. chuni G. indica O. lowii P. guernei C. sloani 0 (0, 0) 0 C. atlantica 1 (2, 3) 0.05 26 (81, 38) 0.99 G. chuni 9 (23, 30) 0.19 0 (0, 0) 0.99 14 (32, 25) 0.80 G. indica 16 (33, 32) 0.49 0 (0, 0) 0.99 20 (41, 25) 0.94 27 (61, 47) 0.80 O. lowii 3 (4, 12) 0 18 (39, 34) 0.66 11 (16, 32) 0.01 0 (0, 0) 0.01 1 (2, 5) 0 P. guernei 0 (0, 0) 0 30 (68, 54) 0.85 7.5 (28, 47) 0.04 0 (0, 0) 0.01 0 (0, 0) 0 11 (23, 20) 0.73 S. affinis 9 (19, 17) 0.70 12 (63, 15) 0.99 11 (33, 17) 0.98 0 (0, 0) 0.15 4 (7, 7) 0.70 21 (100, 27) 1 9 (37, 11) 1 A. cornuta C. sloani C. atlantica G. chuni G. indica O. lowii P. guernei C. sloani 0 (0, 0) 0 C. atlantica 1 (2, 3) 0.05 26 (81, 38) 0.99 G. chuni 9 (23, 30) 0.19 0 (0, 0) 0.99 14 (32, 25) 0.80 G. indica 16 (33, 32) 0.49 0 (0, 0) 0.99 20 (41, 25) 0.94 27 (61, 47) 0.80 O. lowii 3 (4, 12) 0 18 (39, 34) 0.66 11 (16, 32) 0.01 0 (0, 0) 0.01 1 (2, 5) 0 P. guernei 0 (0, 0) 0 30 (68, 54) 0.85 7.5 (28, 47) 0.04 0 (0, 0) 0.01 0 (0, 0) 0 11 (23, 20) 0.73 S. affinis 9 (19, 17) 0.70 12 (63, 15) 0.99 11 (33, 17) 0.98 0 (0, 0) 0.15 4 (7, 7) 0.70 21 (100, 27) 1 9 (37, 11) 1 Numbers in parentheses represent the percent overlap of species A (column) with species B (row) and vice versa. Numbers in bold represent shared overlap >50%. Second column of numbers represents the likelihood of differences in SEAc size. Numbers in bold represent statistically significant differences in SEAc size between the pair of species examined. Table 3. Isotopic niche overlap measured in percentage of shared space (‰) between each pairwise combination of species. A. cornuta C. sloani C. atlantica G. chuni G. indica O. lowii P. guernei C. sloani 0 (0, 0) 0 C. atlantica 1 (2, 3) 0.05 26 (81, 38) 0.99 G. chuni 9 (23, 30) 0.19 0 (0, 0) 0.99 14 (32, 25) 0.80 G. indica 16 (33, 32) 0.49 0 (0, 0) 0.99 20 (41, 25) 0.94 27 (61, 47) 0.80 O. lowii 3 (4, 12) 0 18 (39, 34) 0.66 11 (16, 32) 0.01 0 (0, 0) 0.01 1 (2, 5) 0 P. guernei 0 (0, 0) 0 30 (68, 54) 0.85 7.5 (28, 47) 0.04 0 (0, 0) 0.01 0 (0, 0) 0 11 (23, 20) 0.73 S. affinis 9 (19, 17) 0.70 12 (63, 15) 0.99 11 (33, 17) 0.98 0 (0, 0) 0.15 4 (7, 7) 0.70 21 (100, 27) 1 9 (37, 11) 1 A. cornuta C. sloani C. atlantica G. chuni G. indica O. lowii P. guernei C. sloani 0 (0, 0) 0 C. atlantica 1 (2, 3) 0.05 26 (81, 38) 0.99 G. chuni 9 (23, 30) 0.19 0 (0, 0) 0.99 14 (32, 25) 0.80 G. indica 16 (33, 32) 0.49 0 (0, 0) 0.99 20 (41, 25) 0.94 27 (61, 47) 0.80 O. lowii 3 (4, 12) 0 18 (39, 34) 0.66 11 (16, 32) 0.01 0 (0, 0) 0.01 1 (2, 5) 0 P. guernei 0 (0, 0) 0 30 (68, 54) 0.85 7.5 (28, 47) 0.04 0 (0, 0) 0.01 0 (0, 0) 0 11 (23, 20) 0.73 S. affinis 9 (19, 17) 0.70 12 (63, 15) 0.99 11 (33, 17) 0.98 0 (0, 0) 0.15 4 (7, 7) 0.70 21 (100, 27) 1 9 (37, 11) 1 Numbers in parentheses represent the percent overlap of species A (column) with species B (row) and vice versa. Numbers in bold represent shared overlap >50%. Second column of numbers represents the likelihood of differences in SEAc size. Numbers in bold represent statistically significant differences in SEAc size between the pair of species examined. Figure 5. View largeDownload slide Size-corrected SEAc plotted with mean (± s.e.) δ13C and δ15N values for each species. Figure 5. View largeDownload slide Size-corrected SEAc plotted with mean (± s.e.) δ13C and δ15N values for each species. Mean POM δ13C and δ15N values collected from the meso- and bathypelagic were not significantly different from each other, thus mixing models were run using epipelagic POM data and data combined from the meso- and bathypelagic zones. Mixing model results suggest that all consumers included in this study derive the bulk of their carbon from epipelagic POM (Figure 6). Relative contributions of epipelagic POM ranged from 97.87% ± 1.43 in P. guernei to 73.30% ± 3.19 in G. chuni, while contributions from meso- and bathypelagic POM were much lower, ranging from 26.70% ± 3.19 in G. chuni to 2.13% ± 1.43 in P. guernei. Diagnostic plots of posterior distributions revealed a high negative correlation between the two sources (epipelagic POM and meso-/bathypelagic POM). Considering that the producer data fully constrain consumer data when an appropriate trophic enrichment factor is applied and that model diagnostics (Gelman-Rubin Diagnostic: all variables <1.01; Gweke Diagnostic: <5% of variables outside ±1.96 for each chain) indicate that the model fully converged, the negative correlation is likely caused by the similar δ13C signatures of sources and not from a missing carbon source. Figure 6. View largeDownload slide Estimated relative contributions of POM collected from epipelagic and meso- and bathypelagic depths to (a) A. cornuta, (b) C. sloani, (c) C. atlantica, (d) G. chuni, (e) G. indica, (f) O. lowii, (g) P. guernei, and (h) S. affinis. Figure 6. View largeDownload slide Estimated relative contributions of POM collected from epipelagic and meso- and bathypelagic depths to (a) A. cornuta, (b) C. sloani, (c) C. atlantica, (d) G. chuni, (e) G. indica, (f) O. lowii, (g) P. guernei, and (h) S. affinis. Discussion Trophic structure Trophic positions inferred through stable isotope data suggest that, within this group of fishes, the highest trophic positions are held by the largely piscivorous A. cornuta, G. chuni, G. indica, and S. affinis; intermediate trophic positions are occupied by species preying on mixtures of cephalopods and fishes (C. atlantica and O. lowii), and fishes and crustaceans (C. sloani); and the lowest trophic position is occupied by P. guernei, which eats primarily macrocrustaceans (Hopkins et al., 1996; Sutton and Hopkins, 1996b). Stomach content analysis (SCA) was performed on all samples in this study, and though sample sizes with identifiable food items were relatively small, results agree with findings from previous SCA studies and support the trophic relationships inferred through SIA (Supplementary Table S4). For the species examined, δ15N values spanned 5.91 ‰ or 1.9 TLs, while species mean δ15N values spanned 1.96 ‰ and 0.62 TL (assuming TEF of 3.15). Using mean δ15N values and applying a TEF of 3.15, our observed range of estimated trophic levels (0.62) appears to be in line with other studies examining Mediterranean (1.1 TLs), Pacific (1.6 TLs), and GOM (1.1 TLs) fish assemblages that included both micronektonivores (stomiids, anoplogastrids) and lower trophic level zooplanktivores (myctophids, gonostomatids), which have been shown to be up to 0.6 TLs lower than micronektonivores (Valls et al., 2014; Choy et al., 2015; McClain-Counts et al., 2017). Trophic level estimates determined using a primary consumer to set the isotopic baseline placed each species between the third and fourth trophic levels. Where species-specific comparisons of trophic positions could be made, and applying a δ15N TEF of 3.15 to reported mean δ15N values, our estimated TPs for A. cornuta (3.7) and C. sloani (3.2) were similar to estimates from the Pacific (TP = 3.5 for both species) and to the GOM (C. sloani TP = 2.8) (Choy et al., 2015; McClain-Counts et al., 2017). The observed difference in TP estimates for C. sloani in the GOM was likely caused by the inclusion of smaller C. sloani (<50 mm SL) by McClain-Counts et al. (2017). Estimates of TP for S. affinis (3.4) were similar to Stomias boa collected in the Mediterranean (TP = 3.5) (Valls et al., 2014), while estimates of the three stomiid species [C. sloani (3.2), S. affinis (3.4), and P. guernei (3.1)] were within the estimated worldwide TP range of stomiid fishes (TP = 3.0–3.5) (Choy et al., 2012). This study represents the first descriptions of trophic positions using SIA for C. atlantica, G. chuni, G. indica, O. lowii, and P. guernei, so comparisons to TP estimates in other studies were not possible. Isotopic niche size, estimated using SEAc, was largest for fishes occupying the highest TPs within the guild (A. cornuta, C. chuni, G. indica, S. affinis) and smallest in fishes occupying intermediate and lower TPs (Figure 5). The larger SEAc of the highest TP fishes within this guild could suggest more generalized feeding compared with other species. Differences in SEAc can also be influenced by an organism’s size distribution, which was not equally comprehensive in all species. The small SEAc of C. sloani, for example, was likely affected by samples that only included the largest individuals (>140 mm SL). Isotopic niche overlap was common, although the extent of the overlap was typically non-significant (<50%) (Table 3). In species where isotopic niche overlap was significant (S. affinis and O. lowii), available SCA data suggest prey resource overlap is not as strong as isotopic niche overlap would make them appear (Hopkins et al., 1996; Sutton and Hopkins, 1996b). Samples of POM were more 13 C depleted at eastern longitudes that are closer in proximity to the Mississippi River, while fishes became more 13 C enriched at southern latitudes and western longitudes. Shifts in the isotopic signatures of POM in the GOM have been observed between nearshore and offshore regions and between mesoscale cyclonic and anticyclonic oceanographic features (Wissel and Fry, 2005; Dorado et al., 2012, Wells et al., 2017). Baseline differences in POM isotopic signatures between nearshore and offshore environments of the GOM can be caused by phytoplankton in offshore regions relying more heavily on isotopically light nitrogen produced by diazatrophic cyanobacteria (Trichodesmium spp.) (Holl et al., 2007; Dorado et al., 2012), while baseline differences between cyclonic and anticyclonic regions are driven by upwelling within cyclonic features supplying 15 N enriched N2 to phytoplankton (Wells et al., 2017). The pattern of POM samples becoming 13 C depleted at eastern longitudes and fishes becoming 13 C enriched at lower latitudes and western longitudes is consistent with the idea that organisms collected closer to the continental shelf are more likely to be partially supported by terrestrially derived organic matter from the Mississippi River where the effects of Trichodesmium spp. and upwelling on baseline δ15N values are minimal (Dorado et al., 2012). Ontogenetic shifts in δ13C and δ15N Ontogenetic enrichment in 15 N was documented in six of the eight species examined. While significant relationships between δ15N and body size suggest ontogenetic patterns in feeding ecology, observed trends in some cases were driven by a few points, and the nature of size-based relationships with δ15N could change with the inclusion of different size classes and more samples. Observed enrichment in 15 N with body size could be caused by ontogenetic shifts in prey selection, as has been suggested for A. cornuta, C. sloani, and P. guernei or by ingestion of larger sized prey (Hopkins et al., 1996; Sutton and Hopkins, 1996b). For species such as S. affinis and C. sloani, which have been shown to feed on myctophid fishes across their ontogeny, observed positive relationships between δ15N and body size could be a result of ingestion of larger myctophid fishes, which have been shown to become enriched in 15 N with increasing size (Sutton and Hopkins, 1996b; Cherel et al., 2010; McClain-Counts et al., 2017). The negative relationship between body size and δ15N of O. lowii contrasts with published diet data suggesting that O. lowii undergoes an ontogenetic diet shift from eating fishes as juveniles to feeding primarily on squid and fish as adults (Rofen, 1966; Hopkins et al., 1996). Omosudis lowii are known to have highly distensible stomachs and have been reported to feed on prey much larger than themselves (Rofen, 1966). However, the tendency to feed on large prey appears to occur primarily during juvenile stages, as adults have been found to feed on both large and small prey (Rofen, 1966). Thus, the lack of a relationship between SL and δ15N of O. lowii could be a function of adults and juveniles feeding on similarly sized prey or by switching to prey that occupy lower TPs. The observed relationships between δ15N and body size are not necessarily the result of ontogenetic shifts in diet and can instead reflect spatial and temporal changes in the isotopic signature of nitrogen sources at the base of the foodweb (Wells et al., 2017). Spatial variation in the isotopic signatures of primary producers has been documented in the GOM, but the increased movements and longer tissue turnover rates of fishes likely diminishes spatial variation by increasing the likelihood of an organism integrating the isotopic signatures of multiple isotopic baselines. Relative contributions of epi-, meso-, and bathypelagic POM to deep-pelagic fishes A paradigm of deep-sea ecology is that meso- and bathypelagic organisms feed within foodwebs largely supported by epipelagic POM and that POM suspended at deeper depths contributes little carbon to higher order consumers. Recently, through the use of compound-specific stable isotope analysis (CS-SIA) of amino acids (AAs), that paradigm was challenged by evidence which suggests that zooplankton and micronekton can partly rely on small particle (0.7–53 µm) suspended POM as a carbon source (Hannides et al., 2013; Choy et al., 2015; Gloeckler et al., 2018). Choy et al. (2015) estimated the relative contributions of epipelagic and deep-water POM to the production of four fishes (including A. cornuta) in the North Pacific and found that two meso- and bathypelagic zooplanktivores received contributions from small-particle, deep-pelagic suspended POM ranging between 39 and 81%, while contributions to the micronektonivore, A. cornuta, were far less (0–23%). Gloeckler et al. (2018) examined the δ15N values of source AAs from a micronekton assemblage and found that relative contributions of small, suspended particles to micronekton were greatest in non-migratory species with night-time distributions in the lower mesopelagic and upper bathypelagic. Species with night-time distributions within the epi- and mesopelagic, however, were found to be supported by either surface particles or large, fast-sinking particles (>53 µm) at depth (Hannides et al., 2013; Gloeckler et al. 2018). The results from our mixing-model analyses suggest that the majority of carbon (≥73%) supporting the species examined in this study appears to be derived from epipelagic sources or from fast-sinking particles at depth which carry similar isotopic signatures to particles within the epipelagic (Hannides et al., 2013). These contribution estimates, combined with vertical distribution data which suggest the collective night-time distributions of these predatory fishes span the epi-, meso-, and upper bathypelagic (Sutton and Hopkins, 1996a; Sutton et al., 2010), are in alignment with estimations for micronekton with similar depth distributions made by Choy et al. (2015) and Gloeckler et al. (2018). It should be noted that the relative contribution of small suspended particles at depth to these species cannot be fully assessed without conducting CSIA-AA and that further investigation into the relative importance of small particles to higher trophic level consumers is warranted (Gloeckler et al., 2018). Additional support for the assertion that these species are largely supported by surface derived carbon is provided by diet studies, which suggest that these species consume migratory prey that feed within food chains supported by surface production (Hopkins et al., 1996; Sutton and Hopkins, 1996b), highlighting the extent to which spatially distinct consumers are connected in the northern GOM. Acknowledgements We thank Michael Novotny, Nina Pruzinski, and Matthew Woodstock for their help organizing and processing samples. We thank two anonymous reviewers for their comments, which greatly improved this paper. This research was made possible by a grant from the Gulf of Mexico Research Initiative. Data are publicly available through the Gulf of Mexico Research Initiative Information & Data Cooperative (GRIIDC) at https://data.gulfresearchinitiative.org (doi: 10.7266/N7ZS2V04). References Angel M. V. 1997 . What is the deep sea? Fish Physiology , 16 : 1 – 41 . Google Scholar Crossref Search ADS Angel M. V. , Baker A. 1982 . Vertical distribution of the standing crop of plankton and micronekton at three stations in the northeast Atlantic . Biological Oceanography , 2 : 1 – 30 . Caut S. , Angulo E. , Courchamp F. 2008 . Caution on isotopic model use for analyses of consumer diet . Canadian Journal of Zoology , 86 : 438 – 445 . Google Scholar Crossref Search ADS Cherel Y. , Ducatez S. , Fontaine C. , Richard P. , Guinet C. 2008 . Stable isotopes reveal the trophic position and mesopelagic fish diet of female southern elephant seals breeding on the Kerguelen Islands . Marine Ecology Progress Series , 370 : 239 – 247 . Google Scholar Crossref Search ADS Cherel Y. , Fontaine C. , Richard P. , Labatc J. P. 2010 . Isotopic niches and trophic levels of myctophid fishes and their predators in the Southern Ocean . Limnology and Oceanography , 55 : 324 – 332 . Google Scholar Crossref Search ADS Choy C. A. , Davison P. C. , Drazen J. C. , Flynn A. , Gier E. J. , Hoffman J. C. , McClain-Counts J. P. 2012 . Global trophic position comparison of two dominant mesopelagic fish families (Myctophidae, Stomiidae) using amino acid nitrogen isotopic analyses . PLoS One , 7 : e50133. Google Scholar Crossref Search ADS PubMed Choy C. A. , Popp B. N. , Hannides C. C. S. , Drazen J. C. 2015 . Trophic structure and food resources of epipelagic and mesopelagic fishes in the North Pacific Subtropical Gyre ecosystem inferred from nitrogen isotopic compositions . Limnology and Oceanography , 60 : 1156 – 1171 . Google Scholar Crossref Search ADS Choy C. A. , Portner E. , Iwane M. , Drazen J. C. 2013 . Diets of five important predatory mesopelagic fishes of the central North Pacific . Marine Ecology Progress Series , 492 : 169 – 184 . Google Scholar Crossref Search ADS Clarke T. A. 1982 . Feeding habits of stomiatoid fishes from Hawaiian waters . Fishery Bulletin, US , 80 : 287 – 304 . Clarke T. A. , Wagner P. J. 1976 . Vertical distribution and other aspects of the ecology of certain mesopelagic fishes taken near Hawaii . Fishery Bulletin, US , 74 : 635 – 646 . Danovaro R. , Gambi C. , Dell'Anno A. , Corinaldesi C. , Fraschetti S. , Vanreusel A. , Vincx M. 2008 . Exponential decline of deep-sea ecosystem functioning linked to benthic biodiversity loss . Current Biology , 18 : 1 – 8 . Google Scholar Crossref Search ADS PubMed Dorado S. , Rooker J. , Wissel B. , Quigg A. 2012 . Isotope baseline shifts in pelagic food webs of the Gulf of Mexico . Marine Ecology Progress Series , 464 : 37 – 49 . Google Scholar Crossref Search ADS Drazen J. C. , Sutton T. T. 2017 . Dining in the deep: the feeding ecology of deep-sea fishes . Annual Review of Marine Science , 9 : 337 – 366 . Google Scholar Crossref Search ADS PubMed Fernández-Carrera A. , Rogers K. L. , Weber S. C. , Chanton J. P. , Montoya J. P. 2016 . Deep water horizon oil and methane carbon entered the food web in the Gulf of Mexico . Limnology and Oceanography , 61 : S387 – S400 . Google Scholar Crossref Search ADS Gloeckler K. , Choy C. A. , Hannides C. C. S. , Close H. G. , Goetze E. , Popp B. N. , Drazen J. C. 2018 . Stable isotope analysis of micronekton around Hawaii reveals suspended particles are an important nutritional source in the lower mesopelagic and upper bathypelagic zones . Limnology and Oceanography , 63 : 1168 – 1180 . Google Scholar Crossref Search ADS Gjøsaeter J. , Kawaguchi K. 1980 . A review of the world resources of mesopelagic fish . FAO Fisheries Technical Report No , 193 : 151 pp. Hannides C. C. S. , Popp B. N. , Choy C. A. , Drazen J. C. 2013 . Midwater zooplankton and suspended particle dynamics in the North Pacific Subtropical Gyre: a stable isotope perspective . Limnology and Oceanography , 58 : 1931 – 1946 . Google Scholar Crossref Search ADS Hoffman J. C. , Sutton T. T. 2010 . Lipid correction for carbon stable isotope analysis of deep-sea fishes . Deep-Sea Research Part I: Oceanographic Research Papers , 57 : 956 – 964 . Google Scholar Crossref Search ADS Holl C. M. , Villareal T. A. , Payne C. D. , Clayton T. D. , Hart C. , Montoya J. P. 2007 . Trichodesmium in the western Gulf of Mexico: 15N2-fixation and natural abundance stable isotope evidence . Limnology and Oceanography , 52 : 2249 – 2259 . Google Scholar Crossref Search ADS Hopkins T. L. , Gartner J. V. 1992 . Resource-partitioning and predation impact of a low-latitude myctophid community . Marine Biology , 114 : 185 – 197 . Google Scholar Crossref Search ADS Hopkins T. L. , Sutton T. T. , Lancraft T. M. 1996 . The trophic structure and predation impact of a low latitude midwater fish assemblage . Progress in Oceanography , 38 : 205 – 239 . Google Scholar Crossref Search ADS Hussey N. E. , Macneil M. A. , McMeans B. C. , Olin J. A. , Dudley S. F. J. , Cliff G. , Wintner S. P. et al. 2014 . Rescaling the trophic structure of marine food webs . Ecology Letters , 17 : 239 – 250 . Google Scholar Crossref Search ADS PubMed Irigoien X. , Klevjer T. A. , Røstad A. , Martinez U. , Boyra G. , Acuña J. L. , Bode A. et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean . Nature Communications , 5 : 3271. Google Scholar Crossref Search ADS PubMed Jackson A. L. , Inger R. , Parnell A. C. , Bearhop S. 2011 . Comparing isotopic niche widths among and within communities: SIBER – Stable Isotope Bayesian Ellipses in R . Journal of Animal Ecology , 80 : 595 – 602 . Google Scholar Crossref Search ADS PubMed McCann K. S. 2000 . The diversity–stability debate . Nature , 405 : 228 – 233 . Google Scholar Crossref Search ADS PubMed McClain-Counts J. P. , Demopoulos A. W. J. , Ross S. W. 2017 . Trophic structure of mesopelagic fishes in the Gulf of Mexico revealed by gut content and stable isotope analyses . Marine Ecology , 38 : e12449 – e12423 . Google Scholar Crossref Search ADS McEachran J. D. , Fechhelm J. D. 1998 . Fishes of the Gulf of Mexico, Volume 1: Myxiniformes to Gasterosteiformes . University of Texas Press , Austin . 1120 pp. Mengerink K. J. , Dover C. L. , Van Ardron J. , Baker M. , Escobar-briones E. , Gjerde K. , Koslow J. A. et al. 2014 . A call for deep-ocean stewardship . Science , 344 : 696 – 698 . Google Scholar Crossref Search ADS PubMed Moore J. A. , Vecchione M. , Collette B. B. , Gibbons R. , Hartel K. E. , Galbraith J. K. , Turnipseed M. et al. 2003 . Biodiversity of Bear Seamount, New England Seamount chain: results of exploratory trawling . Journal of Northwest Atlantic Fishery Science , 31 : 363 – 372 . Google Scholar Crossref Search ADS Morato T. , Watson R. , Pitcher T. J. , Pauly D. 2006 . Fishing down the deep . Fish and Fisheries , 7 : 24 – 34 . Google Scholar Crossref Search ADS Moteki M. , Arai M. , Tsuchiya K. , Okamoto H. 2001 . Composition of piscine prey in the diet of large pelagic fish in the eastern tropical Pacific Ocean . Fisheries Science , 67 : 1063 – 1074 . Google Scholar Crossref Search ADS Pakhomov E. A. , Perissinotto R. , McQuaid C. D. 1996 . Prey composition and daily rations of myctophid fishes in the Southern Ocean . Marine Ecology Progress Series , 134 : 1 – 14 . Google Scholar Crossref Search ADS Pauly D. , Trites A. W. , Capuli E. , Christensen V. 1998 . Diet composition and trophic levels of marine mammals . ICES Journal of Marine Science , 55 : 467 – 481 . Google Scholar Crossref Search ADS Peterson B. J. , Fry B. 1987 . Stable isotopes in ecosystem studies . Annual Review of Ecology and Systematics , 18 : 293 – 320 . Google Scholar Crossref Search ADS Polis G. A. , Strong D. R. 1996 . Food web complexity and community dynamics . The American Naturalist , 147 : 813 – 846 . Google Scholar Crossref Search ADS Porteiro F. M. , Sutton T. T. 2007 . Midwater fish assemblages and seamounts. In Seamounts: Ecology, Conservation and Management , pp. 101 – 116 . Ed. by Pitcher T. J. , Morato T. , Hart P. J. B. , Clark M. R. , Haggan N. , Santos R. S. . Blackwell, Fish and Aquatic Resources Series , Oxford . 552 pp. Post D. M. 2002 . Using stable isotopes to estimate trophic position: models, methods, and assumptions . Ecology , 83 : 703 – 718 . Google Scholar Crossref Search ADS Post D. M. , Layman C. A. , Arrington D. A. , Takimoto G. , Quattrochi J. , Montaña C. G. 2007 . Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable isotope analyses . Oecologia , 152 : 179 – 189 . Google Scholar Crossref Search ADS PubMed R Development Core Team. 2016 . R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. www.R-project.org. Raclot T. , Groscolas R. , Cherel Y. 1998 . Fatty acid evidence for the importance of myctophid fishes in the diet of king penguins, Aptenodytes patagonicus . Marine Biology , 132 : 523 – 533 . Google Scholar Crossref Search ADS Ramirez-Llodra E. , Tyler P. A. , Baker M. C. , Bergstad O. A. , Clark M. R. , Escobar E. , Levin L. A. et al. 2011 . Man and the last great wilderness: human impact on the deep sea . PLoS One , doi.org/10.1371/journal.pone.0022588 Robison B. H. 2004 . Deep pelagic biology . Journal of Experimental Marine Biology and Ecology , 300 : 253 – 272 . Google Scholar Crossref Search ADS Robison B. H. 2009 . Conservation of deep pelagic biodiversity. Conservation . Biology , 23 : 847 – 858 . Rofen R. R. , 1966 . Family Omosudidae. In Fishes of the Western North Atlantic. Memoir No. 1. Part 5 , pp. 464 – 481 . Ed. by Bigelow H. G. , Breder C. M. , Olsen Y. H. , Cohen D. M. , Schroeder W. C. , Mead G. H. , Schultz L. P. et al. Sears Foundation for Marine Research, Yale University , New Haven, CT . Stock B. C. , Semmens B. X. 2015 . MixSIAR GUI user manual, version 1.0. http://conserver.iugo-cafe.org/user/brice.semmens. Sutton T. T. 2013 . Vertical ecology of the pelagic ocean: classical patterns and new perspectives . Journal of Fish Biology , 83 : 1508 – 1527 . Google Scholar Crossref Search ADS PubMed Sutton T. T. , Hopkins T. L. 1996a . Species composition, abundance, and vertical distribution of the stomiid (Pisces: Stomiiformes) assemblage in the Gulf of Mexico . Bulletin of Marine Science , 59 : 530 – 542 . Sutton T. T. , Hopkins T. L. 1996b . Trophic ecology of the stomiid (Pisces: Stomiidae) fish assemblage of the eastern Gulf of Mexico: strategies, selectivity and impact of a top mesopelagic predator group . Marine Biology , 127 : 179 – 192 . Google Scholar Crossref Search ADS Sutton T. T. , Porteiro F. M. , Heino M. , Byrkjedal I. , Langhelle G. , Anderson C. I. H. , Horne J. et al. 2008 . Vertical structure, biomass and topographic association of deep-pelagic fishes in relation to a mid-ocean ridge system . Deep Sea Research Part II: Topical Studies in Oceanography , 55 : 161 – 184 . Google Scholar Crossref Search ADS Sutton T. T. , Wiebe P. H. , Madin L. , Bucklin A. 2010 . Diversity and community structure of pelagic fishes to 5000 m depth in the Sargasso Sea . Deep-Sea Research Part II: Topical Studies in Oceanography , 57 : 2220 – 2233 . Google Scholar Crossref Search ADS Sweeting C. J. , Barry J. , Barnes C. , Polunin N. V. C. , Jennings S. 2007 . Effects of body size and environment on diet-tissue15N fractionation in fishes . Journal of Experimental Marine Biology and Ecology , 340 : 1 – 10 . Google Scholar Crossref Search ADS Sweeting C. J. , Barry J. T. , Polunin N. V. C. , Jennings S. 2007 . Effects of body size and environment on diet-tissue13C fractionation in fishes . Journal of Experimental Marine Biology and Ecology , 352 : 165 – 176 . Google Scholar Crossref Search ADS Thurber A. R. , Sweetman A. K. , Narayanaswamy B.E. , Jones D. O. B. , Ingels J. , Hansman R. L. 2014 . Ecosystem function and services provided by the deep sea . Biogeosciences , 11 : 3941 – 3963 . Google Scholar Crossref Search ADS Trueman C. N. , Johnston G. , O'Hea B. , MacKenzie K. M. 2014 . Trophic interactions of fish communities at midwater depths enhance long-term carbon storage and benthic production on continental slopes . Proceedings of the Royal Society B: Biological Sciences , 281 : 20140669 – 20140669 . Google Scholar Crossref Search ADS Valls M. , Olivar M. P. , Fernández de Puelles M. L. , Molí B. , Bernal A. , Sweeting C. J. 2014 . Trophic structure of mesopelagic fishes in the western Mediterranean based on stable isotopes of carbon and nitrogen . Journal of Marine Systems , 138 : 160 – 170 . Google Scholar Crossref Search ADS Webb T. J. , Vanden Berghe E. , O'Dor R. 2010 . Biodiversity’s big wet secret: the global distribution of marine biological records reveals chronic under-exploration of the deep pelagic ocean . PLoS One , 5 : e10223 – e10226 . Google Scholar Crossref Search ADS PubMed Wells R. J. D. , Rooker J. R. , Quigg A. , Wissel B. 2017 . Influence of mesoscale oceanographic features on pelagic food webs in the Gulf of Mexico . Marine Biology , 164 : 1 – 11 . Google Scholar Crossref Search ADS PubMed Wilson R. W. , Millero F. J. , Taylor J. R. , Walsh P. J. , Christensen V. , Jennings S. , Grosell M. 2009 . Contribution of fish to the marine inorganic carbon cycle . Science , 323 : 359 – 362 . Google Scholar Crossref Search ADS PubMed Wissel B. , Fry B. 2005 . Tracing Mississippi River influences in estuarine food webs of coastal Louisiana . Oecologia , 144 : 659 – 672 . Google Scholar Crossref Search ADS PubMed © International Council for the Exploration of the Sea 2018. All rights reserved. For permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Contreras,, Tabit;Olivar, M, Pilar;Hulley, P, Alexander;Fernández, de Puelles, M Luz

doi: 10.1093/icesjms/fsy070pmid: N/A

Abstract We analysed the trophic ecology of the early ontogenetic stages of six mesopelagic fish species (Bathylagoides argyrogaster, Argyropelecus sladeni, Sternoptyx diaphana, Diaphus vanhoeffeni, Hygophum macrochir, and Myctophum affine), which have different morphologies, vertical distributions, and taxonomic affiliations. The larvae and transforming stages of the sternoptychids fed both during the day and at night. However, larvae of the other species fed during the day, as they apparently rely on light for prey capture. The transforming stages of myctophids showed a similar daylight feeding pattern to their larvae, but in D. vanhoeffeni both day and night feeding was evident, thereby indicating the progressive change toward the adult nocturnal feeding pattern. The number of prey and their maximum sizes were linked to predator gut morphology and gape size. Although the maximum prey size increased with predator development, postflexion larvae and transforming stages also preyed on small items, so that the trophic niche breath did not show evidence of specialization. In all the species, copepods dominated the larval diet, but the transforming stages were characterized by increasing diet diversity. Despite the poor development of these early stages, Chesson’s selectivity index calculated for larvae and transforming stages showed positive selection for particular prey. Introduction The mesopelagic zone is generally considered to lie between 200 and 1000 m depth in the water column, although these values may vary slightly in different parts of the World Ocean (Reygondeau et al., 2017), and is characterized by low light conditions. Mesopelagic fishes are one of the most common components in open ocean samples (Gjøsaeter and Kawaguchi, 1980; McGinnis, 1982). Their larvae have also been reported as being the most abundant in ichthyoplankton samples (Moser and Ahlstrom, 1970, 1996). The fishes inhabiting this zone belong to taxa from the Orders Myctophiformes, Stomiiformes, Anguilliformes, Argentiniformes, Aulopiformes, Lophiiformes, and Stephanoberyciformes (Weitzman, 1997). Although all these groups may co-exist at a particular depth in the water column during the day, differential diel vertical migratory behaviours have been reported for most myctophid species, and for certain stomiiforms (families Phosichthyidae and Stomiidae) (Baird, 1971; Merrett and Roe, 1974; Hulley, 1984; Olivar et al., 2017). The migratory fishes follow the nightly zooplankton migration, ascending into the epipelagic layers to feed, and descending to mesopelagic layers during the day to avoid predators and to digest their food (Baird et al., 1975; Hopkins and Baird, 1985; Gartner et al., 1997; Mehner and Kasprzak, 2011; Bernal et al., 2013, 2015; Sutton, 2013). While the adult fishes may have wide ranges in their vertical distributions, their larval stages demonstrate a more limited vertical depth range, mainly between the surface and 200 m. They only perform very restricted vertical displacements, and therefore feed mainly in the upper water layers (Loeb, 1979; Sabatés, 2004; Sassa and Kawaguchi, 2004, Sassa et al., 2007; Olivar et al., 2014, 2018). Feeding ecology and the diets of mesopelagic fishes, based on stomach content analyses, have been mainly investigated for the adult stages, and particularly in myctophids (Clarke, 1980; Kinzer and Schulz, 1985; Hopkins and Gartner, 1992; Rissik and Suthers, 2000; Watanabe et al., 2002; Bernal et al., 2013, 2015; McClain-Counts et al., 2017) and in stomiiform species (Sutton and Hopkins, 1996; Champalbert et al., 2008; Carmo et al., 2015; McClain-Counts et al., 2017). These fishes are mostly opportunistic zooplankton feeders, but the diets of some species also include particulate organic matter and small fish (Palma, 1990; Hopkins and Gartner, 1992; Watanabe and Kawaguchi, 2003; Bernal et al., 2015). Knowledge of larval feeding is limited to fewer species (e.g. Palma, 1990; Sabatés and Saiz, 2000; Conley and Hopkins, 2004; Sassa and Kawaguchi, 2004; Landaeta et al., 2011 for stomiiforms; Bernal et al., 2013; and Contreras et al., 2015, for myctophids). Information on feeding in transforming stages is even more scarce (Contreras et al., 2015). These studies have reported that the larvae of mesopelagic fishes appear to feed on small zooplankton items, and that their diets are related both to availability of prey and to larval development. While prey size is one of the most important factors influencing prey capture, other factors can influence prey capture, such as prey abundance, prey colour, and the swimming behaviour of prey, so indicating that fish larvae might not feed at random but may have selective capacity (Hunter, 1981; Govoni et al., 1986; Llopiz, 2013; Robert et al., 2014). Among those larval features related to feeding, the main constraints are gape size, swimming skill, and the development of sensory organs, in addition to larval behaviour itself (Hubbs and Blaxter, 1986; Browman and O'Brien, 1992). The main environmental factor influencing larval feeding is the light condition, because most fish larvae are visual feeders (Blaxter, 1986; Huse, 1994). Information on the distribution and abundance of mesopelagic fishes in the equatorial and tropical Atlantic is relatively common (Hulley, 1981; Hulley and Krefft, 1985; Hulley and Paxton, 2016a, b; Olivar et al., 2017). Investigations on their larval stages have been focused in regions close to the continents (e.g. Badcock and Merrett, 1976; de Castro et al., 2010; Bonecker et al., 2012; Moyano et al., 2014; Olivar et al., 2016; Namiki et al., 2017), but recent research by Olivar et al., (2018) has analysed the overall distribution and abundance patterns across the Atlantic, showing that larvae of mesopelagic fishes dominate the first 100 m of the water column everywhere. For the present investigation, we analysed the trophic ecology of larval and transforming stages in six mesopelagic species with different larval morphologies, and different vertical distributions: Bathylagoides argyrogaster (Bathylagidae), Argyropelecus sladeni and Sternoptyx diaphana (Sternoptychidae), Diaphus cf. vanhoeffeni, Hygophum macrochir, and Myctophum affine (Myctophidae). Knowledge of the feeding behaviour of the larvae of these species is lacking, and only feeding data on the juvenile stages of A. sladeni and S diaphana have been published (Hopkins and Baird, 1973). The present study compares feeding incidence (FI), size spectra, trophic niche breadth, and diet composition to determine if the larvae and transforming stages of the six species have specific feeding patterns, which can be correlated with their ontogenetic development, vertical distribution, and morphology. Material and methods Sampling In order to characterize the mesopelagic fauna and its environment, a survey comprising a transect of 12 stations was undertaken during April 2015 across the tropical and equatorial Atlantic on board Research Vessel Hesperides (82 m × 15 m). The cruise extended from near the Brazilian coast to south of the Canary Islands, regions where bottom depths range from 3000 to 5200 m (Figure 1) (Olivar et al., 2017, 2018). Fish larvae were collected at 11 stations from 8 to 28 of April. Both day and night plankton samples were obtained at each station within a 24-h period. At each station, oblique tows were undertaken using a MOCNESS-1 net (mouth opening of 1 m2), fitted with 8 nets of 200 μm mesh size. Samples were taken in the following depth strata: 800–600 m, 600–500 m, 500–400 m, 400–300 m, 300–200 m, the lower thermocline layer (ca. 200–100 m), thermocline (ca. 50–100), and the upper mixed layer (ca. 50–0 m). During trawling, the ship’s speed was maintained at 1.5–2.5 knots, and the winch retrieval rate was 20 m/min. The total duration of each haul ranged from 5 to 10 min, except for the deepest layer in which the mean duration was 24 min. The mean volume of water sampled per layer was 470.8 m3 (SD 236.6), ranging between ca. 300 m3 (the shallowest layer) to 870 m3 (the deepest and broadest layer), and with fairly similar volume vs. time ratios between layers (mean 50.7; SD 6.7 m3/min). Figure 1. View largeDownload slide Stations sampled with the MOCNESS-1 net (day sample = circle; night sample = cross). Figure 1. View largeDownload slide Stations sampled with the MOCNESS-1 net (day sample = circle; night sample = cross). In addition to the mesozooplankton samples obtained with the MOCNESS-1 net, microzooplankton samples were collected by vertical hauls with a Calvet net (0.25 m diameter and 0.53 μm mesh size), between 200 m and the surface. Zooplankton samples were preserved in 5% buffered formalin and kept in the dark until later investigation at the laboratory. Laboratory analysis All fishes were sorted and identified to the lowest possible taxon. Larval identifications follow Olivar and Fortuño (1991); Moser and Ahlstrom (1996); Richards (2006); and Fahay (2007). Some 1134 specimens comprising the families Bathylagidae, Sternoptychidae, and Myctophidae were analysed for gut content determination: 93 Bathylagidae (B. argyrogaster), 344 Sternoptychidae (S. diaphana and A. sladeni), and 697 Myctophidae (M. affine, H. macrochir, and Diaphus cf. vanhoeffeni). Due to the low abundance of specimens found below 200 m, data from the region were combined and analysed as two strata: 200—500 and 500—800 m. Previous papers dealing with the main biological and environmental features during the survey (Olivar et al., 2017,, 2018) had differentiated four broad zones across the transect: western sector (from station #2 to station #6); central sector (from station #7 to station #10), upwelling station (#11), and station #12, south of the Canary Islands (Figure 1). Although the actual number of specimens with content in their guts does not allow for detailed comparisons between stations, layers, species, and stages, the overall diets of larvae and transforming stages of the different species, in each of the above zones, were examined through multivariate analysis. Species were grouped according to their developmental stage: larvae (preflexion, flexion, and postflexion, according to the degree of notochordal flexion) and transforming stage (body becomes thicker and the photophores appear, but the squamation has not yet been developed) (Table 1). Specimens were measured using a microscope equipped with an ocular micrometer to an accuracy of 0.1 mm. Before gut dissection, the following measurements were recorded: standard length (SL); lower jaw length (LJL)—measured from the tip of the snout to the junction with the maxilla; upper jaw length (UJL)—measured from the tip of the snout to the posterior end of the maxilla; and mouth width (MW) —measured ventrally as the widest distance between the posterior edges of the maxillae. Table 1. Day and night FI% by developmental stage for the six studied species: Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), Sternoptyx diaphana, Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir, and Myctophum affine. Preflexion larvae Flexion larvae Postflexion larvae Transformation Species %FI day % FI night %FI day % FI night %FI day % FI night %FI day % FI night B. argyrogaster Standard length: <6.1 mm Standard length: 6.1–8.1 mm Standard length: 8.2–12.0 mm N/D 80 0 66.7 0 20 0 N/D N/D (a15; b12) (a14; b0) (a18; b12) (a10; b0) (a5; b1) (a6; b0) A. sladeni. Standard length: <7.5 mm Standard length: 7.5–9.4 mm Standard length: 9.5–12.0 mm Standard length: 7.9–13.0 mm 25 42.9 0 0 0 0 87.5 60 (a4; b1) (a7; b3) (a1; b0) (a8; b0) (a1; b0) (a2; b0) (a8; b7) (a15; b9) S. diaphana Standard length: <6.0 mm Standard length: 6.0–9.7 mm Standard length: 6.3–8.7 mm Standard length: 6.0–14.0 mm 27.3 26.3 42.9 40.9 67.6 20 78.6 86.4 (a11; b3) (a19; b5) (a14; b6) (a22; b9) (a37; b25) (a30; b6) (a28; b22) (a22; b19) D. vanhoeffeni Standard length: ≤4.0 mm Standard length: 4.1–5.0 mm Standard length: 5.1–9.9 mm Standard length: 10.0–14.0 mm 11.1 0 11.1 0 3.5 0 87.2 92.1 (a27; b3) (a2; b0) (a81; b9) (a5; b0) (a85; b3) (a11; b0) (a39; b34) (a35; b38) H. macrochir Standard length: <5.0 mm Standard length: 5.0–6.0 mm Standard length: 6.0–11.0 mm Standard length: 11.1–18.2 mm 28.6 0 21.2 0 3.6 0 14.3 0 (a49; b14) (a21; b0) (a19; b4) (a9; b0) (a28; b1) (a11; b0) (a35; b5) (a11; b0) M. affine Standard length: <4.2 mm Standard length: 4.2–6.0 mm Standard length: 6.1–11.4 mm Standard length: 11.5–15.5 mm 54.5 0 25 0 30 0 100 N/D (a22; b12) (a10; b0) (a28; b7) (a13; b0) (a10; b3) (a7; b0) (a3; b3) Preflexion larvae Flexion larvae Postflexion larvae Transformation Species %FI day % FI night %FI day % FI night %FI day % FI night %FI day % FI night B. argyrogaster Standard length: <6.1 mm Standard length: 6.1–8.1 mm Standard length: 8.2–12.0 mm N/D 80 0 66.7 0 20 0 N/D N/D (a15; b12) (a14; b0) (a18; b12) (a10; b0) (a5; b1) (a6; b0) A. sladeni. Standard length: <7.5 mm Standard length: 7.5–9.4 mm Standard length: 9.5–12.0 mm Standard length: 7.9–13.0 mm 25 42.9 0 0 0 0 87.5 60 (a4; b1) (a7; b3) (a1; b0) (a8; b0) (a1; b0) (a2; b0) (a8; b7) (a15; b9) S. diaphana Standard length: <6.0 mm Standard length: 6.0–9.7 mm Standard length: 6.3–8.7 mm Standard length: 6.0–14.0 mm 27.3 26.3 42.9 40.9 67.6 20 78.6 86.4 (a11; b3) (a19; b5) (a14; b6) (a22; b9) (a37; b25) (a30; b6) (a28; b22) (a22; b19) D. vanhoeffeni Standard length: ≤4.0 mm Standard length: 4.1–5.0 mm Standard length: 5.1–9.9 mm Standard length: 10.0–14.0 mm 11.1 0 11.1 0 3.5 0 87.2 92.1 (a27; b3) (a2; b0) (a81; b9) (a5; b0) (a85; b3) (a11; b0) (a39; b34) (a35; b38) H. macrochir Standard length: <5.0 mm Standard length: 5.0–6.0 mm Standard length: 6.0–11.0 mm Standard length: 11.1–18.2 mm 28.6 0 21.2 0 3.6 0 14.3 0 (a49; b14) (a21; b0) (a19; b4) (a9; b0) (a28; b1) (a11; b0) (a35; b5) (a11; b0) M. affine Standard length: <4.2 mm Standard length: 4.2–6.0 mm Standard length: 6.1–11.4 mm Standard length: 11.5–15.5 mm 54.5 0 25 0 30 0 100 N/D (a22; b12) (a10; b0) (a28; b7) (a13; b0) (a10; b3) (a7; b0) (a3; b3) Numbers in parenthesis indicate the total number of analysed specimens (a), and the number of specimens with gut content (b). N/D = No data. Table 1. Day and night FI% by developmental stage for the six studied species: Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), Sternoptyx diaphana, Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir, and Myctophum affine. Preflexion larvae Flexion larvae Postflexion larvae Transformation Species %FI day % FI night %FI day % FI night %FI day % FI night %FI day % FI night B. argyrogaster Standard length: <6.1 mm Standard length: 6.1–8.1 mm Standard length: 8.2–12.0 mm N/D 80 0 66.7 0 20 0 N/D N/D (a15; b12) (a14; b0) (a18; b12) (a10; b0) (a5; b1) (a6; b0) A. sladeni. Standard length: <7.5 mm Standard length: 7.5–9.4 mm Standard length: 9.5–12.0 mm Standard length: 7.9–13.0 mm 25 42.9 0 0 0 0 87.5 60 (a4; b1) (a7; b3) (a1; b0) (a8; b0) (a1; b0) (a2; b0) (a8; b7) (a15; b9) S. diaphana Standard length: <6.0 mm Standard length: 6.0–9.7 mm Standard length: 6.3–8.7 mm Standard length: 6.0–14.0 mm 27.3 26.3 42.9 40.9 67.6 20 78.6 86.4 (a11; b3) (a19; b5) (a14; b6) (a22; b9) (a37; b25) (a30; b6) (a28; b22) (a22; b19) D. vanhoeffeni Standard length: ≤4.0 mm Standard length: 4.1–5.0 mm Standard length: 5.1–9.9 mm Standard length: 10.0–14.0 mm 11.1 0 11.1 0 3.5 0 87.2 92.1 (a27; b3) (a2; b0) (a81; b9) (a5; b0) (a85; b3) (a11; b0) (a39; b34) (a35; b38) H. macrochir Standard length: <5.0 mm Standard length: 5.0–6.0 mm Standard length: 6.0–11.0 mm Standard length: 11.1–18.2 mm 28.6 0 21.2 0 3.6 0 14.3 0 (a49; b14) (a21; b0) (a19; b4) (a9; b0) (a28; b1) (a11; b0) (a35; b5) (a11; b0) M. affine Standard length: <4.2 mm Standard length: 4.2–6.0 mm Standard length: 6.1–11.4 mm Standard length: 11.5–15.5 mm 54.5 0 25 0 30 0 100 N/D (a22; b12) (a10; b0) (a28; b7) (a13; b0) (a10; b3) (a7; b0) (a3; b3) Preflexion larvae Flexion larvae Postflexion larvae Transformation Species %FI day % FI night %FI day % FI night %FI day % FI night %FI day % FI night B. argyrogaster Standard length: <6.1 mm Standard length: 6.1–8.1 mm Standard length: 8.2–12.0 mm N/D 80 0 66.7 0 20 0 N/D N/D (a15; b12) (a14; b0) (a18; b12) (a10; b0) (a5; b1) (a6; b0) A. sladeni. Standard length: <7.5 mm Standard length: 7.5–9.4 mm Standard length: 9.5–12.0 mm Standard length: 7.9–13.0 mm 25 42.9 0 0 0 0 87.5 60 (a4; b1) (a7; b3) (a1; b0) (a8; b0) (a1; b0) (a2; b0) (a8; b7) (a15; b9) S. diaphana Standard length: <6.0 mm Standard length: 6.0–9.7 mm Standard length: 6.3–8.7 mm Standard length: 6.0–14.0 mm 27.3 26.3 42.9 40.9 67.6 20 78.6 86.4 (a11; b3) (a19; b5) (a14; b6) (a22; b9) (a37; b25) (a30; b6) (a28; b22) (a22; b19) D. vanhoeffeni Standard length: ≤4.0 mm Standard length: 4.1–5.0 mm Standard length: 5.1–9.9 mm Standard length: 10.0–14.0 mm 11.1 0 11.1 0 3.5 0 87.2 92.1 (a27; b3) (a2; b0) (a81; b9) (a5; b0) (a85; b3) (a11; b0) (a39; b34) (a35; b38) H. macrochir Standard length: <5.0 mm Standard length: 5.0–6.0 mm Standard length: 6.0–11.0 mm Standard length: 11.1–18.2 mm 28.6 0 21.2 0 3.6 0 14.3 0 (a49; b14) (a21; b0) (a19; b4) (a9; b0) (a28; b1) (a11; b0) (a35; b5) (a11; b0) M. affine Standard length: <4.2 mm Standard length: 4.2–6.0 mm Standard length: 6.1–11.4 mm Standard length: 11.5–15.5 mm 54.5 0 25 0 30 0 100 N/D (a22; b12) (a10; b0) (a28; b7) (a13; b0) (a10; b3) (a7; b0) (a3; b3) Numbers in parenthesis indicate the total number of analysed specimens (a), and the number of specimens with gut content (b). N/D = No data. The entire gut of each specimen was removed for further investigation. For transforming stages, only the stomach contents were considered for analysis, and prey present in the oesophagus were discarded. Prey items were extracted using a fine needle, placed in a drop of 50% solution of glycerine-distilled water on a glass slide, and were teased out for identification, enumeration, and measurement. The maximum cross-section of each prey item was measured to a precision of 0.001 mm under a stereomicroscope (Leica MZ12, reaching 100× magnification) using a micrometric eye-piece. Identifications were made to coarse taxonomic groups, except for copepods in which identification was to genus level where possible. The identification guides employed were Vives and Shmeleva (2007, 2010) and Rose and Tregouboff (1957). Data analysis Allometric relationships between mouth size and body size were determined by fitting a power function, with the slope of the function representing the allometric coefficient, and confidence intervals of the slope were calculated at the 95% level. The FI was estimated as the percentage of examined specimens containing at least one prey item in the stomach (Arthur, 1976) and was differentiated by day and by night. For each species the trophic niche breadth was analysed according to Pearre (1986) as the standard deviation (SD) of the log 10 transformed maximum prey width, plotted against the SL. The larvae were grouped into 0.12 mm size intervals to produce the maximum number of size classes containing at least three or more prey items. The contribution of the different food categories in the diet of larvae and transforming stages was estimated as their percentage frequency of occurrence (%F) and in terms of their numerical abundance (%N), calculated as the proportion of prey items of a given category to the total number of diet items examined in those larvae with food in their gut. The product of these two values was taken as the percentage index of relative importance of each diet item (%IRI) following Govoni et al. (1986). To assess whether species show selectivity for a particular prey, data from the gut content of individuals collected at station #8 (where all the species occur) were analysed in relation to the abundance of zooplankton (micro- and mesozooplankton, defined as <53 and <200 µm, respectively) obtained at the same station. Selectivity by the larvae was calculated for the two most abundant microzooplankton components, namely nauplii and copepodites of <0.2 mm (4489 and 1560 individuals/m3, respectively). For transforming stages, the most common mesozooplankton prey items in each species were considered, and their abundances in the same MOCNESS-1 layers where the larvae were collected were used. Selectivity was estimated by applying Chesson’s selectivity index (Chesson, 1978) as αi=ri/pi∑i=1mri/pi-1i=1,…,m ⁠, where ri and pi are the respective frequencies of a prey item in the diet and zooplankton collected in the same layer as the fish, and m is the number of zooplankton prey categories considered. Neutral selection would result in a constant α  =  1/m. The diets of the six species were analysed through hierarchical agglomerative and unweighted arithmetic average clustering (CLUSTER procedure; Clarke and Gorley, 2006) of the calculated Bray–Curtis similarity indices. For each fish species caught in each of the four sectors, the average prey abundances per gut were calculated, for both larvae and transforming stages. Only those prey items that appeared at least twice, and only those species-stages occurring twice per sector, were included in the analysis. Data were log-transformed to reduce the influence of very abundant items, and the Bray–Curtis indices were calculated to produce similarity matrices. The significant groups in the cluster dendrogram were determined using the SIMPROF procedure (with 1000 permutations) (Clarke and Gorley, 2006). A SIMPER routine was then followed to identify those prey items that characterise each of the groups. Relevant information on species distribution and ontogenetic changes in morphology related to feeding A brief synopsis of the relevant information