Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Aquatic habitat use in a semi-aquatic mammal: the Eurasian beaver

Aquatic habitat use in a semi-aquatic mammal: the Eurasian beaver Background: Semi-aquatic mammals exploit resources both on land and in water and may require both to meet their habitat requirements including food- and building resources, refuges, and for social interactions with conspe- cifics. Within this, the specific availability of both terrestrial and aquatic resources is expected to impact individual fitness. Beavers are highly dependent on water for movement and protection from predators. They are central place foragers and mostly forage on woody vegetation near water although aquatic vegetation may also be an important food resource. However, little is known about their use of aquatic habitats. We aimed to address this knowledge gap by dead-reckoning fine-scale movement tracks and classifying fine-scale diving events, which we then related to the spatial distribution of aquatic vegetation and habitat components within the territory. Results: Overall, there was a statistically clear decrease in probability that diving would occur at dawn and with increasing distance from territory borders. In addition, the distance from the lodge at which animals dived decreased through the night and during the spring/early summer. There was strong selection for diving habitats located closer to the riverbank, with stronger selection for these areas being observed in individuals with larger home ranges. We saw a higher selection for diving above clay sediment, and within 150 m from the lodge, presumably because mud and clay sediment tended to be located closer to the lodge than sand and rock sediment. Furthermore, we found a clear selection for diving in the presence of quillwort (Isoetes spp.), shoreweed (Littorella uniflora), and stonewort (Nitella spp.). Selection for these focal species was stronger among subordinate individuals. Individuals with lower body condition dived closer to the beaver lodge, and dives located further from the lodge were associated with high densities of aquatic vegetation. Conclusion: We provide new knowledge on the aquatic habitat use in a semi-aquatic mammal and show how energetic constraints may shape how beavers spatially use the aquatic environment, whereby short and shallow dives appear most beneficial. We show how aquatic habitats may have great importance for both foraging, building materi- als and safety, and discuss to how they may affect the fitness of individuals. Keywords: Aquatic foraging, Behavioural ecology, Castor fibre, Dead-reckoning, Habitat selection, Movement ecology, Resource selection functions Background Animal movements are affected by a suite of factors, including the cost of movement [1–6], likelihood of pre- *Correspondence: rmo@usn.no dation [7–11], resource distribution [12–14], reproduc- Faculty of Technology, Natural Sciences, and Maritime Sciences, tion [15, 16], and social interactions [12, 17–19]. This, Department of Natural Sciences and Environmental Health, University in part, explains why there is so much interest in animal of South-Eastern Norway, Bø i Telemark, Norway Full list of author information is available at the end of the article movement ecology, but elucidating causality to explain © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 2 of 19 animal movement is challenging as behavioural ecolo- be combined with fine-scale animal movement deter - gists have to juggle with multiple interacting factors [20, mined by dead-reckoning [49–51] to provide informa- 21]. tion on what animals do in the spaces they inhabit [52]. A major movement delimitator results from competi- Altogether, fine-scale information on behaviour and tion, when animals may exclude potential competitors movement, obtained from the use of acceleration and from an area by being territorial [22, 23]. As a result, dead-reckoning, respectively, are being used increas- some individuals end up defending habitats that do not ingly to study wild animals that are hard to observe have favourable combinations of essential resources, directly and/or without bias [44, 49, 52–56]. and thus have to trade-off resources within their ulti - Beavers are socially monogamous, monomorphic, mate habitat acquisition [24–26]. Consequently, some nocturnal mammals that inhabit various freshwater species benefit by being generalists rather than special - bodies [38, 57]. They live in family groups consist - ists, because the consequences of losing a resource are ing of the dominant breeding pair, kits of the year, and less severe since animals can exploit a greater variety of older non-breeding offspring [33, 58, 59]. Beavers reach resources [27–30]. For example, semi-aquatic mammals sexual maturity during their second winter [59], and have adapted to exploit resources on land and in water, give birth to one to five kits in mid-May at northern and individuals of this group may express considerable latitudes [57]. The kits emerge from the lodge in July plasticity to meet their habitat requirements according to when they start feeding on their own [33]. At around the available food