ICES Journal of Marine Science: Journal du Conseil

Sakuramoto, Kazumi; Kitahara, Takeru; Sugiyama, Hideki

doi: 10.1006/jmsc.1996.0164pmid: N/A

The sandfish (Arctoscopus japonicus) catch in the coastal water off Akita Prefecture have fluctuated greatly over the period 1962–1990. The difference between the maximum and minimum catches over those years is more than 200 times. The aim of this paper was to investigate the relationships between water temperature and the fluctuations of the catch. Using the fuzzy logic approach, the water temperatures measured in the different points off the Prefecture were summarized as an index which would indicate the variation of water temperatures. In this study, the sensitivity tests showed that the index inferred using the water temperatures observed in 100–150 and 200–300 m depth at 9 and 65 km off the coast with time lags 1, 2, 3 and 4 years could explain the fluctuation of sandfish catch. That is, large changes in catch can be simply explained by the fluctuation of water temperatures.

Petrakis, George; Stergiou, Konstantinos I.

doi: 10.1006/jmsc.1996.0172pmid: N/A

In the present study, we compared the size selectivity of the 14 mm (knot-to-knot) diamond-shaped codend (14D), presently used in Hellenic waters, with those of 20 mm square-shaped (20S) and 20 mm diamond-shaped (20D) codends for four commercially important demersal species (Merluccius merluccius, Micromesistius poutassou, Trisopterus minutus capelanus and Lepidorhombus boscii ). Codend selectivity was estimated using the covered codend method. Our results suggested that 14D was not selective for any of the target species and the proportion retained by 14D ranged between 0.88 and 1. In addition, only in the case of M. merluccius did the results suggest that 20S is significantly more selective and retains fewer under-sized fish than 20D. With the exception of M. merluccius, the direct estimation of the 14D L50values was not possible, because of the small number of specimens retained by the cover codend of 14D. The L50value of M. merluccius for 14D was lower than its currently enforced minimum landing size (MLS) and length at 50% maturity. In contrast, for 20S and/or 20D codends, even though the estimated L50values were generally close to, or higher than, the species' MLS, the L50value of 20S for M. merluccius and that of 20D for L. boscii, were much smaller than their lengths at 50% maturity, a fact indicating that even the use of 20S and 20D, respectively, provides small opportunities for reproduction. Hence, the appropriate mesh sizes for these two species must be greater than 20 mm and their MLS must be raised. In contrast, for T. m. capelanus and M. poutassou, the L50values of 20D (but not of 14D) were close to their lengths at 50% maturity, a fact indicating that 20D is appropriate for these two species.The above mentioned facts clearly showed that: (a) 14D is harmful, in terms of size of landed fish, for the Hellenic demersal stocks and hence its use should be discontinued and (b) the use of a single appropriate mesh size for the Mediterranean trawl fishery as a whole is impossible because of its multispecies nature and the different body sizes of species involved. Since the majority of the individuals escaping through the meshes of 20D or 20S are immature and under-sized (i.e. their length is smaller than MLS), the replacement of 14D by either 20D or 20S, the former being the codend mesh size proposed by EU for the entire Mediterranean, will be accompanied by increased yield per recruit.

Van Damme, P. A.; Geets, A.; Hamerlynck, O.; Ollevier, F.

doi: 10.1006/jmsc.1996.0187pmid: N/A

Quantitative samples of the demersal fish fauna of the Oosterschelde were taken between May 1988 and July 1989. The instantaneous mortality rates of the 1988 year classes of whiting (Merlangius merlangus L.), sand goby (Pomatoschistus minutus Pallas) and bib (Trisopterus luscus L.) were estimated at 0.017 d−1, 0.014 d−1and 0.018 d−1, respectively. These three species were examined for the presence of mesoparasitic pennellid copepods. Combination of host population densities and parasite infection intensities allowed for estimation of parasite population densities. Whiting were infected with Lernaeocera branchialis shortly after their arrival (April–May) in the Oosterschelde. By the end of the year approximately 80% of the first year class whiting were infected. Spawning sand gobies were infected with Lernaeocera lusci in May and June. Though the prevalence of L. lusci on juvenile sand gobies in autumn was low, the parasite population density in this period was high. First year class bib were mainly infected by L. lusci after their transition from a pelagic to a demersal behaviour in August. It is suggested that L. branchialis has one generation each year, whereas L. lusci has two. A biological model of the life cycle of both parasite species is presented.