on ontogenetic changes in morphology related to feeding, and a summary of their vertical distribution is given in Table 2 and Figures 2 and 3. Although A. sladeni larvae and transforming stages have been described by Watson (1996), the larval morphological features in preflexion and flexion stages were identical to those of A. hemigymnus, which is also common in the region. Therefore, in this work, the larval stages may include both species, but transformation specimens could be identified as A. sladeni. Similarly, Diaphus cf. vanhoeffeni larvae had the general morphology and pigmentation as described by Moser and Ahlstrom (1974) for Diaphus species, while transforming specimens could be confidently identified as D. vanhoeffeni through adult keys (Hulley and Paxton, 2016b). The six species occurred throughout the study region but presented higher abundances and higher frequencies of occurrence in the central sector. However, S. diaphana was more abundant in western stations (Figure 3). In general, larvae showed shallower distributions than transforming stages (Table 2 and Figure 3). Table 2. Summary of morphological features and vertical distributions of larvae and transforming stages of the studied taxa, and the sources for their descriptions and vertical distributions. Species Body Gut Eyes Mouth Vertical distribution References B. argyrogaster Slender. Straight and long (>80% of SL) Slightly oval. Small Larvae: 50 to 200m, with mean vertical depth 75 m Hermes and Olivar (1987); Olivar and Fortuño (1991); Olivar et al. (2018) A. sladeni Very elongate before flexion. Deep head and trunk region in later stages Relatively short and straight before flexion. Short and balloon like in later stages (<40% SL) Vertically elongate and narrow before flexion. Oval in later stages Relatively large. Larvae: 100–500 m, with main vertical depths from 200 to 300 m. Transforming: 200–500 m Watson(1996); Olivar et al. (2018) S. diaphana Head and gut region deep Shorter than 30% before flexion. Short and balloon like in later stages (<40% SL) Slightly oval in early stages, becoming round with development Relatively small Larvae: 50–800 m. Transforming: 200–800 m Belyanina (1984); Watson (1996); Olivar et al. (2018) D. cf. vanhoeffeni Moderately deep Relatively straight and short (reaching ca. 60% of SL) Slightly round in larvae and round in transforming stages Relatively large Larvae: 0–50 m. Transforming: 50–400 m Olivar et al. (2018) H. macrochir Moderately deep Gut thick in the middle section, but with a very narrow foregut (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Mouth larger than in Diaphus cf. vanhoeffeni and shorter than in M. affine of similar sizes Larvae: 0–100 m. Transforming: 300–600 m Moser and Ahlstrom (1974); Olivar and Fortuño (1991); Olivar et al. (2018) M. affine Body stout, deepest anteriorly, with head very large and wide Gut large and saccular (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Large Larvae: 50–100 m. Transforming: bellow 400 m Moser and Watson, (2006); Olivar et al. (2018) Species Body Gut Eyes Mouth Vertical distribution References B. argyrogaster Slender. Straight and long (>80% of SL) Slightly oval. Small Larvae: 50 to 200m, with mean vertical depth 75 m Hermes and Olivar (1987); Olivar and Fortuño (1991); Olivar et al. (2018) A. sladeni Very elongate before flexion. Deep head and trunk region in later stages Relatively short and straight before flexion. Short and balloon like in later stages (<40% SL) Vertically elongate and narrow before flexion. Oval in later stages Relatively large. Larvae: 100–500 m, with main vertical depths from 200 to 300 m. Transforming: 200–500 m Watson(1996); Olivar et al. (2018) S. diaphana Head and gut region deep Shorter than 30% before flexion. Short and balloon like in later stages (<40% SL) Slightly oval in early stages, becoming round with development Relatively small Larvae: 50–800 m. Transforming: 200–800 m Belyanina (1984); Watson (1996); Olivar et al. (2018) D. cf. vanhoeffeni Moderately deep Relatively straight and short (reaching ca. 60% of SL) Slightly round in larvae and round in transforming stages Relatively large Larvae: 0–50 m. Transforming: 50–400 m Olivar et al. (2018) H. macrochir Moderately deep Gut thick in the middle section, but with a very narrow foregut (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Mouth larger than in Diaphus cf. vanhoeffeni and shorter than in M. affine of similar sizes Larvae: 0–100 m. Transforming: 300–600 m Moser and Ahlstrom (1974); Olivar and Fortuño (1991); Olivar et al. (2018) M. affine Body stout, deepest anteriorly, with head very large and wide Gut large and saccular (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Large Larvae: 50–100 m. Transforming: bellow 400 m Moser and Watson, (2006); Olivar et al. (2018) Table 2. Summary of morphological features and vertical distributions of larvae and transforming stages of the studied taxa, and the sources for their descriptions and vertical distributions. Species Body Gut Eyes Mouth Vertical distribution References B. argyrogaster Slender. Straight and long (>80% of SL) Slightly oval. Small Larvae: 50 to 200m, with mean vertical depth 75 m Hermes and Olivar (1987); Olivar and Fortuño (1991); Olivar et al. (2018) A. sladeni Very elongate before flexion. Deep head and trunk region in later stages Relatively short and straight before flexion. Short and balloon like in later stages (<40% SL) Vertically elongate and narrow before flexion. Oval in later stages Relatively large. Larvae: 100–500 m, with main vertical depths from 200 to 300 m. Transforming: 200–500 m Watson(1996); Olivar et al. (2018) S. diaphana Head and gut region deep Shorter than 30% before flexion. Short and balloon like in later stages (<40% SL) Slightly oval in early stages, becoming round with development Relatively small Larvae: 50–800 m. Transforming: 200–800 m Belyanina (1984); Watson (1996); Olivar et al. (2018) D. cf. vanhoeffeni Moderately deep Relatively straight and short (reaching ca. 60% of SL) Slightly round in larvae and round in transforming stages Relatively large Larvae: 0–50 m. Transforming: 50–400 m Olivar et al. (2018) H. macrochir Moderately deep Gut thick in the middle section, but with a very narrow foregut (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Mouth larger than in Diaphus cf. vanhoeffeni and shorter than in M. affine of similar sizes Larvae: 0–100 m. Transforming: 300–600 m Moser and Ahlstrom (1974); Olivar and Fortuño (1991); Olivar et al. (2018) M. affine Body stout, deepest anteriorly, with head very large and wide Gut large and saccular (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Large Larvae: 50–100 m. Transforming: bellow 400 m Moser and Watson, (2006); Olivar et al. (2018) Species Body Gut Eyes Mouth Vertical distribution References B. argyrogaster Slender. Straight and long (>80% of SL) Slightly oval. Small Larvae: 50 to 200m, with mean vertical depth 75 m Hermes and Olivar (1987); Olivar and Fortuño (1991); Olivar et al. (2018) A. sladeni Very elongate before flexion. Deep head and trunk region in later stages Relatively short and straight before flexion. Short and balloon like in later stages (<40% SL) Vertically elongate and narrow before flexion. Oval in later stages Relatively large. Larvae: 100–500 m, with main vertical depths from 200 to 300 m. Transforming: 200–500 m Watson(1996); Olivar et al. (2018) S. diaphana Head and gut region deep Shorter than 30% before flexion. Short and balloon like in later stages (<40% SL) Slightly oval in early stages, becoming round with development Relatively small Larvae: 50–800 m. Transforming: 200–800 m Belyanina (1984); Watson (1996); Olivar et al. (2018) D. cf. vanhoeffeni Moderately deep Relatively straight and short (reaching ca. 60% of SL) Slightly round in larvae and round in transforming stages Relatively large Larvae: 0–50 m. Transforming: 50–400 m Olivar et al. (2018) H. macrochir Moderately deep Gut thick in the middle section, but with a very narrow foregut (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Mouth larger than in Diaphus cf. vanhoeffeni and shorter than in M. affine of similar sizes Larvae: 0–100 m. Transforming: 300–600 m Moser and Ahlstrom (1974); Olivar and Fortuño (1991); Olivar et al. (2018) M. affine Body stout, deepest anteriorly, with head very large and wide Gut large and saccular (reaching ca. 60% of SL) Elliptical in larvae and round in transforming stages Large Larvae: 50–100 m. Transforming: bellow 400 m Moser and Watson, (2006); Olivar et al. (2018) Figure 2. View largeDownload slide Schematic drawings of the larval morphology of the studied species (note: pigmentation not included). (a) Bathylagoides argyrogaster (4.8 mm SL; modified from Hermes and Olivar,1987); (b) Argyropelecus spp. (9 mm SL; modified from Olivar and Fortuño, 1991), (c) A. sladeni (transforming specimen of 8.2 mm SL; modified from Watson, 1996), (d) Sternoptyx diaphana (9.4 mm SL; modified from Belyanina, 1984), (e) Diaphus cf. vanhoeffeni (4.3 mm SL; present investigation), (f) Hygophum macrochir (7.5 mm SL; modified from Olivar and Fortuño, 1991), and (g) Myctophum affine (5.1 mm SL; modified from Moser and Watson, 2006). Figure 2. View largeDownload slide Schematic drawings of the larval morphology of the studied species (note: pigmentation not included). (a) Bathylagoides argyrogaster (4.8 mm SL; modified from Hermes and Olivar,1987); (b) Argyropelecus spp. (9 mm SL; modified from Olivar and Fortuño, 1991), (c) A. sladeni (transforming specimen of 8.2 mm SL; modified from Watson, 1996), (d) Sternoptyx diaphana (9.4 mm SL; modified from Belyanina, 1984), (e) Diaphus cf. vanhoeffeni (4.3 mm SL; present investigation), (f) Hygophum macrochir (7.5 mm SL; modified from Olivar and Fortuño, 1991), and (g) Myctophum affine (5.1 mm SL; modified from Moser and Watson, 2006). Figure 3. View largeDownload slide Vertical distributions of larval and transforming stages of the species collected with the MOCNESS-1 net. Small black dots denote the centre of each haul. Open symbols indicate day samples and solid symbols night samples. Circles refer to larvae and triangles to transforming stages abundances. Figure 3. View largeDownload slide Vertical distributions of larval and transforming stages of the species collected with the MOCNESS-1 net. Small black dots denote the centre of each haul. Open symbols indicate day samples and solid symbols night samples. Circles refer to larvae and triangles to transforming stages abundances. Results Feeding incidence B. argyrogaster larvae had an exclusively daylight feeding pattern. FI decreased with development from 80% in preflexion to 20% in postflexion stages. No transforming stages specimens were available (Table 1). Both larvae of Argyropelecus spp. and transforming stages of A. sladeni fed throughout the day. Preflexion larvae showed a FI of 25% during daylight hours and 42.9% at night (no prey items were found in the guts of flexion and postflexion larvae). Transforming stages showed a higher FI during the day than at night (87.5% and 60%, respectively: Table 1). S. diaphana showed a similar feeding pattern, with larvae and transforming stages feeding both day and night. An increase in FI was observed with development, from 27.3% in preflexion to 78.6% in transforming stages (Table 1). Larvae of the three myctophids displayed an exclusively daylight feeding pattern. The FI was relatively higher in preflexion than in postflexion stages. M. affine showed the highest FI through its development from 54.5% (preflexion) to 30% (postflexion), followed by H. macrochir with 28.6% (preflexion) to 3.6% (postflexion) and Diaphus cf. vanhoeffeni with 11.1% (preflexion) to 3.5% (postflexion). FI in the transforming stages of M. affine and Diaphus cf. vanhoeffeni was higher than in their larval stages. They showed feeding activity during daylight, although nocturnal feeding was also observed for Diaphus cf. vanhoeffeni, with a night FI of 92.1% (Table 1). Morphometric relationships The species with a large MW in the early stages (i.e. >0.4 mm at 5 mm SL) was M. affine (0.54 mm), followed by Diaphus cf. vanhoeffeni (0.42 mm). B. argyrogaster has the smallest mouth (0.3 mm). MW, length of upper (UJL) and lower jaws (LJL) showed significantly positive allometric relationships in relation to SL in all the studied species, except for MW in B. argyrogaster, which was isometric (allometric coefficient range from 0.877 to 1.099) (Table 3). The species with a relatively fast gape development were S. diaphana, A. sladeni and Diaphus cf. vanhoeffeni and to a lesser extent H. macrochir, and M. affine (Figure 4; Table 3). Table 3. Parameters of the allometric relationships between MW, UJL, LJL, and SL for the studied species. Species n r2 a b 95% CIb B. argyrogaster  MW 68 0.827 0.065 0.988 0.111  UJL 68 0.824 0.072 1.216 0.138  LJL 68 0.868 0.081 1.219 0.117 A. sladeni  MW 44 0.706 0.024 1.573 0.316  UJL 45 0.675 0.034 1.666 0.356  LJL 45 0.707 0.042 1.630 0.323 S. diaphana  MW 183 0.679 0.021 1.724 0.174  UJL 183 0.681 0.033 1.787 0.179  LJL 183 0.674 0.042 1.719 0.175 D. vanhoeffeni  MW 288 0.978 0.044 1.420 0.025  UJL 288 0.970 0.060 1.546 0.030  LJL 288 0.977 0.079 1.464 0.026 H. macrochir  MW 183 0.973 0.043 1.311 0.033  UJL 183 0.970 0.065 1.384 0.036  LJL 183 0.974 0.083 1.322 0.032 M. affine  MW 93 0.898 0.075 1.231 0.086  UJL 93 0.873 0.113 1.324 0.105  LJL 93 0.886 0.141 1.266 0.095 Species n r2 a b 95% CIb B. argyrogaster  MW 68 0.827 0.065 0.988 0.111  UJL 68 0.824 0.072 1.216 0.138  LJL 68 0.868 0.081 1.219 0.117 A. sladeni  MW 44 0.706 0.024 1.573 0.316  UJL 45 0.675 0.034 1.666 0.356  LJL 45 0.707 0.042 1.630 0.323 S. diaphana  MW 183 0.679 0.021 1.724 0.174  UJL 183 0.681 0.033 1.787 0.179  LJL 183 0.674 0.042 1.719 0.175 D. vanhoeffeni  MW 288 0.978 0.044 1.420 0.025  UJL 288 0.970 0.060 1.546 0.030  LJL 288 0.977 0.079 1.464 0.026 H. macrochir  MW 183 0.973 0.043 1.311 0.033  UJL 183 0.970 0.065 1.384 0.036  LJL 183 0.974 0.083 1.322 0.032 M. affine  MW 93 0.898 0.075 1.231 0.086  UJL 93 0.873 0.113 1.324 0.105  LJL 93 0.886 0.141 1.266 0.095 Number of specimens (n), coefficient of determination (r2), intercept (a), allometric coefficient (b), confidence interval of the allometric coefficient (CIb). Table 3. Parameters of the allometric relationships between MW, UJL, LJL, and SL for the studied species. Species n r2 a b 95% CIb B. argyrogaster  MW 68 0.827 0.065 0.988 0.111  UJL 68 0.824 0.072 1.216 0.138  LJL 68 0.868 0.081 1.219 0.117 A. sladeni  MW 44 0.706 0.024 1.573 0.316  UJL 45 0.675 0.034 1.666 0.356  LJL 45 0.707 0.042 1.630 0.323 S. diaphana  MW 183 0.679 0.021 1.724 0.174  UJL 183 0.681 0.033 1.787 0.179  LJL 183 0.674 0.042 1.719 0.175 D. vanhoeffeni  MW 288 0.978 0.044 1.420 0.025  UJL 288 0.970 0.060 1.546 0.030  LJL 288 0.977 0.079 1.464 0.026 H. macrochir  MW 183 0.973 0.043 1.311 0.033  UJL 183 0.970 0.065 1.384 0.036  LJL 183 0.974 0.083 1.322 0.032 M. affine  MW 93 0.898 0.075 1.231 0.086  UJL 93 0.873 0.113 1.324 0.105  LJL 93 0.886 0.141 1.266 0.095 Species n r2 a b 95% CIb B. argyrogaster  MW 68 0.827 0.065 0.988 0.111  UJL 68 0.824 0.072 1.216 0.138  LJL 68 0.868 0.081 1.219 0.117 A. sladeni  MW 44 0.706 0.024 1.573 0.316  UJL 45 0.675 0.034 1.666 0.356  LJL 45 0.707 0.042 1.630 0.323 S. diaphana  MW 183 0.679 0.021 1.724 0.174  UJL 183 0.681 0.033 1.787 0.179  LJL 183 0.674 0.042 1.719 0.175 D. vanhoeffeni  MW 288 0.978 0.044 1.420 0.025  UJL 288 0.970 0.060 1.546 0.030  LJL 288 0.977 0.079 1.464 0.026 H. macrochir  MW 183 0.973 0.043 1.311 0.033  UJL 183 0.970 0.065 1.384 0.036  LJL 183 0.974 0.083 1.322 0.032 M. affine  MW 93 0.898 0.075 1.231 0.086  UJL 93 0.873 0.113 1.324 0.105  LJL 93 0.886 0.141 1.266 0.095 Number of specimens (n), coefficient of determination (r2), intercept (a), allometric coefficient (b), confidence interval of the allometric coefficient (CIb). Figure 4. View largeDownload slide Relationship between SL and MW for Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), Sternoptyx diaphana, Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir, and Myctophum affine (fitting parameters given in Table 3). Figure 4. View largeDownload slide Relationship between SL and MW for Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), Sternoptyx diaphana, Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir, and Myctophum affine (fitting parameters given in Table 3). Predator–prey relationships: number of prey per gut In B. argyrogaster larvae, an increase in the ingested prey number was observed, mainly between preflexion and flexion, while the number of prey was lower in postflexion stages (Figure 5a). Unfortunately, the restricted vertical distribution (50–100 m) of larvae with prey items in the gut does not allow for the study of differences in the mean prey number as a function of depth (Figure 5b). Figure 5. View largeDownload slide Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), and Sternoptyx diaphana: variation in the number of prey ingested per larva by size classes (a), and mean and standard deviation of the number of prey items ingested during the night and the day, in relation to developmental stage and position in the water column (b). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. Figure 5. View largeDownload slide Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), and Sternoptyx diaphana: variation in the number of prey ingested per larva by size classes (a), and mean and standard deviation of the number of prey items ingested during the night and the day, in relation to developmental stage and position in the water column (b). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. Preflexion larvae of Argyropelecus spp. (≤7.5 mm) had from 2 to 4 prey items, while transforming stages of A. sladeni showed a slight increase in number with size, reaching 10 prey items in specimens of 11.6 mm (Figure 5a). Argyropelecus spp. larvae with prey in their guts came from hauls carried out both day and night between 100 and 200 m in depth, where the mean prey number was from 2 to 4. In transforming stages of A. sladeni prey ingestion was higher during the day, with maxima of 10 prey items between 100 and 200 m depth, and 2.5 prey items at night between 200 and 500 m depth (Figure 5b). The number of prey ingested also showed an increase with development in S. diaphana, from a maximum of 2 items in preflexion, to 4 in flexion, and to 11 in postflexion larvae. In transforming stages, the number of prey also increased with size, reaching 25 prey items in specimens of 13 mm (Figure 5a). An increase in the mean prey number with depth and developmental stage was observed. Prey item maxima were observed in postflexion larvae, between 200 and 500 m during the day (between 2.3 and 5.7 prey items). In transforming stages, the maxima were observed during the day between 100 and 200 m (13 prey items) and at night between 500 and 800 m (9.1 prey items) (Figure 5b). The number of prey ingested by the three myctophids was generally lower than for the above species. In Diaphus cf. vanhoeffeni the number of prey ingested decreased between preflexion (maximum of 4 prey) and flexion and postflexion stages (3 and 1 prey items, respectively). In transforming stages, the number of prey was variable, although it showed an increase with a maximum of 14 prey items in specimens of 13 mm (Figure 6a). The maximum mean number of prey (3.7 prey items per gut) was observed in preflexion larvae caught in the uppermost (0 and 50 m) layers, while postflexion larvae in this layer showed a mean of only 1 prey item per gut. Transforming stages showed a broad vertical distribution in the water column, but specimens from the first 100 m presented the maximum values (ca. 4 prey items), both day and night (Figure 6b). Figure 6. View largeDownload slide Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir, and Myctophum affine: variation in the number of prey ingested per larva by size classes (a), and mean and standard deviation of the number of prey items ingested during the night and the day, in relation to developmental stage and position in the water column (b). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. Figure 6. View largeDownload slide Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir, and Myctophum affine: variation in the number of prey ingested per larva by size classes (a), and mean and standard deviation of the number of prey items ingested during the night and the day, in relation to developmental stage and position in the water column (b). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. There were no changes in the number of prey (1–2 items) ingested by H. macrochir larvae either in relation to development, or with depth of occurrence. The highest number of prey (11 items) appeared in one transforming specimen of 17.8 mm (Figure 6a and b). M. affine larvae showed no clear correlation in the number of prey ingested with development, although preflexion larvae had a maximum of 4 prey items per gut and postflexion and transforming 3 and 2 prey items, respectively (Figure 6a). The mean number of prey was similar in the different layers of the water column and different development stages (1 and 2 preys per gut) (Figure 6b). Predator–prey relationships: prey size and trophic niche breadth B. argyrogaster ate prey of a similar small size (100–300 µm) throughout its larval development (Figure 7a). Thus, trophic niche breadth did not reveal any tendency of prey size specialization with development (Figure 7b). Because the larvae of this species were all caught at the same depths (between 50 and 100 m), no differences in the sizes of the prey with depth were evident (Figure 7c). Figure 7. View largeDownload slide Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), and Sternoptyx diaphana: variation in prey width (a) and trophic niche breadth by size classes (b). Mean and standard deviation of prey width ingested during the night and the day in relation to developmental stage and position in the water column (c). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. Figure 7. View largeDownload slide Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), and Sternoptyx diaphana: variation in prey width (a) and trophic niche breadth by size classes (b). Mean and standard deviation of prey width ingested during the night and the day in relation to developmental stage and position in the water column (c). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. Preflexion and flexion larvae of Argyropelecus spp. fed on small prey, between 60 and 250 µm. Transforming stages of A. sladeni ingested prey of a wider range of sizes (from 80 to 800 µm) and showed an increase of maximum prey size with predator size (Figure 7a). Trophic niche breadth did not show any relationship to SL (Figure 7b). Further, no relationship between larval location in the water column and the size of the prey ingested could be established due to the limited vertical distribution of the larvae with prey items in their guts. A similar mean prey size from different layers of the water column was observed for transforming stages: ca. 400 µm both during the day (from 100 to 500 m) and at night (from 200 to 500 m) (Figure 7c). In S. diaphana maximum prey width showed an increasing trend with development. Larvae ingested prey between 78 and 500 µm; and transforming stages between 100 and 1700 µm (Figure 7a). The trophic niche breadth did not vary with SL (Figure 7b). There was a slight increase in mean prey width with depth within each development stage (Figure 7c). Diaphus cf. vanhoeffeni larvae showed an increase in prey size with development stage and fed on prey between 100 to 340 µm. Transforming stages preyed on a wider range of sizes, from 160 µm to 800 µm (Figure 8a). Therefore, trophic niche breadth appeared to be independent of the SL (Figure 8b). The main differences in prey sizes from different layers of the water column correlated more to developmental stage than to depth. The most noticeable result was the larger size of prey ingested by transforming stages at night in the upper layers (from surface to 100 m) compared with the prey size during day feeding, both in this layer and in greater depths (Figure 8c). Figure 8. View largeDownload slide Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir and Myctophum affine: variation in prey width (a) and trophic niche breadth by size classes (b). Mean and standard deviation of prey width ingested during the night and the day in relation to developmental stage and position in the water column (c). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. Figure 8. View largeDownload slide Diaphus vanhoeffeni (larval stages D. cf. vanhoeffeni), Hygophum macrochir and Myctophum affine: variation in prey width (a) and trophic niche breadth by size classes (b). Mean and standard deviation of prey width ingested during the night and the day in relation to developmental stage and position in the water column (c). In (a) solid symbols correspond to the transforming stages and open symbols correspond to larval stages. H. macrochir showed no relationship of prey size to development, with prey widths between 50 and 250 µm in the larval stages. Prey items reached a slightly larger size in transforming stages with a maximum of 850 µm in a 14.5 mm specimen (Figure 8a). However, the trophic niche breadth did not show a relationship with SL (Figure 8b). Differences in prey sizes in relation to depth within larval stages were also not observed (Figure 8c). In M affine larvae, prey sizes increased between 60 and 400 µm from preflexion to postflexion larvae, with a subsequent increase in transforming stages, from 230 to 520 µm (Figure 8a). The relation between trophic niche breadth and SL did not show any significant trend (Figure 8b). There is a general increase in prey size with depth, reflecting the deeper locating of older developmental stages (Figure 8c). Diet The diet of B. argyrogaster larvae was mostly composed of copepods, and was dominated by copepodite stages in preflexion larvae (IRI 91.7%). In flexion larvae unidentified copepodites and adults of the genus Oncaea were the main diet items (IRI 52.15 and 47.1%, respectively). Larger copepods of the genus Paracalanus were the only prey represented in postflexion larvae (Table 4). Table 4. Diets of Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), Sternoptyx diaphana, Diaphus vanhoeffeni (larval stages: D. cf. vanhoeffeni), H. macrochir, and M. affine. B. argyrogaster A. sladeni S. diaphana D. vanhoeffeni H. macrochir M. affine Pre Flex Post Pre Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Copepod eggs 0.90 0.20 11.70 16.70 Copepod nauplii 0.20 0.10 7.60 0.01 100.00 73.50 0.01 26.30 16.70 100.00 12.80 Copepodites 91.70 52.20 99.00 26.60 92.50 97.60 21.70 0.01 26.50 0.20 59.10 66.60 0.90 32.80 28.40 Calanoida:  Acartia 0.16 0.10 3.90 0.04 0.90  Calanus 0.10 7.70 2.60 1.50 7.40 14.30  Centropages 0.10 14.30  Paracalanus 0.10 100.00 5.60 1.50 0.70 5.00 33.30 1.10  Pleuromamma 0.10 Cyclopoida:  Oithona 0.04 2.20 0.02 0.90 14.30 Harpacticoida:  Microsetella 0.10 0.01 0.90 2.10 45.70 Poecilostomatoida: 3.90  Oncaea 3.90 47.10 7.70 69.10 61.00 33.30 89.30 91.70 7.14 57.10  Corycaeus 0.70 11.00 0.04  Sapphirina 0.10 Unidentified Copepods 2.50 0.04 0.30 0.01 Chaetognaths 7.10 Hyperiids 0.80 0.20 Polychaetes 0.60 Molluscs 0.10 0.20 0.20 0.90 51.00 25.70 Euphausiids 0.20 0.10 0.30 Ostracods 45.40 0.70 5.00 9.90 33.30 1.30 2.90 0.51 25.70 64.30 Appendicularians 0.01 0.10 Unidentified prey 0.20 0.03 3.70 0.51 2.90 B. argyrogaster A. sladeni S. diaphana D. vanhoeffeni H. macrochir M. affine Pre Flex Post Pre Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Copepod eggs 0.90 0.20 11.70 16.70 Copepod nauplii 0.20 0.10 7.60 0.01 100.00 73.50 0.01 26.30 16.70 100.00 12.80 Copepodites 91.70 52.20 99.00 26.60 92.50 97.60 21.70 0.01 26.50 0.20 59.10 66.60 0.90 32.80 28.40 Calanoida:  Acartia 0.16 0.10 3.90 0.04 0.90  Calanus 0.10 7.70 2.60 1.50 7.40 14.30  Centropages 0.10 14.30  Paracalanus 0.10 100.00 5.60 1.50 0.70 5.00 33.30 1.10  Pleuromamma 0.10 Cyclopoida:  Oithona 0.04 2.20 0.02 0.90 14.30 Harpacticoida:  Microsetella 0.10 0.01 0.90 2.10 45.70 Poecilostomatoida: 3.90  Oncaea 3.90 47.10 7.70 69.10 61.00 33.30 89.30 91.70 7.14 57.10  Corycaeus 0.70 11.00 0.04  Sapphirina 0.10 Unidentified Copepods 2.50 0.04 0.30 0.01 Chaetognaths 7.10 Hyperiids 0.80 0.20 Polychaetes 0.60 Molluscs 0.10 0.20 0.20 0.90 51.00 25.70 Euphausiids 0.20 0.10 0.30 Ostracods 45.40 0.70 5.00 9.90 33.30 1.30 2.90 0.51 25.70 64.30 Appendicularians 0.01 0.10 Unidentified prey 0.20 0.03 3.70 0.51 2.90 Index of relative importance (%IRI) determined for each developmental stage (Pre, preflexion; Flex, flexion; Post, postflexion; Trans, transforming). Table 4. Diets of Bathylagoides argyrogaster, Argyropelecus sladeni (larval stages: Argyropelecus spp.), Sternoptyx diaphana, Diaphus vanhoeffeni (larval stages: D. cf. vanhoeffeni), H. macrochir, and M. affine. B. argyrogaster A. sladeni S. diaphana D. vanhoeffeni H. macrochir M. affine Pre Flex Post Pre Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Copepod eggs 0.90 0.20 11.70 16.70 Copepod nauplii 0.20 0.10 7.60 0.01 100.00 73.50 0.01 26.30 16.70 100.00 12.80 Copepodites 91.70 52.20 99.00 26.60 92.50 97.60 21.70 0.01 26.50 0.20 59.10 66.60 0.90 32.80 28.40 Calanoida:  Acartia 0.16 0.10 3.90 0.04 0.90  Calanus 0.10 7.70 2.60 1.50 7.40 14.30  Centropages 0.10 14.30  Paracalanus 0.10 100.00 5.60 1.50 0.70 5.00 33.30 1.10  Pleuromamma 0.10 Cyclopoida:  Oithona 0.04 2.20 0.02 0.90 14.30 Harpacticoida:  Microsetella 0.10 0.01 0.90 2.10 45.70 Poecilostomatoida: 3.90  Oncaea 3.90 47.10 7.70 69.10 61.00 33.30 89.30 91.70 7.14 57.10  Corycaeus 0.70 11.00 0.04  Sapphirina 0.10 Unidentified Copepods 2.50 0.04 0.30 0.01 Chaetognaths 7.10 Hyperiids 0.80 0.20 Polychaetes 0.60 Molluscs 0.10 0.20 0.20 0.90 51.00 25.70 Euphausiids 0.20 0.10 0.30 Ostracods 45.40 0.70 5.00 9.90 33.30 1.30 2.90 0.51 25.70 64.30 Appendicularians 0.01 0.10 Unidentified prey 0.20 0.03 3.70 0.51 2.90 B. argyrogaster A. sladeni S. diaphana D. vanhoeffeni H. macrochir M. affine Pre Flex Post Pre Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Pre Flex Post Trans Copepod eggs 0.90 0.20 11.70 16.70 Copepod nauplii 0.20 0.10 7.60 0.01 100.00 73.50 0.01 26.30 16.70 100.00 12.80 Copepodites 91.70 52.20 99.00 26.60 92.50 97.60 21.70 0.01 26.50 0.20 59.10 66.60 0.90 32.80 28.40 Calanoida:  Acartia 0.16 0.10 3.90 0.04 0.90  Calanus 0.10 7.70 2.60 1.50 7.40 14.30  Centropages 0.10 14.30  Paracalanus 0.10 100.00 5.60 1.50 0.70 5.00 33.30 1.10  Pleuromamma 0.10 Cyclopoida:  Oithona 0.04 2.20 0.02 0.90 14.30 Harpacticoida:  Microsetella 0.10 0.01 0.90 2.10 45.70 Poecilostomatoida: 3.90  Oncaea 3.90 47.10 7.70 69.10 61.00 33.30 89.30 91.70 7.14 57.10  Corycaeus 0.70 11.00 0.04  Sapphirina 0.10 Unidentified Copepods 2.50 0.04 0.30 0.01 Chaetognaths 7.10 Hyperiids 0.80 0.20 Polychaetes 0.60 Molluscs 0.10 0.20 0.20 0.90 51.00 25.70 Euphausiids 0.20 0.10 0.30 Ostracods 45.40 0.70 5.00 9.90 33.30 1.30 2.90 0.51 25.70 64.30 Appendicularians 0.01 0.10 Unidentified prey 0.20 0.03 3.70 0.51 2.90 Index of relative importance (%IRI) determined for each developmental stage (Pre, preflexion; Flex, flexion; Post, postflexion; Trans, transforming). Preflexion larvae of Argyropelecus spp. fed almost exclusively on copepodites, while in transforming stages of A. sladeni, ostracods and copepodites constitute the main food (IRI 45.4% and 26.6%, respectively) (Table 4). In S. diaphana, copepods were the most important prey throughout larval development, both in preflexion and flexion stages (IRI > 90%). Postflexion and transforming stages exhibited a more diverse diet, although copepods of genus Oncaea were the most common prey (IRI > 60%) (Table 4). In addition to this, ostracods and chaetognaths acquired certain relevance (IRI 10 and 7%, respectively) in the diets of transforming stages. Preflexion and flexion Diaphus cf. vanhoeffeni larvae feed mainly on copepod nauplii (IRI > 70%); while in postflexion larvae, copepods of genus Paracalanus and Oncaea, and ostracods were also consumed. Transforming stages of D. cf. vanhoeffeni possessed a more diverse diet composition, with copepods of genus Oncaea being the dominant prey (IRI 89.3%) (Table 4). In all larval stages, the diet H. macrochir consisted of early copepod stages (eggs, nauplii, and copepodites). In transforming stages, copepods of the genus Oncaea were their main prey (IRI > 90%) (Table 4). The diet of M. affine larvae was more diverse than in the other myctophids. Molluscs and copepodites were the more important prey items in preflexion larvae (IRI 51% and 32.8%). In flexion larvae, the diet was a mixture of copepods of genus Microsetella (IRI 45.7%), molluscs (IRI 25.7%), and ostracods (IRI 25.7%). In postflexion larvae ostracods were the most important prey (IRI 64.3%) followed by copepodites (IRI 28.4%). The diet of transforming stages consisted of small-sized copepods of the genus Oncaea (IRI 57.1%), or larger specimens of the genera Calanus, Centropages, and Oithona (IRI 14%) (Table 4). Larval selectivity was calculated for specimens collected at station #8. Chesson’s selectivity index for the two main microzooplankton components, nauplii and copepodites <0.2 mm, showed significant positive selection for copepodites and negative for nauplii in B. argyrogaster (preflexion and flexion), S. diaphana (preflexion, flexion, and postflexion), and H. macrochir (flexion). The only positive selection for nauplii was found in preflexion larvae of D. cf. vanhoeffeni but flexion stages showed neutral selection for both prey types, as preflexion larvae of M. affine (Table 5). In transforming stages selectivity for mesozooplankton components could be estimated for A. sladeni, S. diaphana, and D. vanhoeffeni. A significantly positive selection was detected in A. sladeni for copepodites >0.2 mm; and in S. diaphana for the copepod Corycaeus spp. Transforming stages of D. vanhoeffeni showed positive selection for the copepod Oncaea spp. (Table 5), while the selective index was negative for Paracalanus spp. and Ostracoda. Table 5. Mean Chesson’s selectivity index α (±95% confidence interval) for the most common prey items of larvae and transforming stages of Bathylagoides argyrogaster, Argyropelecus sladeni, Sternoptyx diaphana, Diaphus vanhoeffeni, Hygopum macrochir, and Myctophum affine from station #8. N 1/m Nauplii Copepodites <0.2 mm Copepodites >0.2 mm Calanoida Paracalanus Oithona Oncaea Corycaeus Chaetognatha Ostracoda Preflexion larvae B. argyrogaster 8 0.5 0.125 (0.245) 0.875 (0.245)a S. diaphana 3 0.5 0 1.000 (0.000)a D. vanhoeffeni 3 0.5 1.000 (0.000)a 0 H. macrochir 11 0.5 0.343 (0.258) 0.657 (0.258) M. affine 7 0.5 0.429 (0.396) 0.571 (0.396) Flexion larvae B. argyrogaster 8 0.5 0.032 (0.063) 0.968 (0.063)a S. diaphana 6 0.5 0.167 (0.327) 0.833 (0.327)a D. vanhoeffeni 9 0.5 0.461 (0.336) 0.539 (0.336) H. macrochir 2 0.5 0.016 (0.032) 0.984 (0.032)a Postflexion S. diaphana 9 0.5 0 1.000 (0.000)a H. macrochir 1 0.5 0.063 0.937 M. affine 1 0.5 0 1.000 Transforming A. sladeni 15 0.2 0.515 (0.253)a 0.139 (0.177) 0.125 (0.150) 0.004 (0.007) 0.216 (0.196) S. diaphana 39 0.2 0.205 (0.112) 0.198 (0.115) 0.332 (0.143)a 0.097 (0.071) 0.167 (0.098) D. vanhoeffeni 111 0.3 0.151 (0.063) 0.093 (0.051) 0.657 (0.084)a 0.098 (0.054) N 1/m Nauplii Copepodites <0.2 mm Copepodites >0.2 mm Calanoida Paracalanus Oithona Oncaea Corycaeus Chaetognatha Ostracoda Preflexion larvae B. argyrogaster 8 0.5 0.125 (0.245) 0.875 (0.245)a S. diaphana 3 0.5 0 1.000 (0.000)a D. vanhoeffeni 3 0.5 1.000 (0.000)a 0 H. macrochir 11 0.5 0.343 (0.258) 0.657 (0.258) M. affine 7 0.5 0.429 (0.396) 0.571 (0.396) Flexion larvae B. argyrogaster 8 0.5 0.032 (0.063) 0.968 (0.063)a S. diaphana 6 0.5 0.167 (0.327) 0.833 (0.327)a D. vanhoeffeni 9 0.5 0.461 (0.336) 0.539 (0.336) H. macrochir 2 0.5 0.016 (0.032) 0.984 (0.032)a Postflexion S. diaphana 9 0.5 0 1.000 (0.000)a H. macrochir 1 0.5 0.063 0.937 M. affine 1 0.5 0 1.000 Transforming A. sladeni 15 0.2 0.515 (0.253)a 0.139 (0.177) 0.125 (0.150) 0.004 (0.007) 0.216 (0.196) S. diaphana 39 0.2 0.205 (0.112) 0.198 (0.115) 0.332 (0.143)a 0.097 (0.071) 0.167 (0.098) D. vanhoeffeni 111 0.3 0.151 (0.063) 0.093 (0.051) 0.657 (0.084)a 0.098 (0.054) N, number of individuals used to estimate the index. 1/m, indicates neutral selectivity (m, number of prey). a Significant positive selection. Table 5. Mean Chesson’s selectivity index α (±95% confidence interval) for the most common prey items of larvae and transforming stages of Bathylagoides argyrogaster, Argyropelecus sladeni, Sternoptyx diaphana, Diaphus vanhoeffeni, Hygopum macrochir, and Myctophum affine from station #8. N 1/m Nauplii Copepodites <0.2 mm Copepodites >0.2 mm Calanoida Paracalanus Oithona Oncaea Corycaeus Chaetognatha Ostracoda Preflexion larvae B. argyrogaster 8 0.5 0.125 (0.245) 0.875 (0.245)a S. diaphana 3 0.5 0 1.000 (0.000)a D. vanhoeffeni 3 0.5 1.000 (0.000)a 0 H. macrochir 11 0.5 0.343 (0.258) 0.657 (0.258) M. affine 7 0.5 0.429 (0.396) 0.571 (0.396) Flexion larvae B. argyrogaster 8 0.5 0.032 (0.063) 0.968 (0.063)a S. diaphana 6 0.5 0.167 (0.327) 0.833 (0.327)a D. vanhoeffeni 9 0.5 0.461 (0.336) 0.539 (0.336) H. macrochir 2 0.5 0.016 (0.032) 0.984 (0.032)a Postflexion S. diaphana 9 0.5 0 1.000 (0.000)a H. macrochir 1 0.5 0.063 0.937 M. affine 1 0.5 0 1.000 Transforming A. sladeni 15 0.2 0.515 (0.253)a 0.139 (0.177) 0.125 (0.150) 0.004 (0.007) 0.216 (0.196) S. diaphana 39 0.2 0.205 (0.112) 0.198 (0.115) 0.332 (0.143)a 0.097 (0.071) 0.167 (0.098) D. vanhoeffeni 111 0.3 0.151 (0.063) 0.093 (0.051) 0.657 (0.084)a 0.098 (0.054) N 1/m Nauplii Copepodites <0.2 mm Copepodites >0.2 mm Calanoida Paracalanus Oithona Oncaea Corycaeus Chaetognatha Ostracoda Preflexion larvae B. argyrogaster 8 0.5 0.125 (0.245) 0.875 (0.245)a S. diaphana 3 0.5 0 1.000 (0.000)a D. vanhoeffeni 3 0.5 1.000 (0.000)a 0 H. macrochir 11 0.5 0.343 (0.258) 0.657 (0.258) M. affine 7 0.5 0.429 (0.396) 0.571 (0.396) Flexion larvae B. argyrogaster 8 0.5 0.032 (0.063) 0.968 (0.063)a S. diaphana 6 0.5 0.167 (0.327) 0.833 (0.327)a D. vanhoeffeni 9 0.5 0.461 (0.336) 0.539 (0.336) H. macrochir 2 0.5 0.016 (0.032) 0.984 (0.032)a Postflexion S. diaphana 9 0.5 0 1.000 (0.000)a H. macrochir 1 0.5 0.063 0.937 M. affine 1 0.5 0 1.000 Transforming A. sladeni 15 0.2 0.515 (0.253)a 0.139 (0.177) 0.125 (0.150) 0.004 (0.007) 0.216 (0.196) S. diaphana 39 0.2 0.205 (0.112) 0.198 (0.115) 0.332 (0.143)a 0.097 (0.071) 0.167 (0.098) D. vanhoeffeni 111 0.3 0.151 (0.063) 0.093 (0.051) 0.657 (0.084)a 0.098 (0.054) N, number of individuals used to estimate the index. 1/m, indicates neutral selectivity (m, number of prey). a Significant positive selection. Ontogenetic and spatial variations in diet Cluster analysis performed on the mean prey numbers per species, per stage, and per sector, identified two significant clusters: Group A (with 42.2% similarity) includes the transforming stages of all the species and regions; and Group B (with 36.0% similarity) includes the larval stages of all the species and regions, together with transforming A. sladeni from the central region (Figure 9). In terms of the relative prey contributions within each group, Oncaea spp. (60.5%), calanoids (17.7%), and Paracalanus spp. (6.6%) are the main indicators for the transforming group, while unidentified copepodites (71.3%), and nauplii (9%) are those for the larval group. Within the larval group, the main difference between the first subgroup (composed by myctophid larvae) and the second subgroup (sternoptychids and B. argyrogaster) was the higher contribution of nauplii in the diet of the myctophid subgroup. Figure 9. View largeDownload slide Dendrogram obtained after cluster analysis applied on the Bray–Curtis similarity matrix of abundance of the main prey in diets of the six studied species. Significant (p < 0.05) groups were defined by the SIMPROF procedure. Key symbols indicate the zone where samples were obtained: Western, from station #2 to station #6; Central, from station #7 to station #10; and Station #12. Species names abbreviated as the first letter of genus and species. Stages abbreviations: L for larvae and T for transforming stages. Figure 9. View largeDownload slide Dendrogram obtained after cluster analysis applied on the Bray–Curtis similarity matrix of abundance of the main prey in diets of the six studied species. Significant (p < 0.05) groups were defined by the SIMPROF procedure. Key symbols indicate the zone where samples were obtained: Western, from station #2 to station #6; Central, from station #7 to station #10; and Station #12. Species names abbreviated as the first letter of genus and species. Stages abbreviations: L for larvae and T for transforming stages. Discussion Daily feeding pattern Our analyses showed that larval feeding of B. argyrogaster, Diaphus cf. vanhoeffeni, H. macrochir, and M. affine occurred only during daylight hours, thereby confirming that they are visual feeders, as are the majority of fish larvae (Blaxter, 1963; Arthur, 1976; Hunter, 1981; Young and Davis, 1990; Sánchez-Velasco et al., 1999; Sabatés and Saiz, 2000; Morote et al., 2008a, b, 2010). Light does not seem to be an important factor for larval feeding in sternoptychids (Argyropelecus spp. and S. diaphana) since prey items were present both during the day and at night in all the early developmental stages analysed. Similarly, juvenile and adults of S. diaphana may feed both day and night (Hopkins and Baird, 1973), as has also been reported for other sternoptychids (Merrett and Roe, 1974; Hopkins and Baird, 1985). While nocturnal feeding is well known in adult myctophids, when fish migrate from the mesopelagic layers to the near-surface to feed on migrating zooplankton (Sutton, 2013), feeding patterns for transforming stages are not clearly established due to the lack of studies devoted to these stages (Sassa and Kawaguchi, 2004; Contreras et al., 2015). In the western Mediterranean Sea, Contreras et al. (2015) reported that transforming stages of Benthosema glaciale, Ceratoscopelus maderensis, Hygophum benoiti (Myctophidae), and A. hemigymnus (Sternoptychidae) do not show a well-defined feeding pattern in terms of the light conditions, with prey items in a similar digested condition both from day and night samples. Likewise in the present study, transforming stages of D. cf. vanhoeffeni fed both during the day and at night, while those of H. macrochir fed during the day. Transforming stages represent the transitional phase from a larval daylight feeding pattern to an adult nocturnal feeding pattern. In M. asperum, the transition from a day to a crepuscular/nocturnal feeding pattern has been reported to occur just before the final transformation to the juvenile stage (Sassa and Kawaguchi, 2004). Feeding incidence Larval FI and the number of prey items in the gut tend to be related to gut morphology and prey digestibility, notwithstanding the influence that fishing procedures (duration and speed of hauls) may have in the gut’s prey retention (Pepin et al., 2014). Because the results presented here come from the same survey, and follow the same protocols at all the stations, differences in the frequency of empty guts are likely related to regurgitation or evacuation processes associated with gut morphology. There is a large body of literature which has reported lower incidences for straight guts (i.e. those that tend to evacuate gut content during collection) as compared with coiled guts or prominent guts (i.e. those with greater retention capacity) (Govoni et al., 1983; Coombs et al., 1992; Canino and Bailey, 1995; Sassa and Kawaguchi, 2004; Morote et al., 2008a, b, 2010; Landaeta et al., 2011). This has also been observed in the present study for the larval stages of sternoptychids, and of the myctophids D. cf. vanhoeffeni and H. macrochir. M. affine larvae, which have a large and saccular gut, had a high FI. B. argyrogaster larvae, with a straight but long gut, was the species showing the highest FI in preflexion and flexion stages. Other investigators have also reported high FIs in larvae with straight and long guts, such as Sardinella aurita (Kurtz and Matsuura, 2001; Morote et al., 2008b). The higher FI in M. affine and B. argyrogaster when compared with D. cf. vanhoeffeni and H. macrochir, which were all collected in the same layers, points to gut morphology as the reason for these differences. In the case of D. cf. vanhoeffeni, with straight and short gut, it is likely that both regurgitation and evacuation occur. However, in the case of H. macrochir, with its very narrow foregut, evacuation could be more prevalent than regurgitation. The conspicuous change in gut morphology from larvae to transforming stages in A. sladeni and S. diaphana, i.e. from a short and relatively straight gut to a more compact and balloon-like gut, can be related to the higher prey retention in transforming than in larval stages. In the present study, prey numbers only showed an increase with larval size in B. argyrogaster, S. diaphana, and M. affine. However, in transforming stages prey numbers increased notably in A. sladeni, S. diaphana, and D. vanhoeffeni, but not in M. affine. The general increase in FI and prey number with larval size can be attributed to an increasing efficiency in prey capture, brought about by the greater swimming and sensory capacities acquired during development (Hunter, 1981; Ozawa, 1986; Sassa and Kawaguchi, 2004; Morote et al., 2010; Robert et al., 2014; Moteki et al., 2017). In our study, this tendency was observed between larvae and transforming stages of the three myctophids. However, within larval stages, a higher incidence was observed in preflexion than in postflexion larvae. This can probably be related to difficulties in prey capture when switching from very small prey items (nauplii and small copepodites) to larger prey, which may involve a learning period (Hunter, 1981). Predator–prey relationships As with the larvae of many other fish species, those studied here showed a faster growth rate for the mouth size than for body length (Sabatés and Saiz, 2000; Conley and Hopkins 2004; Rodríguez-Graña et al., 2005; Morote et al., 2008a, b). As gape size increases, larvae can ingest larger prey (Arthur, 1976; Anderson, 1994; Conway et al., 1994; Voss et al., 2003; Dickmann et al., 2007). Maximum prey size tended to increase with body length in all the studied species, except for larvae of B. argyrogaster. In this species the prey size is constant, a fact which is probably related to the small gape size throughout all larval stages. The analysis of trophic niche breath did not show any relationship to SL. This indicates that there is no trophic specialization in relation to prey size throughout early development because, as previously reported in other species, larvae continue ingesting small prey items in addition to the larger ones (Pearre, 1986; Sabatés and Saiz, 2000; Morote, 2008a, b; Llopiz, 2013; Bernal et al., 2013; Vera-Duarte and Landaeta, 2016). At comparable body lengths, S. diaphana was the species ingesting a higher number of prey and of larger sizes. This contrasts with the published results on juvenile and adult feeding behaviour reported for this species. They indicate that S. diaphana is an inefficient predator with limited searching and catching capacity (MacArthur and Pianka, 1966; Schoener, 1969). Diet The overall diet composition in the different species and stages did not show geographic differences, suggesting that developmental stage is more important than geographical zone. However, the low degree of taxonomic resolution for prey identification that could be reached in these early stages may account for the apparent lack of differences between the zones. The most common and abundant component of the zooplankton samples throughout the study region were copepods (M.L. Fernández de Puelles, pers. obs.,) and these emerged as the most common prey items in the early development of all the studied species. During the larval stages, diet was mainly composed of nauplii and of copepodites <0.2 mm, while the greater development in the transforming stages was reflected in their more diverse diet, which was dominated by adults of several copepods. It has been pointed out that fish larvae may exhibit species-specific selectivity for their prey even from their first-feeding stage (Robert et al., 2008). Our selectivity estimations for larval stages are constrained by the limited microplankton data available (nauplii and copepodites <0.2 mm), and are not presented here as the actual selectivity for the overall plankton populations. However results showed that despite the scarce development during preflexion and flexion stages, some species showed positive selection for small copepodites (B. argyrogaster, S. diaphana, and H. macrochir) instead of nauplii, which were more abundant. According to the literature, the diets of juveniles and adults of A. sladeni in the equatorial Atlantic consists of similar proportions of copepods and euphausiids followed by ostracods (Kinzer and Schulz, 1988). However, in our study the diet changed from copepodites <0.2 mm in larvae, to a more diverse diet dominated by several stages of copepods and ostracods in the transforming stages. It is likely that euphausiids, almost absent in the guts of our specimens, swim too fast to be captured by these early developmental stages. The diet of S. diaphana was more diverse than in the other species, although copepods constituted their main preys. Previous investigations on juvenile and adults have also reported that this species feeds on a variety of prey items, which includes larger zooplankton prey (amphipods and euphausiids) (Hopkins and Baird, 1973; 1985; Kinzer and Schulz, 1988; Carmo et al., 2015). In the present study the largest prey found was the copepod Corycaeus spp., for which a positive selection was observed. Myctophid larvae have been reported to feed mostly on several stages of copepods, with some species also including ostracods in their diets (Sabatés et al., 2003; Conley and Hopkins, 2004; Sassa and Kawaguchi, 2004; Bernal et al., 2013; Tanaka et al., 2013; Contreras et al., 2015). Similarly, in the present study, copepods also emerge as the primary component in the diets of both larvae and transforming stages. Preflexion to postflexion larvae of M. affine showed a more diverse diet than D. cf. vanhoeffeni and H. macrochir, which must be related to the wider MW and greater gut volume, in the former species. The presence of prey of large size, such as copepods of genera Paracalanus and Corycaeus, and of ostracods, was observed only in postflexion and transforming stages. To summarize, in the present investigation we approached the study of the trophic ecology of early life stages of mesopelagic fishes through gut content analysis of larvae and transforming stages of six of the most common and abundant mesopelagic species in our samples. The main difference in feeding patterns among the studied species was that bathylagid and myctophid larvae feed during daylight hours, while sternoptychid larvae are able to feed under low light intensity conditions (i.e. at night, and/or in mesopelagic layers), as do their transforming and adult stages. Unlike their adults, the transitional stages of the myctophids did not show a nocturnally defined feeding pattern. Although all the species examined showed an increase in gape size with development, specialization toward larger prey in transforming stages was not observed. They fed both on small and large prey items. As is generally recorded, gape size constrains the maximum prey size. Larvae with the smallest mouth (B. argyrogaster) fed on smaller prey, while species at similar developmental stages with wider mouths (M. affine or S. diaphana) ingest larger prey. The diets of the different species and stages were dominated by several stages of copepods, suggesting that feeding is dependent on the most abundant and most easily attainable zooplankton items, although the positive selection for particular copepod taxa points to a certain capacity to choose between available preys. The coarse identification reached through gut content analyses points to an important diet overlap among species whose early life stages inhabit the upper 100 m of the water column. To assess this diet overlap, data on the actual prey species constituting the diets would be necessary. Therefore, other types of analyses such as DNA metabar coding of gut contents (Albaina et al., 2016) may be of great support. Acknowledgements The authors are very grateful to all their colleagues who participated in the MAFIA-2015 survey and to the technicians of the Unit of Marine Technology whose various contributions were very important during the sampling. T. Contreras acknowledges the predoctoral fellowship support from Conicyt-Becas Chile. This research was funded by project CTM2012-39587-C04-03. References Albaina A. , Aguirre M. , Abad D. , Santos M. , Estonba A. 2016 . 