resources, shelter, and social interac- 2–3.5  years old, beavers tend to disperse from their tions [31–33]. natal territory to establish their own territory [15, 60]. Habitat selection happens at various spatial scales [34] Beavers are central place foragers, mostly foraging near and is described as the use of resources (habitat) in a water and their lodges [61–63]. Their diet consists mainly manner that is disproportionate to their availability [35]. of woody vegetation but varies seasonally, and comprised Critically, resource-use may not necessarily be directly primarily bark from deciduous trees during winter to proportional to resource availability, but it may also be more nutritiously rich deciduous leaves, aquatic vegeta- modulated by other ecological factors such as competi- tion, and herbaceous plants in spring and summer [64– tion and predation [25, 36–38]. Although it has always 72]. In some areas, aquatic plants may seasonally account been a challenge to quantify which habitats animals have for up to 90% of the diet [65, 66, 73]. Aquatic vegetation available to them, and how much they use them, today, may offer some nutritional benefits over terrestrial veg - this information is important to inform space-value dis- etation, including better digestibility, higher crude pro- cussions so that confounding ecological variables affect - tein, and higher sodium and iron content [64, 66, 74, 75]. ing habitat value for the animal can be put into context Low concentrations of secondary compounds might also [39–41]. make aquatic vegetation more palatable [66], but this may In this study, we examine aquatic habitat use in Eura- vary with species [76]. Seasonally, rhizomes of aquatic sian beavers (Castor fibre) to identify important char - plants can provide great nutritional value in winter and acteristics of aquatic habitats within beaver territories spring when plants store nutrients in the rhizomes in and investigate potential differences in aquatic habitat preparation for spring growth [77, 78]. Diet variation may use among individuals. To achieve our aim, we combine depend on nutrient content and digestibility of available sophisticated animal-attached tags (GPS loggers and forage as individual beavers attempt to maximize energy Daily Diary units), that allow determination of animal intake over time [66, 76, 79–81]. Beaver foraging behav- behaviour with locations, with a comprehensive assess- iour varies according to environmental factors that affect ment of aquatic habitat characteristics. the distribution of food items [82], but ecological factors Technological developments have hugely enhanced such as food plant density, human disturbance, presence what animal biotelemetry can do for us, elucidating, of conspecifics, and predator activity may also affect their for example, fine-scale spatiotemporal location data on foraging choices and foraging locations [67, 83, 84]. No an increasing range of animals across various environ- clear dietary differences have been found between sexes, ments [40–43]. In particular, tri-axial accelerometers ages, or social ranks in beavers [65, 69, 85], but several are increasingly being used to study wild animals [44, studies indicate that foraging behaviour may differ sea - 45], because they allow determination of an individual’s sonally as territorial movements vary among individu- behaviour [46]. They have been used to classify behav - als [86–88]. Furthermore, individuals may be affected by iour and activity level patterns in beavers, distinguish- various ecological conditions during their lifetime, such ing seven behaviours with high precision, including as loss and acquisition of territories [15, 89], that can swimming and diving [47, 48]. These behaviours can affect their behavioural time-budgets and consequently their body condition, reproduction, and survival [57, 90]. M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 3 of 19 Water acts as a refuge for beavers [91] and is specifi - Using fine-scaled dead-reckoned animal tracks to cally used to minimize predation risk when foraging on determine spatial and temporal locations of aquatic dives, land [92]. Even though beavers depend on the aquatic we aim to examine important characteristics of aquatic environment for movement and safety [83, 86, 87, 93] habitat use by Eurasian beavers and investigate poten- and may even manipulate the environment to optimize tial individual differences. Assuming that dives indicate aquatic conditions [82, 94], aquatic behaviour and habitat aquatic habitat use in beavers, we hypothesize that habi- use have not been well studied in these animals. Although tat use vary temporally and spatially between individuals beavers use aquatic habitats for foraging [64–66, 71], pro- of various ages, sexes, social ranks, and by the composi- tection from predators [92, 95], and collecting resources tion of biotic and abiotic factors within their individual for lodge- and dam-building [96, 97], aquatic habitats territories. may be particularly important when terrestrial vegetation is difficult to access, or is of low nutritional quality [38, Methods 64, 66, 75, 98]. However, their spatial exploitation