Pech, Nicolas; Laloë, Francis

doi: 10.1006/jmsc.1996.0180pmid: N/A

Principal Component Analysis with respect to Instrumental Variables (PCAIV) is a statistical tool for exploratory analysis combining both principal component analysis and multivariate regression analysis. This tool is used to analyse mean fortnightly catches obtained by Senegalese fishermen in two ports from 1975 to 1991. The aim of the study is to identify significant sources of variation and to present separately the impact of each of them. These descriptions are used to characterize the initial data.

Kimura, Daniel K.; Zenger, Harold H.

doi: 10.1006/jmsc.1996.0181pmid: N/A

From 1979 to 1994, the US National Marine Fisheries Service (NMFS) and the Fisheries Agency of Japan (FAJ, 1979–1988), and later the North Pacific Cooperative Fisheries Company (NPCFC, 1989–1994), have performed joint long-line surveys on chartered Japanese commercial fishing vessels in the Gulf of Alaska. These surveys were designed primarily to monitor the relative abundance of sablefish (Anoplopoma fimbria). To prepare for the future termination of these joint surveys, between 1988 and 1994 NMFS performed similar long-line surveys aboard US vessels in the Gulf of Alaska. These paired (i.e. duplicated) surveys from 1988 to 1994, provide a basis for comparing future NMFS survey results with the past time series of joint US and Japanese surveys (i.e. the joint 1979–1994 surveys). This paper describes how modeling the log-ratio of paired cpue results (i.e. the log[(NMFS cpue)/(joint cpue)]), from a standardized survey design, was used to provide a basis for converting the joint survey cpue values to values directly comparable to those from the NMFS surveys. This was necessary because sablefish long-line surveys will be carried out solely by NMFS in the future.

Kloser, R. J.; Williams, A.; Koslow, J. A.

doi: 10.1006/jmsc.1996.0171pmid: N/A

In situ target strength measurements of fish at 600 to 1200 m depth were made around a spawning aggregation of orange roughy off the east coast of Tasmania in 1992. The target strength data showed many modes, none of which could be definitely and uniquely attributed to orange roughy, partly because the orange roughy avoided the towed body housing the acoustics. Dominant modes at −50 and −55 dB were attributed to myctophid fishes with standard length modes at 8.2 and 5.3 cm; these fish have gas-filled swimbladders and were undisturbed by the towed body. Small modes at −44 dB and −31 dB were attributed to the macrourid Coryphaenoides subserrulatus and the morid Halagyreus johnsonii, respectively. The swimbladder of H. johnsonii is gas-filled, while that of C. subserrulatus contains a gas-filled spongy tissue matrix. No evidence was found of a separate mode at −36 or −41.3 dB, the previously reported target strength values of orange roughy. Modelling and tethered experiments on orange roughy suggested the target strength range for a 35 cm standard length fish was −47.2 to −53 dB. The modelling indicated values at the higher end of the range; measurements taken at depth of a tethered fish indicated the lower end. The dominant mode in the in situ data at −50 dB (which ranges from about −48 to −52 dB) was probably associated with orange roughy as well as myctophids. We concluded that the in situ target strength for a 35 cm standard length orange roughy is between −48 and −52 dB. Such a low target strength (compared to other species from the same depth that have gas-filled swimbladders) makes acoustic assessment techniques using echo integration very sensitive to the number of fish with gas-filled swimbladders.