18S rRNA V9 metabarcoding for diet characterization: a critical evaluation with two sympatric zooplanktivorous fish species . Ecology and Evolution , 6 : 1809 – 1824 . Google Scholar Crossref Search ADS PubMed Anderson J. T. 1994 . Feeding ecology and condition of larval and pelagic juvenile redfish Sebastes spp . Marine Ecology Progress Series , 104 : 211 – 226 . Google Scholar Crossref Search ADS Arthur D. K. 1976 . Food and feeding of larvae of three fishes occurring in the California current, Sardinops sagax, Engraulis mordax, and Trachurus symmetricus . Fishery Bulletin, US , 74 : 517 – 530 . Badcock J. , Merrett N. R. 1976 . Midwater fishes in the eastern North Atlantic-I. Vertical distribution and associated biology in 30°N, 23°W, with developmental notes on certain myctophids . Progress in Oceanography , 7 : 3 – 58 . Google Scholar Crossref Search ADS Baird R. C. 1971 . The systematics, distribution, and zoogeography of the marine hatchetfishes (Fam. Sternoptychidae) . Bulletin of the Museum of Comparative Zoology at Harvard College , 142 : 1 – 128 . Baird R. C. , Hopkins T. L. , Wilson D. F. 1975 . Diet and feeding chronology of Diaphus taaningi (Myctophidae) in the CariacoTrench . Copeia , 1975 : 356 – 365 . Google Scholar Crossref Search ADS Belyanina T. N. 1984 . Developmental sequences of Sternoptyx species (Sternoptychidae) . Journal of Ichthyology , 23 : 73 – 86 . Bernal A. , Olivar M. P. , Fernández de Puelles M. L. 2013 . Feeding patterns of Lampanyctus pusillus (Pisces, Myctophidae) throughout its ontogenetic development . Marine Biology , 160 : 81 – 95 . Google Scholar Crossref Search ADS Bernal A. , Olivar M. P. , Maynou F. , Fernández de Puelles M. L. 2015 . Diet and feeding strategies of mesopelagic fishes in the western Mediterranean . Progress in Oceanography , 135 : 1 – 17 . Google Scholar Crossref Search ADS Blaxter J. H. S. 1963 . The feeding of herring larvae and their ecology in relation to feeding . California Cooperative Oceanic Fisheries Investigations Report , 10 : 79 – 88 . Bonecker A. C. T. , Katsuragawa M. , de Castro M. S. , de A. , Pinto Gomes E. , Namiki C. A. P. , Zani-Teixeira M. L. 2012 . Larval fish of the Campos Basin, southeastern Brazil . Check List , 8 : 1280 – 1291 . Google Scholar Crossref Search ADS Browman H. I. , O'Brien W. J. 1992 . The ontogeny of search behaviour in white grappie, Pomoxis annularis . Environmental Biology of Fishes , 34 : 181 – 195 . Google Scholar Crossref Search ADS Canino M. F. , Bailey K. M. 1995 . Gut evacuation of walleye pollock larvae in response to feeding conditions . Journal of Fish Biololy , 46 : 389 – 403 . Google Scholar Crossref Search ADS Carmo V. , Sutton T. , Menezes G. , Falkenhaug T. , Bergstad O. A. 2015 . Feeding ecology of the Stomiiformes (Pisces) of the northern Mid-Atlantic Ridge. 1. The Sternoptychidae and Phosichthyidae . Progress in Oceanography , 130 : 172 – 187 . Google Scholar Crossref Search ADS Champalbert G. , Kouamé B. , Pagano M. , Marchal E. 2008 . Feeding behavior of adult Vinciguerria nimbaria (Phosichthyidae), in the tropical Atlantic (0°-4° N, 15° W) . Marine Biology , 156 : 79 – 95 . Google Scholar Crossref Search ADS Chesson J. 1978 . Measuring preference in selective predation . Ecology , 59 : 211 – 215 . Google Scholar Crossref Search ADS Clarke T. A. 1980 . Diets of fourteen species of vertically migrating mesopelagic fishes in Hawaiian waters . Fishery Bulletin, US , 78 : 619 – 640 . Clarke K. R. , Gorley R. N. 2006 . PRIMER v6: User Manual/Tutorial. Ed. by PRIMER-E. Plymouth. Conley W. J. , Hopkins T. L. 2004 . Feeding ecology of lanternfish (Pisces: Myctophidae) larvae: prey preferences as a reflection of morphology . Bulletin of Marine Science , 75 : 361 – 379 . Contreras T. , Olivar M. P. , Bernal A. , Sabates A. 2015 . Comparative feeding patterns of early stages of mesopelagic fishes with vertical habitat partitioning . Marine Biology , 162 : 2265–2227. Conway D. V. P. , Coombs S. H. D. , Puelles M. L. F. , Tranter R. P. G. 1994 . Feeding of larval sardine, Sardina pilchardus (Walbaum), off the north coast of Spain . Boletín Instituto Español de Oceanografía , 10 : 165 – 175 . Coombs S. , Nichols J. , Conway D. , Milligan S. , Halliday N. 1992 . Food availability for sprat larvae in the Irish Sea . Journal of the Marine Biological Association, United Kingdom , 72 : 821 – 834 . Google Scholar Crossref Search ADS de Castro M. S. , Richards W. J. , Bonecker A. C. T. 2010 . Occurrence and distribution of larval lanternfish (Myctophidae) from the southwest Atlantic Ocean . Zoologia , 27 : 541 – 553 . Google Scholar Crossref Search ADS Dickmann M. , Mollmann C. , Voss R. 2007 . Feeding ecology of Central Baltic sprat Sprattus sprattus larvae in relation to zooplankton dynamics: implications for survival . Marine Ecology Progress Series , 342 : 277 – 289 . Google Scholar Crossref Search ADS Fahay M. 2007 . Early Stages of Fishes in the Western North Atlantic Ocean (Volumes I-II). Northwest Atlantic Fisheries Organization, Dartmouth, Nova Scotia, Canada. 1692 pp. Gartner J. V. , Crabtree R. E. , Kenneth J. S. 1997 . Feeding at depth. In Deep-Sea Fishes , pp. 115 – 182 . Ed. by Randall D.J. , Farrell A. P. . Academic Press , San Diego . Gjøsaeter J. , Kawaguchi K. 1980 . A review of the world resources of mesopelagic fish . Fisheries Technical Paper , 193 : 1 – 151 . Govoni J. J. , Hoss D. E. , Chester A. J. 1983 . Comparative feeding of three species of larval fishes in the northern Gulf of Mexico: Brevoortia patronus, Leiostomus xanthurus, and Micropogonias undulatus . Marine Ecology Progress Series , 13 : 189 – 199 . Google Scholar Crossref Search ADS Govoni J. J. , Ortner P. , Al-Yamani P. F. , Hill L. C. 1986 . Selective feeding of spot, Leiostomus xanthurus, and Atlantic croaker, Micropogonias undulatus, larvae in the northern Gulf of Mexico . Marine Ecology Progress Series , 28 : 175 – 183 . Google Scholar Crossref Search ADS Hermes R. , Olivar M. P. 1987 . Larval development of Bathylagus argyrogaster Norman 1930 (Teleostei, Bathylagidae) . Investigación Pesquera , 51 : 483 – 489 . Hopkins T. L. , Baird R. C. 1973 . Diet of the hatchetfish Sternoptyx diaphana . Marine Biology , 21 : 34 – 46 . Google Scholar Crossref Search ADS Hopkins T., L. , Baird R. C. 1985 . Feeding ecology of four hatchetfishes (Sternoptychidae) in the eastern Gulf of Mexico . Bulletin of Marine Science , 36 : 260 – 277 . Hopkins T. L. , Gartner J. V. 1992 . Resource-partitioning and predation impact of a low-latitude myctophid community . Marine Biology , 114 : 185 – 197 . Google Scholar Crossref Search ADS Hubbs C. , Blaxter J. H. S. 1986 . Development of sense organs and behaviour of teleost larvae with special reference to feeding and predator avoidance . Transactions of the American Fisheries Society , 115 : 98 – 114 . Google Scholar Crossref Search ADS Hulley P. A. 1981 . Results of the research cruise of FRV ‘‘Walter Herwig” to South America. Family Myctophidae (Osteichthyes, Myctophiformes) . Archives Fischereiwissenschaft , 31 : 1 – 300 . Hulley P. A. 1984 . The South African Museum's Meiring Naude cruises Part 14 Family Myctophidae (Osteichthyes. Myctophiformes) . Annals of the South African Museum , 93 : 53 – 96 . Hulley P. A. , Krefft G. 1985 . A zoogeographic analysis of the fishes of the family Myctophidae (Osteichthyes, Myctophiformes) from the 1979-Sargasso Sea Expedition of RV Anton Dohrn. Annals of the South African Museum, 96: 19 – 53 . Hulley P. A. , Paxton J. R. 2016a . Neoscopelidae. In Bony Fishes, Part 1 (Elopiformes-Scorpaeniformes), the Living Marine Resources of the Eastern Central Atlantic , Vol. 1 , pp. 1855 – 1859 . Ed. by Carpenter K. , De Angelis N. . FAO , Rome . Hulley P. A. , Paxton J. R. 2016b . Myctophidae. In Bony Fishes, Part 1 (Elopiformes-Scorpaeniformes), the Living Marine Resources of the Eastern Central Atlantic , Vol. 3 , pp. 1860 – 1928 . Ed. by Carpenter K. , De Angelis N. . FAO , Rome . Hunter J. R. 1981 . Feeding ecology and predation of marine fish larvae. In Marine Fish Larvae: Morphology, Ecology and Relation to Fisheries , pp. 34 – 37 . Ed. by Lasker R. . Washington Sea Grant Program , Seattle. Huse I. 1994 . Feeding at different illumination levels in larvae of three marine teleosts species: cod, Gadus morhua L., plaice, Pleuronectes platessa L., and turbot, Scophthalmus maximus (L.) . Aquaculture and Fisheries Management , 25 : 687 – 695 . Kinzer J. , Schulz K. 1985 . Vertical distribution and feeding patterns of midwater fish in the central equatorial Atlantic. I . Myctophidae. Marine Biology , 85 : 313 – 322 . Google Scholar Crossref Search ADS Kinzer J. , Schulz K. 1988 . Vertical distribution and feeding patterns of midwater fish in the central equatorial Atlantic. II. Sternoptychidae . Marine Biology , 99 : 261 – 269 . Google Scholar Crossref Search ADS Kurtz F. W. , Matsuura Y. 2001 . Food and feeding ecology of Brazilian sardine (Sardinella brasiliensis) larvae from the southeastern Brazilian Bight . Revista Brasileira de Oceanografia , 49 : 60 – 74 . Google Scholar Crossref Search ADS Landaeta M. F. , Suárez-Donoso N. , Bustos C. A. , Balbontín F. 2011 . Feeding habits of larval Maurolicus parvipinnis (Pisces: sternoptychidae) in Patagonian fjords . Journal of Plankton Research , 33 : 1813 – 1824 . Google Scholar Crossref Search ADS Llopiz J. K. 2013 . Latitudinal and taxonomic patterns in the feeding ecologies of fish larvae: a literature synthesis . Journal of Marine Systems , 109–110 : 69 – 77 . Google Scholar Crossref Search ADS Loeb V. J. 1979 . Vertical distribution and development of larval fishes in the North Pacific central gyre during summer . Fishery Bulletin, US , 77 : 777 – 793 . MacArthur R. H. , Pianka E. 1966 . On optimal use of a patchy environment . The American Naturalist , 100 : 603 – 609 . Google Scholar Crossref Search ADS McClain-Counts J. P. , Demopoulos A. W. J. , Ross S. W. 2017 . Trophic structure of mesopelagic fishes in the Gulf of Mexico revealed by gut content and stable isotope analyses . Marine Ecology , 38 : e12449. Google Scholar Crossref Search ADS McGinnis R. F. 1982 . Biogeography of lanternfishes (Myctophidae) south of 30°S . Antarctic Research Series , 35 : 1 – 110 . Mehner T. , Kasprzak P. 2011 . Partial diel vertical migrations in pelagic fish . Journal of Animal Ecology , 80 : 761 – 770 . Google Scholar Crossref Search ADS PubMed Merrett N. R. , Roe H. S. 1974 . Patterns and selectivity in the feeding of certain mesopelagic fishes . Marine Biology , 28 : 115 – 126 . Google Scholar Crossref Search ADS Morote E. , Olivar M. P. , Pankhurst P. , Villate F. , Uriarte I. 2008a . Trophic ecology of bullet tuna Auxis rochei larvae and ontogeny of feeding-related organs . Marine Ecology Progress Series , 353 : 243 – 254 . Google Scholar Crossref Search ADS Morote E. , Olivar M. P. , Villate F. , Uriarte I. 2008b . Diet of round sardinella, Sardinella aurita, larvae in relation to plankton availability in the NW Mediterranean . Journal of Plankton Research , 30 : 807 – 816 . Google Scholar Crossref Search ADS Morote E. , Olivar M. P. , Villate F. , Uriarte I. 2010 . A comparison of anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) larvae feeding in the Northwest Mediterranean: influence of prey availability and ontogeny . ICES Journal of Marine Science , 67 : 897 – 908 . Google Scholar Crossref Search ADS Moser H. G. , Ahlstrom E. H. 1970 . Development of lanternfishes (family Myctophidae) in the California Current. Part I. Species with narrow-eyed larvae . Bulletin of the Los Angeles County Museum of Natural History Science , 7 : 1 – 145 . Moser H. G. , Ahlstrom E. H. 1974 . Role of larval stages in systematic investigations of marine teleosts: the Myctophidae, a case study . Fishery Bulletin , 72 : 391 – 413 . Moser H. G. , Ahlstrom E. H. 1996 . Myctophidae: Lanternfishes. In The Early Stages of Fishes in the California Current Region. (CalCOFI), Atlas 33 , pp. 387 – 475 . Ed. by Moser H. G. . Allen Press , Kansas . 1505 pp. Moser H. G. , Watson W. 2006 . Myctophidae. In Early Stages of Atlantic Fishes: An Identification Guide for the Western Central North Atlantic , pp. 473 – 589 . Ed. by Richards W. J. , W. J. Taylor & Francis Group. . Moteki M. , Tsujimura E. , Hulley P. A. 2017 . Developmental intervals during the larval and early juvenile stages of the Antarctic myctophid fish Electrona antarctica in relation to changes in feeding and swimming functions . Polar Science , 12 : 88 – 98 . Google Scholar Crossref Search ADS Moyano M. , Rodríguez J. M. , Benítez-Barrios V. M. , Hernández-León S. 2014 . Larval fish distribution and retention in the Canary Current system during the weak upwelling season . Fisheries Oceanography , 23 : 191 – 209 . Google Scholar Crossref Search ADS Namiki C. , Katsuragawa M. , Napolitano D. C. , Zani-Teixeira M. L. D. , Mattos R. A. , Almeida da Silveira I. C. 2017 . Hydrodynamically-driven distribution of lanternfish larvae in the Southeast Brazilian Bight . Journal of Marine Systems , 170 : 115 – 133 . Google Scholar Crossref Search ADS Olivar M. P. , Contreras T. , Hulley P. A. , Emilianov M. , López-Pérez C. , Tuset V. , Castellon A. 2018 . Variation in the diel vertical distributions of larvae and transforming stages of oceanic fishes across the tropical and equatorial Atlantic . Progress in Oceanography , 160 : 83 – 100 . Google Scholar Crossref Search ADS Olivar M. P. , Fortuño D. J. M. 1991 . Guide to Ichthyoplankton of the southeast Atlantic (Benguela current region) . Scientia Marina , 55 : 1 – 383 . Olivar M. P. , Hulley P. A. , Castellón A. , Emelianov M. , López C. , Tuset V. M. , Contreras T. , Molí B. 2017 . Mesopelagic fishes across the tropical and equatorial Atlantic: biogeographical and vertical patterns . Progress in Oceanography , 151 : 116 – 137 . Google Scholar Crossref Search ADS Olivar M. P. , Sabatés A. , Alemany F. , Balbín R. , Fernández de Puelles M. L. , Torres A. P. 2014 . Diel-depth distributions of fish larvae off the Balearic Islands (western Mediterranean) under two environmental scenarios . Journal of Marine Systems , 138 : 127 – 138 . Google Scholar Crossref Search ADS Olivar M. P. , Sabatés A. , Pastor M. V. , Pelegrí J. L. 2016 . Water masses and mesoscale control on latitudinal and cross-shelf variations in larval fish assemblages off NW Africa . Deep Sea Research I , 117 : 120 – 137 . Google Scholar Crossref Search ADS Ozawa T. 1986 . Early life history of the family Myctophidae in the ocean off southern Japan. In Studies on the Oceanic Ichthyoplankton in the Western North Pacific , pp. 114 – 187 . Ed. by Ozawa T. . Kyushu University Press , Hukuoka . 68–73 pp. Palma S. 1990 . Ecologie alimentaire de Cyclothone braueri Jespersen et Taning, 1926 (Gonostomatidae) en mer Ligure, Méditerranée occidentale . Journal of Plankton Research , 12 : 519 – 534 . Google Scholar Crossref Search ADS Pearre S. 1986 . Ratio-based trophic niche breadths of fish, the Sheldon spectrum, and the size-efficiency hypothesis . Marine Ecology Progress Series , 27 : 287 – 314 . Google Scholar Crossref Search ADS Pepin P. , Robert D. , Bouchard C. , Dower J. F. , Falardeau M. , Fortier L. , Jenkins G. P. et al. 2014 . Once upon a larva: revisiting the relationship between feeding success and growth in fish larvae . ICES Journal of Marine Science , 72 : 359 – 373 . Google Scholar Crossref Search ADS Reygondeau G. , Guidi L. , Beaugrand G. , Koubbi P. , MacKenzie B. R. , Sutton T. T. , Fioroni M. et al. 2017 . Global biogeopchemical provinces of the mesopelagic zone . Journal of Biogeography . doi: 10.1111/jbi.13149. Richards W. 2006 . Early Stages of Atlantic Fishes: An Identification Guide for the Western Central North Atlantic (Volumes I-II) . Taylor & Francis Group , London, UK . 1335 pp. Rissik D. , Suthers I. M. 2000 . Enhanced feeding by pelagic juvenile myctophid fishes within a region of island-induced flow disturbance in the Coral Sea . Marine Ecology Progress Series , 203 : 263 – 273 . Google Scholar Crossref Search ADS Robert D. , Castonguay M. , Fortier L. 2008 . Effects of intra- and inter-annual variability in prey field on the feeding selectivity of larval Atlantic mackerel (Scomber scombrus) . Journal of Plankton Research , 30 : 673 – 688 . Google Scholar Crossref Search ADS Robert D. , Murphy H. M. , Jenkins G. P. , Fortier L. 2014 . Poor taxonomical knowledge of larval fish prey preference is impeding our ability to assess the existence of a “critical period” driving year-class strength . ICES Journal of Marine Science , 71 : 2042 – 2052 . Google Scholar Crossref Search ADS Rodríguez-Graña L. , Castro L. , Loureiro M. , González H. E. , Calliari D. 2005 . Feeding ecology of dominant larval myctophids in an upwelling area of the Humboldt Current . Marine Ecology Progress Series , 290 : 119 – 134 . Google Scholar Crossref Search ADS Rose M. , Tregouboff G. 1957 . Manuel de planctonoligie Mediterraneenne. Tome I, II . Centre National de la Recherche Scientifique , Paris . 587 pp. Sabatés A. 2004 . Diel variability of fish larvae distribution during the winter mixing period in the NW Mediterranean . ICES Journal of Marine Science , 61 : 1243 – 1252 . Google Scholar Crossref Search ADS Sabatés A. , Bozzano A. , Vallvey I. 2003 . Feeding pattern and the visual light environment in myctophid fish larvae . Journal of Fish Biology , 63 : 1476 – 1490 . Google Scholar Crossref Search ADS Sabatés A. , Saiz E. 2000 . Intra-and interspecific variability in prey size and niche breadth of myctophiform fish larvae . Marine Ecology Progress Series , 201 : 261 – 271 . Google Scholar Crossref Search ADS Sánchez-Velasco L. , Contreras-Arredondo I. , Esqueda-Escarcega G. 1999 . Diet composition of Euthynnus lineatus and Auxis sp. larvae (Pisces: Scombridae) in the Gulf of California . Bulletin of Marine Science , 65 : 687 – 698 . Sassa C. , Kawaguchi K. 2004 . Larval feeding habits of Diaphus garmani and Myctophum asperum (Pisces: Myctophidae) in the transition region of the western North Pacific . Marine Ecology Progress Series , 278 : 279 – 290 . Google Scholar Crossref Search ADS Sassa C. , Kawaguchi K. , Hirota Y. , Ishida M. 2007 . Distribution depth of the transforming stage larvae of myctophid fishes in the subtropical-tropical waters of the western North Pacific . Deep Sea Research , 54 : 2181 – 2193 . Google Scholar Crossref Search ADS Schoener T. W. 1969 . Models of optimal size for solitary predators . The American Naturalist , 103 : 277 – 313 . Google Scholar Crossref Search ADS Sutton T. T. 2013 . Vertical ecology of the pelagic ocean: classical patterns and new perspectives . Journal of Fish Biology , 83 : 1508 – 1527 . Google Scholar Crossref Search ADS PubMed Sutton T. T. , Hopkins T. L. 1996 . Trophic ecology of the stomiid (Pisces: Stomiidae) fish assemblage of the eastern Gulf of Mexico: strategies, selectivity and impact of a top mesopelagic predator group . Marine Biology , 127 : 179 – 192 . Google Scholar Crossref Search ADS Tanaka H. , Sassa C. , Ohshimo S. , Aoki I. 2013 . Feeding ecology of two lanternfishes Diaphus garmani and Diaphus chrysorhynchus . Journal of Fish Biology , 82 : 1011 – 1031 . Google Scholar Crossref Search ADS PubMed Vera-Duarte J. , Landaeta M. F. 2016 . Diet of labrisomid blenny Auchenionchus variolosus (Valenciennes 1836) (Labrisomidae) during its larval development off central Chile (2012–2013) . Journal of Applied Ichthyology , 32 : 46 – 54 . Google Scholar Crossref Search ADS Vives F. , Shmeleva A. A. 2007 . Crustácea, Copépodos marinos I. Calanoida. Fauna Ibérica, Vol. 29. Museo Nacional de Ciencias Naturales, Consejo Superior de Investigaciones Científicas (CSIC), Madrid. 1156 pp. Vives F. , Shmeleva A. A. 2010 . Crustácea, Copépodos marinos II. Non Calanoida. Fauna Ibérica, Vol. 33. Museo Nacional de Ciencias Naturales, Consejo Superior de Investigaciones Científicas (CSIC), Madrid. 492 pp. Voss R. , Køster F. W. , Dickmann M. 2003 . Comparing the feeding habits of co-occurring sprat (Sprattus sprattus) and cod (Gadus morhua) larvae in the Bornholm Basin, Baltic Sea . Fisheries Research , 63 : 97 – 111 . Google Scholar Crossref Search ADS Watanabe H. , Kawaguchi K. 2003 . Decadal change in the diets of the surface migratory myctophid fish Myctophum nitidulum in the Kuroshio region of the western North Pacific: predation on sardine larvae by myctophids . Fisheries Science , 69 : 716 – 721 . Google Scholar Crossref Search ADS Watanabe H. , Kawaguchi K. , Hayashi A. 2002 . Feeding habits of juvenile surface-migratory myctophid fishes (family Myctophidae) in the Kuroshio region of the western North Pacific . Marine Ecology Progress Series , 236 : 263 – 272 . Google Scholar Crossref Search ADS Watson W. 1996 . Sternoptychidae; hatchetfishes. In The Early Stages of Fishes in the California Current Region. (CalCOFI), Atlas 33, pp. 268–283 . Ed. by Moser H. G. . Allen Press , Kansas . 1505 pp. Weitzman S. H. 1997 . Systematics of deep-sea fishes. In Deep-Sea Fishes , pp. 43 – 74 . Ed. by Randall D. J. , Farrell A. P. . Academic Press , London . Young J. W. , Davis T. L. O. 1990 . Feeding ecology of larvae of southern bluefin, albacore and skipjack tunas (Pisces: Scombridae) in the eastern Indian Ocean . Marine Ecology Progress Series , 61 : 17 – 29 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2018. All rights reserved. For permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Anderson, Thomas R; Martin, Adrian P; Lampitt, Richard S; Trueman, Clive N; Henson, Stephanie A; Mayor, Daniel J

doi: 10.1093/icesjms/fsx234pmid: N/A

Abstract An ecosystem-based flow analysis model was used to study carbon transfer from primary production (PP) to mesopelagic fish via three groups of copepods: detritivores that access sinking particles, vertical migrators, and species that reside in the surface ocean. The model was parameterized for 40°S to 40°N in the world ocean such that results can be compared with recent estimates of mesopelagic fish biomass in this latitudinal range, based on field studies using acoustic technologies, of ∼13 Gt (wet weight). Mesopelagic fish production was predicted to be 0.32% of PP which, assuming fish longevity of 1.5 years, gives rise to predicted mesopelagic fish biomass of 2.4 Gt. Model ensembles were run to analyse the uncertainty of this estimate, with results showing predicted biomass >10 Gt in only 8% of the simulations. The work emphasizes the importance of migrating animals in transferring carbon from the surface ocean to the mesopelagic zone. It also highlights how little is known about the physiological ecology of mesopelagic fish, trophic pathways within the mesopelagic food web, and how these link to PP in the surface ocean. A deeper understanding of these interacting factors is required before the potential for utilizing mesopelagic fish as a harvestable resource can be robustly assessed. Introduction Global demand for food is ever increasing, driven by a rising population that is expected to reach nine billion by the year 2050 (Godfray et al., 2010). Mesopelagic fish (the mesopelagic zone has a depth range of 100–1000 m) have been identified as a potential underexploited resource, providing both fishmeal and nutraceutical products such as omega-3 fatty acid dietary supplements for human consumption and the aquaculture industry (St. John et al., 2016 and references therein). Sustainability is a key requirement for the successful exploitation of mesopelagic resources necessitating, at the very minimum, an accurate estimate of the existing biomass of fish inhabiting the mesopelagic ocean. Early estimates of this biomass, based on micronekton net sampling, were ∼1 gigatonne (Gt) wet weight (Gjösæter and Kawaguchi, 1980). It may be, however, that this estimate is too low because mesopelagic fish can avoid pelagic trawls (Kaartvedt et al., 2012). Acoustic methods have more recently been used to evaluate the biomass of mesopelagic fish, with one recent estimate suggesting that the biomass occurring between 40°S and 40°N in the world ocean is ∼11–15 Gt (average 13 Gt) (Irigoien et al., 2014). Acoustic estimates exceed those of trawls (Koslow et al., 1997; Kloser et al., 2009; Davison et al., 2015a) although the associated uncertainties are nevertheless substantial, arising from difficulties with interpreting the resonance of gas-filled swim bladders that are present not only in fish, but also other organisms such as siphonophores (Davison et al., 2015b; Kloser et al., 2016). Mesopelagic fish are one of the least well-studied components of marine ecosystems (St. John et al., 2016). The most common of the mesopelagic fishes are the lanternfish (myctophids), which are ubiquitous throughout the world oceans, with the exception of the Arctic, and which comprise 33 genera and around 245 species (Catul et al., 2011). They feed primarily on planktonic crustaceans such as copepods (Battaglia et al., 2016) and, in turn, provide a source of food for higher trophic levels including squid, mammals and seabirds (Springer et al., 1999; Pereira et al., 2011; Hoving and Robison, 2016). Mesopelagic fish are also an essential prey item for key fishery stocks such as tuna and billfish (Potier et al., 2007; Karakulak et al., 2009). The mesopelagic food web may also play an important role in the sequestration of carbon (C) in the deep ocean and thereby climate regulation (Hudson et al., 2014; Trueman et al., 2014). Many mesopelagic fish undergo diel vertical migration (DVM) to feed in the epipelagic zone (ocean surface to 100 m) at night, descending back to the mesopelagic during the day (Watanabe et al., 1999; Bernal et al., 2015; Klevjer et al., 2016). This migration represents a direct pathway for export of organic C between the surface and deep ocean, along with the production of large fast-sinking faecal material (Davison et al., 2013). Here, we use an ecosystem-based modelling approach to investigate the trophic pathways that connect primary production (PP) in the surface ocean with C flow through the mesopelagic ecosystem, providing an estimate of the biomass of mesopelagic fish between 40°S and 40°N in the world ocean. The model traces the flows of C from PP through the mesopelagic food web via three groups of copepods: permanent residents of the mesopelagic zone that feed on sinking detritus, vertically migrating organisms, and permanent residents of the epipelagic. Results are discussed in context of the many uncertainties associated with the functioning of the mesopelagic ecosystem, highlighting areas in particular need of future research in order to assess the potential sustainability of mesopelagic fish as a harvestable resource. Model description and methods A flow diagram of the model, illustrating the trophic pathways from PP to mesopelagic fish, is shown in Figure 1 along with a list of variables and parameters in Tables 1 and 2. The model is a flow analysis, assumes steady state and does not calculate stocks, with the exception of mesopelagic fish. It is parameterized to represent a generalized ecosystem between 40°S to 40°N in the world ocean so that results can be directly compared with field-based estimates of mesopelagic fish biomass of ∼13 Gt for this latitudinal range (Irigoien et al., 2014). The model tracks the flows of C from PP to mesopelagic fish via copepods and their carnivorous invertebrate predators. Given that the fish undertake DVM, food sources within both the mesopelagic and epipelagic zones are considered. Three types of copepods are distinguished: detritivores, ZD, that are permanent residents of the mesopelagic zone and which feed on sinking particles (detritus, D), migratory copepods, ZM, which, like the mesopelagic fish, feed in epipelagic waters by night and descend in the water column by day to escape predation (Zaret and Suffern, 1976; Hays, 2003) and permanent residents of the epipelagic, ZR, that feed on phytoplankton and the associated microzooplankton assemblage. All three groups are grazed not only by the fish, but also by invertebrate carnivores (V) which represent a wide variety of organisms including amphipods, chaetognaths and jellyfish (Tönnesson and Tiselius, 2005; Daewel et al., 2014). Both copepods and invertebrate carnivores constitute prey for the mesopelagic fish. Table 1. Model variables. Variable . Definition . Unit of measure . GZ(D) Copepod production: detritivores Gt C year−1 GZ(M) Copepod production: vertical migrators Gt C year−1 GZ(R) Copepod production: resident epipelagic Gt C year−1 RZ Copepod food available to V, F Gt C year−1 GV Invertebrate carnivore production Gt C year−1 GF Mesopelagic fish production Gt C year−1 LF Loss rate of fish Gt C year−1 BF Mesopelagic fish biomass Gt wet weight Variable . Definition . Unit of measure . GZ(D) Copepod production: detritivores Gt C year−1 GZ(M) Copepod production: vertical migrators Gt C year−1 GZ(R) Copepod production: resident epipelagic Gt C year−1 RZ Copepod food available to V, F Gt C year−1 GV Invertebrate carnivore production Gt C year−1 GF Mesopelagic fish production Gt C year−1 LF Loss rate of fish Gt C year−1 BF Mesopelagic fish biomass Gt wet weight Open in new tab Table 1. Model variables. Variable . Definition . Unit of measure . GZ(D) Copepod production: detritivores Gt C year−1 GZ(M) Copepod production: vertical migrators Gt C year−1 GZ(R) Copepod production: resident epipelagic Gt C year−1 RZ Copepod food available to V, F Gt C year−1 GV Invertebrate carnivore production Gt C year−1 GF Mesopelagic fish production Gt C year−1 LF Loss rate of fish Gt C year−1 BF Mesopelagic fish biomass Gt wet weight Variable . Definition . Unit of measure . GZ(D) Copepod production: detritivores Gt C year−1 GZ(M) Copepod production: vertical migrators Gt C year−1 GZ(R) Copepod production: resident epipelagic Gt C year−1 RZ Copepod food available to V, F Gt C year−1 GV Invertebrate carnivore production Gt C year−1 GF Mesopelagic fish production Gt C year−1 LF Loss rate of fish Gt C year−1 BF Mesopelagic fish biomass Gt wet weight Open in new tab Table 2. Model parameters (Group 1: relatively certain; Group 2: relatively uncertain). Parameter . Definition . Default value (range) . Unit of measure . Group 1 (varied ± 50% in the ensemble uncertainty analysis) PP PP 43 (43–63) Gt C year−1 fPP,D Frac. PP exported as detritus 0.11 (0.033–0.199) Dimensionless fPP,Z Frac. PP to copepods 0.32 (0.25–0.47) Dimensionless fZM Frac. fPP, Z due to migrators 0.18 (0.12–0.22) Dimensionless KZ GGE: copepods 0.26 (0.13–0.32) Dimensionless KV GGE: carnivores 0.22 (0.10–0.51) Dimensionless KF GGE: fish 0.2 (0.18–0.21) Dimensionless Group 2 (varied ± 75% in the ensemble uncertainty analysis) tD,Z Transfer eff.: detritus to copepods 0.0145 Dimensionless fM,VF Frac. migrating copepods to V, F 0.88 Dimensionless fR,VF Frac. resident copepods to V, F 0.18 Dimensionless fZ,F Frac. copepods grazed by fish 0.5 Dimensionless fV,F Frac. carnivores grazed by fish 0.8 Dimensionless mF Mesopelagic fish mortality 0.67 Year−1 Parameter . Definition . Default value (range) . Unit of measure . Group 1 (varied ± 50% in the ensemble uncertainty analysis) PP PP 43 (43–63) Gt C year−1 fPP,D Frac. PP exported as detritus 0.11 (0.033–0.199) Dimensionless fPP,Z Frac. PP to copepods 0.32 (0.25–0.47) Dimensionless fZM Frac. fPP, Z due to migrators 0.18 (0.12–0.22) Dimensionless KZ GGE: copepods 0.26 (0.13–0.32) Dimensionless KV GGE: carnivores 0.22 (0.10–0.51) Dimensionless KF GGE: fish 0.2 (0.18–0.21) Dimensionless Group 2 (varied ± 75% in the ensemble uncertainty analysis) tD,Z Transfer eff.: detritus to copepods 0.0145 Dimensionless fM,VF Frac. migrating copepods to V, F 0.88 Dimensionless fR,VF Frac. resident copepods to V, F 0.18 Dimensionless fZ,F Frac. copepods grazed by fish 0.5 Dimensionless fV,F Frac. carnivores grazed by fish 0.8 Dimensionless mF Mesopelagic fish mortality 0.67 Year−1 Ranges of Group 1 parameters, as described in the text, are listed in brackets. No ranges are given for Group 2 parameters because these are highly uncertain (see text). Open in new tab Table 2. Model parameters (Group 1: relatively certain; Group 2: relatively uncertain). Parameter . Definition . Default value (range) . Unit of measure . Group 1 (varied ± 50% in the ensemble uncertainty analysis) PP PP 43 (43–63) Gt C year−1 fPP,D Frac. PP exported as detritus 0.11 (0.033–0.199) Dimensionless fPP,Z Frac. PP to copepods 0.32 (0.25–0.47) Dimensionless fZM Frac. fPP, Z due to migrators 0.18 (0.12–0.22) Dimensionless KZ GGE: copepods 0.26 (0.13–0.32) Dimensionless KV GGE: carnivores 0.22 (0.10–0.51) Dimensionless KF GGE: fish 0.2 (0.18–0.21) Dimensionless Group 2 (varied ± 75% in the ensemble uncertainty analysis) tD,Z Transfer eff.: detritus to copepods 0.0145 Dimensionless fM,VF Frac. migrating copepods to V, F 0.88 Dimensionless fR,VF Frac. resident copepods to V, F 0.18 Dimensionless fZ,F Frac. copepods grazed by fish 0.5 Dimensionless fV,F Frac. carnivores grazed by fish 0.8 Dimensionless mF Mesopelagic fish mortality 0.67 Year−1 Parameter . Definition . Default value (range) . Unit of measure . Group 1 (varied ± 50% in the ensemble uncertainty analysis) PP PP 43 (43–63) Gt C year−1 fPP,D Frac. PP exported as detritus 0.11 (0.033–0.199) Dimensionless fPP,Z Frac. PP to copepods 0.32 (0.25–0.47) Dimensionless fZM Frac. fPP, Z due to migrators 0.18 (0.12–0.22) Dimensionless KZ GGE: copepods 0.26 (0.13–0.32) Dimensionless KV GGE: carnivores 0.22 (0.10–0.51) Dimensionless KF GGE: fish 0.2 (0.18–0.21) Dimensionless Group 2 (varied ± 75% in the ensemble uncertainty analysis) tD,Z Transfer eff.: detritus to copepods 0.0145 Dimensionless fM,VF Frac. migrating copepods to V, F 0.88 Dimensionless fR,VF Frac. resident copepods to V, F 0.18 Dimensionless fZ,F Frac. copepods grazed by fish 0.5 Dimensionless fV,F Frac. carnivores grazed by fish 0.8 Dimensionless mF Mesopelagic fish mortality 0.67 Year−1 Ranges of Group 1 parameters, as described in the text, are listed in brackets. No ranges are given for Group 2 parameters because these are highly uncertain (see text). Open in new tab Figure 1. Open in new tabDownload slide Flow diagram of the model showing pathways from PP to fish (with fX parameters specifying fractional division of fluxes). Sinks for C are: S1 remineralisation of C and loss to higher trophic levels in the epipelagic zone; S2 detritus-associated losses; S3 respiration and egestion by ZM, ZR; S4 respiration and egestion by invertebrate carnivores; S5 non-fish carnivore losses to higher trophic levels; S6 respiration and egestion by fish (zooplankton as food); S7 respiration and egestion by fish (carnivores); S8 fish mortality. Flows (Gt C year−1) are shown in red for the steady state solution of the model (see “Results” section). The starting point of the analysis is PP, which is specified as a fixed rate, PP (Gt C year−1). Satellite-based estimates of PP between 40°S and 40°N were extracted from original published fields for a range of algorithms, giving 46.3 (Behrenfeld and Falkowski, 1997), 63.3 (Carr, 2002), 47.9 (Marra et al., 2003), and 43.0 Gt C year−1 (Westberry et al., 2008). We use the last of these, which is based on the most sophisticated and up-to-date methodology, giving PP = 43 Gt C year−1. A fraction of PP, fPP,D, is exported from the surface ocean to the mesopelagic as detritus, providing a food supply for detritivorous zooplankton. Empirical estimates of export ratio (e-ratio = export/PP = fPP,D) averaged between 40°S and 40°N, again extracted from the original data for different algorithms, are variable: 0.091 (Dunne et al., 2005), 0.033 (Henson et al., 2011), and 0.099 (Siegel et al., 2014). Estimates from global biogeochemical models include 0.128 (Moore et al., 2004), 0.198–0.199 (Collins et al., 2011), 0.067–0.077 (Dunne et al., 2012, 2013), and 0.145–0.179 (Séférian et al., 2013). Parameter fPP,D was estimated as an average of these values, giving 0.11. Sinking detritus is acted upon by bacteria and zooplankton in the mesopelagic zone, diminishing the flux with depth and remineralising nutrients and C to their inorganic forms (Steinberg et al., 2008; Giering et al., 2014). It may be that the primary source of nutrition for detritivorous zooplankton is not the non-living detrital matrix, which is refractory in nature but, rather, microbes that colonize the detritus and which are rich in nutrients such as essential fatty acids (Lampitt et al., 1990; Mayor et al., 2014; Anderson et al., 2017). It is therefore difficult to specify a growth efficiency for zooplankton grazing on detritus because of the separate contributions of microbial and refractory substrates and the associated differential ingestion and utilization thereof. Instead, we apply a simple trophic transfer efficiency, tD,Z, which represents the aggregate outcome of detritus utilization, quantifying the fraction of exported detritus, including associated microbes, that is assimilated into the biomass of mesopelagic zooplankton. Copepod growth accruing from the detrital pathway, GZ(D) is then: GZ(D)=tD,ZfPP,DPP.(1) The nutrition of detritivorous zooplankton in the mesopelagic zone was investigated by Anderson et al. (2017) using a stoichiometric model that has C and an essential fatty acid as currencies, and which includes the separate roles of detritus-attached microbes and the raw detrital substrate. Results indicated that feeding on microbes is likely a favourable strategy, despite low microbial biomass. Transfer of C to zooplankton was remarkably low because of trophic losses within the microbial food web prior to ingestion, with a typical value of 1.45%, i.e. tD,Z = 0.0145. This value is much lower than growth efficiencies for zooplankton grazing detritus as applied in previous mesopelagic ecosystem models, e.g. 0.23 (Anderson and Tang, 2010; Giering et al., 2014). We therefore investigate the sensitivity of model-predicted mesopelagic fish biomass to parameter tD,Z. Mesopelagic fish feed on herbivorous, omnivorous and carnivorous copepods, both residents of the epipelagic and those undergoing DVM. The starting point of quantifying these pathways is to calculate the fraction of PP that is consumed by copepods, parameter fPP,Z. It has been estimated that mesozooplankton directly consume 10–15% of PP (Behrenfeld and Falkowski, 1997; Calbet, 2001), giving a mid-range value of 12%. However, this excludes indirect routes, notably via microzooplankton. Steinberg and Landry (2017) constructed a C budget of the global ocean ecosystem and estimated that 66% of PP is consumed by microzooplankton, with a further 10% of PP accounting for consumption of bacterial secondary production, giving a total intake by microzooplankton of 76% of PP. If the gross growth efficiency (GGE) for microzooplankton is 0.3 (Straile, 1997), and mesozooplankton grazing is the primary loss term for microzooplankton, the fraction of PP reaching mesozooplankton via these routes is 0.76 × 0.3 = 0.23. When added to direct consumption of 0.12, this results in mesozooplankton consumption of 35% of PP. Copepod grazing can also be estimated from the predictions of global biogeochemical models, most of which today distinguish between micro- and mesozooplankton. Mesozooplankton grazing of between 19 and 11.2 Gt C year−1 was predicted using different versions of the PISCES model by Buitenhuis et al. (2006) and Aumont et al. (2015), respectively, which convert to fractions of PP of 0.27 and 0.25 (based on PP of 69.7 vs. 44.3 Gt C year−1 for the two models). A similar value can be calculated from the results of the COBALT model as published in Stock et al. (2014), with mesoplankton grazing and PP of 13.4 and 51.9 Gt C year−1, respectively, giving grazing/PP = 0.26. On the other hand, a considerably higher ratio, 0.47, is generated by the MEDUSA model (Yool et al., 2013), caused by a relatively high grazing on non-diatoms. We use a mean of the estimates described earlier, giving fPP,Z = 0.32. The efficiency with which copepod consumption of PP, fPP,Z, is transferred to the mesopelagic ecosystem depends on whether the copepods undergo DVM or not. Migrating copepods are tightly coupled to, and indeed part of, the mesopelagic ecosystem given that many of the invertebrate carnivores and mesopelagic fish themselves undergo DVM. On the other hand, non-migrating copepods may contribute relatively little to the mesopelagic ecosystem if they are largely consumed by epipelagic resident predators. The migrating fraction of total mesozooplankton biomass is commonly estimated as the difference between day and night measurements in surface waters. Variability is inevitably seen between locations and seasons, including estimates of 0.14–0.68, mean 0.37 (equatorial Pacific: Zhang and Dam, 1997), 0.41 (Sargasso Sea: Madin et al., 2001), 0.35-0.53, mean 0.41 (subtropical Pacific: Al-Mutairi and Landry, 2001), 0.29 and 0.44 for copepods (subtropical and subarctic Pacific respectively: Steinberg et al., 2008), and 0.06–0.40 (Mediterranean: Isla et al., 2015). Averaging all these results yields a migrating fraction of 0.37. Parameter fZM in the model is the fraction of total copepod grazing in the epipelagic zone that is attributable to migrating animals. This depends not only on migrator biomass, but also on the duration of feeding in surface waters. The simplest assumption is that migrators spend, on average, 50% of the day (the night hours) feeding in surface waters and so their relative contribution to grazing PP, per unit biomass, is half compared with non-migrating animals. Thus, fZM = 0.37 × 0.5 = 0.18. The growth of migrating and resident epipelagic copepods, GZ(M) and GZ(R), are now: GZ(M)=fPP,ZfZMKZPP,(2) GZ(R)=fPP,Z(1−fZM)KZPP,(3) where KZ is copepod GGE. We use KZ = 0.26 (a typical range is 0.13–0.32; Straile, 1997). Copepods, both migrating and resident in the epipelagic zone, provide food not only for migrating fish and carnivorous invertebrates, but also for higher trophic levels that reside in near-surface waters (Springer et al., 2007; Bachiller et al., 2016). For the purpose of calculating carbon transfer from copepods to higher trophic levels, we assume that the biomass of migrating predators as a fraction of total predators (migrators and epipelagic residents) is 0.37 (no estimates exist), i.e. equal to the corresponding fraction for copepods, and that migrating predators spend an average of 50% of their time in surface waters. The fractional loss of epipelagic resident (non-migratory) copepods that is due to grazing by migrating predators (fish and invertebrates), parameter fR,VF, is then equivalent to fZM, i.e. fR,VF = 0.18. In contrast, the fractional loss of migrating copepods to migrating predators, parameter fM,VF, should be high given the synchrony in behaviour of the two communities. Moreover, if epipelagic resident predators rely on visual cues for feeding then, at least in theory, their grazing on migrating copepods should be zero if the copepods avoid surface waters during daylight hours by DVM. Migrating copepod species do, however, contribute to the diets of epipelagic fish (Beaugrand et al., 2003; Garrido et al., 2015; Bachiller et al., 2016). These fish have evolved complex adaptations to detect prey in dimly lit waters, notably large upward-facing eyes (Gagnon et al., 2013). Feeding may follow a bimodal cycle whereby zooplankton are most effectively captured at dawn and dusk as they migrate to and from surface waters (Allison et al., 1996; Cardinale et al., 2003). As a first approximation, we assume that epipelagic predators access migrating copepods as food for 1 h at either end of the day (dawn and dusk), whereas migrating predators have continuous access. Taking into consideration the relative biomass of the two predator communities (0.37 and 0.63 for migrators and epipelagic residents, respectively), the relative grazing by migrating and epipelagic resident predators on migrating copepods are 0.37 and 0.052 (=0.63 × 2/24), giving fM,VF (the grazing by migrating predators on migrating copepods) as 0.37/(0.37 + 0.052) = 0.88. The total amount of copepod food available to migrating fish and carnivorous invertebrates, RZ, is now: RZ=GZ(D)+fM,VFGZ(M)+fR,VFGZ(R).