of the Study site aquatic components of their territory is poorly under- Our study site was located at the lower reaches of the stood. Research has found, however, that habitat use river Sauar in Vestfold and Telemark County, southeast- may differ between age groups and according to risk lev - ern Norway (Fig. 1). The river drains the lake Heddalsvat - els [87]. Diving behaviour has been studied using accel- net in the north and forms part of the catchment of the erometers, which has highlighted a preference for short lake Norsjø in the south, stretching over approximately (< 30  s) and shallow (up to 4  m, but most < 1  m) dives, 13  km with a width of 45–250  m. The river sections are which indicate some form of aquatic resource selection, generally slow-flowing with stable water levels because of although the link between diving and space use is vague natural lakes and man-made impoundments along part [99]. Being only semi-aquatic [cf. 100], beavers may expe- of its length [103], although flooding events frequently rience a higher cost exploiting aquatic resources [101, occur. The river flows through small villages, farmlands 102] than a fully aquatic equivalent may do. This may and fields interspersed with riparian woodland that com - explain why studies have found them to be diving for less prised mostly Norway spruce (Picea abies), Scots pine than 3% of their nightly activity budget [99]. However, (Pinus sylvestris), birch (Betula spp.), grey alder (Alnus energy requirements have been reported to compare well incana), aspen (Populus tremula) and mountain ash (Sor- with more fully aquatic mammals and birds [101]. bus aucuparia) [62, 103]. Fig. 1 a The location of the study site (red square) in Telemark and Vestfold County, Norway. b Overview of study river with random available sites (yellow circles) and identified beaver diving locations (red triangles) within beaver territories Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 4 of 19 Beavers have inhabited the area since the 1920s when maintain their social rank until they died or disappeared they recolonized the rivers [104]. The population is at from the territory [109]. carrying capacity, as territories of various sizes directly We captured and equipped nine beavers (five males, border each other [33]. Territory borders are identified four females, Table 1) with GPS loggers (Gipsy-5, Techno based on scent mound concentrations, sight observations Smart) and daily diary units (including accelerometer, of known patrolling beavers backed up by GPS data. The magnetometer, thermometer, Wildbyte technologies main river contains ten distinguishable territories inhab- [52]) in the spring and early summer of 2018 (one beaver) ited by approximately 32 individuals [33, 57]. Predation and 2019 (eight beavers). The data loggers were glued pressure is low as wolves (Canis lupus) and bears (Ursus onto the fur on the lower back of the beavers, approxi- arctos) are functionally extinct in the area, and lynx (Lynx mately 15 cm above the tail following the spine, and were lynx) only occur at low densities [84, 105]. removed again after 2 to 3  weeks if they had not fallen The study river is part of a larger monitoring project off by themselves [47]. To extend battery life, GPS log - where beavers in the area have been monitored through gers were programmed to take a fix position every 15 min an extensive capture programme, the Norwegian Beaver between 7  p.m. and 7  a.m. to reduce numbers of unsuc- Project (NBP), since 1997 [57]. The long-term monitor - cessful GPS fix attempts from within the beaver lodges. ing project aims to capture all newcomers (kits and dis- Daily diary units logged continuously at 40 Hz. From the persers from outside the study site) annually, enabling GPS positions, we identified territory borders to esti - identification of individuals at later encounters and fam - mate territory size, expressed in bank length (km) and ily group sizes. calculated 95% autocorrelated kernel density estimates (AKDE) to estimate the overall space use (home range) of Capture and tracking protocol each beaver [110]. Captured beavers were released near Individuals were detected from a motorboat using the capture site within their territory after approximately searchlights and captured at night with large diving- 40 min of handling time [106]. nets in shallow water or with land-nets [106]. Captured individuals were immobilized in cloth sacks, enabling Identification of dives easy handling without anaesthesia, and identified via Accelerometers in tandem with magnetometers can be microchips (PIT tag) and unique combinations of plastic used in dead-reckoning to accurately predict and recon- and metal ear-tags. Beavers were weighed to the near- struct animals’ fine-scale three-dimensional movement est 100  g. Body length was measured following the cur- paths in space and time by sequentially integrating cal- vature of the spine from nose tip to the base of the tail. culated travel vectors [49, 51]. However, the estimated Tail length was measured from the base to the tip of the movement track accumulates error and therefore drifts tail, and tail width was measured from edge