Cotter, A. J. R.; Walker, P.; Coates, P.; Cook, W.; Dare, P. J.

doi: 10.1006/jmsc.1996.0182pmid: N/A

The effects on cockle (Cerastoderma edule L.) populations in Burry Inlet, South Wales of mechanical harvesting using a tractor dredger were investigated with an experimental trial conducted on 29 October 1992. Previously, only hand gathering methods were used, and the trial was intended to assist a licensing decision for mechanical dredging. Six blocks of dredged and undredged (control) plots were set out in each of two areas, one having a low density of cockles, the other high. Approximately 82% of the dredged areas was lifted by the blade of the dredger. The catch consisted almost exclusively of adult cockles (≥2 years old) over 25 mm in length. Appreciable losses of spat and one-year-olds from the dredged plots were also observed even though they were not taken in the catch. Possible reasons are discussed. Counts of damaged individuals rem-classes showed further mortalities attributable to dredging, and changes to shell growth were either minor or absent. Spatfall success in 1993 was depressed by 11% on dredged plots compared to that on control plots in the low density area, but was increased slightly (not significant p≥0.05) in the high density area. It is concluded that delayed effects of the dredging on cockle stocks were negligible.

Ellis, Tim; Nash, Richard D. M.

doi: 10.1006/jmsc.1996.0177pmid: N/A

The locations and timing of plaice Pleuronectes platessa spawning around the Isle of Man were studied from distributions of recently spawned eggs recorded in plankton samples. The area of intense spawning activity south-east of the Isle of Man was found to be further north than indicated in the literature. A previously disregarded spawning ground off the west coast of the Isle of Man was identified in 1991, 1992 and 1993, although egg densities were lower than off the east coast. Production of stage I eggs on the west coast spawning site was estimated at approximately 2% of the total plaice egg production in the Irish Sea. This would be sufficient to supply the local nursery grounds on the west coast of the Isle of Man with plaice larvae. Plaice spawned between early February and early April off the west coast of the Isle of Man, with March as the main period of egg production. Plaice therefore spawn well before the spring plankton bloom in the Irish Sea.

Grioche, A.; Koubbi, P.

doi: 10.1006/jmsc.1996.0162pmid: N/A

Larval fish surveys were made during spring 1993 along the Opale coast (northern France) to describe the hydrological influences on inshore–offshore ichthyoplankton assemblages. The French coast of the north-eastern English Channel is characterised by a coastal water mass, 3 to 5 miles wide, that is separated from offshore water by the presence of an unstable tidal front.Multivariate analysis was used to identify larval fish assemblages. Three kinds of larval distributions were found: (1) offshore larvae, whose distribution was defined by the position of the spawning area, (2) coastal larvae, which were mainly older larvae moving to coastal nursery grounds, and (3) inshore–offshore larvae, depending on the change of spawning areas during the spring.The influence of a tidally-dependent hydrological front was related to inshore–offshore larval exchange and larval development. The changing structure of the frontal zone influenced larval distribution and may be regarded as a regulating factor for the annual recruitment of different species.

Swartzman, G.

doi: 10.1006/jmsc.1996.0160pmid: N/A

The patterns of fish schools, mainly of Pacific hake (Merluccius productus), were studied in the Pacific Eastern Boundary Current during summer 1992. Differences between schooling patterns and depth preferenda of fish schools in the north and south were elucidated through regression analysis of depth, size, and biomass information about fish schools, extracted from acoustic backscatter data. Hake schools appeared to be clustered, based on both visual and statistical evidence. School clusters were larger (more schools) and had more biomass in the north than in the south, but inter-cluster spacing was also larger in the north. Temperature, bottom depth and school depth all significantly affect the biomass and number of schools in these clusters as the product of their effect on individual schools and on school density. The effect of these environmental factors on school cluster biomass and numbers differs from their effect on both individual schools and numbers of schools per arbitrary transect length (e.g. 5 nmi segments). School clusters may be a widespread phenomenon for schooling fish.