(4) There are no data and so we tentatively assume that the fraction of RZ that is consumed by mesopelagic fish, parameter fZ,F, is 0.5, with the remainder, 1−fZ,F, consumed by invertebrate carnivores whose growth, GV, is: GV=(1−fZ,F)KVRZ,(5) where KV is invertebrate carnivore GGE. A number of values of GGE for amphipods and decapods have been published: 0.15–0.18 (Dagg, 1976), 0.16 (Ikeda, 1991) and 0.20–0.51 (Yamada and Ikeda, 2006). Values for gelatinous organisms also show considerable variation, from values around 0.1 (Larson, 1987; Møller and Riisgård, 2007; Møller et al., 2010) to over 0.3 (Reeve et al., 1989; Costello, 1991). We use an average of the above values, KV = 0.22. The fraction of invertebrate carnivore losses to mesopelagic fish was assumed to be high, fV,F = 0.8; we will show that the model is not sensitive to this parameter. The production of mesopelagic fish, GF, can now be calculated as the sum of terms representing ingestion of copepods and invertebrate carnivores: GF=KF(fZ,FRZ+fV,FGV),(6) where KF is fish GGE. Ikeda (1996) investigated the metabolism and energy budget of the mesopelagic fish Maurolicus muelleri and estimated a GGE of 0.18. Transfer efficiency for secondary to tertiary consumers in the pelagic ecosystem of the Oyashio Region was subsequently estimated by Ikeda et al. (2008) to be 0.21. We use KF = 0.20. By assuming that the system is in steady state, the biomass (wet weight) of mesopelagic fish, BF, can be estimated by dividing production by the fish mortality rate, mF: BF=GFσ/mF,(7) where the conversion coefficient, σ, is 11.9 g wet weight per g C dry weight based on conversions of 0.20 for dry: wet weight and 0.42 for C as fraction dry weight (Ikeda et al., 2011). The longevity of mesopelagic fish is typically recorded as being between 1 and 2 years (Bystydzieńska et al., 2010; Linkowski et al., 1993; Takagi et al., 2006; Hosseini-Shekarabi et al., 2015), although sometimes up to 3 or 4 years at high latitudes (Halliday, 1970; Gjösæter, 1973; Greely et al., 1999; Saunders et al., 2015). We use a fish mortality rate of mF = 1/1.5 = 0.67 year−1, but will examine the sensitivity of BF to this parameter. The model was first investigated with default parameter settings (Table 2). The uncertainty of the predicted biomass of mesopelagic fish associated with model parameters was then assessed in two ways. First, a standard sensitivity analysis was carried out in which individual parameters were varied, in turn, ±50%. Second, an ensemble analysis was undertaken in which model solutions were generated throughout the entire 13D parameter space, randomly assigning values for each parameter within specified ranges. For this purpose, model parameters were divided into two groups (Table 2) based on our experiences parameterising the model: Group 1 (relatively certain) could be constrained with some confidence from the literature, whereas Group 2 (relatively uncertain) are poorly known and weakly constrained. In effect, this is a process of expert elicitation that is often used in constructing Bayesian frameworks (Choy et al., 2009; Krueger et al., 2012). Group 1 parameters are: PP, the fractions of PP to detritus and copepods (fPP,D and fPP,Z), the fraction of fPP,Z attributable to migrators (fZM) and the growth efficiencies for copepods, invertebrate carnivores and mesopelagic fish (KZ, KV, and KF). Group 2 parameters are: tD,Z (transfer efficiency of detritus utilization by copepods), trophic pathway partitioning parameters fM,VF (migrating copepods to mesopelagic V, F), fR,VF (epipelagic resident copepods to mesopelagic V, F), fZ,F (copepod fraction to mesopelagic fish), fV,F (invertebrate carnivore fraction to mesopelagic fish) and, finally, mF (the mortality rate of mesopelagic fish). Group 1 parameters were varied ±50% in the ensemble uncertainty analysis, while Group 2 parameters were varied ±75%. These are wide limits and, if anything, should overestimate the total uncertainty associated with predicting BF. The chosen percentages are subjective but, nevertheless, we believe generally representative based on our experience of parameterising the model from the literature. The model is set up in a Microsoft Excel spreadsheet, which is available on request from the first author. Results The model was first run with default parameters (Table 2), including input PP of 43 Gt C year−1. Predicted C fluxes from PP to fish are shown in Figure 2 (see also the steady state solution of the model as shown in Figure 1). The majority of PP is remineralized or lost to epipelagic predators, with only 21% (8.9/43) predicted to supply the mesopelagic ecosystem via sinking detritus (11%; parameter fPP,D), vertically migrating copepods (5.1%; the product fPP,ZfZ,MfM,VF) and resident epipelagic copepods (4.7%; fPP,Z(1−fZM)fR,VF). The total production of the prey of mesopelagic fish, i.e. copepods and invertebrate carnivores, is 1.3 Gt C year−1, most of which is due to migrant and resident epipelagic copepods that have direct access to PP. The predicted contribution via detritivorous copepods is small (GZ(D) = 0.07 Gt C year−1) because of the low trophic transfer efficiency (tD,Z = 0.0145) associated with using refractory detritus as a source of food. Likewise, the predicted contribution of invertebrate carnivores to mesopelagic fish diet is low (GV = 0.13 Gt C year−1) because these animals are one trophic level above the copepods. The mesopelagic fish consume 0.68 Gt C year−1 of the 1.3 Gt C year−1 available to them (the remainder goes to other predators), equivalent to 1.6% of PP. With a growth efficiency, KF, of 0.2, the predicted production of mesopelagic fish is 0.14 Gt C year−1, i.e. 0.32% of PP. Mesopelagic fish biomass can now be estimated as the quotient production/mortality (GF/mF; Equation 7). Using mF = 0.67 year−1 (Table 2), which equates to a fish longevity of 1.5 years, predicted mesopelagic fish biomass is 2.4 Gt C wet weight (11.9 g wet weight per g C dry weight). Figure 2. Open in new tabDownload slide Predicted fluxes of carbon from PP to fish. Columns are (left to right): (1) PP; (2) C supply to the mesopelagic ecosystem; pie chart shows relative contributions via detritus, D, migrating and epipelagic resident copepods (those animals that act as food for mesopelagic invertebrate carnivores and fish), ZM and ZR respectively; (3) mesopelagic fish food sources (growth of); pie chart shows allocation between copepods, ZD (detritivorous copepods), ZM, ZR, and invertebrate carnivores, V; (4) C flux to fish; pie chart shows growth, GF, vs. respiration plus egestion, RF + EF. Figures at top are numerical values for the heights of the bars. Parameter sensitivity analysis was carried out using predicted mesopelagic fish biomass, BF, as the focus because fish represent the apex of the model ecosystem. Each parameter was varied, in turn, ± 50% (Figure 3). Predicted BF shows relatively high sensitivity to several Group 1 (relatively certain) parameters: PP, fPP,Z, KZ, and KF. Model sensitivity to these parameters is unsurprising given that PP provides the source of C entering the system, fPP,Z specifies the fraction of PP utilized by copepods, (the main vector of transfer between the epipelagic and mesopelagic ecosystems) and KZ and KF are the growth efficiencies of copepods and mesopelagic fish, respectively. Regarding Group 2 parameters (relatively uncertain), predicted BF shows moderate sensitivity to parameters fM,VF, and fR,VF, which specify the fractions of migrating and epipelagic resident copepods grazed by migrating predators, i.e. invertebrate carnivores and mesopelagic fish (with the remainder providing food for epipelagic resident predators). Sensitivity is also seen for parameter fZ,F, which quantifies the fraction of that grazing utilized by mesopelagic fish (rather than invertebrate carnivores), but not fV,F because, as noted earlier, invertebrate carnivores are only a minor food source for the fish. A notable result is that the sensitivity analysis indicates low sensitivity for parameter fPP,D, the fraction of PP exported as detritus, because the detrital substrate is utilized for growth by detritivorous copepods with low efficiency (0.0145; parameter tD,Z). These zooplankton therefore contribute only a small fraction of mesopelagic fish diet. Finally, predicted BF is unsurprisingly sensitive to fish mortality rate, mF. For example, halving mF to 0.335 per year, which is equivalent to increasing fish longevity from 1.5 to 3 years, leads to predicted mesopelagic fish biomass increasing from 2.4 to 4.9 Gt. Figure 3. Open in new tabDownload slide Sensitivity of predicted mesopelagic fish biomass (Gt wet weight) to model parameter values ±50% (with maximum bound of 1.0 for parameters fM,VF and fV,F). Parameters are divided between Group 1 (relatively certain) and Group 2 (relatively uncertain). STD is standard run (parameters as in Table 2). Colours show relative contributions of copepods (ZM, migrating copepods; ZR, copepods that reside in the epipelagic; ZD, detritivorous copepods) and carnivorous invertebrates (V) to fish diet. Uncertainty associated with parameter values depends not only on sensitivity, as described above, but also on the intrinsic difficulty in assigning parameter values from the literature, as per the division of parameters into Groups 1 and 2. Of particular note in this regard is parameter tD,Z, the trophic transfer efficiency for copepods utilizing detritus as a food source. Our default value of 0.0145 is based on the recent modelling study of Anderson et al. (2017), whereas previous studies (Anderson and Tang, 2010; Giering et al., 2014) have used a transfer efficiency of 0.23, i.e. more than an 15-fold higher. If this value is used here for parameter tD,Z, exported detritus is then predicted to become a significant source of C fuelling the mesopelagic ecosystem: the growth of detritivorous zooplankton increases from 0.07 to 1.1 Gt C year−1, the contribution of these zooplankton to the diet of mesopelagic fish increases from 5 to 42% and predicted mesopelagic fish biomass increases from 2.4 to 4.6 Gt (Figure 4). Figure 4. Open in new tabDownload slide Model sensitivity to transfer efficiency for copepods utilizing detritus (parameter tD,Z): (a) detritus export, Gt C year−1; (b) growth of detritivorous copepods; (c) mesopelagic fish biomass showing contributions of different food sources: ZR, ZM, ZD, and V to production (colours as in Figure 3). The overall uncertainty in predicted mesopelagic fish biomass due to model parameterization depends on the combined uncertainties associated with the parameters in total. We therefore undertook ensemble analyses of the model, with each ensemble consisting of 107 runs. Parameter values for each run were randomly generated within specified ranges. In the case of Group 1 (relatively certain) parameters, this was ±50%, and for Group 2 parameters (relatively uncertain) it was ±75%. Parameter tD,Z was treated differently based on the analysis shown in Figure 4. When varying Group 2 parameters ±75%, tD,Z was varied between its standard value, 0.0145, and 0.23. Varying parameters fM,VF and fV,F ±75% leads to values >1, which are not permissible; in this event, parameter values were reset equal to 1. Predictions for mesopelagic fish biomass were allocated within 0.1 Gt bins, which were then plotted as a frequency distribution. Results (Figure 5) exhibit strong positive skew, showing that it is possible to predict BF significantly higher than our standard value of 2.4 Gt (default parameter settings: Table 2). Nevertheless, only 27% of predictions exceed 5 Gt biomass, and only 8% exceed 10 Gt. The predicted frequency distribution widens if both Groups 1 and 2 parameters are varied within ranges ±75% (Figure 5), although even here only 26 and 10% of the distribution exceeds 5 and 10 Gt, respectively. Note that the mode of the frequency distribution shifts to the left because the relative contribution of parameter tD,Z to overall uncertainty is diminished (this parameter is only increased, rather than varied ±); the mode of a frequency distribution where numbers are multiplied together (as is the case for model parameters) is generally less than the mean. If, on the other hand, all parameters are varied ±50% (with the range for parameter tD,Z scaled in proportion), the frequency distribution narrows somewhat, with 23 and 2% of predictions for BF exceeding 5 and 10 Gt, respectively. Figure 5. Open in new tabDownload slide Frequency distribution (shaded yellow area) of predictions for mesopelagic fish biomass from an ensemble of 107 model runs with parameters varied ±50% (Group 1 parameters) and ±75% (Group 2 parameters, except for tD,Z, which was varied between 0.0145 and 0.23). Parameters fM,VF and fV,F were assigned a maximum bound of 1.0. Also shown are frequency distributions for parameters varied ±75% throughout (red line) and ±50% throughout (blue line; with equivalent scaling for tD,Z). The vertical dashed line represents the standard model prediction of 2.4 Gt. Finally, the relative contributions of the least well constrained parameters (those of Group 2) to overall uncertainty in predicted mesopelagic fish biomass was investigated in greater detail. Returning to the baseline analysis in which parameters in Groups 1 and 2 were varied ±50 and ±75% respectively, the values of parameters in Group 2 were, in turn, fixed at their default settings (Table 2). Maintaining 107 model runs in each ensemble, the resulting frequency distributions show greatest departure from the baseline when parameters tD,Z and mF were assigned fixed values (Figure 6). The positive skew of the frequency distributions shifts left in both cases, with only 4 and 2% of predictions for BF exceeding 10 Gt for parameters tD,Z and mF, respectively (compared with 8% when all parameters are varied). The analysis thus re-emphasizes the analysis shown in Figure 4, namely the importance of understanding the efficiency of detritus utilization by mesopelagic copepods, as well as highlighting the importance of mesopelagic fish mortality (longevity). Figure 6. Open in new tabDownload slide Effect of fixing Group 2 (relatively uncertain) parameters, in turn, to their standard values (Table 2) on the ensemble predictions. Yellow-shaded area is the frequency distribution with all parameters included (as in Figure 5). Frequency distributions with parameters excluded are: tD,Z (pink line), mF (green), fZ,F (blue), fR,VF (orange), fM,VF (grey), and fV,F (cyan). Discussion A simple food web model was constructed to investigate the flows of C from PP to the mesopelagic ecosystem, providing predictions for the production and biomass of mesopelagic fish. The model was parameterized for the world ocean between 40°S and 40°N, thereby permitting comparison with contemporary field estimates of mesopelagic fish biomass, based on acoustic data, of 11–15 Gt w.w for this latitudinal range (Irigoien et al., 2014). Using default parameter values, our model estimate is substantially lower, 2.4 Gt w.w. The uncertainty in this estimate associated with model parameter values was assessed by undertaking ensembles of model runs throughout the 13D parameter space. Results indicated that it is possible for the model to generate predicted mesopelagic fish biomass >10 Gt, but this only occurred in 8% of the ensemble predictions. Despite the relatively short food chain, model results indicated that just 1.6% of PP ends up being ingested by mesopelagic fish and, combined with a gross growth efficiency of 0.2, 0.32% of PP accrues as mesopelagic fish production. In fact, this percentage is relatively high compared with estimates of fish catches of commercial fish species such as herring, plaice and sole, which are often between 0.05 and 0.1% of PP (Nielsen and Richardson, 1996; Sommer et al., 2002; Chassot et al., 2007, 2010). Pauly and Christensen (1995) estimated global fish catch to be 0.21% of PP, although this estimate showed wide variation when separated into different ecosystem types with values, for example, of 0.01 and 2.3% for open ocean and upwelling ecosystems, respectively. Large variations in fish catch as a percentage of PP are to be expected between ecosystems because different food webs have different size structures and numbers of trophic levels (Ryther, 1969; Pauly and Christensen, 1995; Jennings and Collingridge, 2015; Stock et al., 2017). Furthermore, fish catch need not necessarily be a good indicator of population biomass because of discards, the efficiency of nets, selective catch of different species, etc. Even if the production of mesopelagic fish is only 0.32% of PP, this may provide potential as a harvestable resource. Quantifying mesopelagic fish biomass is also a useful indicator, especially as the model estimate can be compared with those from field studies. Additional factors that need to be taken into consideration when considering harvestable potential include a knowledge of controlling mechanisms, a holistic understanding of the community, and ecological concerns associated with harvesting a significant, but poorly understood, component of marine food webs (St. John et al., 2016). The modelling work presented herein emphasizes the importance of migrating animals in the transfer of C from the surface ocean to the mesopelagic zone. It also highlights major gaps in our knowledge of the ecology and trophodynamics of mesopelagic ecosystems and their associated links to PP, gaps that are not easy to fill because food web interactions in the mesopelagic zone are difficult to measure (Robinson et al., 2010). The knowledge gaps may be divided into four areas: (i) trophic linkages between copepods and both migrating and non-migrating (epipelagic resident) predators, (ii) mesopelagic fish diet, (iii) mesopelagic fish growth efficiency and mortality (longevity), and (iv) the efficiency of detritus utilization as a food source by mesopelagic zooplankton. These are now discussed in turn: (i) The relative extent to which migrating copepods are consumed by migrating predators vs. epipelagic predators depends on the relative abundances of the two predator groups and the amount of time that predators and prey are co-located. The predator groups represent a diverse array of organisms, including fish and invertebrate carnivores, and so specifying their relative abundances is difficult. For simplicity, we assumed that the ratio of migrating to non-migrating predators is the same as for copepods, which could be estimated from data. Most of the mesopelagic fauna undergo diel migration (Angel and Pugh, 2000; Siegelman-Charbit and Planque, 2016), including the fish and invertebrate carnivores (Clarke, 1980; Catul et al., 2011; Bernal et al., 2015). We therefore assumed that these migrating predators have continuous access to migrating copepod prey, the two communities operating in concert. Quantifying the extent to which migrating copepods are grazed by the epipelagic resident predator community is, however, considerably more problematic. Grazing would be zero if DVM was a perfect predator avoidance strategy, but it is not. Migrating copepods are known to contribute significantly to the diets of epipelagic fish (Beaugrand et al., 2003; Garrido et al., 2015; Bachiller et al., 2016). Feeding may occur primarily during the hours of dawn and dusk as the migrators travel to and from surface waters (Allison et al., 1996; Cardinale et al., 2003). We made the simplifying assumption in the model that epipelagic predators gain access to the migrating copepods for an average of 1 h at either end of each day. One interesting complication, for example, is that moonlight may expose migrating organisms to predation. Indeed, it appears that they may counter this threat by avoiding surface waters at night during full moon (Last et al., 2016). (ii) Having determined the extent to which copepods are grazed by the migrating predator community as a whole, there is still the difficulty of specifying the fraction that is exploited by the mesopelagic fish (vs. the invertebrate carnivores). We made the simple assumption that mesopelagic fish are responsible for 50% of copepod losses (parameter fZ,F), along with 80% of invertebrate carnivore losses (parameter fV,F), highlighting the difficulty in deriving quantitative estimates of trophic pathway parameters. Based on this assumption, our results indicated that copepods account for 85% of the mesopelagic fish diet. The predicted relative contributions were 5% from mesopelagic detritivorous copepods, 41% from migrating copepods and 39% from epipelagic resident copepods. It is well known that mesopelagic fish, including myctophids, feed on crustaceans (Clarke, 1980; Van Noord et al., 2016), often with a predominance of copepods (Kawaguchi and Mauchline, 1982; Pepin, 2013; Bernal et al., 2015; Saunders et al., 2015). Other prey items include gelatinous zooplankton (Bystydzieńska et al., 2010; Hudson et al., 2014). Estimates of myctophid fish trophic position derived from stable isotope analysis of nitrogen (δ15 N) in bulk tissues (Cherel et al., 2010) and individual amino acids (Choy et al., 2015; Hetherington et al., 2017) vary from 3.3 to 4.2, indicating that these fish are secondary and tertiary consumers. Model results were thus generally consistent with this trophic positioning. (iii) Model predictions for the biomass of mesopelagic fish are directly influenced by fish growth efficiency and mortality rate (KF and mF, respectively) and so it is unsurprising that results are sensitive to these parameters. We used a fish GGE of 0.20 (Ikeda, 1996; Ikeda et al., 2008). Similar values of growth efficiency have been recorded in commercial fish species such as cod, haddock and herring (Peck et al., 2003, 2015; Bernreuther et al., 2013), although higher values nearer 0.3 are also seen (e.g., Björnsson and Tryggvadóttir, 1996; Sogard and Olla, 2001; Bernreuther et al., 2013). Predicted mesopelagic fish biomass is particularly sensitive to the parameterization of fish longevity. We used a default mortality of 0.67 year−1 which is equivalent to life span of 1.5 years (Bystydzieńska et al., 2010; Lingkowski et al., 1993; Tagaki et al., 2006; Hosseini-Shekarabi et al., 2015). If longevity is increased to 3 years (Halliday, 1970; Gjösæter, 1973; Greely et al., 1999; Saunders et al., 2015), predicted fish biomass doubles from 2.4 to 4.9 Gt. Note that uncertainty in mortality rate does not influence predicted mesopelagic fish production. (iv) A perhaps surprising outcome of our analysis is that detritivorous zooplankton are predicted to contribute only 5% to the diet of mesopelagic fish. This occurred because of the low trophic transfer efficiency for copepods grazing on detritus, 1.45% (parameter tD,Z = 0.0145). Detritus is made up of refractory compounds that are depleted in micronutrients including amino acids and fatty acids (Mann, 1988; Cowie and Hedges, 1996; Mayor et al., 2011). It is however colonized by micro-organisms that provide substrates that are readily absorbed and which are rich in these micronutrients (Phillips, 1984). A favourable strategy for detritivorous copepods may therefore be to selectively utilize these micro-organisms as a source of nutrition (Mayor et al., 2014). A recent stoichiometric analysis of this phenomenon (Anderson et al. 2017) indicated that overall transfer efficiencies for detritus utilization by zooplankton may be very low, e.g. 1.45%, because of the low biomass of (nutritious) microbes present within the detrital matrix. This means that, when using this parameter value for trophic transfer, the predicted production of detritivorous copepods in the mesopelagic zone was only 0.16% of PP. As such, detritivorous copepods are only a minor source of food for mesopelagic fish. If the transfer efficiency is increased more than 15-fold to 0.23, a value used in previous models (Anderson and Tang, 2010, Giering et al., 2014), the predicted share of detritivorous copepods in the diet of mesopelagic fish increases from 5% to 42% and predicted fish biomass increases from 2.4 to 4.6 Gt. Anderson et al. (2017) highlighted the need for more information on the physiological and ecological interactions between zooplankton, microbes and detritus. Our work here serves to re-emphasize that need. The model was parameterized as a general representation of the flows of C between PP and the mesopelagic ecosystem between 40°S to 40°N in the world ocean. A simple approach is justified at this stage given the many uncertainties in our understanding of the mesopelagic ecosystem. The structure and function of marine ecosystems are in reality variable in space and time. For example, the export of sinking detritus as a fraction of PP is variable (Dunne et al., 2005; Henson et al., 2011; Siegel et al., 2014) which will in turn mean that the relationship between mesopelagic fish biomass and PP is not necessarily linear. Likewise, the contribution of copepods to overall grazing of PP is also variable in space and time (Calbet, 2001). Recent studies of mesopelagic biogeography have defined as many as 33 biomes throughout the ocean (Proud et al., 2017; Sutton et al., 2017). These studies were, however, limited by the sparsity and spatiotemporally biased data sets available such that much work remains in order to derive biomes that are comprehensive and robust (Sutton et al., 2017). As and when robust biomes are defined, it may then be possible to extend our model to investigate spatiotemporal variability in the dynamics of the mesopelagic ecosystem and the associated production of fish biomass. If the model was parameterized beyond the latitudinal range 40°S to 40°N, a factor to consider is that high latitude, lipid-rich zooplankton undergo seasonal migrations and thereby transfer significant amounts of labile C to the mesopelagic zone via the “lipid pump” (Jónasdóttir et al., 2015). Estimates of mesopelagic fish biomass at the global scale should take into consideration this C as an additional source of energy and nutrition to mesopelagic fish. In conclusion, we used a flow analysis model to study C fluxes from PP to the mesopelagic ecosystem, giving a prediction for mesopelagic fish biomass of 2.4 Gt w.w. (between 40°S and 40°N in the world ocean). Defining the mesopelagic food web interactions in the model was problematic and many of the associated parameter values were poorly constrained. Our model analysis highlights how little is known about the physiological ecology of mesopelagic fish, trophic pathways within the mesopelagic food web, and how these link to PP in the surface ocean. A deeper understanding of these interacting factors is required before the potential for utilizing mesopelagic fish as a harvestable resource can be considered as a viable proposition. Acknowledgements The work was funded by the Natural Environment Research Council programmes COMICS (NE/M020835/1) and DIAPOD (NE/P0063513/1). We also acknowledge National Capability funding from NERC. We wish to thank two anonymous reviewers for their comments on the article. References Al-Mutairi H. , Landry M. R. 2001 . Active export of carbon and nitrogen at Station ALOHA by diel migrant zooplankton . Deep-Sea Research II , 48 : 2083 – 2103 . Google Scholar Crossref Search ADS WorldCat Allison E. H. , Irvine K., Thompson A. B., Ngatunga B. P. 1996 . Diets and food consumption rates of pelagic fish in Lake Malawi, Africa . Freshwater Biology , 35 : 489 – 515 . Google Scholar Crossref Search ADS WorldCat Anderson T. R. , Tang K. W. 2010 . Carbon cycling and POC turnover in the mesopelagic zone of the ocean: insights from a simple model . Deep-Sea Research II , 57 : 1581 – 1592 . Google Scholar Crossref Search ADS WorldCat Anderson T. R. , Pond D. W., Mayor D. J. 2017 . The role of microbes in the nutrition of detritivorous invertebrates: a stoichiometric analysis . Frontiers in Microbiology , 7 : 2113. Google Scholar Crossref Search ADS PubMed WorldCat Angel M. V. , Pugh P. R. 2000 . Quantification of diel vertical migration by micronektic taxa in the northeast Atlantic . Hydrobiologia , 440 : 161 – 179 . Google Scholar Crossref Search ADS WorldCat Aumont O. , Ethé C., Tagliabue A., Bopp L., Gehlen M. 2015 . PISCES-v2: an ocean biogeochemical model for carbon and ecosystem studies . Geoscientific Model Development , 8 : 2465 – 2513 . Google Scholar Crossref Search ADS WorldCat Bachiller E. , Skaret G., Nøttestad L., Slotte A., Stergiou K. I. 2016 . Feeding ecology of northeast Atlantic mackerel, Norwegian spring-spawning herring and blue whiting in the Norwegian Sea . Plos One , 11 : e0149238 . Google Scholar Crossref Search ADS PubMed WorldCat Battaglia P. , Andaloro F., Esposito V., Granata A., Guglielmo L., Guglielmo R., Musolino S., Romeo T., Zagami G. 2016 . Diet and trophic ecology of the lanternfish Electrona risso (Cocco 1829) in the Strait ofMessina (central Mediterranean Sea) and potential resource utilization from the Deep Scattering Layer (DSL) . Journal of Marine Systems , 159 : 100 – 108 . Google Scholar Crossref Search ADS WorldCat Beaugrand G. , Brander K. M., Lindley J. A., Souissi S., Reid P. C. 2003 . Plankton effect on cod recruitment in the North Sea . Nature , 426 : 661 – 664 . Google Scholar Crossref Search ADS PubMed WorldCat Behrenfeld M. J. , Falkowski P. G. 1997 . Photosynthetic rates derived from satellite-based chlorophyll concentration . Limnology and Oceanography , 42 : 1 – 20 . Google Scholar Crossref Search ADS WorldCat Bernal A. , Olivar M. P., Maynou F., Fernández de Puelles M. L. 2015 . Diet and feeding strategies of mesopelagic fishes in the western Mediterranean . Progress in Oceanography , 135 : 1 – 17 . Google Scholar Crossref Search ADS WorldCat Bernreuther M. , Herrmann J.-P., Peck M. A., Temming A. 2013 . Growth energetics of juvenile herring, Clupea harengus L.: food conversion efficiency and temperature dependency of metabolic rate . Journal of Applied Ichtyology , 29 : 331 – 340 . Google Scholar Crossref Search ADS WorldCat Björnsson B. , Tryggvadóttir S. V. 1996 . Effects of size and optimal temperature for growth and growth efficiency of immature Atlantic halibut (Hippoglossus hippoglossus L.) . Aquaculture , 142 : 33 – 42 . Google Scholar Crossref Search ADS WorldCat Buitenhuis E. , Le Quéré C., Aumont O., Beaugrand G., Bunker A., Hirst A., Ikeda T., O'Brien T., Piontkovski S., Straile D. 2006 . Biogeochemical fluxes through mesozooplankton . Global Biogeochemical Cycles , 20 : GB2003. Google Scholar Crossref Search ADS WorldCat Bystydzieńska Z. E. , Phillips A. J., Linkowski T. B. 2010 . Larval stage duration, age and growth of blue lanternfish Tarletonbeania crenularis (Jordan and Gilbert, 1880) derived from otolith microstructure . Environmental Biology of Fishes , 89 : 493 – 503 . Google Scholar Crossref Search ADS WorldCat Calbet A. 2001 . Mesozooplankton grazing effect on primary production: a global comparative analysis in marine ecosystems . Limnology and Oceanography , 46 : 1824 – 1830 . Google Scholar Crossref Search ADS WorldCat Cardinale M. , Casini M., Arrhenius F., Häkansson N. 2003 . Diel spatial distribution and feeding activity of herring (Clupea harengus) and sprat (Sprattus sprattus) in the Baltic Sea . Aquatic Living Resources , 16 : 283 – 292 . Google Scholar Crossref Search ADS WorldCat Carr M.-E. 2002 . Estimation of the potential productivity in Eastern Boundary Currents using remote sensing . Deep-Sea Research II , 49 : 59 – 80 . Google Scholar Crossref Search ADS WorldCat Catul V. , Gauns M., Karuppasamy P. K. 2011 . A review on mesopelagic fishes belonging to family Myctophidae . Reviews in Fish Biology and Fisheries , 21 : 339 – 354 . Google Scholar Crossref Search ADS WorldCat Chassot E. , Mélin F., Le Pape O., Gascuel D. 2007 . Bottom-up control regulates fisheries production at the scale of eco-regions in European seas . Marine Ecology Progress Series , 343 : 45 – 55 . Google Scholar Crossref Search ADS WorldCat Chassot E. , Bonhommeau S., Dulvy N. K., Mélin F., Watson R., Gascuel D., Le Pape O. 2010 . Global marine primary production constrains fisheries catches . Ecology Letters , 13 : 495 – 505 . Google Scholar Crossref Search ADS PubMed WorldCat Cherel Y. , Fontaine C., Richard P., Labat J.-P. 2010 . Isotopic niches and trophic levels of myctophid fishes and their predators in the Southern Ocean . Limnology and Oceanography , 55 : 324 – 332 . Google Scholar Crossref Search ADS WorldCat Choy C. A. , Popp B. N., Hannides C. C. S., Drazen J. C. 2015 . Trophic structure and food resources of epipelagic and mesopelagic fishes in the North Pacific Subtropical Gyre ecosystem inferred from nitrogen isotopic compositions . Limnology and Oceanography , 60 : 1156 – 1171 . Google Scholar Crossref Search ADS WorldCat Choy S. L. , O'Leary R., Mengersen K. 2009 . Elicitation by design in ecology: using expert opinion to inform priors for Bayesian statistical models . Ecology , 90 : 267 – 277 . Google Scholar Crossref Search ADS WorldCat Clarke T. A. 1980 . Diets of fourteen species of vertically migrating mesopelagic fishes in Hawaiian waters . Fishery Bulletin , 78 : 619 – 640 . Google Scholar OpenURL Placeholder Text WorldCat Collins W. J. , Bellouin N., Doutriaux-Boucher M., Gedney N., Halloran P., Hinton T., Hughes J. et al. 2011 . Development and evaluation of an Earth-System model – HadGEM2 . Geoscientific Model Development , 4 : 1051 – 1075 . Google Scholar Crossref Search ADS WorldCat Costello J. 1991 . Complete carbon and nitrogen budgets for the hydromedusa Cladonema californicum (Anthomedusa: Cladonemidae) . Marine Biology , 108 : 119 – 128 . Google Scholar Crossref Search ADS WorldCat Cowie G. L. , Hedges J. I. 1996 . Digestion and alteration of the biochemical constituents of a diatom (Thalassiosira weissflogii) ingested by an herbivorous zooplankton (Calanus pacificus) . Limnology and Oceanograhy , 41 : 581 – 594 . Google Scholar Crossref Search ADS WorldCat Daewel U. , Hjøllo S. S., Huret M., Ji R., Maar M., Niiranen S., Travers-Trolet M., Peck M. A., van de Wolfshaar K. E. 2014 . Predation control of zooplankton dynamics: a review of observations and models . ICES Journal of Marine Science , 71 : 254 – 271 . Google Scholar Crossref Search ADS WorldCat Dagg M. J. 1976 . Complete carbon and nitrogen budgets for the carnivorous amphipod, Calliopius laeviusculus (Krøyer) . International Revue der Gesamten Hydrobiologie , 61 : 297 – 357 . Google Scholar Crossref Search ADS WorldCat Davison P. C. , Checkley D. M. Jr., Koslow J. A., Barlow J. 2013 . Carbon export mediated by mesopelagic fishes in the northeast Pacific Ocean . Progress in Oceanography , 116 : 14 – 30 . Google Scholar Crossref Search ADS WorldCat Davison P. , Lara-Lopez A., Koslow J. A., 2015a . Mesopelagic fish biomass in the southern California current ecosystem . Deep-Sea Research II , 1 . 129 – 142 . Google Scholar Crossref Search ADS WorldCat Davison P. C. , Koslow J. A., Kloser R. J., 2015b . Acoustic biomass estimation of mesopelagic fish: backscattering from individuals, populations, and communities . ICES Journal of Marine Science , 7 . 1413 – 1424 . Google Scholar Crossref Search ADS WorldCat Dunne J. P. , Armstrong R. A., Gnanadesikan A., Sarmiento J. L. 2005 . Empirical and mechanistic models for the particle export ratio . Global Biogeochemical Cycles , 19 : GB4026. Google Scholar Crossref Search ADS WorldCat Dunne J. P. , John J. G., Adcroft A. J. 2012 . GFDL's ESM2 Global coupled climate-carbon Earth System Models. Part I: physical formulation and baseline simulation characteristics . Journal of Climate , 25 : 6646 – 6665 . Google Scholar Crossref Search ADS WorldCat Dunne J. P. , John J. G., Shevliakova E., Stouffer R. J., Krasting J. P., Malyshev S. L., Milly P. C. D. et al. 2013 . GFDL's ESM2 Global coupled climate-carbon Earth System Models. Part II: Carbon system formulation and baseline simulation characteristics . Journal of Climate , 26 : 2247 – 2267 . Google Scholar Crossref Search ADS WorldCat Gagnon Y. L. , Sutton T. T., Johnsen S. 2013 . Visual acuity in pelagic fishes and mollusks . Vision Research , 92 : 1 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Garrido S. , Silva A., Pastor J., Dominguez R., Silva A. V., Santos A. M. 2015 . Trophic ecology of pelagic fish species off the Iberian coast: diet overlap, cannibalism and intraguild predation . Marine Ecology Progress Series , 539 : 271 – 286 . Google Scholar Crossref Search ADS WorldCat Giering S. L. C. , Sanders R., Lampitt R. S., Anderson T. R., Tamburini C., Boutrif M., Zubkov M. V. et al. 2014 . Reconciliation of the carbon budget in the ocean’s twilight zone . Nature , 507 : 480 – 483 . Google Scholar Crossref Search ADS PubMed WorldCat Gjösæter J. 1973 . Age, growth, and mortality of the myctophid fish, Benthosema glaciale (Reinhardt), from western Norway . Sarsia , 52 : 1 – 14 . Google Scholar Crossref Search ADS WorldCat Gjösæter J. , Kawaguchi K. 1980 . A review of the world resources of mesopelagic fish . Food and Agriculture Organization Fisheries Technical Paper , 193 : 151. Google Scholar OpenURL Placeholder Text WorldCat Godfray H. C. J. , Beddington J. R., Crute I. R., Haddad L., Lawrence D., Muir J. F., Pretty J. et al. 2010 . Food security: the challenge of feeding 9 billion people . Science , 327 : 812 – 818 . Google Scholar Crossref Search ADS PubMed WorldCat Greely T. M. , Gartner J. V. Jr, Torres J. J. 1999 . Age and growth of Electrona antarctica (Pisces: Myctophidae), the dominant mesopelagic fish of the Southern Ocean . Marine Biology , 133 : 145 – 158 . Google Scholar Crossref Search ADS WorldCat Halliday R. G. 1970 . Growth and vertical distribution of the glacier lanternfish, Benthosema glaciale, in the northwestern Atlantic . Journal of the Fisheries Research Board of Canada , 27 : 105 – 116 . Google Scholar Crossref Search ADS WorldCat Hays G. C. 2003 . A review of the adaptive significance and ecosystem consequences of zooplankton diel vertical migrations . Hydrobiologia , 503 : 163 – 170 . Google Scholar Crossref Search ADS WorldCat Henson S. A. , Sanders R., Madsen E., Morris P. J., Le Moigne F., Quartly G. D. 2011 . A reduced estimate of the strength of the ocean’s biological carbon pump . Geophysical Research Letters , 38 : L04606 . Google Scholar Crossref Search ADS WorldCat Hetherington E. D. , Olson R. J., Drazen J. C., Lennert-Cody C. E., Ballance L. T., Kaufmann R. S., Popp B. N. 2017 . Spatial food-web structure in the eastern tropical Pacific Ocean based on compound-specific nitrogen isotope analysis of amino acids . Limnology and Oceanography , 62 : 541 – 560 . Google Scholar Crossref Search ADS WorldCat Hosseini-Shekarabi S. P. , Valinassab T., Bystydzieńska Z., Linkowski T. 2015 . Age and growth of Benthosema pterotum (Alcock, 1890) (Myctophidae) in the Oman Sea . Journal of Applied Ichthyology , 31 : 51 – 56 . Google Scholar Crossref Search ADS WorldCat Hoving H. J. T. , Robison B. H. 2016 . Deep-sea in situ observations of gonatid squid and their prey reveal high occurrence of cannibalism . Deep-Sea Research I , 116 : 94 – 98 . Google Scholar Crossref Search ADS WorldCat Hudson J. M. , Steinberg D. K., Sutton T. T., Graves J. E., Latour R. J. 2014 . Myctophid feeding ecology and carbon transport along the northern Mid-Atlantic Ridge . Deep-Sea Research, I , 93 : 104 – 116 . Google Scholar Crossref Search ADS WorldCat Ikeda T. 1991 . Assimilated carbon budget for the hyperiid amphipod Themisto japonica (Bovallius) from the Japan Sea as influenced by temperature . Journal of the Oceanographical Society of Japan , 47 : 7 – 16 . Google Scholar Crossref Search ADS WorldCat Ikeda T. 1996 . Metabolism, body composition and energy budget of the mesopelagic fish Maurolicus muelleri in the Sea of Japan . Fishery Bulleting , 94 : 49 – 58 . Google Scholar OpenURL Placeholder Text WorldCat Ikeda T. , Shiga N., Yamaguchi A. 2008 . Structure, biomass distribution and trophodynamics of the pelagic ecosystem in the Oyashio region, western subarctic Pacific . Journal of Oceanography , 64 : 339 – 354 . Google Scholar Crossref Search ADS WorldCat Ikeda T. , McKinnon A. D., Doherty P. J. 2011 . Metabolism and chemical composition of small teleost fishes from tropical inshore waters . Marine Ecology Progress Series , 435 : 197 – 207 . Google Scholar Crossref Search ADS WorldCat Irigoien X. , Klevjer T. A., Røstad A., Martinez U., Boyra G., Acuña J. L., Bode A. et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean . Nature Communications , 5 : 3271 . Google Scholar Crossref Search ADS PubMed WorldCat Isla A. , Scharek R., Latasa M. 2015 . Zooplankton diel vertical migration and contribution to deep active carbon flux in the NW Mediterranean . Journal of Marine Systems , 143 : 86 – 97 . Google Scholar Crossref Search ADS WorldCat Jennings S. , Collingridge K., MacKenzie B. R. 2015 . Predicting consumer biomass, size-structure, production, catch potential, responses to fishing and associated uncertainties in the world’s marine ecosystems . PLoS ONE , 10 : e0133794 . Google Scholar Crossref Search ADS PubMed WorldCat Jónasdóttir S. H. , Visser A. W., Richardson K., Heath M. R. 2015 . Seasonal copepod lipid pump promotes carbon sequestration in the deep North Atlantic . Proceedings of the National Academy of Sciences of the United States of America , 112 : 12122 – 12126 . Google Scholar Crossref Search ADS PubMed WorldCat Kaartvedt S. , Staby A., Aksnes D. L. 2012 . Efficient trawl avoidance by mesopelagic fishes causes large underestimation of their biomass . Marine Ecology Progress Series , 456 : 1 – 6 . Google Scholar Crossref Search ADS WorldCat Karakulak F. S. , Salman A., Oray I. K. 2009 . Diet composition of bluefin tuna (Thunnus thynnus L. 1758) in the Eastern Mediterranean Sea, Turkey . Journal of Applied Ichthyology , 25 : 757 – 761 . Google Scholar Crossref Search ADS WorldCat Kawaguchi K. , Mauchline J. 1982 . Biology of Myctophid fishes (family Myctophidae) in the Rockall Trough, northeastern Atlantic Ocean . Biological Oceanography , 1 : 337 – 373 . Google Scholar OpenURL Placeholder Text WorldCat Klevjer T. A. , Irigoien X., Røstad A., Fraile-Nuez E., Benítez-Barrios V. M., Kaartvedt S. 2016 . Large scale patterns in vertical distribution and behaviour of mesopelagic scattering layers . Nature Scientific Reports , 6 : 19873. Google Scholar Crossref Search ADS WorldCat Kloser R. J. , Ryan T. E., Young J. W., Lewis M. E. 2009 . Acoustic observations of micronekton fish on the scales of an ocean basin: potential and challenges . ICES Journal of Marine Science , 66 : 998 – 1006 . Google Scholar Crossref Search ADS WorldCat Kloser R. J. , Ryan T. E., Keith G., Gershwin L. 2016 . Deep-scattering layer, gas-bladder density, and size estimates using a two-frequency acoustic and optical probe . ICES Journal of Marine Science , 73 : 2037 – 2048 . Google Scholar Crossref Search ADS WorldCat Koslow J. A. , Kloser R. J., Williams A. 1997 . Pelagic biomass and community structure over the mid-continental slope off southeastern Australia based upon acoustic and midwater trawl sampling . Marine Ecology Progress Series , 146 : 21 – 35 . Google Scholar Crossref Search ADS WorldCat Krueger T. , Page T., Hubacek K., Smith L., Hiscock K. 2012 . The role of expert opinion in environmental modelling . Environmental Modelling and Software , 36 : 4 – 18 . Google Scholar Crossref Search ADS WorldCat Lampitt R. S. , Noji T., von Bodungen B. 1990 . What happens to zooplankton faecal pellets? Implications for material flux . Marine Biology , 104 : 15 – 23 . Google Scholar Crossref Search ADS WorldCat Larson R. J. 1987 . A note on the feeding, growth, and reproduction of the epipelagic Scyphomedusa Pelagia noctiluca (Forskål) . Biological Oceanography , 4 : 447 – 454 . Google Scholar OpenURL Placeholder Text WorldCat Last K. S. , Hobbs L., Berge J., Brierley A. S., Cottier F. 2016 . Moonlight drives ocean-scale mass vertical migration of zooplankton during the Arctic winter . Current Biology , 26 : 244 – 251 . Google Scholar Crossref Search ADS PubMed WorldCat Linkowski T. B. , Radtke R. L., Lenz P. H. 1993 . Otolith microstructure, age and growth of two species of Ceratoscopelus (Osteichthyes: Myctophidae) from the eastern North Atlantic . Journal of Experimental Marine Biology and Ecology , 167 : 237 – 260 . Google Scholar Crossref Search ADS WorldCat Madin L. P. , Horgan E. F., Steinberg D. K. 2001 . Zooplankton at the Bermuda Atlantic Time-series Study (BATS) station: diel, seasonal and interannual variation in biomass, 1994}1998 . Deep-Sea Research II , 48 : 2063 – 2082 . Google Scholar Crossref Search ADS WorldCat Mann K. H. 1988 . Production and use of detritus in various freshwater, estuarine, and coastal marine ecosystems . Limnology and Oceanography , 33 : 910 – 930 . Google Scholar OpenURL Placeholder Text WorldCat Marra J. , Ho C., Trees C. 2003 . An alternative algorithm for the calculation of primary production from remote sensing data. LDEO Technical Report, #LDEO-2003-1, Lamont-Doherty Earth Observatory, Palisades, New York. Mayor D. J. , Cook K., Thornton B., Walsham P., Witte U. F. M., Zuur A. F., Anderson T. R. 2011 . Absorption efficiencies and basal turnover of C, N and fatty acids in a marine calanoid copepod . Functional Ecology , 25 : 509 – 518 . Google Scholar Crossref Search ADS WorldCat Mayor D. J. , Sanders R., Giering S. L. C., Anderson T. R. 2014 . Microbial gardening in the ocean’s twilight zone: detritivorous metazoans benefit from fragmenting, rather than ingesting, sinking detritus . Bioessays , 36 : 1132 – 1137 . Google Scholar Crossref Search ADS PubMed WorldCat Møller L. F. , Riisgård H. U. 2007 . Feeding, bioenergetics and growth in the common jellyfish Aurelia aurita and two hydromedusae, Sarsia tubulosa and Aequorea vitrine . Marine Ecology Progress Series , 346 : 167 – 177 . Google Scholar Crossref Search ADS WorldCat Møller L. F. , Canon J. M., Tiselius P. 2010 . Bioenergetics and growth in the ctenophore Pleurobrachia pileus . Hydrobiologia , 645 : 167 – 178 . Google Scholar Crossref Search ADS WorldCat Moore K. J. , Doney S. C., Lindsay K. 2004 . Upper ocean ecosystem dynamics and iron cycling in a global three-dimensional model . Global Biogeochemical Cycles , 18 : GB4028. Google Scholar Crossref Search ADS WorldCat Nielsen E. , Richardson K. 1996 . Can changes in the fisheries yield in the Kattegat (1950–1992) be linked to changes in primary production? ICES Journal of Marine Science , 53 : 988 – 994 . Google Scholar Crossref Search ADS WorldCat Pauly D. , Christensen V. 1995 . Primary production required to sustain global fisheries . Nature , 374 : 255 – 257 . Google Scholar Crossref Search ADS WorldCat Peck M. A. , Buckley L. J., Caldarone E. M., Bengtson D. A. 2003 . Effects of food consumption and temperature on growth rate and biochemical-based indicators of growth in early juvenile Atlantic cod Gadus morhua and haddock Melanogrammus aeglefinus . Marine Ecology Progress Series , 251 : 233 – 243 . Google Scholar Crossref Search ADS WorldCat Peck M. A. , Ewest B., Herrmann J.-P., Al-Misned F. A., Mortuza M. G., Stäcker S., Temming A. 2015 . Relationships between feeding, growth and swimming activity of European sprat (Sprattus sprattus L.) post-larvae in the laboratory . Environmental Biology of Fishes , 98 : 1117 – 1127 . Google Scholar Crossref Search ADS WorldCat Pepin P. 2013 . Distribution and feeding of Benthosema glaciale in the western Labrador Sea: fish–zooplankton interaction and the consequence to calanoid copepod populations . Deep Sea Research I , 75 : 119 – 134 . Google Scholar Crossref Search ADS WorldCat Pereira J. N. , Neves V. C., Prieto R., Silva M. A., Cascão I., Oliveira C., Cruz M. J. et al. 2011 . Diet of mid-Atlantic Sowerby’s beaked whales Mesoplondon bidens . Deep-Sea Research I , 58 : 1084 – 1090 . Google Scholar Crossref Search ADS WorldCat Phillips N. W. 1984 . Role of different microbes and substrates as potential supplies of specific, essential nutrients to marine detritivores . Bulletin of Marine Science , 35 : 283 – 298 . Google Scholar OpenURL Placeholder Text WorldCat Potier M. , Marsac F., Cherel Y., Lucas V., Sabatié R., Maury O., Ménard F. 2007 . Forage fauna in the diet of three large pelagic fishes (lancetfish, swordfish and yellowfin tuna) in the western equatorial Indian Ocean . Fisheries Research , 83 : 60 – 72 . Google Scholar Crossref Search ADS WorldCat Proud R. , Cox M. J., Brierley A. S. 2017 . Biogeography of the global ocean’s mesopelagic zone . Current Biology , 27 : 113 – 119 . Google Scholar Crossref Search ADS PubMed WorldCat Reeve M. R. , Syms M. A., Kremer P. 1989 . Growth dynamics of a ctenophore (Mnemiopsis) in relation to variable food supply, I. Carbon biomass, feeding, egg production, growth and assimilation efficiency . Journal of Plankton Research , 11 : 535 – 552 . Google Scholar Crossref Search ADS WorldCat Robinson C. , Steinberg D. K., Anderson T. R., Arístegui J., Carlson C. A., Frost J. R., Ghiglione J.-F. et al. 2010 . Mesopelagic zone ecology and biogeochemistry – a synthesis . Deep-Sea Research II , 57 : 1504 – 1518 . Google Scholar Crossref Search ADS WorldCat Ryther J. H. 1969 . Photosynthesis and fish production in the sea . Nature 166 : 72 – 76 . Google Scholar OpenURL Placeholder Text WorldCat Saunders R. A. , Collins M. A., Ward P., Stowasser G., Shreeve R., Tarling G. A. 2015 . Trophodynamics of Protomyctophum (Myctophidae) in the Scotia Sea (Southern Ocean) . Journal of Fish Biology , 87 : 1031 – 1058 . Google Scholar Crossref Search ADS PubMed WorldCat Séférian R. , Bopp L., Gehlen M., Orr J. C., Ethé C., Cadule P., Aumont O. et al. 2013 . Skill assessment of three earth system models with common marine biogeochemistry . Climate Dynamics , 40 : 2549 – 2573 . Google Scholar Crossref Search ADS WorldCat Siegel D. A. , Buesseler K. O., Doney S. C., Sailley S. F., Behrenfeld M. J., Boyd P. W. 2014 . Global assessment of ocean carbon export by combining satellite observations and food-web models . Global Biogeochemical Cycles , 28 : 181 – 196 . Google Scholar Crossref Search ADS WorldCat Siegelman-Charbit L. , Planque B. 2016 . Abundant mesopelagic fauna at oceanic high latitudes . Marine Ecology Progress Series , 546 : 277 – 282 . Google Scholar Crossref Search ADS WorldCat Sogard S. M. , Olla B. L. 2001 . Growth and behavioral responses to elevated temperatures by juvenile sablefish Anoplopoma fimbria and the interactive role of food availability . Marine Ecology Progress Series , 217 : 121 – 134 . Google Scholar Crossref Search ADS WorldCat Sommer U. , Stibor H., Katechakis A., Sommer F., Hansen T. 2002 . Pelagic food web configurations at different levels of nutrient richness and their implications for the ratio fish production: primary production . Hydrobiologia , 484 : 11 – 20 . Google Scholar Crossref Search ADS WorldCat Springer A. M. , Piatt J. F., Shuntov V. P., Van Vliet G. B., Vladimirov V. L., Kuzin A. E., Perlov A. S. 1999 . Marine birds and mammals of the Pacific subarctic Gyres . Progress in Oceanography , 43 : 443 – 487 . Google Scholar Crossref Search ADS WorldCat Springer A. M. , Byrd G. V., Iverson S. J. 2007 . Hot oceanography: planktivorous seabirds reveal ecosystem responses to warming of the Bering Sea . Marine Ecology Progress Series , 352 : 289 – 297 . Google Scholar Crossref Search ADS WorldCat St. John M. A. , Borja A., Chust G., Heath M., Grigorov I., Mariani P., Martin A. P. et al. 2016 . A dark hole in our understanding of marine ecosystems and their services: perspectives from the mesopelagic community . Frontiers in Marine Science , 3 : 31 . Google Scholar Crossref Search ADS WorldCat Steinberg D. K. , Cope J. S., Wilson S. E., Kobari T. 2008 . A comparison of mesopelagic mesozooplankton community structure in the subtropical and subarctic North Pacific Ocean . Deep-Sea Research II , 55 : 1615 – 1635 . Google Scholar Crossref Search ADS WorldCat Steinberg D. K. , Landry M. R. 2017 . Zooplankton and the ocean carbon cycle . Annual Review of Marine Science , 9 : 413 – 444 . Google Scholar Crossref Search ADS PubMed WorldCat Stock C. A. , Dunne J. P., John J. G. 2014 . Global-scale carbon and energy flows through the marine planktonic food web: an analysis with a coupled physical–biological model . Progress in Oceanography , 120 : 1 – 28 . Google Scholar Crossref Search ADS WorldCat Stock C. A. , John J. G., Rykaczewski R. R., Asch R. G., Cheung W. W. L., Dunne J. P., Friedland K. D. et al. 2017 . Reconciling fisheries catch and ocean productivity . Proceedings of the National Academy of Sciences of the United States of America , 114 : E1441 – E1449 . Google Scholar Crossref Search ADS PubMed WorldCat Straile D. 1997 . Gross growth efficiencies of protozoan and metazoan zooplankton and their dependence on food concentration, predator-prey weight ratio, and taxonomic group . Limnology and Oceanography , 42 : 1375 – 1385 . Google Scholar Crossref Search ADS WorldCat Sutton T. T. , Clark M. R., Dunn D. C., Halpin P. N., Rogers A. D., Guinotte J., Bograd S. J. et al. 2017 . A global biogeographic classification of the mesopelagic zone . Deep-Sea Research I , 126 : 85 – 102 . Google Scholar Crossref Search ADS WorldCat Takagi K. , Yatsu A., Moku M., Sassa C. 2006 . Age and growth of lanternfishes, Symbolophorus californiensis and Ceratoscopelus warmingii (Myctophidae), in the Kuroshio–Oyashio Transition Zone . Ichthyological Research , 53 : 281 – 289 . Google Scholar Crossref Search ADS WorldCat Tönnesson K. , Tiselius P. 2005 . Diet of the chaetognaths Sagitta setosa and S. elegans in relation to prey abundance and vertical distribution . Marine Ecology Progress Series , 289 : 177 – 190 . Google Scholar Crossref Search ADS WorldCat Trueman C. N. , Johnston G., O’Hea B., MacKenzie K. M. 2014 . Trophic interactions of fish communities at midwater depths enhance long-term carbon storage and benthic production on continental slopes . Proceedings of the Royal Society B , 281 : 20140669 . Google Scholar Crossref Search ADS PubMed WorldCat Van Noord J. E. , Olson R. J., Redfern J. V., Duffy L. M., Kaufmann R. S. 2016 . Oceanographic influences on the diet of 3 surface-migrating myctophids in the eastern tropical Pacific Ocean . Fishery Bulletin , 114 : 274 – 287 . Google Scholar Crossref Search ADS WorldCat Watanabe H. , Moku M., Kawaguchi K., Ishimaru K., Ohno A. 1999 . Diel vertical migration of myctophid fishes (family Myctophidae) in the transitional waters of the western North Pacific . Fisheries Oceanography , 8 : 115 – 127 . Google Scholar Crossref Search ADS WorldCat Westberry T. , Behrenfeld M. J., Siegel D. A., Boss E. 2008 . Carbon-based primary productivity modelling with vertically resolved photoacclimation . Global Biogeochemical Cycles , 22 : GB2024. Google Scholar Crossref Search ADS WorldCat Yamada Y. , Ikeda T. 2006 . Production, metabolism and trophic importance of four pelagic amphipods in the Oyashio region, western subarctic Pacific . Marine Ecology Progress Series , 308 : 155 – 163 . Google Scholar Crossref Search ADS WorldCat Yool A. , Popova E. E., Anderson T. R. 2013 . MEDUSA-2.0: an intermediate complexity biogeochemical model of the marine carbon cycle for climate change and ocean acidification studies . Geoscientific Model Development , 6 : 1767 – 1811 . Google Scholar Crossref Search ADS WorldCat Zaret T. M. , Suffern J. S. 1976 . Vertical migration in zooplankton as a predator avoidance mechanism . Limnology and Oceanography , 21 : 804 – 813 . Google Scholar Crossref Search ADS WorldCat Zhang Z. , Dam H. G. 1997 . Downward export of carbon by diel migrant mesozooplankton in the central equatorial Pacific . Deep-Sea Research II , 44 : 2191 – 2202 . Google Scholar Crossref Search ADS WorldCat © International Council for the Exploration of the Sea 2018. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. © International Council for the Exploration of the Sea 2018.