to edge of over time, so it needs to be corrected through ground- the dorsal surface at the midpoint between tail base and truthing, e.g. correcting the track according to GPS fixes tip. Measurements of body length and tail proportions in [49]. cm were used to calculate tail fat index ((tail length × tail We calibrated the daily diary data in the software width)/body length), representing the body condition of DDMT (Daily Diary Multiple Trace, Wildbyte Technolo- beavers [57, 59]. gies). Using the acceleration and magnetism data, we Individuals were sexed based on the colour and vis- dead-reckoned the movement track of each beaver in the cosity of their anal gland secretion [107] and assigned a software Framework4 [111]. The dead-reckoned move - minimum age based on body mass at first capture [57, ment tracks were hereafter corrected using the GPS posi- 108]; minimum 2  years (subadult) when body mass was tions as ground-truthing [49]. GPS positions were filtered between 17 to 19.5  kg inclusive, and minimum 3  years to remove positions with horizontal dilution of precision (adult) when body mass was above 19.5  kg. Territorial (HDOP) values above five and with less than four avail - dominance was in most cases attributed to adult terri- able satellites to reduce the effects of imprecise GPS posi - torial residents of each sex. Territorial dominance was tions [112, 113]. verified by eventual dispersal of the alternative candidate, To identify diving locations, we divided the dead-reck- greatest body weight among same-sex group members oned movement tracks into ten second bursts and used or lactation in females (large nipples, i.e. > 0.5 cm). Indi- the acceleration to assign behavioural activities to each viduals dispersing into a territory were posited to have burst based on the acceleration-based behavioural classi- achieved the dominant breeding position when the pre- fication model by Graf et al. [47]. The classification model vious dominant same-sex individual had disappeared, or can clearly differentiate the acceleration between seven evidence outlined above was applicable. Unless proven behaviours: swimming, diving, sleeping, feeding, stand- otherwise, dominant individuals were assumed to ing, walking, and grooming. To furthermore filter out M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 5 of 19 Table 1 Overview of tracked individuals with daily diaries and GPS loggers in a population of Eurasian beavers in southeastern Norway Beaver Tracking period Nights Dives Territory Home range, ha Sex Age, years Social rank Mass, kg Body Tail fat index size, km length, cm Anna April 2–April 15 2019 13 92 4.9 1.3 F 7 Dominant 23.3 85.5 4.3 Ceasar April 2–April 9 2019 7 143 4.9 1.1 M 9 Dominant 17.7 81.0 3.9 Dylan May 9–May 11 2018 2 110 6.6 10.3 M 6 Dominant 21.2 82.0 4.0 Laurits April 26–May 7 2019 12 37 2.2 0.4 M 13 Dominant 19.0 77.5 3.8 Mason April 2–April 10 2019 7 49 3.4 2.2 M 6 Subordiante 17.9 72.0 4.2 Mattanja April 25–May 4 2019 9 89 2.2 4.0 M 3 Subordinate 18.0 72.0 3.9 Maximus April 25–April 29 2019 4 9 6.6 6.2 M 2 Subordinate 17.1 79.0 3.6 Tanja May 28–June 13 2019 17 269 4.4 1.1 F 16 Dominant 18.6 75.0 3.1 Tatjana April 2–April 6 2019 4 47 3.4 0.7 F 3 Subordinate 16.4 70.0 4.0 Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 6 of 19 potentially falsely classified dives, we focused on diving diving event was identified within a given hour, 0 = no sections where the previous 10-s bursts were additionally diving events were identified within a given hour) varied predicted to be swimming. through the night and between individuals. We analysed the effects of spatiotemporal variables (date, hour of the night, distance to riverbank, lodge, and territory border), Assessment of aquatic habitats individual differences (sex, social rank, age, body sizes, We visited and assessed the aquatic habitat of all div- and home range size), and between components of the ing locations as well as random aquatic locations avail- territories (territory size, mean water depth, mean veg- able within the territories of each beaver between June etation cover, and mean species richness). and October 2019 (Fig.  1b). We only sampled sites with We used a detrended correspondence analysis (DCA) a water depth of less than 10  m since vegetation growth from the R package ‘vegan’ v. 2.5–7 [117] to assess the here is more abundant due to light conditions [114], general distribution pattern of the aquatic vegetation in whilst also accounting for known diving depths in bea- the study river in relation to physical (water depth and vers [99]. sediment type), spatial (distance to riverbank, beaver For vegetation sampling at each location, sites were lodge, and territory border) and vegetation characteris- sampled using a 1 × 1  m quadrat with an aluminium tics (vegetation cover and species richness) of the sites. A frame. The quadrat was placed as close to the location matrix including aggregated abundance per species was as wind and currents allowed. The frame construction used for the DCA. However, species that only occurred had a pyramidal shape, enabling a GoPro camera (GoPro in five sites or less were removed from the matrix. The Hero5) to be attached to the top, 0.8  m above the sur- correlation between the aquatic species compositions face, which would keep the quadrat within the camera and physical, spatial, and vegetation characteristics of view. The quadrat was left at the bottom of the site after the sites were assessed by passively fitting them to the sediment settled, and pictures and films were recorded. ordination (permutations = 999). From the ordination, When water depth allowed it, aquatic plants and physi- comparing species with use, we could furthermore iden- cal characteristics were recorded in  situ using an aqua tify species that appeared to be of potential importance scope. Aquatic plants were collected with a rake when to the beavers when diving. For the subsequent resource identification required closer inspection. Plant species selection functions, we included variables for the number identification followed the database of Artsdatabanken of focal species present at a site and vegetation cover of [115]. Species abundance was quantified as coverage in these focal species. percentage, rounding to the nearest 5% [116]. Plants with We investigated the aquatic habitat selection within less than 5% cover were registered as 1% per species. territories using GLMMs with Bernoulli distribution, We categorized each site according to physical charac- logit link, and beaver ID as a random effect (1 = diving teristics [water depth and sediment type (clay, mud, sand, site, 0 = random available site within the territory) [34, and rock)], spatial characteristics (distance to riverbank, 35]. We analysed whether aquatic habitats located at beaver lodge, and territory border), and characteristics varying water depths, sediment types, distances to river- of the aquatic vegetation. We characterized the aquatic bank, beaver lodge, and territory border, or with varying vegetation by cover and species richness (number of spe- vegetation cover (overall and focal species), and spe- cies) to evaluate importance of quantity and diversity, cies richness (overall and focal species) were used more respectively. than was generally available in the territories [35, 118]. Additionally, we analysed the variations in diving selec- Statistical analysis tion among individuals in univariate models weighted by We used generalized linear mixed-effects models number of identified dives by fitting the resource selec - (GLMMs) with Poisson distribution, log link, and beaver tion function to each individual [119]. This enabled us to ID as a random effect to investigate how the number of analyse how the individual diving selection coefficients identified dives per night varied between individuals and varied between beavers of different age, sex, body size, components of the territories. We analysed the effects of and social rank that furthermore inhabit territories of sex (male, female), social rank (dominant, subordinate), different size and with varying amount of available water age (years), body size (body mass, body length, and tail fat depth, vegetation cover (overall and focal species), and index), and home range size (95% AKDE), and the effects species richness (overall and focal species) [13, 25, 119]. of territory size (bank length in km), mean water depth As dives may have different purposes according to their (m) and mean vegetation cover and species richness. spatial location, we furthermore investigated how log Using GLMMs with Bernoulli distribution, logit link, transformed distance to the lodge and riverbank varied with beaver ID and tracking night as random effects, we temporally (date and hour of the night), by environmental also investigated how hourly diving probability (1 = a M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 7 of 19 characteristics (depth, sediment, vegetation cover, spe- out using the R packages ‘glmmTMB’ v. 1.0.2.1 [122] and cies richness), and among individuals of different age, ‘MuMIn’ v. 1.43.17 [123]. The most parsimonious mod - sex, body size, social rank, territory size, and home range els within ΔAICc < 2 were chosen as the best models to by using GLMMs with gaussian distribution and beaver describe the variation [121, 124]. In each model, variables ID as a random effect. To examine characteristics associ - that included zero within their 95% confidence inter - ated to the proximate area of the beaver lodge, we fur- val (CI) were considered uninformative and reported as thermore analysed how dives within 150 m of the beaver unclear effects [124]. The best models were visually vali - lodge varied according to the above variables using uni- dated using the R package ‘DHARMa’ v. 0.4.1 [125] to variate GLMMs with Gaussian distribution and beaver plot standardized model residuals against the fitted val - ID as a random effect. ues [120] and, when relevant, furthermore checked for In all analyses, a list of candidate models was created zero-inflation. Top candidate models for all analyses can using ecologically relevant combinations of fixed effects be found in the supplemental material (Additional file  1). to account for variability in endogenous (such as sex, age, All analyses were conducted in R 4.0.3 [126]. and