Pakhomov, Evgeny, A;Podeswa,, Yasha;Hunt, Brian P, V;Kwong, Lian, E

doi: 10.1093/icesjms/fsy134pmid: N/A

Abstract Pelagic decapods were sampled during August 2011 in the central North Pacific Subtropical Gyre (NPSG). Depth-stratified samples using a MOCNESS-10 (10 m2 Multiple Opening/Closing Net and Environmental Sensing System) were collected at two stations to the west and north of the Hawaiian island of Oahu: Station Kahe: 21°20.6′N–158°16.4′W and Station ALOHA: 22°45′N–158°00′W. Total decapod abundance and biomass were 4.3 ind. m−2 and 0.71 gDW m−2. While 40 decapod taxa were identified, only 22 species were sampled sufficiently to study quantitatively their vertical migrations. All species were classified into three migration groups: full migrators (6 species); partial migrators (13 species); and non-migrators (3 species). Using measured local temperature profiles along with decapod densities and published models of respiration, excretion and mortality as well as gut fullness data, the individual and total active downward carbon flux was calculated. Active carbon flux of migrating decapods ranged from 383 to 625 µgC m−2 day−1. This active flux was equal to 4.8–7.8% of passive flux at the mean nighttime residence depth of ∼711 m), 2.1–3.4% of passive flux at the mean daytime residence depth (∼262 m), and 1.5–2.4% of passive flux at the base of the euphotic zone (∼173 m). Mortality flux accounted for ∼70% of total active flux, followed by gut flux—∼18%. Introduction The oceans are estimated to have absorbed approximately half of total anthropogenic CO2 emissions since the beginning of the industrial revolution (Sabine et al., 2004; Le Quéré et al., 2018). The physical and biological processes that mediate the transfer of carbon from the surface of the ocean to the ocean’s interior are a key component of the global carbon cycle and the biological pump is one of the most important pathways for carbon transported vertically in the ocean and buried in the sediments. The biological pump includes the passive sinking of particulate organic matter (POM), diffusion and advection of dissolved organic matter (DOM), and active transport by the vertical migration of animals (Hidaka et al., 2001). In the past, passive POM sinking, also known as “gravitational flux” or “passive flux”, and diffusion and advection of DOM were considered the most important processes mediating vertical transport (Longhurst, 1991). However, since the 1990s the importance of active transport of carbon by vertical migrators has been increasingly recognized (Dam et al., 1995, Hansen and Visser, 2016). This mechanism involves the transfer of organic matter consumed by zooplankton and nekton at the surface to their daytime residence depths through a combination of respiration, excretion, defecation, and mortality (Lampert, 1989; Longhurst, 1991). To date however, the majority of empirical studies assessing active carbon transport concentrated on the mesozooplankton and estimated flux out of the upper mixing layer only (Supplementary Table S1). Micronekton are actively swimming marine organisms, generally larger than drifting mesozooplankton < 2 cm), but smaller than larger nekton (>10 cm) (Brodeur et al., 2005). While they can be defined precisely based on Reynolds numbers, operationally micronekton include taxa too small to be caught by most large meshed pelagic trawls, but too mobile to be caught efficiently by conventional plankton gears (Brodeur et al., 2005). Their patchy distribution and high mobility makes them very difficult to sample without bias (Pakhomov and Yamamura, 2010). For these reasons, micronekton tend to be poorly sampled and understood (Brodeur and Yamamura, 2005). Micronekton nevertheless are the most conspicuous members of the mesopelagic community (Brodeur et al., 2005), and significant nighttime residents of the epipelagic (Brodeur and Yamamura, 2005; Pakhomov and Yamamura, 2010). Functionally, micronekton are a primary food source for a wide variety of nektonic species, including commercially harvested species, and their active vertical migrators link smaller zooplankton with both large epipelagic (such as tuna, swordfish, and sharks) and meso/bathypelagic predators (Brodeur et al., 2005; Brodeur and Yamamura, 2005). Pelagic decapods are an abundant and important component of the micronekton community throughout many regions of the world’s oceans (Maynard et al., 1975; Hopkins et al., 1989; Flock and Hopkins, 1992; Hopkins and Sutton, 1998). In the central North Pacific Subtropical Gyre (NPSG), previous studies have shown penaeid and caridean shrimps to be the second and fifth most abundant micronekton groups in deep (0–1200 m) net tows, and the first and fifth most abundant micronekton groups in shallow (0–400 m) nighttime tows (Maynard et al., 1975; Hopkins et al., 1994). Despite the high local abundance and functional importance of micronekton, very little is known about the diets or trophic role of pelagic decapods in the central NPSG. Active carbon flux attributed to migrant zooplankton in various areas of the world’s oceans has been comparable to local gravitational fluxes (Longhurst et al., 1990; Longhurst and Williams, 1992; Dam et al., 1995; Le Borgne and Rodier, 1997; Zhang and Dam, 1997; Steinberg et al., 2000, 2002; Al-Mutairi and Landry, 2001) suggesting a potentially important role for abundant pelagic decapods. Longhurst et al. (1990) were among first to quantify the active flux attributed to migratory zooplankton. Using data from tropical and subtropical stations in the northwestern Atlantic and eastern Pacific, they showed that respiratory carbon flux due to zooplankton migrations across the pycnocline was equal to 13–58% of gravitational flux. A significant component of downward carbon flux had thus been completely missed in previous carbon models (Longhurst et al., 1990). That study also indicated that active flux was highly variable between locations, and depended strongly upon the zooplankton community composition (Longhurst et al., 1990, see also Supplementary Table S1). Most previous estimates of active flux were made using sampling gears well suited to sample mesozooplankton. Only a handful studies have attempted to assess the contribution of micronekton, or its main taxonomic groups, to vertical carbon flux pointing that pelagic decapods and myctophids may support significant active carbon transport to depths exceeding 300 m (Hidaka et al., 2001; Davison et al., 2013; Schukat et al., 2013). Previous studies of micronekton active carbon flux have been conducted in ecosystems where decapods contributed modestly to total micronekton density. During summer 2004, micronekton communities were sampled off the west coast of Oahu Island and both pelagic decapod diversity and contribution to the micronektonic community were found to be comparable to euphausiids, myctophids, and other fish (Kwong et al., 2018). Stations ALOHA and Kahe have been regularly sampled since October 1988, under the auspices of the Hawaii Ocean Time-series (HOT) program (Karl and Lukas, 1996). These stations were chosen as sites representative of the NPSG, which is the earth’s largest contiguous biome (Karl, 1999; Karl and Lukas, 1996), but until now, the carbon flux through micronekton vertical migrations has not been adequately measured. The aims of this study were twofold: (a) to describe the composition, vertical distribution, and diel vertical migrations of pelagic decapods in the central NPSG, and (b) assess decapod contribution to local vertical carbon flux. Material and methods Samples were collected between August 19 and 25 of 2011 aboard the R/V Kilo Moana at two stations near the Hawaiian Island of Oahu, Station Kahe and Station ALOHA (Figure 1). Figure 1. View largeDownload slide Sampling positions during August 2011 R/V Kilo Moana voyage in the NPSG off Hawaii. Figure 1. View largeDownload slide Sampling positions during August 2011 R/V Kilo Moana voyage in the NPSG off Hawaii. Station Kahe was located ∼10 km from land (at 21°20.6′N 158°16.4′W), with a bottom depth of ∼1500 m, while Station ALOHA is considered an open ocean station and located ∼100 km from land (at 22°45′N 158°00′W) with a bottom depth of approximately 4800 m (Karl and Lukas, 1996). Micronekton were collected using a MOCNESS-10 gear (Multiple Opening/Closing Net and Environmental Sensing System): a frame trawl with a 10 m2 mouth opening and 6 mm mesh that was towed at a speed of 2 kts, and equipped with a SeaBird CTD to measure the physical properties of the water column. The MOCNESS-10 was outfitted with six nets: five of these nets were used to sample five discrete depth intervals, while the sixth net performed an oblique tow on the way down over the entire depth range. In this article, all decapods from the nets sampling discrete depths were identified to the lowest taxonomic level possible, counted and their carapace lengths measured. Carapace length was measured from the posterior middorsal margin to the posterior edge of the orbit using digital calipers with a resolution of 0.1 mm. Six depth-stratified tows were conducted, four at Station ALOHA and two at Station Kahe. Tows were made to depths of up to 1500 m during the day and up to 2500 m at night (Figure 2a). Due to limited sampling effort, data from Kahe and ALOHA stations were pooled. This was justified because the physical setting as well as decapod species composition at both stations were highly similar (Figure 2b). Nevertheless, daytime sampling in the top 500 m only yielded one sample (Figure 2a). While it may be considered a low effort, recent study showed that daytime micronekton decapod density in the upper 500 m layer is insignificant (Kwong et al., 2018). Figure 2. View largeDownload slide Stations Kahe and ALOHA combined sample size at each depth interval (a) and typical temperature and salinity profiles (b) during the 2011 R/V Kilo Moana voyage. Figure 2. View largeDownload slide Stations Kahe and ALOHA combined sample size at each depth interval (a) and typical temperature and salinity profiles (b) during the 2011 R/V Kilo Moana voyage. Additional samples used to calculate length–weight relationships and carbon weight to dry weight ratios were collected at the Kahe Station during the 2004 Oscar Elton Sette cruise (Podeswa and Pakhomov, 2015). From the samples preserved in 6% buffered formalin roughly 75 individuals per species were used to determine species-specific carapace length to dry weight relationships. Individuals were rinsed thoroughly, carapace lengths were measured as described above, and wet weights were measured to 0.1 mg after blotting each individual with KimWipes to remove excess water. The decapods were then dried in an oven at 50°C for 24–72 h (24 h for small individuals, 48 h for moderate-sized individuals, 72 h for large individuals), then re-weighed on the same balance to determine their dry weights. It was necessary to determine wet to dry weight ratios for decapods for two reasons. First, for species used for gut content analysis the dry weight of stomach content and the decapod was directly available. However, for species where no gut content analysis was performed, carapace lengths and wet weights were measured. In such situations, dry weights were not directly measured to avoid unnecessary destruction of specimens. Instead, wet weights were converted to dry weights mathematically using predetermined wet to dry weight ratios. Second, the wet to dry weight ratios are affected by dry weight loss during formalin preservation, which has been well documented for zooplankton and micronekton in the literature (Pakhomov, 2003). Dry weight loss could be estimated based on body water content through a parabolic equation (Pakhomov, 2003): Dryweightloss=0.045* (body water content)2−6.898* (body water content)+289.4 (1) Dry weight loss is taken as a percentage of the mass of the unpreserved individual, while body water content is equal to the percentage of mass lost when an individual is dried completely in an oven. Relationships between carapace length and corrected dry weight were then derived for each species (Table 1). “Corrected dry weights” refer to both (a) dry weights that have been corrected for dry mass loss and (b) to wet weights that first have been converted to dry weights, and then corrected for dry weight loss. Table 1. Relationship between carapace length (CL, mm) and corrected dry weights (DW, mg) for all pelagic decapod species found to perform diel vertical migrations. Species Mass units a B p-value R2 Acanthephyra smithi* DW 0.2862 2.7956 <0.0001 0.968 Allosergestes pectinatus WW 0.1415 2.7634 <0.0001 0.849 Allosergestes sargassi WW 0.4947 2.0937 <0.0001 0.950 Deosergestes erectus WW 0.0575 2.8279 <0.0001 0.997 Gennadas spp.** DW 0.7724 2.2225 <0.0001 0.949 Janicella spinicauda DW 0.3375 2.6935 <0.0001 0.708 Neosergestes consobrinus WW 0.0450 3.2939 <0.0001 0.955 Neosergestes orientalis DW 0.2956 2.3369 <0.0001 0.792 Notostomus elegans*** DW 0.0121 3.4175 <0.0001 0.987 Parasergestes armatus DW 0.0743 2.8101 <0.0001 0.893 Sergestes atlanticus WW 0.0421 3.2794 <0.0001 0.993 Sergia bigemmeus WW 0.1345 2.8011 <0.0001 0.993 Sergia gardineri DW 0.6103 2.1930 <0.0001 0.820 Sergia scintillans DW 0.2389 2.5427 <0.0001 0.778 Stylopandalus richardi DW 0.1108 3.1685 <0.0001 0.907 Systellaspis debilis WW 0.0985 3.1415 <0.0001 0.990 Species Mass units a B p-value R2 Acanthephyra smithi* DW 0.2862 2.7956 <0.0001 0.968 Allosergestes pectinatus WW 0.1415 2.7634 <0.0001 0.849 Allosergestes sargassi WW 0.4947 2.0937 <0.0001 0.950 Deosergestes erectus WW 0.0575 2.8279 <0.0001 0.997 Gennadas spp.** DW 0.7724 2.2225 <0.0001 0.949 Janicella spinicauda DW 0.3375 2.6935 <0.0001 0.708 Neosergestes consobrinus WW 0.0450 3.2939 <0.0001 0.955 Neosergestes orientalis DW 0.2956 2.3369 <0.0001 0.792 Notostomus elegans*** DW 0.0121 3.4175 <0.0001 0.987 Parasergestes armatus DW 0.0743 2.8101 <0.0001 0.893 Sergestes atlanticus WW 0.0421 3.2794 <0.0001 0.993 Sergia bigemmeus WW 0.1345 2.8011 <0.0001 0.993 Sergia gardineri DW 0.6103 2.1930 <0.0001 0.820 Sergia scintillans DW 0.2389 2.5427 <0.0001 0.778 Stylopandalus richardi DW 0.1108 3.1685 <0.0001 0.907 Systellaspis debilis WW 0.0985 3.1415 <0.0001 0.990 “a” and “b” are constants derived by regression analysis for the equation DW = a * CLb. * Also used for Acanthephyra curtirostris; ** used for all Gennadas species (bouvieri, clavicarpus, capensis, incertus, and tinayrei); *** also used for Notostomus gibbosus. Table 1. Relationship between carapace length (CL, mm) and corrected dry weights (DW, mg) for all pelagic decapod species found to perform diel vertical migrations. Species Mass units a B p-value R2 Acanthephyra smithi* DW 0.2862 2.7956 <0.0001 0.968 Allosergestes pectinatus WW 0.1415 2.7634 <0.0001 0.849 Allosergestes sargassi WW 0.4947 2.0937 <0.0001 0.950 Deosergestes erectus WW 0.0575 2.8279 <0.0001 0.997 Gennadas spp.** DW 0.7724 2.2225 <0.0001 0.949 Janicella spinicauda DW 0.3375 2.6935 <0.0001 0.708 Neosergestes consobrinus WW 0.0450 3.2939 <0.0001 0.955 Neosergestes orientalis DW 0.2956 2.3369 <0.0001 0.792 Notostomus elegans*** DW 0.0121 3.4175 <0.0001 0.987 Parasergestes armatus DW 0.0743 2.8101 <0.0001 0.893 Sergestes atlanticus WW 0.0421 3.2794 <0.0001 0.993 Sergia bigemmeus WW 0.1345 2.8011 <0.0001 0.993 Sergia gardineri DW 0.6103 2.1930 <0.0001 0.820 Sergia scintillans DW 0.2389 2.5427 <0.0001 0.778 Stylopandalus richardi DW 0.1108 3.1685 <0.0001 0.907 Systellaspis debilis WW 0.0985 3.1415 <0.0001 0.990 Species Mass units a B p-value R2 Acanthephyra smithi* DW 0.2862 2.7956 <0.0001 0.968 Allosergestes pectinatus WW 0.1415 2.7634 <0.0001 0.849 Allosergestes sargassi WW 0.4947 2.0937 <0.0001 0.950 Deosergestes erectus WW 0.0575 2.8279 <0.0001 0.997 Gennadas spp.** DW 0.7724 2.2225 <0.0001 0.949 Janicella spinicauda DW 0.3375 2.6935 <0.0001 0.708 Neosergestes consobrinus WW 0.0450 3.2939 <0.0001 0.955 Neosergestes orientalis DW 0.2956 2.3369 <0.0001 0.792 Notostomus elegans*** DW 0.0121 3.4175 <0.0001 0.987 Parasergestes armatus DW 0.0743 2.8101 <0.0001 0.893 Sergestes atlanticus WW 0.0421 3.2794 <0.0001 0.993 Sergia bigemmeus WW 0.1345 2.8011 <0.0001 0.993 Sergia gardineri DW 0.6103 2.1930 <0.0001 0.820 Sergia scintillans DW 0.2389 2.5427 <0.0001 0.778 Stylopandalus richardi DW 0.1108 3.1685 <0.0001 0.907 Systellaspis debilis WW 0.0985 3.1415 <0.0001 0.990 “a” and “b” are constants derived by regression analysis for the equation DW = a * CLb. * Also used for Acanthephyra curtirostris; ** used for all Gennadas species (bouvieri, clavicarpus, capensis, incertus, and tinayrei); *** also used for Notostomus gibbosus. To estimate the abundance of each decapod species at depth intervals, the catch was divided by the volume of water filtered. Due to the logistical challenge of accommodating various research group’s needs, somewhat different depth intervals were sampled during different tows (Figure 2a). For example, in different series either 600–700 m or 700–800 m or 600–800 m depth intervals were sampled. It was thus necessary to either interpolate or extrapolate when estimating abundance for each depth interval and interpolation was deemed more appropriate. For example, data from the 600–700 m, 700–800 m, and 600–800 m depth intervals were analysed using the following equations: Estimated 600−700m abund.=2/3 *(600−700mabund.)+1/3 *(600–800mabund.) (2) Estimated 700−800m abund.=2/3 *(700−800mabund.)+1/3 *(600–800mabund.) (3) It has been well established that nets underestimate the densities of zooplankton, micronekton, and nekton. The catch efficiency of a net should therefore be taken into account when deriving abundance from net data (Aron and Collard, 1969; Misund et al., 1999; Itaya et al., 2001, 2007; Wiebe et al., 2004). The catch efficiency of the MOCNESS 10 was conservatively assumed 33.3% due to the following considerations. First, Itaya et al. (2007) tested the effect of towing speed on catch per unit effort (CPUE) for a variety of frame trawls. They found that a frame trawl with a 4 m2 mouth opening had a CPUE for two myctophid species (Benthosema suborbitale and Diogenichthys atlanticus) 2.8–3.4 fold higher at a towing speed of 3–4 kts versus 2 kts (Itaya et al., 2007). Since it is unlikely that a 4 m2 frame trawl towed even at 4 kts has a 100% catch efficiency, the assumed catch efficiency of 33.3% was likely an overestimate. Second, the CUPE of 4 and 12.3 m2 frame trawls towed at 4 kts was similar for D. atlanticus, while for B. suborbitale it was approximately double in the latter trawl compared with the 4 m2 trawl (Itaya et al., 2007). While this would suggest that, for certain species, the 10 m2 net used in this study may somewhat offset the net avoidance incurred by the lower tow speed, the assumed 33.3% efficiency is still on the conservative side. Third, the 33.3% sampling efficiency could be arrived at based on studies that used a strobe light attached to the front of a MOCNESS net to “stun” euphausiids increasing catch by up to three times (Sameoto et al., 1993; Wiebe et al., 2004). Finally, direct MOHT and MOCNESS 10 inter-comparison in the northwestern Pacific showed that the latter net sampled only ∼10% and ∼20% of myctophids and euphausiids, respectively (Pakhomov and Yamamura, 2010, Table 3.2). Based on the available evidence, applying a catch efficiency of 33.3% for micronektonic decapods sampled with a 10 m2 MOCNESS towed at 2 kts seemed conservative and acceptable. Finally, it was necessary to account for reduced catches during the day. Due to increased visibility of the net during the day, avoidance is expected to be greater. It has been suggested that nighttime estimated abundances often exceed daytime densities due to visual net avoidance, and thus daytime abundances should be assumed equal to nighttime abundances. During our study, large day/night difference in density was only observed for 1 out of 19 migratory decapod species (e.g. Sergia bigemmeus), so overall this was a minor adjustment. To determine whether a population is performing diel vertical migrations, the weighted mean depth (WMD) of each species must be calculated during both the day and the night as follows: WMD=∑(ni *zi *di)/∑(ni *zi), (4) where, di is the depth of a sample i (the depth at the centre of the depth interval, in m), zi the thickness of the interval (in m), and ni is the abundance of individuals at that depth (ind. 1000 m−3) (Andersen et al., 2001). Welch’s two-sample t-test was performed to determine statistically significant differences between day and night WMDs of individual species (Sawilowsky, 2002). In general, t-tests are highly robust in terms of both Type I and II errors to departures from the assumption of normality. Therefore, non-normal distributions for some species should not have been a significant issue (Sawilowsky, 2002). Species with significantly different day and night WMDs were classified as vertical migrators. Some species did not show significant migration toward the surface at night, while others performed partial migration. Partially migrating species were identified based on a bimodal distribution at night, with one peak at the daytime depth (non-migrating portion of population) and another peak (migrating portion) closer to the surface. For such species, the upward migrating portion was expressed as the percentage of the total nighttime population and its WMD calculated. Estimates of active flux (see below) were calculated based only upon the migratory portion of the population. Active flux of carbon by migrating micronekton is composed of four major components: respiratory (RC), excretory (EC), mortality (Mflux), and gut (Gflux) flux. Each of these fluxes was calculated using individual rate processes that were scaled up to the migrating densities, accounting for the amount of time migrating decapods resided in the mesopelagic depths (see below). Respiratory flux was calculated using an empirical allometric relationship derived by Ikeda (1985), which predicts the respiration rate of a zooplankton/micronekton based on ambient temperature and the organism’s biomass. lnRO=−0.2512 + 0.7886 *ln DWmg+0.0490*T, (5) where RO is the respiratory oxygen uptake (µl O2 ind.−1 h−1), DWmg is the dry weight of the organism (mg), and T is the environmental temperature (oC). This hourly respiration rate was then converted to a daily respiration rate by the number of hours of daylight, which at Station ALOHA over the study period was 12.6 h. Once rates of oxygen uptake were determined, it was necessary to convert these rates into respiratory equivalents following Al-Mutairi and Landry (2001) and scale up to total decapod density: RC=RO*RQ *12/22.4 * TD *N, (6) where RC is the respiratory carbon flux (µg C m−2 day−1), RQ is the respiratory quotient (assumed 0.97 after Gnaiger, 1983), which is the molar ratio of carbon produced to oxygen utilized, 12 is the molecular weight of carbon, and 22.4 is the molar volume of an ideal gas at standard pressure and temperature, TD is the number of hours per day the organism spends at depth (12.6 h) and N is the decapod migrating density (ind. m−2). Active excretory flux was calculated based on the findings of Steinberg et al. (2000). The authors measured CO2 respiration and DOC excretion of a wide variety of diel migratory crustacean species at the US JGOFS Bermuda Atlantic Time-series Study. Respiration and excretion rates were found to vary similarly, with both rates depending on environmental temperature and the dry weight of the organism (Steinberg et al., 2000). In the case of diel migratory decapods, DOC excretion averaged 32% of CO2 respiration in terms of µg C respired or excreted per mg dry weight (Steinberg et al., 2000). Thus, for this study, excretory DOC (excretory flux, EC) was assumed equal to 32% of respiratory CO2. Decapod mortality was calculated by estimating the hourly weight-specific mortality rate using the organism’s dry weight, based on the model of Peterson and Wroblewski (1984). HM=2.196*10−4*DWg−0.25, (7) where HM is the hourly weight-specific mortality rate (h−1), and DWg is the dry weight of the animal (g). This was converted into a mortality flux using the following equation from Zhang and Dam (1997): Mflux =HM*DWµg*CR*TD, (8) where, Mflux is the daily mortality flux, DWµg is the dry weight of the animal (µg), CR is the carbon weight to dry weight ratio, and TD is the number of hours (12.6 h) per day the organism spends at depth. Carbon to dry weight ratio for pelagic decapods captured at Kahe Station was measured to be 0.42 (CR) during the 2004 Oscar Elton Sette cruise (Podeswa and Pakhomov, 2015). Gut flux (Gflux) was estimated primarily using Equation (9), empirically derived using decapods captured during the 2004 Oscar Elton Sette cruise, predicting stomach content dry weight based on organism dry weight and stomach fullness (Podeswa and Pakhomov, 2015). FBDW =0.020*OrgDW *Fullness, (9) where FBDW is the food ball dry weight (mg), OrgDW is the organism’s dry weight (mg), and Fullness is the visually estimated stomach fullness. The food ball dry weight for all migratory individuals (Gflux) was determined using Equation (9) and species-specific values for peak nighttime stomach fullness. For a few species where no stomach content data was obtained, the mean peak nighttime stomach fullness of 0.52 averaged across all species was used. To convert food ball dry weights to food ball carbon, a mean ratio of 0.46 (obtained for fish, squid, euphausiids and copepods) was employed (Podeswa and Pakhomov, 2015). It was assumed that the highest nighttime stomach content of a migrant shrimp was carried from the euphotic to the mesopelagic zone, digested and remains egested at depth (Clarke, 1980). An assimilation efficiency of 88% was adopted and thus max average food ball weight (in mg C) was multiplied by 0.12 to estimate the carbon weight of the egested material (Hopkins and Baird, 1977). Once active fluxes of all species were summed, a weighted average taken to determine the mean depth of carbon transport. Weighting was based on the size of the active flux. For example, if “Species A” migrated to a daytime WMD of 800 m, with a total active flux of 90 µgC m−2 day−1, while “Species B” migrated to a daytime WMD of 700 m, with a total active flux of 10 µgC m−2 day−1, there would be a total of 100 µgC m−2 day−1 transported to a WMD of 790 m. Once active flux was determined, it was related to the previously estimated passive flux at the same depth. A 5-year time-series study performed at Station ALOHA found that, below the base of the euphotic zone, long-term passive carbon flux could be accurately modelled based on depth using the following equation (Karl et al., 1996): PC−FLUX(Z)=28.7*(Z/150)−0.818, (10) where PC-FLUX(Z) is the particulate carbon flux to a depth of Z meters, in mgC m−2 day−1. The same study also found the base of the euphotic zone to be located at 173 ± 7 m. Results Temperature and salinity Temperatures at Station ALOHA remained steady at ∼25.5°C in the mixed layer (∼0–60 m), before rapidly declining in the thermocline to ∼6.3°C at the base of the thermocline (∼530 m) (Figure 2b). Temperatures continued to decline below the thermocline reaching ∼2.9°C at 1500 m (Figure 2b). Salinity remained constant at ∼35.1 in the mixed layer, rising to ∼35.3 at ∼100 m, before falling rapidly to ∼34.0 at the base of the halocline (∼500 m) (Figure 2b). Temperature and salinity profiles at Station Kahe were very similar to those found at Station ALOHA. The main differences were lower salinity in the mixed layer at Station Kahe (34.5 at Kahe vs. 35.1 at ALOHA), and a shallower base of the thermocline (490 m at Kahe vs. 530 m at ALOHA) (Figure 2b). Pelagic decapod diversity The diversity of pelagic decapods was high at both sites with 40 taxonomic groups identified to species or genus level (Table 2). In total, 21 genera and 5 families were represented: Oplophoridae, Pasiphaeidae, Penaeidae, Sergestidae, and Sicyoniidae. All species ≥20 mm for which fewer than 10 individuals were caught were excluded from the analysis of fluxes. In total, 22 species/groups were included in the analysis. Table 2. Pelagic decapod species composition and density during the 2011 R/V Kilo Moana voyage. Species No Mean DW mg ind.−1 Abundance ind. m−2 Biomass mgDW m−2 Acanthephyra curtirostris 41 209.16 0.256 56.438 Acanthephyra smithi 17 626.34 0.0934 59.287 Allosergestes pectinatus 38 12.10 0.152 1.771 Allosergestes sargassi 10 31.32 0.049 1.571 Deosergestes erectus 26 154.26 0.146 25.253 Gennadas bouvieri 95 84.96 0.481 41.773 Gennadas capensis 48 94.27 0.331 35.780 Gennadas clavicarpus 33 67.76 0.121 8.265 Gennadas incertus 13 50.70 0.071 3.694 Gennadas spp. 80 58.98 0.372 22.496 Gennadas tinayrei 10 51.45 0.048 2.753 Janicella spinicauda 18 40.15 0.105 4.378 Neosergestes consobrinus 26 10.56 0.132 1.426 Neosergestes orientalis 29 29.76 0.146 4.202 Notostomus gibbosus 23 1746.01 0.166 275.861 Parasergestes armatus 44 67.76 0.278 19.117 Sergestes atlanticus 19 25.55 0.073 1.719 Sergia bigemmeus 17 77.72 0.118 10.544 Sergia gardineri 117 35.62 0.560 20.376 Sergia scintillans 24 45.60 0.092 4.120 Stylopandalus richardi 28 65.76 0.167 12.171 Systellaspis debilis 28 362.29 0.151 54.920 Total 4.108 667.915 Total non-migrating biomass 423.421 Total migrating biomass 244.494 Acanthephyra prionota 4 212.83 0.012 2.553 Acanthephyra sp. 7 53.22 0.015 0.798 Bentheogennema sp. 4 164.97 0.012 1.979 Funchalia taaningi 3 401.31 0.010 4.013 Glyphus sp. 2 35.35 0.005 0.176 Heterocarpus ensifer parvispina 1 113.15 0.004 0.452 Meningodora marptocheles 1 283.94 0.004 1.135 Oplophoridae sp. larvae 3 15.48 0.010 0.154 Oplophorus gracilirostris 4 824.93 0.012 9.899 Parapasiphae sulcatifrons 1 476.44 0.004 1.905 Parasergestes vigilax 4 16.40 0.012 0.196 Penaeidae sp. 7 40.22 0.015 0.603 Petalidium sp. 3 31.08 0.010 0.311 Sergia bisulcatus 3 534.03 0.010 5.340 Sergia inequalis 4 272.13 0.012 3.265 Sergia spp. (<20 mm) 12 56.50 0.050 2.825 Sergia tenuiremis 4 475.53 0.012 5.706 Sicyonia sp. 1 148.50 0.004 0.594 Total (excluded) 0.213 41.911 Species No Mean DW mg ind.−1 Abundance ind. m−2 Biomass mgDW m−2 Acanthephyra curtirostris 41 209.16 0.256 56.438 Acanthephyra smithi 17 626.34 0.0934 59.287 Allosergestes pectinatus 38 12.10 0.152 1.771 Allosergestes sargassi 10 31.32 0.049 1.571 Deosergestes erectus 26 154.26 0.146 25.253 Gennadas bouvieri 95 84.96 0.481 41.773 Gennadas capensis 48 94.27 0.331 35.780 Gennadas clavicarpus 33 67.76 0.121 8.265 Gennadas incertus 13 50.70 0.071 3.694 Gennadas spp. 80 58.98 0.372 22.496 Gennadas tinayrei 10 51.45 0.048 2.753 Janicella spinicauda 18 40.15 0.105 4.378 Neosergestes consobrinus 26 10.56 0.132 1.426 Neosergestes orientalis 29 29.76 0.146 4.202 Notostomus gibbosus 23 1746.01 0.166 275.861 Parasergestes armatus 44 67.76 0.278 19.117 Sergestes atlanticus 19 25.55 0.073 1.719 Sergia bigemmeus 17 77.72 0.118 10.544 Sergia gardineri 117 35.62 0.560 20.376 Sergia scintillans 24 45.60 0.092 4.120 Stylopandalus richardi 28 65.76 0.167 12.171 Systellaspis debilis 28 362.29 0.151 54.920 Total 4.108 667.915 Total non-migrating biomass 423.421 Total migrating biomass 244.494 Acanthephyra prionota 4 212.83 0.012 2.553 Acanthephyra sp. 7 53.22 0.015 0.798 Bentheogennema sp. 4 164.97 0.012 1.979 Funchalia taaningi 3 401.31 0.010 4.013 Glyphus sp. 2 35.35 0.005 0.176 Heterocarpus ensifer parvispina 1 113.15 0.004 0.452 Meningodora marptocheles 1 283.94 0.004 1.135 Oplophoridae sp. larvae 3 15.48 0.010 0.154 Oplophorus gracilirostris 4 824.93 0.012 9.899 Parapasiphae sulcatifrons 1 476.44 0.004 1.905 Parasergestes vigilax 4 16.40 0.012 0.196 Penaeidae sp. 7 40.22 0.015 0.603 Petalidium sp. 3 31.08 0.010 0.311 Sergia bisulcatus 3 534.03 0.010 5.340 Sergia inequalis 4 272.13 0.012 3.265 Sergia spp. (<20 mm) 12 56.50 0.050 2.825 Sergia tenuiremis 4 475.53 0.012 5.706 Sicyonia sp. 1 148.50 0.004 0.594 Total (excluded) 0.213 41.911 Abundance and biomass values listed for each species are the mean of daytime and nighttime estimates. Species in bold accounted for < 6% of total abundance and biomass and were not included in the active carbon transport calculations. Table 2. Pelagic decapod species composition and density during the 2011 R/V Kilo Moana voyage. Species No Mean DW mg ind.−1 Abundance ind. m−2 Biomass mgDW m−2 Acanthephyra curtirostris 41 209.16 0.256 56.438 Acanthephyra smithi 17 626.34 0.0934 59.287 Allosergestes pectinatus 38 12.10 0.152 1.771 Allosergestes sargassi 10 31.32 0.049 1.571 Deosergestes erectus 26 154.26 0.146 25.253 Gennadas bouvieri 95 84.96 0.481 41.773 Gennadas capensis 48 94.27 0.331 35.780 Gennadas clavicarpus 33 67.76 0.121 8.265 Gennadas incertus 13 50.70 0.071 3.694 Gennadas spp. 80 58.98 0.372 22.496 Gennadas tinayrei 10 51.45 0.048 2.753 Janicella spinicauda 18 40.15 0.105 4.378 Neosergestes consobrinus 26 10.56 0.132 1.426 Neosergestes orientalis 29 29.76 0.146 4.202 Notostomus gibbosus 23 1746.01 0.166 275.861 Parasergestes armatus 44 67.76 0.278 19.117 Sergestes atlanticus 19 25.55 0.073 1.719 Sergia bigemmeus 17 77.72 0.118 10.544 Sergia gardineri 117 35.62 0.560 20.376 Sergia scintillans 24 45.60 0.092 4.120 Stylopandalus richardi 28 65.76 0.167 12.171 Systellaspis debilis 28 362.29 0.151 54.920 Total 4.108 667.915 Total non-migrating biomass 423.421 Total migrating biomass 244.494 Acanthephyra prionota 4 212.83 0.012 2.553 Acanthephyra sp. 7 53.22 0.015 0.798 Bentheogennema sp. 4 164.97 0.012 1.979 Funchalia taaningi 3 401.31 0.010 4.013 Glyphus sp. 2 35.35 0.005 0.176 Heterocarpus ensifer parvispina 1 113.15 0.004 0.452 Meningodora marptocheles 1 283.94 0.004 1.135 Oplophoridae sp. larvae 3 15.48 0.010 0.154 Oplophorus gracilirostris 4 824.93 0.012 9.899 Parapasiphae sulcatifrons 1 476.44 0.004 1.905 Parasergestes vigilax 4 16.40 0.012 0.196 Penaeidae sp. 7 40.22 0.015 0.603 Petalidium sp. 3 31.08 0.010 0.311 Sergia bisulcatus 3 534.03 0.010 5.340 Sergia inequalis 4 272.13 0.012 3.265 Sergia spp. (<20 mm) 12 56.50 0.050 2.825 Sergia tenuiremis 4 475.53 0.012 5.706 Sicyonia sp. 1 148.50 0.004 0.594 Total (excluded) 0.213 41.911 Species No Mean DW mg ind.−1 Abundance ind. m−2 Biomass mgDW m−2 Acanthephyra curtirostris 41 209.16 0.256 56.438 Acanthephyra smithi 17 626.34 0.0934 59.287 Allosergestes pectinatus 38 12.10 0.152 1.771 Allosergestes sargassi 10 31.32 0.049 1.571 Deosergestes erectus 26 154.26 0.146 25.253 Gennadas bouvieri 95 84.96 0.481 41.773 Gennadas capensis 48 94.27 0.331 35.780 Gennadas clavicarpus 33 67.76 0.121 8.265 Gennadas incertus 13 50.70 0.071 3.694 Gennadas spp. 80 58.98 0.372 22.496 Gennadas tinayrei 10 51.45 0.048 2.753 Janicella spinicauda 18 40.15 0.105 4.378 Neosergestes consobrinus 26 10.56 0.132 1.426 Neosergestes orientalis 29 29.76 0.146 4.202 Notostomus gibbosus 23 1746.01 0.166 275.861 Parasergestes armatus 44 67.76 0.278 19.117 Sergestes atlanticus 19 25.55 0.073 1.719 Sergia bigemmeus 17 77.72 0.118 10.544 Sergia gardineri 117 35.62 0.560 20.376 Sergia scintillans 24 45.60 0.092 4.120 Stylopandalus richardi 28 65.76 0.167 12.171 Systellaspis debilis 28 362.29 0.151 54.920 Total 4.108 667.915 Total non-migrating biomass 423.421 Total migrating biomass 244.494 Acanthephyra prionota 4 212.83 0.012 2.553 Acanthephyra sp. 7 53.22 0.015 0.798 Bentheogennema sp. 4 164.97 0.012 1.979 Funchalia taaningi 3 401.31 0.010 4.013 Glyphus sp. 2 35.35 0.005 0.176 Heterocarpus ensifer parvispina 1 113.15 0.004 0.452 Meningodora marptocheles 1 283.94 0.004 1.135 Oplophoridae sp. larvae 3 15.48 0.010 0.154 Oplophorus gracilirostris 4 824.93 0.012 9.899 Parapasiphae sulcatifrons 1 476.44 0.004 1.905 Parasergestes vigilax 4 16.40 0.012 0.196 Penaeidae sp. 7 40.22 0.015 0.603 Petalidium sp. 3 31.08 0.010 0.311 Sergia bisulcatus 3 534.03 0.010 5.340 Sergia inequalis 4 272.13 0.012 3.265 Sergia spp. (<20 mm) 12 56.50 0.050 2.825 Sergia tenuiremis 4 475.53 0.012 5.706 Sicyonia sp. 1 148.50 0.004 0.594 Total (excluded) 0.213 41.911 Abundance and biomass values listed for each species are the mean of daytime and nighttime estimates. Species in bold accounted for < 6% of total abundance and biomass and were not included in the active carbon transport calculations. Abundance, biomass, and diel vertical migrations A highly significant linear relationship between dry weight (DW) and wet weight (WW) using all migrating decapod species collected at Kahe Station was observed (DW = 0.179 * WW; p < 0.0001 [t = 140.9, df = 479], R2 = 0.976). Equation (1) implied a dry mass loss of 26.4% when water content is 82.1%. Thus before preservation in formalin, the decapods would have had a dry weight 35.9% higher than measured after preservation, as 1/(1 − 0.264) = 1.359. After all dry weights were corrected for dry mass loss, length weight relationships for 22 pelagic decapod species were calculated (Table 1). Total decapod abundance and biomass were 4.3 ind. m−2 and 0.71 gDW m−2 and rare species in total accounted for ∼5% of both abundance and biomass (Table 2). Using Welch’s two-sample t-test, it was determined that there was no evidence of significant diel vertical migrations for three of the species, indicated that they resided in mesopelagic zone through the day (Table 3). For six species the entire population migrated upwards at night, and the remaining 13 species migrated partially with 50–92% of the population migrating upwards during the nighttime (Table 3). Among the 22 species that were included in the analysis, migrating biomass (migrators as well as the portion of partial migrators) was 0.24 gDW m−2, accounting for ∼37% of their total biomass (Table 2). Table 3. Summary of diel migratory data for all pelagic decapod species in this study. Species DWMD (m) NWMD (m) t df p M (%) NWMDm (m) Species for which tde entire population migrates Acanthephyra smithi 678 370 3.53 15.45 0.003 100 n/a Allosergestes sargassi 475 235 5.02 7.89 0.001 100 n/a Gennadas bouvieri 803 459 11.00 81.33 <0.001 100 n/a Gennadas incertus 898 248 12.61 8.55 <0.001 100 n/a Neosergestes orientalis 508 113 18.26 7.89 <0.001 100 n/a Parasergestes armatus 562 281 3.86 17.31 0.001 100 n/a Species for which part of the population migrates Allosergestes pectinatus 543 314 3.17 15.23 0.006 69.4 110 Deosergestes erectus 671 255 4.86 14.57 <0.001 92.2 167 Gennadas capensis 1103 541 6.69 39.68 <0.001 71.0 324 Gennadas clavicarpus 700 234 9.18 27.7 <0.001 78.7 114 Gennadas spp. 726 420 5.73 50.82 <0.001 49.6 107 Gennadas tinayrei 759 249 13.92 6.88 <0.001 80.4 121 Janicella spinicauda 483 186 24.88 9.00 <0.001 87.8 104 Neosergestes consobrinus 564 394 2.18 23.71 0.040 55.4 108 Sergestes atlanticus 592 194 3.95 3.99 0.017 88.2 116 Sergia bigemmeus 900 224 7.90 6.22 <0.001 89.7 149 Sergia gardineri 748 275 10.19 100.74 <0.001 76.4 113 Sergia scintillans 660 261 6.55 19.00 <0.001 76.7 105 Stylopandalus richardi 572 195 5.84 14.96 <0.001 88.1 116 Non-migratory species Acanthephyra curtirostris 974 923 0.50 13.13 0.627 0 n/a Notostomus gibbosus 904 955 0.52 18.61 0.606 0 n/a Systellaspis debilis 685 681 0.06 6.25 0.953 0 n/a Species DWMD (m) NWMD (m) t df p M (%) NWMDm (m) Species for which tde entire population migrates Acanthephyra smithi 678 370 3.53 15.45 0.003 100 n/a Allosergestes sargassi 475 235 5.02 7.89 0.001 100 n/a Gennadas bouvieri 803 459 11.00 81.33 <0.001 100 n/a Gennadas incertus 898 248 12.61 8.55 <0.001 100 n/a Neosergestes orientalis 508 113 18.26 7.89 <0.001 100 n/a Parasergestes armatus 562 281 3.86 17.31 0.001 100 n/a Species for which part of the population migrates Allosergestes pectinatus 543 314 3.17 15.23 0.006 69.4 110 Deosergestes erectus 671 255 4.86 14.57 <0.001 92.2 167 Gennadas capensis 1103 541 6.69 39.68 <0.001 71.0 324 Gennadas clavicarpus 700 234 9.18 27.7 <0.001 78.7 114 Gennadas spp. 726 420 5.73 50.82 <0.001 49.6 107 Gennadas tinayrei 759 249 13.92 6.88 <0.001 80.4 121 Janicella spinicauda 483 186 24.88 9.00 <0.001 87.8 104 Neosergestes consobrinus 564 394 2.18 23.71 0.040 55.4 108 Sergestes atlanticus 592 194 3.95 3.99 0.017 88.2 116 Sergia bigemmeus 900 224 7.90 6.22 <0.001 89.7 149 Sergia gardineri 748 275 10.19 100.74 <0.001 76.4 113 Sergia scintillans 660 261 6.55 19.00 <0.001 76.7 105 Stylopandalus richardi 572 195 5.84 14.96 <0.001 88.1 116 Non-migratory species Acanthephyra curtirostris 974 923 0.50 13.13 0.627 0 n/a Notostomus gibbosus 904 955 0.52 18.61 0.606 0 n/a Systellaspis debilis 685 681 0.06 6.25 0.953 0 n/a DWMD and NWMD: daytime and nighttime weighted mean depth for the entire population; M (%): proportion of the population migrating. For the partially migratory species, the nighttime WMD for only the portion that migrates is also provided (NWMDm). Results of Welch’s two-sample t-tests are provided for each species (t, df, p), with the test comparing daytime WMDs to night time WMDs for the whole populations. View Large Table 3. Summary of diel migratory data for all pelagic decapod species in this study. Species DWMD (m) NWMD (m) t df p M (%) NWMDm (m) Species for which tde entire population migrates Acanthephyra smithi 678 370 3.53 15.45 0.003 100 n/a Allosergestes sargassi 475 235 5.02 7.89 0.001 100 n/a Gennadas bouvieri 803 459 11.00 81.33 <0.001 100 n/a Gennadas incertus 898 248 12.61 8.55 <0.001 100 n/a Neosergestes orientalis 508 113 18.26 7.89 <0.001 100 n/a Parasergestes armatus 562 281 3.86 17.31 0.001 100 n/a Species for which part of the population migrates Allosergestes pectinatus 543 314 3.17 15.23 0.006 69.4 110 Deosergestes erectus 671 255 4.86 14.57 <0.001 92.2 167 Gennadas capensis 1103 541 6.69 39.68 <0.001 71.0 324 Gennadas clavicarpus 700 234 9.18 27.7 <0.001 78.7 114 Gennadas spp. 726 420 5.73 50.82 <0.001 49.6 107 Gennadas tinayrei 759 249 13.92 6.88 <0.001 80.4 121 Janicella spinicauda 483 186 24.88 9.00 <0.001 87.8 104 Neosergestes consobrinus 564 394 2.18 23.71 0.040 55.4 108 Sergestes atlanticus 592 194 3.95 3.99 0.017 88.2 116 Sergia bigemmeus 900 224 7.90 6.22 <0.001 89.7 149 Sergia gardineri 748 275 10.19 100.74 <0.001 76.4 113 Sergia scintillans 660 261 6.55 19.00 <0.001 76.7 105 Stylopandalus richardi 572 195 5.84 14.96 <0.001 88.1 116 Non-migratory species Acanthephyra curtirostris 974 923 0.50 13.13 0.627 0 n/a Notostomus gibbosus 904 955 0.52 18.61 0.606 0 n/a Systellaspis debilis 685 681 0.06 6.25 0.953 0 n/a Species DWMD (m) NWMD (m) t df p M (%) NWMDm (m) Species for which tde entire population migrates Acanthephyra smithi 678 370 3.53 15.45 0.003 100 n/a Allosergestes sargassi 475 235 5.02 7.89 0.001 100 n/a Gennadas bouvieri 803 459 11.00 81.33 <0.001 100 n/a Gennadas incertus 898 248 12.61 8.55 <0.001 100 n/a Neosergestes orientalis 508 113 18.26 7.89 <0.001 100 n/a Parasergestes armatus 562 281 3.86 17.31 0.001 100 n/a Species for which part of the population migrates Allosergestes pectinatus 543 314 3.17 15.23 0.006 69.4 110 Deosergestes erectus 671 255 4.86 14.57 <0.001 92.2 167 Gennadas capensis 1103 541 6.69 39.68 <0.001 71.0 324 Gennadas clavicarpus 700 234 9.18 27.7 <0.001 78.7 114 Gennadas spp. 726 420 5.73 50.82 <0.001 49.6 107 Gennadas tinayrei 759 249 13.92 6.88 <0.001 80.4 121 Janicella spinicauda 483 186 24.88 9.00 <0.001 87.8 104 Neosergestes consobrinus 564 394 2.18 23.71 0.040 55.4 108 Sergestes atlanticus 592 194 3.95 3.99 0.017 88.2 116 Sergia bigemmeus 900 224 7.90 6.22 <0.001 89.7 149 Sergia gardineri 748 275 10.19 100.74 <0.001 76.4 113 Sergia scintillans 660 261 6.55 19.00 <0.001 76.7 105 Stylopandalus richardi 572 195 5.84 14.96 <0.001 88.1 116 Non-migratory species Acanthephyra curtirostris 974 923 0.50 13.13 0.627 0 n/a Notostomus gibbosus 904 955 0.52 18.61 0.606 0 n/a Systellaspis debilis 685 681 0.06 6.25 0.953 0 n/a DWMD and NWMD: daytime and nighttime weighted mean depth for the entire population; M (%): proportion of the population migrating. For the partially migratory species, the nighttime WMD for only the portion that migrates is also provided (NWMDm). Results of Welch’s two-sample t-tests are provided for each species (t, df, p), with the test comparing daytime WMDs to night time WMDs for the whole populations. View Large Acanthephyra curtirostris, Notostomus gibbosus, and Systellaspis debilis belonged to the non-migratory group (Figure 3). All non-migratory species were relatively large (A. curtirostris being the smallest with an average mass of 220.4 mgDW ind.