social rank) and exogenous factors (such as territory size, vegetation composition) that may be important in describing the ecology of beavers (Fig. 2). Because of the Results sample size, individual effects (sex, age, social rank, ter - Nine beavers were tracked with data loggers (Wild- ritory size, and home range size) should be interpreted byte technologies, Daily Diaries [52]) and GPS loggers with care as they only imply possible ecological effects (Techno Smart) affixed to the lower back for a total of that should be investigated with more individuals in 77 nights. Identified diving events lasted between 10 future studies. We included spatiotemporal interactions and 110  s, with the majority (80%) lasting 10  s or less. (between hour and distance to riverbank, lodge, and We identified on average (mean ± SD) 9.5 ± 3.1 dives per territory border, respectively) in the analysis for diving night for each beaver. We found no clear differences in probability, but excluded interactions in all other analyses the number of dives per night between males and females because of the limited sample size. Individual selection (10.9 ± 3.3 and 8.8 ± 3.0, respectively), among dominants coefficients were similarly analysed in univariate models and subordinates (12.9 ± 3.6 and 6.4 ± 2.5, respectively) because of the limited sample size. The fixed effects used or as a function of age, body size and tail fat index (Addi- in all analyses were not correlated (Pearson r coefficients tional file  1). Furthermore, territory size, home range less than 0.5) and variance inflation factor values were size, mean water depth in territory, and mean vegeta- less than 3 [120]. tion cover and species richness in territory did not have Model selection was based on Akaike’s information cri- a clear effect on the number of identified dives per night terion corrected for small sample size [121], and carried (Additional file 1). Fig. 2 Alluvial diagram showing how covariates included in the analyses (left) may relate to various ecological mechanisms (middle) which ecologically may have physical, behavioural, and social consequences (right) for the fitness of beavers Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 8 of 19 Table 3 Detrended correspondence analysis results for aquatic vegetation sites in a population of Eurasian beaver in southeastern Norway DCA1 DCA2 DCA3 DCA4 Eigenvalues 0.62 0.60 0.43 0.42 Axis lengths 6.15 4.94 4.30 3.51 Proportion explained % 9.63 9.39 6.78 6.52 Cumulative prop. explained % 9.63 19.02 25.80 32.32 Fig. 3 The predicted relationship ± 95% confidence interval between hourly diving probability, time of night, and distance from territory border among nine individuals in a Eurasian beaver population in southeastern Norway. Yellow boxes represent time of sunset and sunrise through the tracking period. Red boxes represent time of dusk and dawn through the tracking period Table 2 Eec ff t size (β), standard error (SE), lower (LCI) and upper (UCI) 95% confidence interval of explanatory variables for the analysis of nightly diving probability among nine individuals in a Eurasian beaver population in southeastern Norway Fig. 4 Detrended correspondence analysis (DCA) of aquatic Variable Estimate β SE LCI UCI vegetation sites (points) within territories of nine individuals in a Eurasian beaver population in southeastern Norway. Arrows represent Intercept − 2.970 0.652 − 4.247 − 1.692 passively fitted environmental gradients and red labels represent Hour − 0.114 0.002 − 0.119 − 0.110 environmental centroids. The green ellipse encircles species that may Log (distance to − 0.184 0.007 − 0.197 − 0.171 represent important resources at diving locations of the beavers territory border) Hour × log − 0.064 0.002 − 0.069 − 0.059 (distance to ter- ritory border) depth in territory and mean vegetation diversity within 2 2 Marginal R : 0.01 Conditional R : 0.84 the territory on nightly diving probability (Table 2, Addi- Effects were modelled using a GLMM with Bernoulli distribution. Beaver ID and tional file 1). tracking night were included as random effects. Informative parameters are given in bold Characterization of aquatic vegetation The DCA ordination described up to 32.3% of the vari - Diving probability ation in the aquatic vegetation composition within the Hourly diving probability varied through the night, beaver territories with DCA1 and DCA2 describing 9.6 decreasing over the final hours of the night (Fig.  3) and and 9.4% of the variation, respectively (Table  3). The with increasing distance from the territory borders aquatic vegetation within the territories showed great (Table 2, Fig. 3). We found no differences in hourly diving variation and differed clearly with increasing water depth, probability between sexes, social ranks, age, or relative to vegetation cover, species richness, and sediment type body size and tail fat index. Furthermore, we found no (Fig.  4). Diving sites and random available sites within clear effect of date, distance to riverbank, distance from the beaver territories clearly differed in species composi the lodge, territory size, home range size, mean water tion. Diving sites were especially