−1), and all belonged to the family Oplophoridae (Table 2). A. curtirostris and N. gibbosus resided at similar depths, with WMDs ranging from 904 to 974 m during both the day and nighttime, while S. debilis resided at shallower depths, with daytime and nighttime WMD being 685 m and 681 m, respectively (Figure 3, Table 3). Of the non-migratory decapods, A. curtisostris was the most abundant with a mean abundance of 0.256 ind. m−2 throughout the water column, while N. gibbosus made up the most biomass, mean biomass of 275.9 mgDW m−2 (Table 2). Figure 3. View largeDownload slide Abundance and biomass depth profiles of non-migrating decapods Acanthephyra curtirostris (a), Notostomus gibbosus (b), and Systellaspis debilis (c). For the abundance profile, daytime and nighttime WMDs are shown as solid lines, with dotted lines indicating standard errors. Figure 3. View largeDownload slide Abundance and biomass depth profiles of non-migrating decapods Acanthephyra curtirostris (a), Notostomus gibbosus (b), and Systellaspis debilis (c). For the abundance profile, daytime and nighttime WMDs are shown as solid lines, with dotted lines indicating standard errors. Acanthephyra smithi, Allosergestes sargassi, Gennadas bouvieri, Gennadas incertus, Neosergestes orientalis, and Parasergestes armatus belonged to fully migratory decapod group (Table 3). For all six species, the entire population migrated toward the surface at night. Unlike the non-migratory species, which all belonged to the family Oplophoridae, the fully migratory species were taxonomically diverse, with one species from the family Oplophoridae (A. smithi), two species from the family Benthesicymidae (G. bouvieri and G. incertus), and three species from the family Sergestidae (A. sargassi, N. orientalis, and P. armatus). Other than A. smithi, all were of a small to moderate size, ranging from 29 mgDW ind−1 for N. orientalis, to 87 mgDW ind−1 for G. bouvieri (Table 2). Among this group, G. bouvieri and G. incertus resided at the deepest depths during the day, at 803 m and 898 m WMD, respectively (Figure 4, Supplementary Figure S2). A. smithi resided at a shallower daytime depth of 678 m (Supplementary Figure S2), while A. sargassi, N. orientalis, and P. armatus resided at the shallowest daytime depths, at 475 m, 508 m and 562 m, respectively (Figure 4, Supplementary Figure S2). Of all six fully migratory species, only N. orientalis migrated into the euphotic zone (shallower than 173 m) during the night, with a nighttime WMD of ∼113 m (Figure 4). G. incertus, A. sargassi, and P. armatus migrated to just below the base of the euphotic zone, with daytime WMDs of 235 m, 248 m, and 281 m, respectively (Figure 4, Supplementary Figure S2). Of the fully migratory species, G. bouvieri was the most abundant (0.481 ind. m−2), while A. smithi had the highest biomass, 59.3 µgDW m−2, as well as the highest biomass per individual, 634.5 µgDW ind−1 (Table 2). Figure 4. View largeDownload slide Abundance and biomass depth profiles of fully migrating decapods Gennadas incertus (a), Neosergestes orientalis (b), and Parasergestes armatus (c). For the abundance profile, daytime and nighttime WMDs are shown as solid lines, with dotted lines indicating standard errors. Figure 4. View largeDownload slide Abundance and biomass depth profiles of fully migrating decapods Gennadas incertus (a), Neosergestes orientalis (b), and Parasergestes armatus (c). For the abundance profile, daytime and nighttime WMDs are shown as solid lines, with dotted lines indicating standard errors. Allosergestes pectinatus, Deosergestes erectus, Gennadas capensis, Gennadas clavicarpus, Gennadas spp., Gennadas tinayrei, Janicella spinicauda, Neosergestes consobrinus, Sergestes atlanticus, Sergia bigemmeus, Sergia gardineri, Sergia scintillans, and Stylopandalus richardi were all found to be migratory species (Table 3). All were partial migrants showing bimodal nighttime distributions (Figure 5, Supplementary Figures S3–S5). Of these, only G. capensis migrated to a nighttime depth below the base of the euphotic zone (WMD 324 m) (Figure 5). The remaining 12 species migrated to a WMD within the epipelagic (<173 m) zone (Figure 5, Supplementary Figures S3–S5). Figure 5. View largeDownload slide Abundance and biomass depth profiles of partially migrating decapods Allosergestes pectinatus (a), Deosergestes erectus (b), and Gennadas capensis (c). For the abundance profile, daytime, and nighttime WMDs are shown as solid lines, with dotted lines indicating standard errors, while the dashed line indicates the nighttime WMD for only the portion of the population that migrates to the surface. Figure 5. View largeDownload slide Abundance and biomass depth profiles of partially migrating decapods Allosergestes pectinatus (a), Deosergestes erectus (b), and Gennadas capensis (c). For the abundance profile, daytime, and nighttime WMDs are shown as solid lines, with dotted lines indicating standard errors, while the dashed line indicates the nighttime WMD for only the portion of the population that migrates to the surface. The other three members of the genus Gennadas, G. clavicarpus, G. tinayrei, and Gennadas spp. (damaged individuals) showed similar migratory patterns, migrating from daytime depths of 700–759 m to nighttime depths of 107–121 (Supplmentary Figure S3). Decapods of the now defunct genus Sergestes displayed similar migratory distributions. The genus Sergestes has been recently re-classified into six separate but closely linked genera: Allosergestes, Deosergestes, Eusergestes, Neosergestes, and Parasergestes (Judkins and Kensley, 2008), which appears to be reflected in their migratory distributions. A. pectinatus, N. consobrinus, and S. atlanticus were particularly similar in their vertical distributions: all migrated to 108–116 m WMD during the night and to depths of 543–592 m WMD during the day (Figure 5, Supplementary Figure S4). In contrast, D. erectus was found at deeper depths during both the day and night (671 m and 167 m, respectively) (Figure 5). It should be noted that D. erectus is a relatively large species (Table 2). Three members of the genus Sergia displayed differing migratory distributions. While all three were partial migrators, ascending to similar depths within the euphotic zone at night, S. bigemmeus resided at 900 m during the day, while S. gardineri and S. scintillans resided at 748 m and 660 m WMD, respectively (Supplementary Figure S5). J. spinicauda, of the family Ophlophoridae, and S. richardi, of the family Pandalidae, also displayed similar migratory distributions with daytime and nighttime WMDs of 483–572 m and 104–116 m, respectively (Supplementary Figures S4 and S5). Both species were of similar size (Table 2). S. gardineri was the most abundant in this group but A. pectinatus made up highest biomass (Table 2). There was a positive significant (R2 = 0.473, p = 0.00228, F = 13.46, df = 1 and 15) correlation between decapod mean dry weight and the WMD to which they migrate to during the daytime, for < 100 mg individuals. However, when species with a mean dry weight > 100 mg ind.−1 were included (D. erectus and A. smithi), the correlation was no longer significant (R2 = 0.0121, p = 0.654, F = 0.209, df = 1 and 17). Active carbon flux In total, decapods transported 0.692 mgC m−2 day−1 through active carbon flux, from a mean depth of 262 m at night to a mean depth of 711 m during the day (Table 4). Mortality flux accounted on average for ∼70% of total active flux. This was followed by the gut flux (∼18%) and respiration flux (∼9%) (Table 4). The excretion flux did not exceed 3% of total active flux. G. bouvieri and A. smithi were the greatest contributors to all four classes of active flux, followed by D. erectus, G. capensis, Gennadas spp., P. armatus, S. gardineri, and S. richardi (Table 4). Table 4. Active downward carbon flux components for all diel migratory decapod species, from a mean nighttime depth of 262 m to a mean daytime depth of 711 m. Species Respiratory flux (µgC.m-2.day-1) Excretory flux (µgC.m-2.d-1) Mortality flux (µgC.m-2.d-1) Gut flux (µgC.m-2.d-1) Acanthephyra smithi 3.11 1.01 76.44 30.91 Allosergestes pectinatus 1.38 0.45 4.30 0.64 Allosergestes sargassi 0.92 0.30 4.28 0.82 Deosergestes erectus 3.58 1.16 41.55 12.14 Gennadas bouvieri 11.0 3.57 88.57 21.78 Gennadas capensis 5.57 1.81 51.01 13.25 Gennadas clavicarpus 2.08 0.67 14.65 3.39 Gennadas incertus 1.42 0.46 8.80 1.93 Gennadas spp. 3.90 1.27 25.92 5.83 Gennadas tinayrei 0.80 0.26 5.21 1.15 Janicella spinicauda 1.84 0.60 9.80 1.53 Neosergestes consobrinus 0.92 0.30 2.82 0.41 Neosergestes orientalis 2.61 0.85 11.75 1.89 Parasergestes armatus 6.13 1.99 42.93 9.97 Sergestes atlanticus 1.06 0.34 4.45 0.79 Sergia bigemmeus 2.43 0.79 19.92 4.93 Sergia gardineri 7.89 2.6 41.03 5.84 Sergia scintillans 1.38 0.45 7.90 1.97 Stylopandalus richardi 3.30 1.07 23.75 6.40 Total flux for all species 61.31 19.90 485.16 125.56 Species Respiratory flux (µgC.m-2.day-1) Excretory flux (µgC.m-2.d-1) Mortality flux (µgC.m-2.d-1) Gut flux (µgC.m-2.d-1) Acanthephyra smithi 3.11 1.01 76.44 30.91 Allosergestes pectinatus 1.38 0.45 4.30 0.64 Allosergestes sargassi 0.92 0.30 4.28 0.82 Deosergestes erectus 3.58 1.16 41.55 12.14 Gennadas bouvieri 11.0 3.57 88.57 21.78 Gennadas capensis 5.57 1.81 51.01 13.25 Gennadas clavicarpus 2.08 0.67 14.65 3.39 Gennadas incertus 1.42 0.46 8.80 1.93 Gennadas spp. 3.90 1.27 25.92 5.83 Gennadas tinayrei 0.80 0.26 5.21 1.15 Janicella spinicauda 1.84 0.60 9.80 1.53 Neosergestes consobrinus 0.92 0.30 2.82 0.41 Neosergestes orientalis 2.61 0.85 11.75 1.89 Parasergestes armatus 6.13 1.99 42.93 9.97 Sergestes atlanticus 1.06 0.34 4.45 0.79 Sergia bigemmeus 2.43 0.79 19.92 4.93 Sergia gardineri 7.89 2.6 41.03 5.84 Sergia scintillans 1.38 0.45 7.90 1.97 Stylopandalus richardi 3.30 1.07 23.75 6.40 Total flux for all species 61.31 19.90 485.16 125.56 Total active carbon flux was 0.692 mgC m−2 day−1. Table 4. Active downward carbon flux components for all diel migratory decapod species, from a mean nighttime depth of 262 m to a mean daytime depth of 711 m. Species Respiratory flux (µgC.m-2.day-1) Excretory flux (µgC.m-2.d-1) Mortality flux (µgC.m-2.d-1) Gut flux (µgC.m-2.d-1) Acanthephyra smithi 3.11 1.01 76.44 30.91 Allosergestes pectinatus 1.38 0.45 4.30 0.64 Allosergestes sargassi 0.92 0.30 4.28 0.82 Deosergestes erectus 3.58 1.16 41.55 12.14 Gennadas bouvieri 11.0 3.57 88.57 21.78 Gennadas capensis 5.57 1.81 51.01 13.25 Gennadas clavicarpus 2.08 0.67 14.65 3.39 Gennadas incertus 1.42 0.46 8.80 1.93 Gennadas spp. 3.90 1.27 25.92 5.83 Gennadas tinayrei 0.80 0.26 5.21 1.15 Janicella spinicauda 1.84 0.60 9.80 1.53 Neosergestes consobrinus 0.92 0.30 2.82 0.41 Neosergestes orientalis 2.61 0.85 11.75 1.89 Parasergestes armatus 6.13 1.99 42.93 9.97 Sergestes atlanticus 1.06 0.34 4.45 0.79 Sergia bigemmeus 2.43 0.79 19.92 4.93 Sergia gardineri 7.89 2.6 41.03 5.84 Sergia scintillans 1.38 0.45 7.90 1.97 Stylopandalus richardi 3.30 1.07 23.75 6.40 Total flux for all species 61.31 19.90 485.16 125.56 Species Respiratory flux (µgC.m-2.day-1) Excretory flux (µgC.m-2.d-1) Mortality flux (µgC.m-2.d-1) Gut flux (µgC.m-2.d-1) Acanthephyra smithi 3.11 1.01 76.44 30.91 Allosergestes pectinatus 1.38 0.45 4.30 0.64 Allosergestes sargassi 0.92 0.30 4.28 0.82 Deosergestes erectus 3.58 1.16 41.55 12.14 Gennadas bouvieri 11.0 3.57 88.57 21.78 Gennadas capensis 5.57 1.81 51.01 13.25 Gennadas clavicarpus 2.08 0.67 14.65 3.39 Gennadas incertus 1.42 0.46 8.80 1.93 Gennadas spp. 3.90 1.27 25.92 5.83 Gennadas tinayrei 0.80 0.26 5.21 1.15 Janicella spinicauda 1.84 0.60 9.80 1.53 Neosergestes consobrinus 0.92 0.30 2.82 0.41 Neosergestes orientalis 2.61 0.85 11.75 1.89 Parasergestes armatus 6.13 1.99 42.93 9.97 Sergestes atlanticus 1.06 0.34 4.45 0.79 Sergia bigemmeus 2.43 0.79 19.92 4.93 Sergia gardineri 7.89 2.6 41.03 5.84 Sergia scintillans 1.38 0.45 7.90 1.97 Stylopandalus richardi 3.30 1.07 23.75 6.40 Total flux for all species 61.31 19.90 485.16 125.56 Total active carbon flux was 0.692 mgC m−2 day−1. Using Equation (10), the passive particulate carbon flux at the mean daytime depth (711 m), mean nighttime depth (262 m), and at the base of the euphotic zone (173 m, Karl and Lukas, 1996), was calculated to be 8.05, 18.2, and 25.54 mgC m−2 day−1, respectively. Therefore, total decapod-mediated active carbon flux were equal to 8.6% of the passive flux to the mean daytime depth (711 m), 3.8% of the passive flux to the mean nighttime depth (262 m), and 2.7% of passive flux at the base of the euphotic zone. Discussion Density and vertical migrations Decapod community density estimated by this study (4.3 ind. m−2 in the upper 1000 m) in the NPSG near Hawaii was close to the middle range of similar estimates 0.9–10 ind. m−2 in various parts of the world ocean (see below). It was 2–4-fold higher than previous estimates (0.1–1.8 ind. m−2) near Hawaii and in the Gulf of Mexico (Maynard et al., 1975; Walters, 1976; Flock and Hopkins, 1992), comparable to abundances (3.3–3.9 ind. m−2) recorded for the Arabian Sea and near Azores (Domanski, 1986; Mincks et al., 2000; Ariza et al., 2015), and about twofold lower than densities (∼10 ind. m−2) found in the Benguela upwelling system (Schukat et al., 2013). As in previous studies, the majority of decapod species undertook extensive vertical migrations and generally showed very similar migration patterns to those reported near Hawaii and elsewhere. An exception was Systellaspis debilis, which was classified as a non-migrating species in our study but a strong vertical migrator in other studies (e.g. Maynard et al., 1975; Hopkins et al., 1989,, 1994; Schukat et al., 2013). This suggests that, possibly due to low sampling effort, we may have underestimated the migrating portion of the decapod community. Nevertheless, the proportion of migrating biomass based on individual species (50–92% of the total population in partially migrating species) compared reasonably well with overall micronekton migrating biomass (range 20–90% of total) estimated using acoustics in major basins of the world ocean (Klevjer et al., 2016). Comparison to previous active flux estimates It is difficult to reconcile decapod active flux with published values that are generally estimates for total zooplankton. Three available respiratory flux assessments for this group were 2 to >40-fold higher, e.g. ∼0.1 (this study) vs. <0.5–4.4 mgC m−2 day−1 (Hidaka et al., 2001; Schukat et al., 2013; Ariza et al., 2015), than estimates in our study. These discrepancies are much higher than differences in the decapod total density in various parts of the world ocean (see above) and can only be explained by the inconsistencies in the respiratory flux estimates adopted in different studies. The study that attempted measuring excretion, mortality, and gut flux yielded active carbon transport of 1–6 mgC m−2 day−1 (Angel and Pugh, 2000), which was 1.5–9.5-fold higher than our estimate and that could be explained by the differences in the decapod standing stock in the regions investigated. Our decapod respiratory flux estimates [0.06 mgC m−2 day−1 or 0.2, 0.3, and 0.7% of total passive flux at 173, 262, and 700 m depths, respectively (Supplementary Table S1)] are on the low end of similar assessments in the North Pacific and near the Azores (0.5–1.2%, Hidaka et al., 2001; Ariza et al., 2015). However, they are significantly lower that similar estimates (26–36%) in the Benguela upwelling system (Schukat et al., 2013). Previous studies reported that decapod biomass makes up 3.2–32.1% of the total migratory biomass (corrected for a sampling efficiency of 14%; Hidaka et al., 2001; Pakhomov and Yamamura, 2010; Ariza et al., 2015). Within our study region, Pakhomov and Yamamura (2010) collected micronekton using three different sampling gears with variable mouth and mesh sizes as part of the Micronekton Intercalibration Experiment (MIE-1). On average, they found that decapod biomass made up ∼12 ± 9% of the total migratory biomass. The Isaacs-Kidd Midwater Trawl (IKMT) used in MIE-1 had the same mesh size as the MOCNESS-10 used in this study. IKMT migratory biomass during MIE-1 was dominated by cephalopods (32%), myctophids (14%), euphausiids (10%), stomatopods (9%), and decapods (8%) (Kwong et al., 2018). Based on the above values, during our study we estimated total active carbon flux to ∼700 m mediated by micronekton to be within 5.1–7.4 mgC m−2 day−1. This would be comparable to 20–29%, 28–41%, and 63–92% of passive flux at 173, 262, and 711 m, respectively. While it is a coarse estimate (simple scaling-up from the decapod contribution to biomass assuming that other groups migrate/behave similarly), it provides a starting point for comparison with other assessments published in the literature (Supplementary Table S1). Published active and passive carbon flux estimates at the base of the euphotic zone, plotted ignoring the enormous diversity of assessment methods, geographical and taxonomic variability, show that there is a general positive linear relationship between both flux types, and between active transport and its proportion of passive transport (Figure 6a and b). Steinberg and Landry (2017) found a similar pattern (likely compiling data from similar sources) but also showed that active carbon transport is positively related to migratory biomass. Finally, within the literature the active flux proportion of total carbon flux (active and passive) appeared to be stable, averaging ∼22%, with most estimates falling in the range of 15–40% (Figure 6c). It is notable that the total active flux estimated in this study fell close to the average literature values (Supplementary Table S1, Ariza et al., 2015) as well as to modelling approximations of respiratory fluxes assessed by Bianchi et al. (2013a). Nevertheless, it should be acknowledged that most active flux measurements were based on respiratory and excretion fluxes alone, and very few accounted for mortality or gut flux (Supplementary Table S1). Assuming a dominant contribution of both mortality and gut flux, our estimates would have fallen into the lower end (likely < 10%) of other estimates (Figure 6c). Figure 6. View largeDownload slide Importance of active carbon flux by zooplankton and micronekton at the base of mixed layer from published sources: (a) active vs. passive carbon flux; (b), active carbon flux vs. percentage active flux of passive flux; and (c) active carbon flux contribution to total flux (combined passive and active flux). Data are combined from Table 3 in Ariza et al. (2015) and Supplementary Table S1 (this study). Solid circle belongs to the mean estimate obtained in this study. Figure 6. View largeDownload slide Importance of active carbon flux by zooplankton and micronekton at the base of mixed layer from published sources: (a) active vs. passive carbon flux; (b), active carbon flux vs. percentage active flux of passive flux; and (c) active carbon flux contribution to total flux (combined passive and active flux). Data are combined from Table 3 in Ariza et al. (2015) and Supplementary Table S1 (this study). Solid circle belongs to the mean estimate obtained in this study. In the past, there have been large (orders of magnitude) discrepancies in studies reconciling carbon demands of the mesopelagic fauna and its availability via passive carbon flux pointing to high uncertainty in both estimates (e.g. Boyd et al., 1999; Steinberg et al., 2008; Burd et al., 2010). These could be improved by removing from the mesopelagic fauna active vertical migrators (Giering et al., 2014). Currently, it is believed that besides passive carbon flux, mesopelagic plankton demands could be met through suspended (0.7–52 µm) particles and active vertical carbon flux (e.g. Hannides et al., 2009,, 2013,, 2015; Steinberg et al., 2000,, 2002; Choy et al., 2015; Gloeckler et al., 2018). Despite high uncertainty, active flux mediated by zooplankton is an important contributor to the total carbon flux (Steinberg and Landry, 2017). Yet, most estimates were made using migrating tropical and temperate mesozooplankton and generally assessed active transport out of the upper mixed (<200 m) layer of water (Supplementary Table S1, Steinberg and Landry, 2017). However, (a) mesozooplankton generally do not migrate below 300 m, (b) they have short gut passage time, and (c) their mortality was not taken into account. Because in the majority of tropical regions production and consumption cycles are tightly aligned, the contribution of mesozooplankton active transport to total flux is substantial (Valencia et al., 2018). Studies estimating similar contribution of large zooplankton and micronekton migrating below 300–500 m are rare and highly uncertain (Hidaka et al., 2001; Davison et al., 2013; Schukat et al., 2013). Micronekton indeed migrate much deeper than mesozooplankton, where passive flux is insignificant (Karl et al., 1996). The importance of active transport is therefore expected to be greater at mesopelagic depths (400–800 m) across most of the world ocean (Bianchi et al., 2013b). Few previous studies that attempted to assess active carbon flux to mesopelagic layers (Supplementary Table S1) have shown that it can account for >70% of passive flux in the deeper part of the low mesopelagic zone (Steinberg et al., 2000; Davison et al., 2013; Schukat et al., 2013; Hudson et al., 2014). Those estimates are in line with modelling assessments made by Bianchi et al. (2013a, b), which only assessed respiratory active flux. The results of our study show that active carbon transport is not only important for nutrient regeneration and oxygen consumption in the mesopelagic zone but also provides critical food supply (reflected in high mortality contribution to total active flux) to the resident mesopelagic community. Decapods are an important part of active carbon transport, as they appear to migrate deeper than the majority of other macroplankton and micronekton (e.g. myctophids) (Davison et al., 2013; Bianchi et al., 2013a). Hence, when calculating active flux to mesopelagic layer it is critical to sample throughout the depth range of the organisms’ migrations, to calculate the depth to which carbon is transported. It is worth noting that this study provides quantitative evidence for the Vinogradov’s ladder of vertical migrations (Vinogradov, 1962). This theory proposed that vertical migrations could be an important source of energy and materials to the mesopelagic, bathypelagic, and abyssopelagic zone residents (Allison et al., 1996). While migrations from the bathypelagic to the surface are very rare, some deep living organisms perform upward migrations to mesopelagic depths (Vinogradov, 1962). The quantification of fluxes due to this vertical transfer ladder has proven difficult due to challenging bathypelagic and abyssopelagic sampling, and low fauna abundance (Haedrich and Henderson, 1974; Yamamura et al., 1993; Allison et al., 1996). It appears however, that active carbon flux may be at least as important in the mesopelagic realm as sinking phytoplankton and other detritus originating from the upper water column (Allison et al., 1996). Samples for this study were collected in the central NPSG, mostly at station ALOHA and secondly at station Kahe, two oceanographic locations that have been regularly monitored by the Hawaii Ocean Time-Series (HOT) program since October 1988 (Karl and Lukas, 1996). One of the central goals of the HOT program is to understand local biological processes and particulate matter fluxes (Karl and Lukas, 1996). However, up until now the contribution of diel migratory micronekton flux to the total downward carbon flux in the area has not been assessed. The central NPSG is a study site with very low passive gravitational flux typical of the oligotrophic oceanic realm and indicative of a weak traditional biological pump (Francois et al., 2002). It is thus not surprising that active flux in this region is particularly significant contributor to the overall local downward carbon flux. An interesting finding of this study points to a lower than expected variability of active flux proportion of total (active and passive) flux at the base of the euphotic zone over a wide range of plankton communities globally (Figure 6c). Such measurements should however be more methodologically consistent and extended to temperate as well as Polar Regions of the world ocean. Potential sources of error A wide variety of models and assumptions were used in this study, which could lead to potential sources of error and uncertainty. The largest potential uncertainly likely relates to the net catch efficiency. It is generally recognized that acoustic biomass estimates may exceed net estimates by an order of magnitude and only fast towing large nets could provide biomass values a few folds lower than acoustics (Pakhomov and Yamamura, 2010). Catch efficiency assumptions thus remain a significant source of uncertainty in this and other studies. While net catch efficiency represents a potentially significant source of error, the uncertainty due to the models used to estimate respiratory flux, excretory flux, mortality flux, and gut flux should also be discussed. Respiratory and excretion rates were estimated as a function of body mass and temperature, based on data from a wide variety of zooplankton and micronekton, and could explain >90% of their variation (Ikeda, 1985; Steinberg et al., 2000). In the gut flux estimates, the main source of error would relate to the locally obtained relationship that predicted food ball dry weight from dry weight of the organism and stomach fullness. However, while highly variable, ∼82% of the variance in the data was explained by the model. On the one hand, the gut flux estimate could be an underestimate, because only prey mass in the stomach was taken into account, ignoring material in the intestines. On the other hand, it could be an overestimate if deep water feeding occurred (Podeswa and Pakhomov, 2015). Lastly, mortality flux estimates were based on a size-dependent mortality rate model developed for particles in the size range of fish eggs (∼0.1 mg DW) to adult fish (∼1000 g DW) (Peterson and Wroblewski, 1984). There is a considerable error associated with predictions made by this model for the smallest and largest organisms. However, for the mass range of the individuals in this study (roughly 0.01–0.6 gDW ind.−1) the model predictions tend to be closest to empirical values. Furthermore, sensitivity analysis conducted by Zhang and Dam (1997) showed that mortality flux is highly sensitive to the residence time of migrators below of the euphotic zone and may vary by a factor of two. Lastly, low number of samples (N = 1) in the top 500 m during daytime should also be considered as a shortcoming of the current study. However, general micronekton absence in the upper layers during daytime is well documented (e.g. Kwong et al., 2018). Concluding remarks The estimated active flux due to the diel vertical migrations of decapods was low compared with most previous estimates for zooplankton and micronekton communities as a whole, but comparable to previous estimates of decapod specific active flux. The relatively high abundance of migratory decapods, combined with the low passive flux in the central NPSG, suggests that this estimate for active flux due to migrant decapods may be more important relative to local passive flux in this region than in other areas of the world’s oceans. Despite differences in community structure between the study area and other locations, this study emphasizes the importance of local micronekton migrations to vertical carbon flux on a global scale. The deep daytime migrations as well as the relatively high prevalence of species that did not migrate all the way up to the euphotic zone, highlights the importance of sampling of the entire water column, not just migrations into and out of the euphotic zone, as has been the case in many previous studies. Overall, the active flux estimates produced by this study show that total micronekton, including migrant decapods, are a sizable contributor to downward carbon flux in the open ocean. Currently, active flux is still poorly defined and requires new and standardized approaches (e.g. biomass spectra combined with acoustics and/or modelling; Kwong and Pakhomov, 2017) to narrow down uncertainties in empirical estimates and reliably quantify regional and global micronekton-mediated carbon contributions to the mesopelagic realm. Acknowledgements This research was supported by the NSERC DG to PEA and the University of British Columbia. We would like to thank research team and crew who helped to collect samples during August 2011 research voyage aboard the R/V Kilo Moana to the Station ALOHA. We would particularly like to thank Brian Popp and Jeff Drazen for inviting YP and EAP to participate in this voyage. We are grateful to Svenja Kruse, Roger Francois, Kang Wang, Brock Ramshaw, and Désirée Tommasi for their helpful advice, feedback, and overall great assistance throughout this project. References Allison E. H. , Irvine K. , Thompson A. B. 1996 . Lake flies and the deep-water demersal fish community of Lake Malawi . Journal of Fish Biology , 48 : 1006 – 1010 . Google Scholar Crossref Search ADS Al-Mutairi H. , Landry M. R. 2001 . Active export of carbon and nitrogen at station ALOHA by diel migrant zooplankton . Deep-Sea Research Part II: Topical Studies in Oceanography , 48 : 2083 – 2103 . Google Scholar Crossref Search ADS Andersen V. , Gubanova A. , Nival P. , Ruellet T. 2001 . Zooplankton community during the transition from spring bloom to oligotrophy in the open NW Mediterranean and effects of wind events. 2. Vertical distributions and migrations . Journal of Plankton Research , 23 : 243 – 261 . Google Scholar Crossref Search ADS Angel M. V. , Pugh P. R. 2000 . Quantification of diel vertical migration by micronektonic taxa in the northeast Atlantic . Hydrobiologia , 440 : 161 – 179 . Google Scholar Crossref Search ADS Ariza A. , Garijo J. C. , Landeira J. M. , Bordes F. , Hernández-León S. 2015 . Migrant biomass and respiratory carbon flux by zooplankton and micronekton in the subtropical northeast Atlantic Ocean (Canary Islands) . Progress in Oceanography , 134 : 330 – 342 . Google Scholar Crossref Search ADS Aron W. , Collard S. 1969 . A study of the influence of net speed on catch . Limnology and Oceanography , 14 : 242 – 249 . Google Scholar Crossref Search ADS Bianchi D. , Galbraith E. D. , Carozza D. A. , Mislan M. A. S. , Stock C. 2013 . Intensification of open-ocean oxygen depletion by vertically migrating animals . Nature Geoscience , 6 : 545 – 548 . Google Scholar Crossref Search ADS Bianchi D. , Stock C. , Galbraith E. D. , Sarmiento L. 2013 . Diel vertical migration: ecological controls and impacts on biological pump in a one-dimensional ocean model . Global Biogeochemical Cycles , 27 : 478 – 491 . Google Scholar Crossref Search ADS Boyd P. W. , Sherry N. D. , Berges J. A. , Bishop J. K. B. , Calvert S. E. , Charette M. A. , Giovannoni S. J. , et al. 1999 . Transformations of biogenic particulates from the pelagic to the deep ocean realm . Deep Sea Research Part II: Topical Studies in Oceanography , 46 : 2761 – 2792 . Google Scholar Crossref Search ADS Brodeur R. , Yamamura O. 2005 . Micronekton of the North Pacific. PICES Scientific Report No. 30. Brodeur R. D. , Seki M. P. , Pakhomov E. A. , Suntsov A. V. 2005 . Micronekton - what are they and why are they important? PICES Press , 14 : 7 – 11 . Burd A. B. , Hansell D. A. , Steinberg D. K. , Anderson T. R. , Arístegui J. , Baltar F. , Beaupré S. R. , et al. 2010 . Assessing the apparent imbalance between geochemical and biochemical indicators of meso- and bathypelagic biological activity: what the @$♯! is wrong with present calculations of carbon budgets? . Deep Sea Research Part II: Topical Studies in Oceanography , 57 : 1557 – 1571 . Google Scholar Crossref Search ADS Choy C. A. , Popp B. N. , Hannides C. S. S. , Drazen J. S. 2015 . Trophic structure and food resources of epipelagic and mesopelagic fishes in the North Pacific Subtropical Gyre ecosystem inferred from nitrogen isotopic compositions . Limnology and Oceanography , 60 : 1156 – 1171 . Google Scholar Crossref Search ADS Clarke T. A. 1980 . Diets of fourteen species of vertically migrating mesopelagic fishes in Hawaiian waters . Fishery Bulletin , 78 : 619 – 640 . Dam H. G. , Roman M. R. , Youngbluth M. J. 1995 . Downward export of respiratory carbon and dissolved inorganic nitrogen by diel-migrant mesozooplankton at the JGOFS Bermuda time-series station . Deep-Sea Research Part I: Oceanographic Research Papers , 42 : 1187 – 1197 . Google Scholar Crossref Search ADS Davison P. C. , Checkley D. M. , Koslow J. A. , Barlow J. 2013 . Carbon export mediated by mesopelagic fishes in the northeast Pacific Ocean . Progress in Oceanography , 116 : 14 – 30 . Google Scholar Crossref Search ADS Domanski P. A. 1986 . The Azores Front: A zoogeographical boundary ? UNESCO Technical Papers in Marine Science , 49 : 73 – 83 . Flock M. E. , Hopkins T. L. 1992 . Species composition, vertical distribution, and food habits of the Sergestid shrimp assemblage in the Eastern Gulf of Mexico . Journal of Crustacean Biology , 12 : 210 – 223 . Google Scholar Crossref Search ADS Francois R. , Honjo S. , Krishfield R. , Manganini S. 2002 . Factors controlling the flux of organic carbon to the bathypelagic zone of the ocean . Global Biogeochemical Cycles , 16 : 34-1 – 20 . Google Scholar Crossref Search ADS Giering S. L. C. , Sanders R. , Lampitt R. S. , Anderson T. A. , Tamburini C. , Boutrif M. , Zubkov M. V. , et al. 2014 . Reconciliation of the carbon budget in the ocean’s twilight zone . Nature , 507 : 480 – 483 . Google Scholar Crossref Search ADS PubMed Gloeckler K. , Choy C. A. , Hannides C. C. S. , Close H. G. , Goetze E. , Popp B. N. , Drazen J. C. 2018 . Stable isotope analysis of micronekton around Hawaii reveals suspended particles are an important nutritional source in the lower mesopelagic and upper bathypelagic zones . Limnology and Oceanography , 63 : 1168 – 1180 . Google Scholar Crossref Search ADS Gnaiger E. 1983 . Calculation of energetic and biochemical equivalents of respiratory oxygen consumption. In Polarorgraphic Oxygen Sensors , pp. 337 – 345 . Ed. by Gnaiger E. , Forstner H. . Springer , Berlin, Heidelberg . 370 pp. Haedrich R. L. , Henderson N. R. 1974 . Pelagic food of Corypahenoides armatus, a deep benthic rattail . Deep-Sea Research Part I: Oceanographic Research Papers , 21 : 739 – 744 . Hannides C. C. S. , Drazen J. C. , Popp B. N. 2015 . Mesozooplankton zooplankton metabolic demand in the North Pacific Subtropical Gyre . Limnology and Oceanography , 60 : 419 – 428 . Google Scholar Crossref Search ADS Hannides C. C. S. , Landry M. R. , Benitez-Nelson C. R. , Styles R. M. , Montoya J. P. , Karl D. M. 2009 . Export stoichiometry and migrant-mediated flux of phosphorus in the North Pacific Subtropical Gyre . Deep Sea Research Part I: Oceanographic Research Papers , 56 : 73 – 88 . Google Scholar Crossref Search ADS Hannides C. C. S. , Popp B. N. , Choy C. A. , Drazen J. C. 2013 . Midwater zooplankton and suspended particle dynamics in the North Pacific Subtropical Gyre: a stable isotope perspective . Limnology and Oceanography , 58 : 1931 – 1946 . Google Scholar Crossref Search ADS Hansen A. N. , Visser A. W. 2016 . Carbon export by vertically migrating zooplankton: an optimal behavior model . Limnology and Oceanography , 61 : 701 – 710 . Google Scholar Crossref Search ADS Hidaka K. , Kawaguchi K. , Murakami M. , Takahashi M. 2001 . Downward transport of organic carbon by diel migratory micronekton in the western equatorial pacific: its quantitative and qualitative importance . Deep-Sea Research Part I: Oceanographic Research Papers , 48 : 1923 – 1939 . Google Scholar Crossref Search ADS Hopkins T. L. , Baird R. C. , 1977 . Aspects of the feeding ecology of oceanic midwater fishes, In Oceanic Sound Scattering Prediction , pp. 325 – 360 . Ed. by Andersen N. R. , Zahuranec B. J. . Plenum Press , New York . Hopkins T. L. , Flock M. E. , Gartner J. V. , Torres J. J. 1994 . Structure and trophic ecology of a low latitude midwater decapod and mysid assemblage . Marine Ecology Progress Series , 109 : 143 – 156 . Google Scholar Crossref Search ADS Hopkins T. L. , Gartner J. V. , Flock M. E. 1989 . The Caridean shrimp (Decapoda: Natantia) assemblage in the mesopelagic zone of the eastern Gulf of Mexico . Bulletin of Marine Science , 45 : 1 – 14 . Hopkins T. L. , Sutton T. T. 1998 . Midwater fishes and shrimps as competitors and resource partitioning in low latitude oligotrophic ecosystems . Marine Ecology Progress Series , 164 : 37 – 45 . Google Scholar Crossref Search ADS Hudson J. M. , Steinberg D. K. , Sutton T. T. , Graves J. E. , Latour R. J. 2014 . Myctophid feeding ecology and carbon transport along the northern Mid-Atlantic Ridge . Deep-Sea Research Part I: Oceanographic Research Papers , 93 : 104 – 116 . Google Scholar Crossref Search ADS Ikeda T. 1985 . Metabolic rates of epipelagic marine zooplankton as a function of body mass and temperature . Marine Biology , 85 : 1 – 11 . Google Scholar Crossref Search ADS Itaya K. , Fujimori Y. , Shimizu S. , Komatsu T. , Miura T. 2007 . Effect of towing speed and net mouth size on catch efficiency in framed midwater trawls . Fisheries Science , 73 : 1007 – 1016 . Google Scholar Crossref Search ADS Itaya K. , Fujimori Y. , Shiode D. , Aoki I. , Yonezawa T. , Shimizu S. , Miura T. 2001 . Sampling performance and operational quality of a frame trawl used to catch juvenile fish . Fisheries Science , 67 : 436 – 443 . Google Scholar Crossref Search ADS Judkins D. C. , Kensley B. 2008 . New genera in the family Sergestidae (Crustacea: Decapoda: Penaeidea) . Proceedings of the Biological Society of Washington , 121 : 72 – 84 . Google Scholar Crossref Search ADS Karl D. M. 1999 . A sea of change: biogeochemical variability in the North Pacific Subtropical Gyre . Ecosystems , 2 : 181 – 214 . Google Scholar Crossref Search ADS Karl D. M. , Christian J. R. , Dore J. E. , Hebel D. V. , Letelier R. M. , Tupas L. M. , Winn C. D. 1996 . Seasonal and interannual variability in primary production and particle flux at Station ALOHA . Deep-Sea Research Part II: Topical Studies in Oceanography , 43 : 539 – 568 . Google Scholar Crossref Search ADS Karl D. M. , Lukas R. 1996 . The Hawaii Ocean Time-series (HOT) program: background, rationale and field implementation . Deep-Sea Research Part II: Topical Studies in Oceanography , 43 : 129 – 156 . Google Scholar Crossref Search ADS Klevjer T. A. , Irigoien X. , Rostad A. , Fraile-Nuez E. , Benitez-Barrios V. M. , Kaartvedt S. 2016 . Large scale patterns in vertical distribution and behavior of mesopelagic scattering layers . Scientific Reports , 6 : 19873. Google Scholar Crossref Search ADS PubMed Kwong L. E. , Pakhomov E. A. 2017 . Assessment of active vertical carbon transport: new methodology. Uchenye Zapiski Kazanskogo Universiteta . Seriya Estestvennye Nauki , 159 : 492 – 509 . Kwong L. E. , Pakhomov E. A. , Suntsov A. V. , Seki M. P. , Brodeur R. D. , Pakhomova L. G. , Domokos R. 2018 . An intercomparison of the taxonomic and size composition of tropical macrozooplankton and micronekton collected using three sampling gears . Deep-Sea Research Part I: Oceanographic Research Papers , 135 : 34 – 45 . Google Scholar Crossref Search ADS Le Borgne R. , Rodier M. 1997 . Net zooplankton and the biological pump: a comparison between the oligotrophic and mesotrophic equatorial Pacific . Deep-Sea Research Part II: Topical Studies in Oceanography , 44 : 2003 – 2023 . Google Scholar Crossref Search ADS Le Quéré C. , Andrew R. M. , Friedlingstein P. , Sitch S. and others. 2018 . Global carbon budget 2017 . Earth System Science Data , 10 : 1551–1448. Lampert W. 1989 . The adaptive significance of diel vertical migration of zooplankton . Functional Ecology , 3 : 21 – 27 . Google Scholar Crossref Search ADS Longhurst A. R. 1991 . Role of the marine biosphere in the global carbon cycle . Limnology and Oceanography , 36 : 1507 – 1526 . Google Scholar Crossref Search ADS Longhurst A. , Williams R. 1992 . Carbon flux by seasonal vertical migrant copepods is a small number . Journal of Plankton Research , 14 : 1495 – 1509 . Google Scholar Crossref Search ADS Longhurst A. R. , Bedo A. W. , Harrison W. G. , Head E. J. , Sameoto D. 1990 . Vertical flux of respiratory carbon by oceanic diel migrant biota . Deep-Sea Research , 37 : 685 – 694 . Google Scholar Crossref Search ADS Maynard S. D. , Riggs F. V. , Walters J. F. 1975 . Mesopelagic micronekton in Hawaiian waters: faunal composition, standing stock, and diel vertical migration . Fishery Bulletin , 73 : 726 – 736 . Mincks S. L. , Bollens S. M. , Madin L. P. , Horgan E. , Butler M. , Kremer P. M. , Craddock J. E. 2000 . Distribution, abundance, and feeding ecology of decapods in the Arabian Sea, with implications for vertical flux . Deep Sea Research Part II: Topical Studies in Oceanography , 47 : 1475 – 1516 . Google Scholar Crossref Search ADS Misund O. , Luyeye N. , Coetzee J. , Boyer D. 1999 . Trawl sampling of small pelagic fish off Angola: effects of avoidance, towing speed, tow duration, and time of day . ICES Journal of Marine Science , 56 : 275 – 283 . Google Scholar Crossref Search ADS Pakhomov E. A. 2003 . Correction of zooplankton and benthos biomass under-estimations from formaldehyde-preserved samples . Archive of Fishery and Marine Research , 50 : 141 – 148 . Pakhomov E. , Yamamura O. 2010 . Report of the advisory panel on micronekton sampling inter-calibration experiment . PICES Scientific Report 38 , 1 – 108 . Peterson I. , Wroblewski J. S. 1984 . Mortality rate of fishes in the pelagic ecosystem . Canadian Journal of Fisheries and Aquatic Sciences , 41 : 1117 – 1120 . Google Scholar Crossref Search ADS Podeswa Y. , Pakhomov E. A. 2015 . Feeding ecology of pelagic decapods in the North Pacific Subtropical Gyre: implications for active carbon transport . Arthropoda Selecta , 24 : 317 – 334 . Sabine C. L. , Feely R. A. , Gruber N. , Key R. M. , Lee K. , Bullister J. L. , Wanninkhof R. , et al. 2004 . The oceanic sink for anthropogenic CO2 . Science , 305 : 367 – 371 . Google Scholar Crossref Search ADS PubMed Sameoto D. D. , Cochrane N. , Herman A. 1993 . Convergence of acoustic, optical, and net-catch estimates of euphausiid abundance: use of artificial light to reduce net avoidance . Canadian Journal of Fisheries and Aquatic Sciences , 50 : 334 – 336 . Google Scholar Crossref Search ADS Sawilowsky S. S. 2002 . Fermat, Schubert, Einstein, and Behrens-Fisher: the probable difference between two means when variances are not equal . Journal of Modern Applied Statistical Methods , 1 : 461 – 472 . Google Scholar Crossref Search ADS Schukat A. , Bode M. , Auel H. , Carballo R. , Martin B. , Koppelmann R. , Hagen W. 2013 . Pelagic decapods in the northern Benguela upwelling system: distribution, ecophysiology and contribution to active carbon flux . Deep-Sea Research Part I: Oceanographic Research Papers , 75 : 146 – 156 . Google Scholar Crossref Search ADS Steinberg D. K. , Carlson C. A. , Bates N. R. , Goldthwait S. A. , Madin L. P. , Michaels A. F. 2000 . Zooplankton vertical migration and the active transport of dissolved organic and inorganic carbon in the Sargasso Sea . Deep-Sea Research Part I: Oceanographic Research Papers , 47 : 137 – 158 . Google Scholar Crossref Search ADS Steinberg D. K. , Cope J. S. , Wilson S. E. , Kobari T. 2008 . A comparison of mesopelagic mesozooplankton community structure in the subtropical and subarctic North Pacific Ocean . Deep-Sea Research Part II: Topical Studies in Oceanography , 55 : 1615 – 1635 . Google Scholar Crossref Search ADS Steinberg D. K. , Goldthwait S. A. , Hansell D. A. 2002 . Zooplankton vertical migration and the active transport of dissolved organic and inorganic nitrogen in the Sargasso Sea . Deep-Sea Research Part I: Oceanographic Research Papers , 49 : 1445 – 1461 . Google Scholar Crossref Search ADS Steinberg D. K. , Landry M. R. 2017 . Zooplankton and ocean carbon cycle . Annual Review of Marine Science , 9 : 413 – 444 . Google Scholar Crossref Search ADS PubMed Valencia B. , Decima M. , Landry M. R. 2018 . Environmental effects on mesozooplankton size structure and export flux at Station ALOHA, North Pacific . Global Biochemical Cycles , 32 : 289 – 305 . Google Scholar Crossref Search ADS Vinogradov M. E. 1962 . Feeding of the deep-sea zooplankton. Rapports et Procées-verbaux . Rèunions du Conseil International pour l’Exploration de la Mer , 53 : 114 – 120 . Walters J. F. 1976 . Ecology of Hawaiian sergestid shrimps (Penaeidea: Sergestidae) . Fishery Bulletin , 74 : 799 – 836 . Wiebe P. H. , Ashjian C. J. , Gallager S. M. , Davis C. S. , Lawson G. L. , Copley N. J. 2004 . Using a high-powered strobe light to increase the catch of Antarctic krill . Marine Biology , 144 : 493 – 502 . Google Scholar Crossref Search ADS Yamamura O. , Inada T. , Shimazaki K. 1993 . Predation on firefly squid, Watasenia scintillans by demersal fishes off Sendai Bay, North Japan. In Recent Advances in Cephalopod Fishery Biology , pp. 633 – 639 . Ed. by Okuntani T. , O’Dor R. K. , Kubodera T. Tokai University Press , Tokyo . Zhang X. , Dam H. G. 1997 . Downward export of carbon by diel migrant mesozooplankton in the central equatorial Pacific . Deep-Sea Research Part II: Topical Studies in Oceanography , 44 : 2191 – 2202 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2018. All rights reserved. For permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Proud, Roland; Handegard, Nils Olav; Kloser, Rudy J; Cox, Martin J; Brierley, Andrew S

doi: 10.1093/icesjms/fsy037pmid: N/A

Abstract The mesopelagic community is important for downward oceanic carbon transportation and is a potential food source for humans. Estimates of global mesopelagic fish biomass vary substantially (between 1 and 20 Gt). Here, we develop a global mesopelagic fish biomass model using daytime 38 kHz acoustic backscatter from deep scattering layers. Model backscatter arises predominantly from fish and siphonophores but the relative proportions of siphonophores and fish, and several of the parameters in the model, are uncertain. We use simulations to estimate biomass and the variance of biomass determined across three different scenarios; S1, where all fish have gas-filled swimbladders, and S2 and S3, where a proportion of fish do not. Our estimates of biomass ranged from 1.8 to 16 Gt (25–75% quartile ranges), and median values of S1 to S3 were 3.8, 4.6, and 8.3 Gt, respectively. A sensitivity analysis shows that for any given quantity of fish backscatter, the fish swimbladder volume, its size distribution and its aspect ratio are the parameters that cause most variation (i.e. lead to greatest uncertainty) in the biomass estimate. Determination of these parameters should be prioritized in future studies, as should determining the proportion of backscatter due to siphonophores. Introduction In this article, we consider, from the standpoint of available acoustic survey data, what the global biomass of mesopelagic fish (fish in the 200–1000 m depth range, including myctophids or “lantern fish”) might be. The importance of the mesopelagic community The mesopelagic community plays an important role in global biogeochemical cycling and the biological carbon pump, and is attracting increasing attention from commercial fishers (St John et al., 2016). Biogeochemical and ecosystem models which simulate the biological carbon pump require validation of the mesopelagic component to provide confidence in their predictions of vertical carbon flux, which itself feeds into climate/Earth-system models (Giering et al., 2014). To cast light on the “dark hole in our understanding” of the mesopelagic (St John et al., 2016), to progress towards ecosystem-based management in advance of developing fisheries, and to make headway on conservation of water-column habitat in areas beyond national jurisdiction (Roberts et al., 2017), a robust estimate of mesopelagic fish biomass is required. A substantial amount of mesopelagic biomass in the 1 mm+ size fraction is contained, during the day, within deep scattering layers (DSLs), primarily made up of fish, zooplankton, squid, and jellyfish. DSLs are detected using echosounders, which emit sound waves and record backscatter (see Chu, 2011, for a review). Echosounder observations (summarized pictorially as echograms) can be analysed and biological features such as layers, schools, and swarms can be identified and quantified (Holliday, 1972; Coetzee, 2000; Cox et al., 2011; Proud et al., 2015). DSL depth varies globally between ∼200 and 1000 m and is driven, at least in part, by environmental conditions, e.g. light intensity, oxygen concentration, temperature, and mixing (Bianchi et al., 2013; Klevjer et al., 2016; Aksnes et al., 2017; Proud et al., 2017). There is often more than one single scattering layer (e.g. Andreeva et al., 2000). DSL backscattering intensity (a proxy for biomass) also varies globally and is correlated with primary production (PP) at the surface and temperature at DSL depth (Netburn and Koslow, 2015; Proud et al., 2017). Typically, DSL backscatter reduces during the night, as a proportion of the community migrates to the surface to feed (Brierley, 2014). Present estimates of mesopelagic fish biomass It has been estimated recently that the global biomass of mesopelagic fish could be around 11–15 gigatonnes (Gt) (Irigoien et al., 2014). That estimate arises from analysis of an acoustic survey (38 kHz) transect run west to east around the world through the mid latitudes (the Malaspina 2010 Circumnavigation Expedition). Under that analysis, the acoustic backscatter is attributed 100% to fish and the resulting biomass estimate is ∼11–15 times higher than an historic estimate of 1 Gt based on net sampling (Gjøsaeter and Kawaguchi, 1980). Although it is recognized that trawl avoidance by mesopelagic fish may lead to a large underestimation of their biomass by net sampling (Kaartvedt et al., 2012), acoustic energy is not necessarily directly proportional to fish biomass and the acoustic energy is not only from fish and so the disagreement between “historic” and “new” estimates could be due to inaccuracy in both assumptions rather than just to a failing of one. Proud et al. (2017) compiled a quasi-global database of daytime 38 kHz mesopelagic backscattering intensity, and used it to define a mesopelagic biogeography. The biogeography was compiled from completely different acoustic data to the Malaspina study, with a much wider geographic extent, and was based on characteristics of objectively identified DSLs. Total backscatter arising from DSLs in the mesopelagic zone (200–1000 m) was predicted to be 6.02 ± 1.4 × 109 m2 (mean area-backscattering coefficient multiplied by surface area). In previous work, we scaled this prediction to global fish biomass (70° N and 70° S), estimating a value of 9 Gt (Proud, 2016). Irigoien et al. (2014) suggest that their estimate (11–15 Gt for the geographic region between latitudes 40° N and 40° S) could be ∼30% higher (i.e. 14.3–19.5 Gt) if expanded to the area between 70° N and 70° S, which is similar to the extent of the Proud et al. (2017) biogeography. That scaled estimate is startlingly high, and begs the question “can it even be sustained given known PP?” Back-of-the envelope trophic calculations reveal that “yes, it can”: for a given value of PP, temperature and trophic efficiency (TE), the biomass of mesopelagic fish (trophic level = 3.2, www.fishbase.org) can be predicted (Gascuel et al., 2008). For TE values of 5, 10, and 20%, a global mean temperature at the depth of the DSL of 7.2°C and average open-ocean PP of 0.312 g C d−1 (Proud et al., 2017), global mesopelagic fish biomasses are predicted to be c. 0.73, 3.4, and 15.5 Gt, respectively. Recent high mesopelagic estimates are therefore at least energetically possible if TEs to mesopelagic fish are of the order of 20% (Davison et al., 2013; Heymans et al., 2014; Rosenberg et al., 2014; Jennings and Collingridge, 2015). However, a recent food-web model (Anderson et al., 2019) and a macroecological model (Jennings and Collingridge, 2015), predict global mesopelagic fish biomass of 2.4 and <1.4 Gt (median biomass for all consumers including mesopelagic fish), respectively: these are closer to the historic mesopelagic fish biomass estimate of 1 Gt (Gjøsaeter and Kawaguchi, 1980). Gas-bladdered mesopelagic siphonophores and fish The mesopelagic zone is occupied by a range of taxa, including crustacean and gelatinous zooplankton and cephalopods. Some of these can be numerically very abundant (e.g. copepods) and some have a high acoustic target strength (TS). Gas-bearing siphonophores (Physonects and Cystonects) and teleost fish, with gas-filled swimbladders are strong acoustic targets (Barham, 1963, 1966; Warren, 2001; Scoulding et al., 2015) and resonant at depth at 38 kHz (Kloser et al., 2016). To estimate plausible ranges of mesopelagic fish biomass with acoustic data the effect of resonant scatter from siphonophores with gas-filled pneumatophores (gas bladders) and fish with gas-filled swimbladders (gas bladders) needs to be considered (Davison et al., 2015). The TS from individual gas bladders (gas-filled organs of fish and siphonophores), produces >95% of the organism's total TS (Foote, 1987) and is often used to approximate the TS of a gas-bladdered fish or siphonophore (Warren, 2001; Scoulding et al., 2015). Gas bladder TS can be predicted using acoustic scattering models, typically assuming that the shape of the gas bladder can be approximated by a prolate spheroid (Scoulding et al., 2015). Modelled gas bladder TS increases linearly with size in the Rayleigh zone and plateaus as the size of the gas bladder approaches a value of λ/2π in the geometric zone (Figure 1), where λ is the wavelength of the incident sound wave. In the resonant region (Figure 1), the gas bladder vibrates in sympathy with the acoustic wave and re-radiates more energy than predicted by commonly used log-linear TS-to-length relationships (e.g. Foote, 1987). The degree of the re-radiation is dependent somewhat on the tension of the gas bladder wall and the tissue viscosity (Love, 1978; Baik, 2013; Scoulding et al., 2015). In this study, acoustic data were available at the commonly used fisheries acoustic frequency of 38 kHz. The wavelength at 38 kHz in seawater is ∼3.9 cm, and 38 kHz sound scattering by a mesopelagic fish or siphonophore at 500 m, with an in situ gas bladder ∼1 mm in length, falls well within the resonant region. Resonant scattering will be provoked routinely in surveys at 38 kHz of mesopelagic fish and siphonophores when the equivalent spherical radius (the radius of a sphere equal in volume to the prolate spheroid) of the gas bladder is within ∼0.4–1 mm (Davison et al., 2015; Kloser et al., 2016). Resonant scattering need not be confined just to “small” fish or siphonophores as the investment of fat in the swimbladders of older and larger fish (Neighbors and Nafpaktitis, 1982) may result in swimbladders of larger fish actually containing smaller volumes of gas than smaller fish; a reduction in body density with length may also contribute to this effect. Furthermore, compression of the gas bladders of downward-migrating fish, according to Boyle’s Law, may result in the gas bladder volume being substantially smaller than would be suggested by fish length alone, and so diel variability in scattering type/intensity may occur. The issue of mesopelagic fish TS is clearly vexed. Figure 1. Open in new tabDownload slide Predicted 38 kHz target strength (TS) of a gas bladder (gas-filled organs of fish and siphonophores) at 500 m over a range of sizes. The gas bladder was modelled as a prolate spheroid using a resonance acoustic scattering model (Scoulding et al., 2015) and the equivalent spherical radius (aesr) was calculated using aesr =  ab21/3 where a and b are the semi-major and semi-minor axes of the prolate spheroid, respectively. Rayleigh and geometric scattering zones are indicated along with consequences to TS of resonance. Figure 1. Open in new tabDownload slide Predicted 38 kHz target strength (TS) of a gas bladder (gas-filled organs of fish and siphonophores) at 500 m over a range of sizes. The gas bladder was modelled as a prolate spheroid using a resonance acoustic scattering model (Scoulding et al., 2015) and the equivalent spherical radius (aesr) was calculated using aesr =  ab21/3 where a and b are the semi-major and semi-minor axes of the prolate spheroid, respectively. Rayleigh and geometric scattering zones are indicated along with consequences to TS of resonance. Approach We take the approach here of attempting to attribute received echo energy in plausible proportions to fish and other potential sound-scatterers, and to scale the likely echo energy from fish to biomass in a manner that takes in to account non-linearity due to resonant scattering (Love, 1978). We do not seek formally to solve the “inverse” problem (Holliday et al., 1989), rather to determine a realistically bounded indication of the possible ranges of global mesopelagic fish biomass. This approach is necessary because collectively science does not yet have, and indeed is not likely to have anytime soon, detailed information on the abundance and size of the community of organisms inhabiting the mesopelagic realm globally. That community is made up of a diverse range of taxa spanning a size range of several orders of magnitude, and no single sampling approach will yield an unbiased view. Net sampling, for example, delivers only slow-moving and physically robust organisms because the better swimmers (which tend to be larger) can evade nets, and fragile, gelatinous organisms are mangled by them (Pakhomov and Yamamura, 2010). Objectives The objective of this article is to estimate the likely range of global mesopelagic fish biomass and the drivers of uncertainty. Our method was as follows: (i) define a generalized acoustic biomass model, (ii) obtain a global mean value of mesopelagic backscatter from the literature (Proud et al., 2017), (iii) with reference to acoustic scattering models, review the dominant sources of backscatter found within the mesopelagic zone, (iv) define mesopelagic fish biomass model and identify unknown parameters and possible confounding animal behaviours, (v) define plausible statistical distributions to capture the full range of uncertainty in the unknown parameters and develop scenarios to simulate a range of animal behaviours, (vi) quantify uncertainty across all scenarios of mesopelagic fish biomass estimates, and (vii) run a global sensitivity analysis to quantify the contribution of each parameter to overall uncertainty in the mesopelagic fish biomass model. Methods The method presented here follows a generalized approach. We first define general equations used for the conversion of acoustic backscattering intensity to biomass (see also Equation 1 in Kloser et al., 2009 and Equation 9.11 in Simmonds and MacLennan, 2005), and then parameterize them for estimation of mesopelagic fish biomass. For a given aggregation of organisms, comprising G groups (e.g. taxonomic, functional, or anatomical), each made up of Mg members, over a common area A, the mean area-backscattering coefficient for all groups, sa (m2 m−2), is given by sa=1AΣg=1GΣm=1Mgσbsg,m,(1) where g is the group index, m is the group member index and σbsg,m (m2) is the backscattering cross-section for member m of group g. Group biomass, Bg (kg), is then calculated by Bg=Apgsaσbsg¯Wg¯, (2) where Wg¯ (kg) and σbsg¯ are the mean member weight and mean area-backscattering coefficient of group g and pg, the proportion of sa that is produced by group g, is given by pg=ngσbsg¯Σg=1Gngσbsg¯,(3) where ng is the relative proportion by number of the aggregation represented by group g. To estimate Bg, the following steps are taken: (i) define the group of interest (g), establish the region and depth range z1 to z2 (volume) which contains the group of interest, and define all other known scattering groups found within the volume; (ii) predict or measure sa over the volume for a given incident frequency; (iii) define acoustic scattering models to predict σbsg¯ of each group; (iv) determine which groups contribute substantially to sa and estimate pg; (v) solve (2) for the group of interest, and identify unknown parameters and animal behaviours; (vi) define distributions for unknown parameters and scenarios for unknown animal behaviours; (vii) estimate uncertainty in Bg over the parameter space for each scenario; and (viii) determine the sensitivity of Bg to input parameters. The remainder of the method follows this procedure for the case of the putative global population of mesopelagic fish. Model definitions and global mesopelagic area-backscattering coefficient In this study, we define our target group as fish in mesopelagic (200–1000 m) DSLs (Figure 2) during the daytime in the open ocean (seabed depth > 1000 m). Globally, there are c. 900 species of mesopelagic fish and the most abundant and diverse family group is Myctophidae (lanternfish) with c. 250 species (Bone et al., 1995); other fish families which include large numbers of mesopelagic fish are Gonostomatidae (bristlemouths), Phosichthyidae (lightfishes), and Sternoptychidae (e.g. marine hatchetfishes) (Bone et al., 1995). The global open ocean has, following Proud et al. (2017), a total surface area, A, of 3.11 × 1014 m2 and total global daytime 38 kHz mesopelagic backscatter of 6.02 ± 1.4 × 109 m2; at 38 kHz, the signal-to-noise ratio is such that the observable range is at least 1000 m. The global sa value, determined by dividing the total mesopelagic backscatter by A, is 1.94 ± 0.44 × 10−5 m2 m−2. We contend that the following taxonomically based scattering groups make possible substantial contributions to mesopelagic sa: copepods; euphausiids; squid; jellyfish; and siphonophores (Physonects and Cystonects, referred to collectively as just siphonophores from here on for simplicity). Figure 2. Open in new tabDownload slide DSL day–night variability in three oceans. Mean linear volume backscattering coefficient (sv, m−1) in 10 m depth bins to 1000 m for day (yellow) and black (night) for 38 kHz acoustic survey data (source: www.imos.org.au). Shading represents the standard deviation of the sv values. Dusk and dawn defined as 1 h prior to and after sunset and sunrise, respectively. Figure 2. Open in new tabDownload slide DSL day–night variability in three oceans. Mean linear volume backscattering coefficient (sv, m−1) in 10 m depth bins to 1000 m for day (yellow) and black (night) for 38 kHz acoustic survey data (source: www.imos.org.au). Shading represents the standard deviation of the sv values. Dusk and dawn defined as 1 h prior to and after sunset and sunrise, respectively. Acoustic scattering models The scattering groups defined here fall into two categories, gas-bearing organisms (mesopelagic fish and siphonophores) and weakly scattering fluid-filled organisms (copepods, euphausiids, squid, and jellyfish). The σbs of the fluid-filled group was predicted using the distorted-wave-born approximation (DWBA) model (Chu et al., 1993) using parameters from Lavery et al. (2007). The gas-filled swimbladders of fish (gas bladders) and gas-filled pneumatophores of siphonophores (gas bladders) produce >95% of the organisms' backscatter at 38 kHz (Foote, 1980). This figure is likely to be closer to 99% for mesopelagic fish: Foote (1980) refers to much larger and denser epipelagic fish for which backscatter from body tissue makes up a larger proportion of the total. The σbs of the gas bladders of fish and siphonophores can be predicted using the resonance model of Love (1978), including adaptations for shape (Ye, 1998) and directivity (Stanton, 1988). Generally, our resonance model formulation followed that of the prolate spheroid model described by Scoulding et al. (2015), apart from the calculation of the resonant frequency (5) which was taken directly from Love (1978). Resonant scattering is dependent on wavelength of incident frequency, depth range, viscosity of tissue, and size of the gas bladder (Love, 1978; Davison et al., 2015; Kloser et al., 2016). The backscattering cross-section is given by σbs=aesr2ρw/ρf2ω02/ω2-12+δ2ω,a,b,ξ;ΩDk,a,θ,σ, (4) where ω0 is the angular resonant frequency found by solving ω0aesr2=Ce2(a,b)3γaPzρf+2sρfaesr3γa-1, (5)aesr is the equivalent spherical radius given by aesr=ab21/3, (6) where a and b are the semi-major and semi-minor axes of the prolate spheroid, respectively, and are related by the b-scaling parameter, β, given by a=bβ, (7) ω is the angular incident frequency, δ is a damping factor, ρf is flesh density, ρw is water density, s is the surface tension, D is the directivity function (Stanton, 1988) averaged over a normal distribution of orientation angles, N(θ, σ), k is the wave number, Ce is a spheroidal elongation factor (Strasberg, 1953; Weston, 1967), γa is the specific heat ratio for gas, Pz is pressure at depth (Pz = z/10 + 1, where z is depth in meters), ξ is the dynamic viscosity and Ω is a set of damping constants (Table 1). All constants used in this study are given in Table 1. Table 1. Resonance model parameter values for gas-filled swimbladders and pneumatophores (gas bladders). Symbol . Description . Unit . Value . ω Incident frequency Hz 38000 Damping constants (Ω) ρa Density of air kg m−3 1.3 (Love, 1978) γa Ratio of specific heat for air – 1.4 (Love, 1978) cpa Specific heat at constant pressure for air cal kg−1°C−1 240 (Love, 1978) κa Thermal conductivity of air cal m−1 s−1°C−1 5.5 × 10−3 (Love, 1978) cw Sound speed in sea water m s−1 1500 ρw Density of sea water kg m−3 1027 Gas bladder parameters Swimbladder Pneumatophore θ Mean orientation angle degrees 0a 0a σ Standard deviation of orientation angle degrees 30a 30a z Depth m Variable 525 (Proud et al., 2017) ρf Tissue density kg m−3 1050 (Love, 1978) 1030 (Lavery et al., 2007) ξ Dynamic viscosity kg m−1 s−1 Variable 4/3b (Scoulding et al., 2015) β b-scaling parameter – 0.64 (Yasuma et al., 2010) 0.36 (Barham, 1963) s Surface tension at gas cavity-tissue interface N m−1 32 (Love, 1978) 0.074c (Love, 1978) Symbol . Description . Unit . Value . ω Incident frequency Hz 38000 Damping constants (Ω) ρa Density of air kg m−3 1.3 (Love, 1978) γa Ratio of specific heat for air – 1.4 (Love, 1978) cpa Specific heat at constant pressure for air cal kg−1°C−1 240 (Love, 1978) κa Thermal conductivity of air cal m−1 s−1°C−1 5.5 × 10−3 (Love, 1978) cw Sound speed in sea water m s−1 1500 ρw Density of sea water kg m−3 1027 Gas bladder parameters Swimbladder Pneumatophore θ Mean orientation angle degrees 0a 0a σ Standard deviation of orientation angle degrees 30a 30a z Depth m Variable 525 (Proud et al., 2017) ρf Tissue density kg m−3 1050 (Love, 1978) 1030 (Lavery et al., 2007) ξ Dynamic viscosity kg m−1 s−1 Variable 4/3b (Scoulding et al., 2015) β b-scaling parameter – 0.64 (Yasuma et al., 2010) 0.36 (Barham, 1963) s Surface tension at gas cavity-tissue interface N m−1 32 (Love, 1978) 0.074c (Love, 1978) Comments are referred to using letters. a At 38 kHz, orientation does not significantly affect backscatter of small targets (Scoulding et al., 2015). b In the absence of any measurements, we used the mean value from Scoulding et al. (2015). c Surface tension of a gas bubble. Open in new tab Table 1. Resonance model parameter values for gas-filled swimbladders and pneumatophores (gas bladders). Symbol . Description . Unit . Value . ω Incident frequency Hz 38000 Damping constants (Ω) ρa Density of air kg m−3 1.3 (Love, 1978) γa Ratio of specific heat for air – 1.4 (Love, 1978) cpa Specific heat at constant pressure for air cal kg−1°C−1 240 (Love, 1978) κa Thermal conductivity of air cal m−1 s−1°C−1 5.5 × 10−3 (Love, 1978) cw Sound speed in sea water m s−1 1500 ρw Density of sea water kg m−3 1027 Gas bladder parameters Swimbladder Pneumatophore θ Mean orientation angle degrees 0a 0a σ Standard deviation of orientation angle degrees 30a 30a z Depth m Variable 525 (Proud et al., 2017) ρf Tissue density kg m−3 1050 (Love, 1978) 1030 (Lavery et al., 2007) ξ Dynamic viscosity kg m−1 s−1 Variable 4/3b (Scoulding et al., 2015) β b-scaling parameter – 0.64 (Yasuma et al., 2010) 0.36 (Barham, 1963) s Surface tension at gas cavity-tissue interface N m−1 32 (Love, 1978) 0.074c (Love, 1978) Symbol . Description . Unit . Value . ω Incident frequency Hz 38000 Damping constants (Ω) ρa Density of air kg m−3 1.3 (Love, 1978) γa Ratio of specific heat for air – 1.4 (Love, 1978) cpa Specific heat at constant pressure for air cal kg−1°C−1 240 (Love, 1978) κa Thermal conductivity of air cal m−1 s−1°C−1 5.5 × 10−3 (Love, 1978) cw Sound speed in sea water m s−1 1500 ρw Density of sea water kg m−3 1027 Gas bladder parameters Swimbladder Pneumatophore θ Mean orientation angle degrees 0a 0a σ Standard deviation of orientation angle degrees 30a 30a z Depth m Variable 525 (Proud et al., 2017) ρf Tissue density kg m−3 1050 (Love, 1978) 1030 (Lavery et al., 2007) ξ Dynamic viscosity kg m−1 s−1 Variable 4/3b (Scoulding et al., 2015) β b-scaling parameter – 0.64 (Yasuma et al., 2010) 0.36 (Barham, 1963) s Surface tension at gas cavity-tissue interface N m−1 32 (Love, 1978) 0.074c (Love, 1978) Comments are referred to using letters. a At 38 kHz, orientation does not significantly affect backscatter of small targets (Scoulding et al., 2015). b In the absence of any measurements, we used the mean value from Scoulding et al. (2015). c Surface tension of a gas bubble. Open in new tab Using the constants in Table 1, σbs for a gas-bladdered fish or siphonophore, can be predicted when aesr and z of the gas bladder are known. Swimbladder volume of a fish, Vswb, is related to aesr by aesr=3Vswb4π1/3. (8) The proportion of fish body volume, pswb, can be used to calculate Vswb=pswbVf, (9) where Vf is the volume of the fish, given by Vf=4πlf2lf2α23. (10) Here α is the fish aspect ratio α=lfwf, (11) where fish shape has been approximated by a prolate spheroid and lf (mm) and wf (mm) are the length and width of the fish, respectively. Therefore, for a given lf, α and pswb value, aesr can be estimated using (8–10), converted to a and b values, using (6 and 7), for a given b-scaling factor (β, Table 1), and used to predict σbsa,b,z,ξ≅σbslf,pswb,α,z,ξ, (12) from (4). Note that in (12), the constants defined in Table 1 have been omitted for clarity. Fish weight (Wf, kg) can then be calculated by multiplying fish volume by density and is given by Wf=Vfρf, (13) and is therefore a function of length and aspect ratio. Mesopelagic echoes The gas bladders of fish and siphonophores produce backscatter that contributes to 95% or more of the organisms’ target strength [TS = 10log10(σbs), dB re 1 m2, Maclennan et al., 2002] when insonified at 38 kHz, the frequency used in this study. This proportion may be substantially >95% at certain specific depth-size combinations when resonant backscattering is provoked (Davison et al., 2015) increasing the organisms' TS by a factor of 10 or more (Figure 3, see also Kloser et al., 2016). Figure 3. Open in new tabDownload slide Dominant mesopelagic scatterers and resonance (a) A resonance model (Love, 1978) was used to predict resonant frequency of gas-filled prolate spheroids (ps—approximate shape of inflated fish swimbladders and siphonophore pneumatophores) over a range of sizes and depths; (b) TS values predicted using the resonance model for a prolate spheroid over a range of depths, where size (equivalent spherical radius) was selected to produce resonant backscattering at 38 kHz. Damping (ξ) for prolate spheroid was set to 0, except for the 0.85 mm ps at 1000 m (dashed line) where ξ = 20 (Love, 1978). A DWBA model (Chu et al., 1993) was used to predict frequency response of a fish body (width = 1.63 cm, density contrast g = 1.023, sound-speed contrast h = 1.032), squid (width = 1.2 cm, g = 1.043, h = 1.053), medusae (g = 1.009, h = 1.0004), copepod (g = 1.058, h = 1.02) and euphausiid (g = 1.016, h = 1.019). Sizes (lengths) are given in the plot. Sound-speed and density contrast values taken from Lavery et al. (2007). Figure 3. Open in new tabDownload slide Dominant mesopelagic scatterers and resonance (a) A resonance model (Love, 1978) was used to predict resonant frequency of gas-filled prolate spheroids (ps—approximate shape of inflated fish swimbladders and siphonophore pneumatophores) over a range of sizes and depths; (b) TS values predicted using the resonance model for a prolate spheroid over a range of depths, where size (equivalent spherical radius) was selected to produce resonant backscattering at 38 kHz. Damping (ξ) for prolate spheroid was set to 0, except for the 0.85 mm ps at 1000 m (dashed line) where ξ = 20 (Love, 1978). A DWBA model (Chu et al., 1993) was used to predict frequency response of a fish body (width = 1.63 cm, density contrast g = 1.023, sound-speed contrast h = 1.032), squid (width = 1.2 cm, g = 1.043, h = 1.053), medusae (g = 1.009, h = 1.0004), copepod (g = 1.058, h = 1.02) and euphausiid (g = 1.016, h = 1.019). Sizes (lengths) are given in the plot. Sound-speed and density contrast values taken from Lavery et al. (2007). Other large mesopelagic organisms such as squid, which are likely to have similar global abundances to fish (Clarke, 1996), and medusae, have similar, low TSs as the bodies (flesh/bone) of fish (Figure 3) and, as with the bodies of fish, are likely to produce a relatively small proportion of total sa (fish bodies might contribute just 5% or less: Foote, 1980; Forland et al., 2014). Smaller organisms such as copepods, although much more numerically abundant than fish, have TS values that are up to c. 9 orders of magnitude below that of gas-filled structures (Figure 3) so even huge densities of these organisms will not contribute significantly to the total backscatter. For example, even a preposterously high mean global copepod density of 1 million individuals per m2 and a size distribution of N(µ = 2 mm, σ = 0.5 mm) would equate to a contribution of <1% of the predicted global mesopelagic sa. In summary, we assume that backscatter from gas bladders of fish and siphonophores (gas-bearing organisms) produces close to 100% of mesopelagic sa (Lavery et al., 2010; Irigoien et al., 2014; Davison et al., 2015). This agrees with measurements from a lowered probe that attribute 95% of the scattering at 38 kHz to gas bladders, many in resonance (Kloser et al., 2016). From individuals to populations The previously defined equations for σbs are applicable only to individuals. To estimate global mesopelagic fish biomass, mean fish population σbs values are needed, which require fish-length distributions. We determined these as follows. We first defined a log-normal distribution: X∼lnNμ=0,σX2, (14) where X is a random variable for which the mean µ = 0 and the variance σ2 =  σX2 ⁠. This distribution describes the shape of the fish-length distribution and can be varied by changing a single parameter, σX2 ⁠. A number sequence was used to define N equal-width fish length–frequency-distribution classes spanning the minimum and maximum fish-length values Lmin and Lmax: L=rangea=Lmin,b=Lmax,c=N, (15) where range(a, b, c) is a function, producing a sequence of numbers starting from a and ending with b, with total length c. Similarly, for a given σX2 ⁠, the log-normal distribution range was defined: Ld=rangeppfX0.001,ppfX0.999,N, (16) where ppfX is the percent point function of X, and Ld and L are equal in length. Backscattering cross-section, σbs, values were calculated for each length class and by integrating over the probability density function (pdf) of distribution X, using the trapezium rule, the population mean σbs was estimated σbsf¯=Σi=1N-1σbsi+σbsi+12pXXi+pXXi+12Xi+1-Xiφi, (17) where pX is the pdf of X and σbsi ⁠, and φi are the σbs value and statistical weight of the ith length class, respectively. The length class weight, φi, is included to enable the relative contribution of each length class (i.e. proportion of class with gas bladders) to σbsf¯ to be varied. Similarly, mean population weight is given by Wf¯=Σi=1N-1Wfi+Wfi+12pXXi+pXXi+12Xi+1-Xi. (18) Here, the statistical weighting is absent because all fish will contribute to the mean population weight, regardless of whether they possess a gas bladder or not. Mesopelagic fish biomass model The backscatter from fish and siphonophores was assumed to produce the majority of mesopelagic sa—a reasonable assumption given Figure 3. Therefore, simplified total area-backscattering coefficient sa* is the sum of the contributions of backscattering from siphonophores and fish: sa*=pf+psiphsa, (19) where pf and psiph are the proportion of sa produced by fish and siphonophores, respectively. Substituting (17 and 18) into (2), mesopelagic fish biomass, for a global population, is estimated using Bf=Asafσbsf¯σX2,α,pswb,z,ξWf¯σX2,α, (20) where saf is the amount of sa produced by fish (pfsa). Model scenarios All myctophids (a very common and abundant mesopelagic fish) are thought to develop swimbladders during development (Bone et al., 1995; Moser, 1996) and some species of mesopelagic fish are known to keep their swimbladders throughout their lifecycles (e.g. marine hatchetfish). Myctophids caught in the Tasman sea region were found mostly to have gas bladders (Flynn and Pogonoski, 2012) but mesopelagic fish in general have often been reported to have varied swimbladder states, including absent, uninflated, and inflated (Butler and Pearcy, 1972; Neighbors and Nafpaktitis, 1982; Bardarson, 2013), often reported to be linked to ontogeny, where juveniles or young adults possess uninflated or absent swimbaldders (Yasuma et al., 2010) and late-stage adults have reduced (fat invested) swimbladders (Butler and Pearcy, 1972; Neighbors and Nafpaktitis, 1982). To examine the impact of the observed variability in swimbladder state on acoustically inferred biomass of a global mesopelagic fish population, three scenarios were investigated. Scenario 1 (S1), had equally weighted length classes (φ = 1, see Equation 17), i.e. all fish had gas bladders. In scenario 2 (S2), φ followed a cosine function (i.e. the smallest length classes always had gas bladders), given by φi=φmin+ cos Li-Lmin90/L97.5-Lmin1-φmin, Li<L97.5φmin, Li≥L97.5, (21) where i is the length class index, φmin is a constant ranging between 0 and 1 and denotes the minimum proportion of fish per length class with gas bladders (i.e. the proportion of fish that do not lose their gas bladders with age) and L97.5 is the length at the 97.5th percentile of the cumulative distribution function (Figure 4a); the 97.5th percentile was chosen to avoid fitting the cosine curve to extensive tails in the log-normal distributions. In effect, the rate of decay of the curve was controlled by the value of φmin. For scenario 3 (S3), φ was scaled to a sine curve, similar to the shape observed by Yasuma et al. (2010) for Myctophum asperum, over the population length range: φi=sinLi-Lmin90/Lcent-Lmin, (22) where the maximum value of the sine function angle was set to 180 degrees and Lcent is the length at the centre of the pdf where the cumulative distribution function equalled 0.5 (Figure 4b). This ensured that a large proportion of small (young) and large (old) fish were without gas bladders and that the proportion with gas bladders increased towards the centre of the distribution. Figure 4. Open in new tabDownload slide Length class weighting (φ; proportion of gas-bladdered fish by length class, shown by dashed line) for scenarios 2 (a) and 3 (b). (a) Single log-normal distribution plotted (µ= 0, σX2=0.3), φ is plotted over a range of φmin values between 0 and 1. (b) Log-normal distributions scaled to fish length class, where the mean was set to 0 and the variance (⁠ σX2 ⁠) ranged from 0.3 to 1. φ plotted for each distribution. Maximum value of x-axis set to L = 212 mm, which is the length class at the 97.5th percentile of the broadest log-normal distribution (µ = 0, σX2 = 0.3). Figure 4. Open in new tabDownload slide Length class weighting (φ; proportion of gas-bladdered fish by length class, shown by dashed line) for scenarios 2 (a) and 3 (b). (a) Single log-normal distribution plotted (µ= 0, σX2=0.3), φ is plotted over a range of φmin values between 0 and 1. (b) Log-normal distributions scaled to fish length class, where the mean was set to 0 and the variance (⁠ σX2 ⁠) ranged from 0.3 to 1. φ plotted for each distribution. Maximum value of x-axis set to L = 212 mm, which is the length class at the 97.5th percentile of the broadest log-normal distribution (µ = 0, σX2 = 0.3). Model input parameters To predict mesopelagic fish biomass (Bf, 20) for scenarios 1–3, uniform distributions were assumed for each model variable (⁠ saf ⁠, σX2 ⁠, α, pswb, z, ξ, Table 2). Scenario 2 also included φmin, the minimum proportion of gas-bladdered fish per length class, which was also assumed to have a uniform distribution, with minimum and maximum values of 0 and 1, respectively (Figure 4a). The parameter values have the potential to be widely variable but at present we do not have enough knowledge to predict accurately their global distributions. Here, we use uniform distributions to ensure that we capture the full range of variability that they could potentially contribute to global mesopelagic fish biomass. These parameters and their corresponding distribution ranges were chosen based on the following reasoning. Table 2. Assumed fish population input parameter statistical distributions. Symbol . Description . Unit . Distribution . saf Mesopelagic fish area-backscattering coefficient m2 m−2 U(0, 2.38 × 10−5) (Proud et al., 2017) σX2 Variance of length distribution – U(0.3, 1) α Aspect ratio of fish body – U(4, 12) pswb Swimbladder volume as a proportion of fish volume – U(0.0001, 0.0263) (Yasuma et al., 2010) z Depth, z m U(200, 1000) ξ Dynamic viscosity kg m−1 s−1 U(0, 20) (Love, 1978) φmin Minimum proportion of gas-bladdered fish per length class used in scenario 2 – U(0, 1) Symbol . Description . Unit . Distribution . saf Mesopelagic fish area-backscattering coefficient m2 m−2 U(0, 2.38 × 10−5) (Proud et al., 2017) σX2 Variance of length distribution – U(0.3, 1) α Aspect ratio of fish body – U(4, 12) pswb Swimbladder volume as a proportion of fish volume – U(0.0001, 0.0263) (Yasuma et al., 2010) z Depth, z m U(200, 1000) ξ Dynamic viscosity kg m−1 s−1 U(0, 20) (Love, 1978) φmin Minimum proportion of gas-bladdered fish per length class used in scenario 2 – U(0, 1) Distribution of each parameter is given (U, uniform distribution). Open in new tab Table 2. Assumed fish population input parameter statistical distributions. Symbol . Description . Unit . Distribution . saf Mesopelagic fish area-backscattering coefficient m2 m−2 U(0, 2.38 × 10−5) (Proud et al., 2017) σX2 Variance of length distribution – U(0.3, 1) α Aspect ratio of fish body – U(4, 12) pswb Swimbladder volume as a proportion of fish volume – U(0.0001, 0.0263) (Yasuma et al., 2010) z Depth, z m U(200, 1000) ξ Dynamic viscosity kg m−1 s−1 U(0, 20) (Love, 1978) φmin Minimum proportion of gas-bladdered fish per length class used in scenario 2 – U(0, 1) Symbol . Description . Unit . Distribution . saf Mesopelagic fish area-backscattering coefficient m2 m−2 U(0, 2.38 × 10−5) (Proud et al., 2017) σX2 Variance of length distribution – U(0.3, 1) α Aspect ratio of fish body – U(4, 12) pswb Swimbladder volume as a proportion of fish volume – U(0.0001, 0.0263) (Yasuma et al., 2010) z Depth, z m U(200, 1000) ξ Dynamic viscosity kg m−1 s−1 U(0, 20) (Love, 1978) φmin Minimum proportion of gas-bladdered fish per length class used in scenario 2 – U(0, 1) Distribution of each parameter is given (U, uniform distribution). Open in new tab There are few reported open-ocean observations of siphonophore density or pneumatophore size distributions. Pneumatophores have been reported to have mean lengths ranging between 0.15 mm (Lavery et al., 2007) and 3.27 mm (Barham, 1963), and densities from <1 to >1000 individuals m−2 (Mackie et al., 1988). Considering the lack of information, psiph might reasonably be considered to vary from place to place anywhere between 0 and 1, i.e. siphonophores potentially could produce almost all or none of the total mesopelagic sa. Mesopelagic fish sa, saf ⁠, was drawn from a uniform distribution, where the minimum value was set to 0 (for the case where psiph = 1, Equation 19) and maximum value of 2.38 × 10−5 m2 m−2, the upper bound (mean plus RMSE) taken from Proud et al. (2017). The variance of the log-normal distributions, σX2 ⁠, was varied uniformly between 0.3 and 1. The distribution ranges were then matched with the fish population length range (15), where Lmin and Lmax were set to 8 mm (approximately the size of a newly developed juvenile myctophid; Moser, 1996) and 315 mm (maximum reported length of any species in the family Myctophidae, www.fishbase.org), respectively, to yield length distributions (Figure 4). As σX2 was increased from 0.3 to 1, the length distribution shifts from a gaussian-like distribution (μ = c. 88 mm, equal to the median asymptotic length of 219 species of myctophid from www.fishbase.org) to a population dominated by smaller (more likely to produce resonant backscatter) fish, as commonly observed (e.g. Davison et al., 2015). Although Lmax was set to 315 mm, the effective maximum length of the fish population was closer to 212 mm (97.5th percentile of broadest log-normal distribution, see Figure 4), which is similar to the 97.5 percentile of the asymptotic lengths of all documented myctophid species (c. 211 mm, www.fishbase.org). For the length distributions used in this study (Figure 4), the majority of fish were smaller than 88 mm, which is consistent with the sizes of common taxa of mesopelagic fish (e.g. Gonostomatidae, Sternoptychidae, Myctophidae, and Phosichthyidae) known to have gas bladders (Marshall, 1971; Bone et al., 1995; Flynn and Pogonoski, 2012). The aspect ratio, α, a representation of variability in species morphology within the population, was varied uniformly between 4 and 12 (Flynn and Pogonoski,, 2012). Changes in α impact both fish mean weight and swimbladder volume (9–13). During diel vertical migration (DVM), the gas bladders of fish and siphonophores undergo compression on descent and expansion on ascent, following Boyle’s Law. Some species inflate their gas bladders and follow a “constant buoyancy strategy,” whilst others do not, and swim to maintain depth (Denton, 1961; Hersey et al., 1962; Kalish et al., 1986; Thompson and Love, 1996; Love et al., 2003, 2004; Scoulding et al., 2015); this behaviour is likely to vary between and within species, and ontogenetically. Our uncertainty around gas bladder function during and between bouts of DVM stems from the difficulty in making measurements of gas bladder volume at the surface. Bladders of fish brought to the surface from depth may be excessively distended or ruptured following rapid decompression and the measured volumes may not be good indicators of actual volumes in situ at depth. Because of this uncertainty, gas bladder volume as a proportion of body size, pswb, was varied between 0.0001 and 0.0263, equivalent to 0.01 and 2.63% of body volume (Yasuma et al., 2010). The maximum value of pswb is likely to be limited by neutral buoyancy (body density), which does vary with length for many species (Butler and Pearcy, 1972; Neighbors and Nafpaktitis, 1982; Davison, 2011b). The range of values chosen, result in a broad range of Vswb values (9), which include proportions, both small, representative of gas bladders that do not re-inflate when compressed at depth, and larger proportions, which are consistent with gas bladder sizes required to maintain neutral buoyancy. Fish (or DSL) depth, z, was varied uniformly in the model between 200 and 1000 m. This parameter affects the resonant frequency (5). For a given gas bladder size, the resonant frequency increases with depth; during DVM, a change in depth will cause a change in gas bladder volume, and this variability is captured by other model parameters (primarily by pswb) not by depth. Dynamic viscosity, ξ, (i.e. the viscosity of the gas bladder wall) was set to vary uniformly between 0 kg m−1 s−1, i.e. no damping, resulting in a sharp resonance peak (as might be expected from a gas bubble) up to a value of 20 kg m−1 s−1, which completely dampens the resonance peak (Figure 3). The latter value is as suggested by Love (1978) for midwater fish. There is some contention around this parameter and its validity in modelling the TS of resonant gas bladders (Baik, 2013). Here, we included a broad range of ξ values to ensure that we captured both undamped and damped resonant behaviour in our estimates of global mesopelagic fish biomass. Estimates of global mesopelagic fish biomass A joint probability distribution (J) was defined from the six (seven in the case of scenario 2) uniform distributions of the input parameters defined in Table 2. For a single model run, 5000 samples were generated from J using the quasi-random Sobol sequence (Sobol, 1967): this provides a more evenly distributed sample set in parameter space than a purely random sample. Sample estimates of mean fish population σbs, mean fish W, and global mesopelagic fish biomass were generated using (17, 18, and 20), respectively, for all three scenarios. Uncertainty in biomass estimates was quantified using summary statistics, and is presented by box plots in Figure 5. Figure 5. Open in new tabDownload slide Summarized mesopelagic fish biomass model results for each scenario. (a) Mean fish TS = 10log10(⁠ σbsf¯ ⁠), (b) global fish biomass, and (c) mean individual fish weight. Scenario 1 (S1) all fish have gas bladders, whereas in scenarios 2 (S2) and 3 (S3), a proportion of fish are without gas bladders. Horizontal line within each box is the median value, box limits are the inter-quartile range, i.e. 25 and 75% quantiles. The whiskers (vertical lines) are 1.5 times the inter-quartile range. Outliers not plotted. Figure 5. Open in new tabDownload slide Summarized mesopelagic fish biomass model results for each scenario. (a) Mean fish TS = 10log10(⁠ σbsf¯ ⁠), (b) global fish biomass, and (c) mean individual fish weight. Scenario 1 (S1) all fish have gas bladders, whereas in scenarios 2 (S2) and 3 (S3), a proportion of fish are without gas bladders. Horizontal line within each box is the median value, box limits are the inter-quartile range, i.e. 25 and 75% quantiles. The whiskers (vertical lines) are 1.5 times the inter-quartile range. Outliers not plotted. Three additional model runs were performed in which saf was set to the upper (2.38 × 10−5 m2 m−2), lower (1.5 × 10−5 m2 m−2) and mean (1.94 × 10−5 m2 m−2) values of global mesopelagic sa from Proud et al. (2017), to determine a range of maximum mesopelagic fish biomass estimates, i.e. the contribution of siphonophores to global mesopelagic sa was assumed to be zero. The maximum values of Bf were scaled between 0 and 100% and compared with potential contribution of siphonophores to mesopelagic sa (psiph, 19), predicted over a range of siphonophore mean global densities (ρsiph, 0.1–1000 inds m−2, uniform) and gas bladder size distributions {asiph (mm) ∼ N[µ = U(0.3, 5), σ = µ/3.5]}, where psiph is given by psiph=ρsiphσbs(asiph,b,z,ξ)¯saf, (23) where z = 525 m the global mean DSL value (Proud et al., 2017), ξ = 4/3 (mean from Scoulding et al., 2015) and b was calculated using βasiph (7), using a β value of 0.36 (Barham, 1963). Sensitivity of input parameters A global sensitivity analysis was conducted using a variance based sensitivity metric (Saltelli et al., 2010) to investigate how the different input parameters of the biomass model affected the total biomass. The total effect index is a sensitivity metric that captures both the first order effect as well as higher order effects (interactions). The total effect index for parameter Xi is given as STi=EX∼iVXiBf|X∼i V(Bf), (24) where VXiBf|X∼i is the variance of the biomass estimate when changing the input parameter, Xi, EX∼i is the mean of VXiBf|X∼i and V(Bf) is the total variance of the model. The inner variance estimator captures the variance in the biomass while varying Xi and the outer mean operator takes the mean of these variances. We used simulations to estimate the sensitivity indices. The total effect index STi=12NΣj=1NBfCj-BfCDij2V(Bf) (25) is estimated using the Jansen (1999) estimator (Saltelli et al., 2010, Table 2). Here, N is the number of simulations, C is a set of N sets of parameters drawn from a Sobol sequence (rows are realizations j and columns are the parameter i), CDi is identical to C except that parameter i is replaced from a similar but independent resampling set D ⁠. STi was calculated for 100 model runs, where for each run, N was set to 5000. Results A fish biomass model was constructed and parametrized by seven input factors (⁠ saf ⁠, σX2 ⁠, α, pswb, z, ξ, φmin, the latter of which was used only in scenario 2, see Table 2 for definitions) and run for three different scenarios: S1, which assumed the fish population was comprised solely of fish with gas bladders; S2 where all fish had gas bladders as juveniles, and a minimum proportion of fish kept their gas bladders throughout their life, whilst a growing proportion (following a cosine curve) lost their gas bladder with increasing length, and S3, a population with a large proportion of small and large fish without gas bladders. Five thousand biomass estimates were generated for each model run to capture the range of possible variability and so illustrate uncertainty (Figure 5). Fish biomass uncertainty Model results for S1–S3 were summarized using box plots (Figure 5). Median values of TS decreased from −53.8 dB re 1 m2 (lower quartile, Q1 = −55.6; upper quartile, Q3 = −52.4) to −56.8 dB re 1 m2 (Q1 = −59.1; Q3 = −55) and median biomass increased from 3.833 Gt (Q1 = 1.812; Q3 = 7.374) to 8.292 Gt (Q1 = 3.670; Q3 = 15.962) from S1 to S3. Since the proportion of gas-bladdered fish per length class has no impact on fish weight, it was constant for all three scenarios and had a median value of 4.51 grams (Q1 = 2.25; Q3 = 8.64). Maximum fish biomass and contributions of fish and siphonophores to global mesopelagic backscatter Maximum mesopelagic fish biomass was estimated for each scenario. The minimum lower (25%) and maximum upper (75%) quartiles from the maximum fish biomass estimates for all three scenarios were used to represent the range of maximum mesopelagic fish biomass assuming that siphonophores made no contribution towards global daytime mesopelagic DSL backscatter. The maximum values were then scaled between 0 and 100% to yield fish biomass estimates for different global mean densities and gas bladder size distributions of siphonophores (Figure 6 and Table 3). Fish biomass values were calculated for TE values of 5, 10, and 20% per trophic level to be 0.732, 3.363, and 15.453 Gt, respectively, and were plotted in Figure 6. Table 3. Median mesopelagic fish biomass predictions (Gt) by siphonophore contribution for each scenario. Siphonophore contribution (%) . . . Scenario . 0 . 10 . 20 . 30 . 40 . 50 . 60 . 70 . 80 . 90 . 99 . 1 7.082 6.373 5.665 4.957 4.249 3.541a 2.832a 2.124b 1.416c 0.708c 0.071 2 8.588 7.729 6.870 6.012 5.153 4.294 3.435a 2.576a,b 1.718c 0.859c 0.086 3 15.255 13.729 12.204 10.678 9.153 7.627 6.102 4.576a 3.051a,b 1.525c 0.153c Siphonophore contribution (%) . . . Scenario . 0 . 10 . 20 . 30 . 40 . 50 . 60 . 70 . 80 . 90 . 99 . 1 7.082 6.373 5.665 4.957 4.249 3.541a 2.832a 2.124b 1.416c 0.708c 0.071 2 8.588 7.729 6.870 6.012 5.153 4.294 3.435a 2.576a,b 1.718c 0.859c 0.086 3 15.255 13.729 12.204 10.678 9.153 7.627 6.102 4.576a 3.051a,b 1.525c 0.153c a Value range that includes mesopelagic fish biomass calculated for a trophic efficiency of 10% per trophic level (3.363 Gt, see Figure 6). b Value closest to a food-web model estimate of mesopelagic fish biomass (2.4 Gt, Anderson et al., 2019). c Value range that includes mesopelagic fish biomass predicted by ocean trawls (1 Gt, Gjøsaeter and Kawaguchi, 1980) and total consumer biomass by a macroecological model (1.4 Gt, Jennings and Collingridge, 2015). Open in new tab Table 3. Median mesopelagic fish biomass predictions (Gt) by siphonophore contribution for each scenario. Siphonophore contribution (%) . . . Scenario . 0 . 10 . 20 . 30 . 40 . 50 . 60 . 70 . 80 . 90 . 99 . 1 7.082 6.373 5.665 4.957 4.249 3.541a 2.832a 2.124b 1.416c 0.708c 0.071 2 8.588 7.729 6.870 6.012 5.153 4.294 3.435a 2.576a,b 1.718c 0.859c 0.086 3 15.255 13.729 12.204 10.678 9.153 7.627 6.102 4.576a 3.051a,b 1.525c 0.153c Siphonophore contribution (%) . . . Scenario . 0 . 10 . 20 . 30 . 40 . 50 . 60 . 70 . 80 . 90 . 99 . 1 7.082 6.373 5.665 4.957 4.249 3.541a 2.832a 2.124b 1.416c 0.708c 0.071 2 8.588 7.729 6.870 6.012 5.153 4.294 3.435a 2.576a,b 1.718c 0.859c 0.086 3 15.255 13.729 12.204 10.678 9.153 7.627 6.102 4.576a 3.051a,b 1.525c 0.153c a Value range that includes mesopelagic fish biomass calculated for a trophic efficiency of 10% per trophic level (3.363 Gt, see Figure 6). b Value closest to a food-web model estimate of mesopelagic fish biomass (2.4 Gt, Anderson et al., 2019). c Value range that includes mesopelagic fish biomass predicted by ocean trawls (1 Gt, Gjøsaeter and Kawaguchi, 1980) and total consumer biomass by a macroecological model (1.4 Gt, Jennings and Collingridge, 2015). Open in new tab Figure 6. Open in new tabDownload slide Relative contributions of fish and siphonophores to global mesopelagic area-backscattering coefficient, over a range of siphonophore mean global densities and gas bladder size distributions. (a) Each line represents a separate mean gas bladder equivalent spherical radius (aesr, labelled in mm). Shaded region calculated from global mesopelagic area-backscattering coefficient RMSE from Proud et al. (2017). Mean aesr values larger than 1 mm can be approximated by the line labelled “>1.” (b) Shaded region spans the upper (75%) quartile of S3 and the lower (25%) quartile of S1 (3.495–29.975 Gt), calculated using results of the maximum fish biomass model run. Coloured lines are median values for each scenario. TE (per trophic level) between phytoplankton and mesopelagic fish estimated from Gascuel et al. (2008) using a mean water temperature of 7.2°C and PP value of 0.312 g C d−1 (Proud et al., 2017) are indicated by dashed lines. Figure 6. Open in new tabDownload slide Relative contributions of fish and siphonophores to global mesopelagic area-backscattering coefficient, over a range of siphonophore mean global densities and gas bladder size distributions. (a) Each line represents a separate mean gas bladder equivalent spherical radius (aesr, labelled in mm). Shaded region calculated from global mesopelagic area-backscattering coefficient RMSE from Proud et al. (2017). Mean aesr values larger than 1 mm can be approximated by the line labelled “>1.” (b) Shaded region spans the upper (75%) quartile of S3 and the lower (25%) quartile of S1 (3.495–29.975 Gt), calculated using results of the maximum fish biomass model run. Coloured lines are median values for each scenario. TE (per trophic level) between phytoplankton and mesopelagic fish estimated from Gascuel et al. (2008) using a mean water temperature of 7.2°C and PP value of 0.312 g C d−1 (Proud et al., 2017) are indicated by dashed lines. For any given siphonophore density and gas bladder size distribution, global fish biomass values can be predicted for each gas bladder scenario using Figure 6, e.g. for a global population of siphonophores with normally distributed gas bladder lengths with a mean < 0.6 mm (e.g. as in Lavery et al., 2007), equivalent to a aesr of c. 0.15 mm (6 and 8), fish biomass would make up close to 100% of mesopelagic sa for any given mean global density of siphonophores (<1000 individuals m−2). Conversely, for a mean gas bladder aesr larger than 1 mm (e.g. Barham, 1963; Pickwell, 1966), siphonophores are dominant, contributing almost 100% of mesopelagic sa for a given mean open-ocean siphonophore density, larger than c. 6.5 individuals per m2 (e.g. Robison et al., 1998). At 10% TE per trophic level, mesopelagic fish biomass is c. 3.363 Gt (Figure 6)—a value which falls between a fish contribution of 40 and 50% under S1, between 30 and 40% for S2 and between 20 and 30% for S3 calculated from model median fish biomass values (Table 3). Non-acoustic estimates of global mesopelagic fish biomass (Gjøsaeter and Kawaguchi, 1980; Jennings and Collingridge, 2015; Anderson et al., 2019) suggest that the contribution of siphonophores to total mesopelagic backscatter may be 50% or more (Table 3). Sensitivity analysis The total effect index STi was calculated for input parameters used in the maximum fish biomass model run for each scenario (Figure 7).The area-backscattering coefficient for mesopelagic fish, saf was set to the global mean value, 1.94 × 10−5 m2 m−2 (Proud et al., 2017). This constraint was applied because the model would be very sensitive to a distribution that varies the total saf value between 0 and the maximum value, 2.38 × 10−5 m2 m−2; including saf as a distribution rather than a constant in the sensitivity analysis leads to unclear results, as STi for saf tends to 1 and the values for the other parameters are very small. For scenarios S1 and S2, the results show that swimbladder volume (as a proportion of body size), pswb, is the most important parameter, followed by the aspect ratio, α, and length distribution parameter, σX2 ⁠. Fish depth (z) and viscosity (ξ), affecting the resonant frequency (5) and damping of the resonant peak (Equation 4 and Figure 3), respectively, contribute relatively little to the overall model uncertainty. At an individual level, these parameters are very important (Scoulding et al., 2015) but this importance reduces substantially when considering the full range of the parameter space, which includes, for example, variability in population structure (e.g. shape of length-frequency distribution). For S3, pswb, and σX2 are the most important parameters, i.e. α has a reduced impact on model uncertainty. Figure 7. Open in new tabDownload slide The total effect index STi of the mesopelagic fish biomass model input parameters for scenarios S1–S3: distribution variance (⁠ σX2 ⁠), swimbladder volume as a proportion of body volume (pswb), depth (z), aspect ratio (α), dynamic viscosity of gas bladder wall (ξ), and minimum proportion of gas-bladdered fish (φmin). In this model run, mesopelagic fish sa, saf ⁠, is constant, set to the global mean value of 1.94 × 10−5 m2 m−2. Figure 7. Open in new tabDownload slide The total effect index STi of the mesopelagic fish biomass model input parameters for scenarios S1–S3: distribution variance (⁠ σX2 ⁠), swimbladder volume as a proportion of body volume (pswb), depth (z), aspect ratio (α), dynamic viscosity of gas bladder wall (ξ), and minimum proportion of gas-bladdered fish (φmin). In this model run, mesopelagic fish sa, saf ⁠, is constant, set to the global mean value of 1.94 × 10−5 m2 m−2. Discussion By exploration of likely echo energy levels arising from mesopelagic organisms, characterized using acoustic scattering models, it became apparent that siphonophores and fish were likely to be the dominant scatterers in the mesopelagic zone during the daytime (Figure 3). Our model results predict a range of global mesopelagic fish biomass values between 1.812 (lower quartile of S1) and 15.962 Gt (upper quartile of S3). The median biomass value of S1, 3.833 Gt, is our equivalent of a previous median acoustic biomass estimate of between 14.3 and 19.5 Gt (Irigoien et al., 2014, extrapolated from those authors’ 40° S to 40° N geographical range, to “our” 70° S to 70° N range). Our lower values are a consequence of acknowledging that a proportion of the total acoustic backscatter is resonant (high intensity echoes from low biomass targets), that siphonophores contribute to the total backscatter, and the uncertainty in population characteristics (i.e. species’ morphology and length distribution). For S3, where the proportion of gas-bladdered fish is reduced for small and larger fish, our prediction of median biomass of 8.292 Gt differs from Irigoien et al. (2014) by only a factor of two. Due to escapement and avoidance, the global biomass estimate by trawling of 1 Gt (Gjøsaeter and Kawaguchi, 1980) could be out by factor of seven or more (Koslow et al., 1997; Kloser et al., 2009; Yasuma and Yamamura, 2010; Davison, 2011a), which allows a prediction of c. 7 Gt or more: that is very close to the median value of S3. Conversely, a recent simple food-web model predicted mesopelagic fish biomass to be just 2.4 Gt (Anderson et al., 2019), which is within the biomass ranges of S1 and S2. Considering that S2 is probably the more likely of our scenarios (Butler and Pearcy, 1972; Neighbors and Nafpaktitis, 1982; Davison et al., 2015), the uncertainty in our acoustic derived estimate could be reduced to between 2.091 and 8.903 Gt (lower quartile to upper quartile). This range also overlaps with the range derived using a macroecological model, i.e. a median biomass of 1.4 Gt for all consumers, with 95th percentile of 8.1 Gt (Jennings and Collingridge, 2015). Model sensitivity Given that the relative proportions of fish and siphonophore backscatter are known, swimbladder volume (as a proportion of body size), pswb, was the most sensitive parameter in the mesopelagic fish biomass model for all three scenarios (Figure 7). This is not surprising, since the gas bladders of mesopelagic fish are predominantly smaller than the wavelength of sound at 38 kHz (c. 4 cm) and therefore volume, not shape or orientation, drives TS (i.e. Rayleigh scattering, see Figure 1). Other parameters in the model, such as the aspect ratio (α) and distribution variance (⁠ σX2 ⁠) were also important (Figure 7). To reduce the uncertainty caused by pswb in the model, we must first resolve the issue of compression during DVM, i.e. do fish inflate their bladders at depth (constant buoyancy strategy), to remain neutrally buoyant, or not? Evidence in the literature is mixed (Denton, 1961; Hersey et al., 1962; Kalish et al., 1986; Thompson and Love, 1996; Love et al., 2003, 2004; Scoulding et al., 2015). The problem can only be solved by making more observations of the mesopelagic community at depth, using for example, paired optical/acoustic systems (e.g. Marouchos et al., 2016), which will improve our knowledge and help narrow the distributions of the other model parameters. Model caveats The mesopelagic fish biomass model (20) derived here assumes that siphonophores are the only other significant contributor to mesopelagic sa. In most cases, this is probably a valid assumption (Barham, 1963; Lavery et al., 2007; Kloser et al., 2016) but, where fish densities are low, other scatterers such as squid and jellyfish will become more prominent (Clarke, 1996; Pauly et al., 2009; Haraldsson et al., 2012). In some instances, e.g. the polar regions, zooplankton populations may dominate (Murphy et al., 2007). At a global scale, the assumption is reasonable but to follow a similar approach as made here at regional or smaller scales, the contribution of other scatterers will need to be considered. Fortunately, our approach can readily incorporate contributions from other scatterers, indeed we started with multiple scatterers (Figure 3) and eliminated groups that contributed little to total scattering. We recommend this approach is adopted at regional scales and that results arising from such analyses are not extrapolated to other regions that likely contain different mesopelagic communities. We made a number of assumptions when selecting model parameters and distributions (Tables 1 and 2). We depended heavily on the literature which, in most cases, describes observations made only over small spatial ranges (i.e. local studies; e.g. Neighbors and Nafpaktitis, 1982; Yasuma et al., 2010). This reflects the paucity of data concerning mesopelagic fish and siphonophores, and highlights the pressing requirement for increased sampling at depth. The validity of our analysis results is also dependent upon our present level of knowledge, which is relatively poor (St John et al., 2016). For example, the density of mesopelagic fish (ρf) was assumed to be 1050 kgm−3, a reasonable median value taken from Love (1978). The value of ρf for mesopelagic fish varies between 1030 and 1080 kgm−3 (Capen, 1967; Butler and Pearcy, 1972; Davison, 2011b; Davison et al., 2015). Inputting this range of ρf values into our biomass model did not substantially change estimates of fish biomass (±2.5%). However, fish with higher densities are more likely to have inflated swimbladders, and that possible interaction is not considered in our analysis. As our understanding develops and we obtain more observational data, we may need to include additional parameters into our model framework. The largest source of uncertainty in the mesopelagic fish biomass model is the unknown contribution of siphonophore to mesopelagic sa. It is likely that fish produce most of the backscatter at 38 kHz but the proportion remains largely unknown. Siphonophores A range of siphonophore densities and gas bladder sizes have been observed, from very small siphonophores in the Gulf of Maine (0.15 mm mean gas bladder length, Lavery et al., 2007) and low-density populations <1 individuals m−2 (Mackie et al., 1988), to large Nanomia bijuga in the San Diego Trough (3.27 mm length, Barham, 1963) and high densities in the east Indian Ocean (>1000 individuals m−2, Musayeva, 1976). Siphonophores are often less abundant in open ocean than in neritic regions, for example Kloser et al. (2016) observed, using a lowered probe, just 2.5 individuals m−2 in the open Southern Ocean (Kloser et al., 2016). The total size of siphonophores varies from a few cm to several m but there is presently no understanding of how this size relates to gas bladder size at depth: variability in siphonophore size by cryptic species, population genetic variation, seasonality, or ecological conditions are also unknown (C. Dunn, pers. comm.). Against this background of uncertainty, to attribute legitimately large proportions of global mesopelagic sa to fish biomass, one of the following must be true: (i) in the open ocean, siphonophore densities are relatively low in the mesopelagic zone; (ii) after descent to depth, during DVM, a lot of gas-bearing siphonophores are not able to re-inflate their gas bladders at the now-high ambient pressure, or (iii) the majority of siphonophore gas bladders do not produce resonant backscatter at 38 kHz in the 200–1000 m depth range (e.g. 0.4 mm > aesr > 1.0 mm, Kloser et al., 2016). Presently, we are limited to small-scale visual estimates of siphonophores, from SCUBA to ROVs and submersibles (e.g. Rogers, 1978; Robison et al., 1998); the Monterey Bay Aquarium Research Institute does have an extensive database called the Video Annotation and Reference System (VARS), which contains records of siphonophores observed in ROV dives, since before 2000 (Schlining and Stout, 2006). To reduce uncertainty in estimates of fish biomass, more data on global variability in siphonophore density and size distribution are needed. Swimbladders The swimbladder state (present/absent/reduced/inflated) and volume (with respect to body size) of mesopelagic fish is highly variable between species (Butler and Pearcy, 1972; Neighbors and Nafpaktitis, 1982; Yasuma et al., 2010) and within species (Scoulding et al., 2015). Swimbladder function during DVM is also not well understood (Denton, 1961; Hersey et al., 1962; Kalish et al., 1986; Thompson and Love, 1996; Love et al., 2003, 2004; Scoulding et al., 2015). It is likely that some species adopt the “constant buoyancy strategy,” as observed by Barham (1971) from a submersible, where fish reside in a torpid state and remain neutrally buoyant during the daytime. The alternative is the “tread-water strategy,” whereby fish maintain depth by swimming (Love et al., 2004). The trade-off between the two strategies is the energetic cost of absorption and secretion of gas, which can be high (Bone et al., 1995) vs. the energetic cost of swimming to maintain depth. In addition, a fish in a torpid state, vs. a fish in constant movement, may be more difficult to detect visually, and hence, at lower risk of predation from deep-diving predators (e.g. King penguins and Elephant seals). Strategy may also change during the life cycle of mesopelagic fish, since density reduces with size (via lipid investment, e.g. Gee, 1983) and therefore, older, larger fish, are more likely to opt for the constant buoyancy strategy (Butler and Pearcy, 1972; Neighbors and Nafpaktitis, 1982; Davison, 2011b). An additional complication is that “resident” DSLs have often been observed (Figure 2) presumably containing some species that do not migrate, and the proportion of species that migrate do so seasonally with different proportions and length classes (Koslow et al., 1997; Flynn and Kloser, 2012). In the absence of any known environmental drivers of swimbladder volume and state in mesopelagic fish, our method is useful because it provides a mean view of a likely very complex system. To move forward and reduce uncertainty in estimates of mesopelagic fish biomass, a better understanding of variability at the individual level is required (e.g. TS variation with depth). If DVM behaviour, swimbladder state, and volume can be related to the environment, we will not only make more accurate estimates of mesopelagic fish biomass but also will gain a better understanding of how community-scale properties, such as vertical depth structure, emerge from the behaviour of individuals. Wider implications for ecosystem models The analysis framework developed here could be used to build an acoustic observation model (Handegard et al., 2013), to predict the expected acoustic “views” of simulated ecosystems, e.g. Atlantis, SEAPODYM, MIZER, and size-based ecosystem models (Lehodey et al., 2008, 2014; Fulton et al., 2011; Trebilco et al., 2013; Scott et al., 2014), and to compare those predictions with actual acoustic observations. This would serve to provide ecosystem modellers with a method to validate the mesopelagic component of their models. This is of particular importance for ecological/biogeochemical models that simulate the biological carbon pump and provide carbon fluxes for coupled climate models (Giering et al., 2014). Moving forward The predicted global mesopelagic sa used here was based on 38 kHz observations (Proud et al., 2017). A lot of data are available at a frequency of 38 kHz, but use of single frequency data alone does not enable frequency–response analyses that can identify scattering type (e.g. Kloser et al., 2002; Lavery et al., 2010). Data at 18 kHz, the only other commonly used frequency that has a high enough signal-to-noise ratio to provide useful observations from the entire mesopelagic zone, are often collected alongside 38 kHz data (www.imos.org.au). Using 18 and 38 kHz data together could enable resonance peaks to be identified, and the mean size of the target to be predicted. Performed at a global scale, this would at least provide some information concerning regional-scale size structure of gas-bearing organisms. Increased in situ optical and acoustic sensing in the ocean will advance the understanding of the depth distribution and abundance of siphonophores. As an example, profiling probes are being proposed (Handegard et al., 2010) and developed (Kloser et al., 2016). The gelatinous community (including siphonophores) is woefully under sampled and the incorporation of cameras on profiling probes will greatly increase our understanding of their distribution and abundance. In the future, combining such probes with acoustic and optical sensors could be done on a global scale in an ARGO float style of approach (Handegard et al., 2010). Concluding remarks We used predicted global 38 kHz DSL backscattering intensity (from Proud et al., 2017) to estimate global mesopelagic fish biomass. Our range of possible estimates spanned 1.812–15.962 Gt (lower and upper quartile). This range of values lends credence to the idea that there may be a substantial biomass of fish in the mesopelagic zone. Such a biomass could play a substantial role in the biological carbon pump, and could potentially bolster future food security. Uncertainty in mesopelagic fish biomass estimates could be reduced by (i) including more frequencies in the analysis to aid in determining size structure of resonant scatterers; (ii) development of an individual-based model to link DVM behaviour, weight/condition, swimbladder state, and volume to the environment, and (iii) obtaining more information on the size and depth distribution and density of siphonophores, both by collation of existing data and through the use of new technologies such as profiling acoustic optical systems (Marouchos et al., 2016). Acknowledgements This study has received support from the European H2020 International Cooperation project Mesopelagic Southern Ocean Prey and Predators (MESOPP, http://www.mesopp.eu/). The authors thank Ben Scoulding for providing code, Philip Pugh for helpful discussions concerning siphonophores, Gareth Lawson for help with jellyfish modelling, and Samuele Lo Piano and Sindre Vatnehol for assistance with the sensitivity analysis. We also thank the reviewers for helpful and informative comments. Funding Horizon 2020 Framework Programme, (Grant/Award Number: “692173”). References Aksnes D. L. , Røstad A., Kaartvedt S., Martinez U., Duarte C. M., Irigoien X. 2017 . Light penetration structures the deep acoustic scattering layers in the global ocean . Science Advances , 3 : 1 – 6 . Google Scholar Crossref Search ADS WorldCat Anderson T. R. , Martin A. P., Lampitt R. S., Trueman C. N., Henson S. A., Mayor D. J. 2019 . Quantifying carbon fluxes from primary production to mesopelagic fish using a simple food web model . ICES Journal of Marine Science , 76 : 690 – 701 . Google Scholar Crossref Search ADS WorldCat Andreeva I. B. , Galybin N. N., Tarasov L. L. 2000 . Vertical structure of the acoustic characteristics of deep scattering layers in the ocean . Acoustical Physics , 46 : 505 – 510 . Google Scholar Crossref Search ADS WorldCat Baik K. 2013 . Comment on ‘Resonant acoustic scattering by swimbladder-bearing fish’ [J. Acoust. Soc. Am. 64, 571–580 (1978)] (L) . The Journal of the Acoustical Society of America , 133 : 5 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Bardarson B. 2013 . Modelled Target Strengths of Three Lanternfish (Family: Myctophidae) in the North East Atlantic Based on Swimbladder and Body Morphology . University of St Andrews, Scotland . 97 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Barham E. G. 1963 . Siphonophores and the deep scattering layer . Science , 140 : 826 – 828 . Google Scholar Crossref Search ADS PubMed WorldCat Barham E. G. 1966 . Deep scattering layer migration and composition: observations from a diving saucer . Science , 151 : 1399 – 1403 . Google Scholar Crossref Search ADS PubMed WorldCat Barham E. G. 1971 . Deep-sea fishes: lethargy and vertical orientation. In Proceedings of an international symposium on biological sound scattering in the ocean, pp. 100–118. Ed. by G. B. Farquhar. U.S. Government Printing Office, Washington, DC. Bianchi D. , Galbraith E. D., Carozza D. A., Mislan K. A. S., Stock C. A. 2013 . Intensification of open-ocean oxygen depletion by vertically migrating animals . Nature Geoscience , 6 : 545 – 548 . Google Scholar Crossref Search ADS WorldCat Bone Q. , Marshall N. B., Blaxter J. H. S. 1995 . Biology of Fishes . Blackie , New York . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Brierley A. S. 2014 . Diel vertical migration . Current Biology , 24 : R1074 – R1076 . Google Scholar Crossref Search ADS PubMed WorldCat Butler J. L. , Pearcy W. G. 1972 . Swimbladder morphology and specific gravity of myctophids off oregon . Journal of the Fisheries Research Board of Canada , 29 : 1145 – 1150 . Google Scholar Crossref Search ADS WorldCat Capen R. L. 1967 . Swimbladder Morphology of Some Mesopelagic Fishes in Relation to Sound Scattering . Research Report 1447. US Navy Electronics Laboratory, San Diego, California . 1–31 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chu D. 2011 . Technology evolution and advances in fisheries acoustics . Journal of Marine Science and Technology , 19 : 245 – 252 . Google Scholar OpenURL Placeholder Text WorldCat Chu D. , Foote K. G., Stanton T. K. 1993 . Further analysis of target strength measurements of Antarctic krill at 38 and 120 kHz: comparison with deformed cylinder model and inference of orientation distribution . The Journal of the Acoustical Society of America , 93 : 2985 – 2988 . Google Scholar Crossref Search ADS WorldCat Clarke M. R. 1996 . The role of cephalopods in the world’s oceans: general conclusion and the future . Philosophical Transactions of the Royal Society B: Biological Sciences , 351 : 1105 – 1112 . Google Scholar Crossref Search ADS WorldCat Coetzee J. 2000 . Use of a shoal analysis and patch estimation system (SHAPES) to characterise sardine schools . Aquatic Living Resources , 13 : 1 – 10 . Google Scholar Crossref Search ADS WorldCat Cox M. J. , Watkins J. L., Reid K., Brierley A. S. 2011 . Spatial and temporal variability in the structure of aggregations of Antarctic krill (Euphausia superba) around South Georgia, 1997–1999 . ICES Journal of Marine Science , 68 : 489 – 498 . Google Scholar Crossref Search ADS WorldCat Davison P. C. 2011a . The Export of Carbon Mediated by Mesopelagic Fishes in the Northeast Pacific Ocean . University of California , San Diego . 149 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Davison P. C. 2011b . The specific gravity of mesopelagic fish from the northeastern Pacific Ocean and its implications for acoustic backscatter . ICES Journal of Marine Science , 68 : 2064 – 2074 . Google Scholar Crossref Search ADS WorldCat Davison P. C. , Checkley D. M., Koslow J. A., Barlow J. 2013 . Carbon export mediated by mesopelagic fishes in the northeast Pacific Ocean . Progress in Oceanography , 116 : 14 – 30 . Google Scholar Crossref Search ADS WorldCat Davison P. C. , Koslow J. A., Kloser R. J. 2015 . Acoustic biomass estimation of mesopelagic fish: backscattering from individuals, populations, and communities . ICES Journal of Marine Science , 72 : 1413 – 1424 . Google Scholar Crossref Search ADS WorldCat Denton E. J. 1961 . The buoyancy of fish and cephalopods . Progress in Biophysics and Molecular Biology , 11 : 177 – 234 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Flynn A. J. , Kloser R. J. 2012 . Cross-basin heterogeneity in lanternfish (family Myctophidae) assemblages and isotopic niches (δ 13C and δ 15N) in the southern Tasman Sea abyssal basin . Deep-Sea Research Part I: Oceanographic Research Papers , 69 : 113 – 127 . Google Scholar Crossref Search ADS WorldCat Flynn A. J. , Pogonoski J. J. 2012 . Guide to Mesopelagic Fishes of the Southern Tasman Sea . CSIRO Marine and Atmospheric Research, Hobart, Australia . 221 pp. Foote K. G. 1980 . Importance of the swimbladder in acoustic scattering by fish: a comparison of gadoid and mackerel target strengths . The Journal of the Acoustical Society of America , 67 : 2084 – 2089 . Google Scholar Crossref Search ADS WorldCat Foote K. G. 1987 . Fish target strengths for use in echo integrator surveys . The Journal of the Acoustical Society of America , 82 : 981 – 987 . Google Scholar Crossref Search ADS WorldCat Forland T. N. , Hobæk H., Korneliussen R. J. 2014 . Scattering properties of Atlantic mackerel over a wide frequency range . ICES Journal of Marine Science , 71 : 1904 – 1912 . Google Scholar Crossref Search ADS WorldCat Fulton E. A. , Link J. S., Kaplan I. C., Savina-Rolland M., Johnson P., Ainsworth C., Horne P. et al. 2011 . Lessons in modelling and management of marine ecosystems: the Atlantis experience . Fish and Fisheries , 12 : 171 – 188 . Google Scholar Crossref Search ADS WorldCat Gascuel D. , Morissette L., Palomares M. L. D., Christensen V. 2008 . Trophic flow kinetics in marine ecosystems: toward a theoretical approach to ecosystem functioning . Ecological Modelling , 217 : 33 – 47 . Google Scholar Crossref Search ADS WorldCat Gee J. H. 1983 . Ecologic implications of buoyancy control in fish. In Fish Biomechanics, pp. 140–176 . Ed. by Webb P. W., Weihs D.. Praeger , New York . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Giering S. L. C. , Sanders R., Lampitt R. S., Anderson T. R., Tamburini C., Boutrif M., Zubkov M. V. et al. 2014 . Reconciliation of the carbon budget in the ocean’s twilight zone . Nature , 507 : 480 – 483 . Google Scholar Crossref Search ADS PubMed WorldCat Gjøsaeter J. , Kawaguchi K. 1980 . A review of the world resources of mesopelagic fish . FAO Fisheries Technical Paper , I – 151 . Google Scholar OpenURL Placeholder Text WorldCat Handegard N. O. , Buisson L. D., Brehmer P., Chalmers S. J., De Robertis A., Huse G., Kloser R. et al. 2013 . Towards an acoustic-based coupled observation and modelling system for monitoring and predicting ecosystem dynamics of the open ocean . Fish and Fisheries , 14 : 605 – 615 . Google Scholar Crossref Search ADS WorldCat Handegard N. O. , Demer D. A., Kloser R., Lehodey P., Maury O., Simrad Y. 2010 . Toward a global ocean ecosystem mid-trophic automatic acoustic sampler (MAAS). In Proceedings of OceanObs’09: Sustained Ocean Observations and Information for Society 2. Ed. by J. Hall, D. E. Harrison, and D. Stammer. ESA Publication, Venice. Haraldsson M. , Tönnesson K., Tiselius P., Thingstad T. F., Aksnes D. L. 2012 . Relationship between fish and jellyfish as a function of eutrophication and water clarity . Marine Ecology Progress Series , 471 : 73 – 85 . Google Scholar Crossref Search ADS WorldCat Hersey J. , Backus R., Hellwig J. 1962 . Sound-scattering spectra of deep scattering layers in the western North Atlantic Ocean . Deep Sea Research Part I: Oceanographic Research Papers , 8 : 196 – 210 . Google Scholar OpenURL Placeholder Text WorldCat Heymans J. J. , Coll M., Libralato S., Morissette L., Christensen V. 2014 . Global patterns in ecological indicators of marine food webs: a modelling approach . PLoS One , 9 : e95845. Google Scholar Crossref Search ADS PubMed WorldCat Holliday D. V. 1972 . Resonance structure in echoes from schooled pelagic fish . The Journal of the Acoustical Society of America , 51 : 1322 . Google Scholar Crossref Search ADS WorldCat Holliday D. V. , Pieper R. E., Kleppel G. S. 1989 . Determination of zooplankton size and distribution with multifrequency acoustic technology . ICES Journal of Marine Science , 46 : 52 – 61 . Google Scholar Crossref Search ADS WorldCat Irigoien X. , Klevjer T. A., Røstad A., Martinez U., Boyra G., Acuña J. L., Bode A. et al. 2014 . Large mesopelagic fishes biomass and trophic efficiency in the open ocean . Nature Communications , 5 : 3271 . Google Scholar Crossref Search ADS PubMed WorldCat Jansen M. J. W. 1999 . Analysis of variance designs for model output . Computer Physics Communications , 117 : 35 – 43 . Google Scholar Crossref Search ADS WorldCat Jennings S. , Collingridge K. 2015 . Predicting consumer biomass, size-structure, production, catch potential, responses to fishing and associated uncertainties in the world’s marine ecosystems . PLoS One , 10 : 1 – 28 . Google Scholar OpenURL Placeholder Text WorldCat Kaartvedt S. , Staby A., Aksnes D. L. 2012 . Efficient trawl avoidance by mesopelagic fishes causes large underestimation of their biomass . Marine Ecology Progress Series , 456 : 1 – 6 . Google Scholar Crossref Search ADS WorldCat Kalish J. M. , Greenlaw C. F., Pearcy W. G., Holliday D. V. 1986 . The biological and acoustical structure of sound scattering layers off oregon . Deep Sea Research Part A, Oceanographic Research Papers , 33 : 631 – 653 . Google Scholar Crossref Search ADS WorldCat Klevjer T. A. , Irigoien X., Røstad A., Fraile-Nuez E., Benítez-Barrios V. M., Kaartvedt S. 2016 . Large scale patterns in vertical distribution and behaviour of mesopelagic scattering layers . Scientific Reports , 6 : 19873 . Google Scholar Crossref Search ADS PubMed WorldCat Kloser R. J. , Ryan T., Sakov P., Williams A., Koslow J. A. 2002 . Species identification in deep water using multiple acoustic frequencies . Canadian Journal of Fisheries and Aquatic Sciences , 59 : 1065 – 1077 . Google Scholar Crossref Search ADS WorldCat Kloser R. J. , Ryan T. E., Keith G., Gershwin L. 2016 . Deep-scattering layer, gas-bladder density, and size estimates using a two-frequency acoustic and optical probe. ICES Journal of Marine Science, 73 : 2037 – 2048 . Kloser R. J. , Ryan T. E., Young J. W., Lewis M. E. 2009 . Acoustic observations of micronekton fish on the scale of an ocean basin: potential and challenges . ICES Journal of Marine Science , 66 : 998 – 1006 . Google Scholar Crossref Search ADS WorldCat Koslow J. A. , Kloser R. J., Williams A. 1997 . Pelagic biomass and community structure over the mid-continental slope off southeastern Australia based upon acoustic and midwater trawl sampling . Marine Ecology Progress Series , 146 : 21 – 35 . Google Scholar Crossref Search ADS WorldCat Lavery A. C. , Chu D., Moum J. N. 2010 . Measurements of acoustic scattering from zooplankton and oceanic microstructure using a broadband echosounder . ICES Journal of Marine Science , 67 : 379 – 394 . Google Scholar Crossref Search ADS WorldCat Lavery A. C. , Wiebe P. H., Stanton T. K., Lawson G. L., Benfield M. C., Copley N. 2007 . Determining dominant scatterers of sound in mixed zooplankton populations . The Journal of the Acoustical Society of America , 122 : 3304 – 3326 . Google Scholar Crossref Search ADS PubMed WorldCat Lehodey P. , Conchon A., Senina I., Domokos R., Calmettes B., Jouanno J., Hernandez O. et al. 2014 . Optimization of a micronekton model with acoustic data . ICES Journal of Marine Science , 72 : 1399 – 1412 . Google Scholar Crossref Search ADS WorldCat Lehodey P. , Senina I., Murtugudde R. 2008 . A spatial ecosystem and populations dynamics model (SEAPODYM) – modeling of tuna and tuna-like populations . Progress in Oceanography , 78 : 304 – 318 . Google Scholar Crossref Search ADS WorldCat Love R. H. 1978 . Resonant acoustic scattering by swimbladder-bearing fish . The Journal of the Acoustical Society of America , 64 : 571 . Google Scholar Crossref Search ADS WorldCat Love R. H. , Fisher R. A., Wilson M. A., Nero R. W. 2004 . Unusual swimbladder behavior of fish in the Cariaco Trench . Deep Sea Research Part I: Oceanographic Research Papers , 51 : 1 – 16 . Google Scholar Crossref Search ADS WorldCat Love R. H. , Thompson C. H., Nero R. W. 2003 . Changes in volume reverberation from deep to shallow water in the eastern Gulf of Mexico . The Journal of the Acoustical Society of America , 114 : 2698 . Google Scholar Crossref Search ADS PubMed WorldCat Mackie G. O. , Pugh P. R., Purcell J. E. 1988 . Siphonophore Biology. Advances in Marine Biology, 24 : 97 – 262 . Maclennan D. N. , Fernandes P. G., Dalen J. 2002 . A consistent approach to definitions and symbols in fisheries acoustics . ICES Journal of Marine Science , 59 : 365 – 369 . Google Scholar Crossref Search ADS WorldCat Marouchos A. , Sherlock M., Kloser R. J., Ryan T., Cordell J. 2016 . A profiling acoustic and optical system (pAOS) for pelagic studies; Prototype development and testing. In OCEANS 2016 - Shanghai. IEEE. pp. 1–6. Marshall N. 1971 . Swimbladder development and the life of deep sea fishes. In Proceedings of an International Symposium on Biological Sound Scattering in the Ocean, pp. 69–73. Ed. by G. B. Farquhar. Department of the Navy, Washington, DC. Moser H. G. (Ed). 1996 . The Early Stages of Fishes in the California Current Region . Allen press , Lawrence, Kansas . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Murphy E. J. , Watkins J. L., Trathan P. N., Reid K., Meredith M. P., Thorpe S. E., Johnston N. M. et al. 2007 . Spatial and temporal operation of the Scotia Sea ecosystem: a review of large-scale links in a krill centred food web . Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences , 362 : 113 – 148 . Google Scholar Crossref Search ADS PubMed WorldCat Musayeva E. 1976 . Distribution of siphonophores in the eastern part of the Indian Ocean . Trudy Institufa Oceanologii , 105 : 171 – 197 . Google Scholar OpenURL Placeholder Text WorldCat Neighbors M. A. , Nafpaktitis B. G. 1982 . Lipid compositions, water contents, swimbladder morphologies and buoyancies of nineteen species of midwater fishes (18 myctophids and 1 neoscopelid) . Marine Biology , 66 : 207 – 215 . Google Scholar Crossref Search ADS WorldCat Netburn A. N. , Koslow J. A. 2015 . Dissolved oxygen as a constraint on daytime deep scattering layer depth in the southern California current ecosystem . Deep-Sea Research Part I: Oceanographic Research Papers , 104 : 149 – 158 . Google Scholar Crossref Search ADS WorldCat Pakhomov E. A. , Yamamura O. 2010 . Report of the Advisory Panel on Micronekton Sampling Inter-calibration Experiment. PICES Scientific Report, No 38. pp 108. Pauly D. , Graham W., Libralato S., Morissette L., Deng Palomares M. L. 2009 . Jellyfish in ecosystems, online databases, and ecosystem models . Hydrobiologia , 616 : 67 – 85 . Google Scholar Crossref Search ADS WorldCat Pickwell G. V. 1966 . Physiological dynamics of siphonophores from deep scattering layers: size of gas-filled floats and rate of gas production. U.S. Navy Electronics Laboratory, San Diego, California. 50 pp. Proud R. 2016 . A Biogeography of the Mesopelagic Community . University of St Andrews, Scotland . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Proud R. , Cox M. J., Brierley A. S. 2017 . Biogeography of the global ocean’s mesopelagic zone . Current Biology , 27 : 113 – 119 . Google Scholar Crossref Search ADS PubMed WorldCat Proud R. , Cox M. J., Wotherspoon S., Brierley A. S. 2015 . A method for identifying sound scattering layers and extracting key characteristics . Methods in Ecology and Evolution , 6 : 1190 – 1198 . Google Scholar Crossref Search ADS WorldCat Roberts C. M. , O'Leary B. C, McCauley D. J., Cury P. M., Duarte C. M., Lubchenco J., Pauly D. et al. 2017 . Marine reserves can mitigate and promote adaptation to climate change . Proceedings of the National Academy of Sciences , 114 : 6167 – 6175 . Google Scholar Crossref Search ADS WorldCat Robison B. H. , Reisenbichler K. R., Sherlock R. E., Silguero J. M. B., Chavez F. P. 1998 . Seasonal abundance of the siphonophore, Nanomia bijuga, in Monterey Bay . Deep-Sea Research Part II: Topical Studies in Oceanography , 45 : 1741 – 1751 . Google Scholar Crossref Search ADS WorldCat Rogers C. 1978 . Aggregation of the siphonophore Nanomia cara in the Gulf of Maine: observations from a submersible . Fishery Bulletin , 76 : 281 – 284 . Google Scholar OpenURL Placeholder Text WorldCat Rosenberg A. A. , Fogarty M. J., Cooper A. B., Dickey-Collas M., Fulton E. A., Gutierrez N. L., Hyde K. J. W. et al. 2014 . Developing new approaches to global stock status assessment and fishery production potential of the seas. FAO Fisheries and Aquaculture Circular No. 1086. FAO, Rome. 175 pp. Saltelli A. , Annoni P., Azzini I., Campolongo F., Ratto M., Tarantola S. 2010 . Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index . Computer Physics Communications , 181 : 259 – 270 . Google Scholar Crossref Search ADS WorldCat Schlining B. , Stout N. 2006 . MBARI’s video annotation and reference system. In OCEANS 2006, pp. 1–5. IEEE. Scott F. , Blanchard J. L., Andersen K. H. 2014 . mizer: an R package for multispecies, trait-based and community size spectrum ecological modelling . Methods in Ecology and Evolution , 5 : 1121 – 1125 . Google Scholar Crossref Search ADS PubMed WorldCat Scoulding B. , Chu D., Ona E., Fernandes P. G. 2015 . Target strengths of two abundant mesopelagic fish species . The Journal of the Acoustical Society of America , 137 : 989 – 1000 . Google Scholar Crossref Search ADS PubMed WorldCat Simmonds E. J. , MacLennan D. N. 2005 . Fisheries Acoustics: Theory and Practice . 2nd edn. Blackwell Science, Oxford . 437 pp. OpenURL Placeholder Text WorldCat Sobol I. 1967 . On the distribution of points in a cube and the approximate evaluation of integrals . USSR Computational Mathematics and Mathematical Physics , 7 : 86 – 112 . Google Scholar Crossref Search ADS WorldCat St John M. A. , Borja A., Chust G., Heath M., Grigorov I., Mariani P., Martin A. P. et al. 2016 . A dark hole in our understanding of marine ecosystems and their services: perspectives from the mesopelagic community . Frontiers in Marine Science , 3 : 1 – 6 . Google Scholar Crossref Search ADS WorldCat Stanton T. K. 1988 . Sound scattering by cylinders of finite length. II. Elastic cylinders . The Journal of the Acoustical Society of America , 83 : 64 – 67 . Google Scholar Crossref Search ADS WorldCat Strasberg M. 1953 . The pulsation frequency of nonspherical gas bubbles in liquids . The Journal of the Acoustical Society of America , 25 : 536 . Google Scholar Crossref Search ADS WorldCat Thompson C. H. , Love R. H. 1996 . Determination of fish size distributions and areal densities using broadband low-frequency measurements . ICES Journal of Marine Science , 53 : 197 – 201 . Google Scholar Crossref Search ADS WorldCat Trebilco R. , Baum J. K., Salomon A. K., Dulvy N. K. 2013 . Ecosystem ecology: size-based constraints on the pyramids of life . Trends in Ecology and Evolution , 28 : 423 – 431 . Google Scholar Crossref Search ADS PubMed WorldCat Warren J. 2001 . In situ measurements of acoustic target strengths of gas-bearing siphonophores . ICES Journal of Marine Science , 58 : 740 – 749 . Google Scholar Crossref Search ADS WorldCat Weston D. 1967 . Sound Propagation in the Presence of Bladder Fish, in Underwater Acoustics . Plenum , New York . 55–58 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Yasuma H. , Sawada K., Takao Y., Miyashita K., Aoki I. 2010 . Swimbladder condition and target strength of myctophid fish in the temperate zone of the Northwest Pacific . ICES Journal of Marine Science , 67 : 135 – 144 . Google Scholar Crossref Search ADS WorldCat Yasuma H. , Yamamura O. 2010 . Comparison between acoustic estimates. In Report of the Advisory Panel on Micronekton Sampling Inter-Calibration Experiment , pp. 51 – 57 . Ed. by Pakhomov E., Yamamura O.. North Pacific Marine Science Organization (PICES ), Sydney . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ye Z. 1998 . Low-frequency acoustic scattering by gas-filled prolate spheroids in liquids. II. Comparison with the exact solution . The Journal of the Acoustical Society of America , 103 : 822 . Google Scholar Crossref Search ADS WorldCat © International Council for the Exploration of the Sea 2018. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. © International Council for the Exploration of the Sea 2018.