associated with varying M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 9 of 19 amounts of quillwort (Isoetes spp.), shoreweed (Littorella with decreasing water depths, increasing species rich- uniflora), and stonewort (Nitella spp.) (Fig.  4). Further- ness, and decreasing vegetation cover. A list of all aquatic more, sites were predominantly more used for diving species can be found in Additional file 1. Resource selection of aquatic habitats within the territory Table 4 Eec ff t size (β), standard error (SE), lower (LCI) and upper We found a clear diving selection for aquatic locations (UCI) 95% confidence interval of explanatory variables for the closer to the riverbank but found no clear diving selection analysis of diving location selection among nine individuals in a as a function of distance from lodge, distance from terri- Eurasian beaver population in southeastern Norway tory border, or water depth (Table  4, Fig.  5a, Additional Variable Estimate β SE LCI UCI file  1). Diving selection probability varied as a function of sediment type. Locations having either mud, sand, or Intercept − 0.934 0.216 − 1.358 − 0.510 rock sediment were less selected than locations with clay Log (distance to riverbank) − 0.686 0.086 − 0.855 − 0.517 sediment (Table 4, Fig. 5b). Furthermore, diving selection Sediment 0.857 0.265 0.337 1.376 Clay probability increased when several of either quillwort, Sediment − 0.349 0.308 − 0.952 0.254 Sand shoreweed, and stonewort were present, but we found no Sediment − 0.461 0.286 − 1.022 0.100 Rocks effect of vegetation cover, overall species richness, or veg - Number of focal species 0.340 0.141 0.065 0.616 etation cover of the focal species on the diving selection present 2 2 probability (Table 4, Fig. 5c, Additional file 1). Marginal R : 0.19 Conditional R : 0.26 Individual diving selection coefficients varied among Effects were modelled using a GLMM with Bernoulli distribution. Beaver ID was individuals and territories. Individuals exploiting larger included as random effect. Informative parameters are given in bold home ranges (i.e. 95% AKDE) had a weaker selection for Reference level for sediment = Mud Fig. 5 The predicted relationship ± 95% confidence interval between diving selection probability and a distance to riverbank, b sediment type and c number of focal vegetation species present among nine individuals in a Eurasian beaver population in southeastern Norway Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 10 of 19 diving locations closer to the riverbank than individuals individuals than dominant individuals (Table  5, Fig.  6b). with smaller home ranges (Table 5, Fig. 6a). We found no We found no conditional effects of other variables on the clear context dependent effects on selection coefficients selection coefficients (Additional file 1). for the various sediment types (Additional file  1). Social We found that diving locations further from the riv- rank had an effect on diving selection for number of focal erbank had a deeper water depth and higher vegeta- species present which were stronger among subordinate tion cover of focal species compared to diving locations closer to the riverbank (Table  6, Fig.  7a, b). Diving loca- tions located further from the beaver lodge had a higher vegetation cover than diving locations closer to the bea- Table 5 Eec ff t size (β), standard error (SE), lower (LCI) and ver lodge (Table  6, Fig.  7c). Diving distance to the bea- upper (UCI) 95% confidence interval of explanatory variables ver lodge also increased with increasing tail fat index and for the analysis of individual selection coefficient for distance decreased during the night and during the spring/early to riverbank and number of focal species present among nine individuals in a Eurasian beaver population in southeastern summer (Table  6, Fig.  7d–f ). Focusing on dives within Norway 150 m of the beaver lodge, we found a difference among sediment types with dives located on clay and mud sedi- Variable Estimate β SE LCI UCI ment being closer to the lodge than dives located on sand Selection for distance to riverbank and rock (Table 6, Fig. 8). Intercept − 0.465 0.169 − 0.864 − 0.067 Log (home range) − 0.420 0.142 − 0.757 − 0.084 Discussion 2 2 R : 0.56 R : 0.49 adjusted We provide new knowledge on aquatic habitat use in a Selection for number of focal species present semi-aquatic mammal, the Eurasian beaver, by examin- Intercept 0.175 0.131 − 0.135 0.485 ing finely resolved information on beaver movement and Social rank 0.922 0.228 0.384 1.460 subordinate diving in relation to fine-scaled qualitative assessments of 2 2 R : 0.70 R : 0.66 adjusted aquatic habitat characteristics within beaver territories. Effects were modelled using a GLM with gaussian distribution. Analyses were We observed clear spatiotemporal variations in hourly weighed by number of diving sites for each individual. Informative parameters diving probability but found no differences among indi - are given in bold viduals or territories. Beavers selected for both spatial, Reference level for social rank = Dominant Fig. 6 The predicted relationship ± 95% confidence interval between a selection coefficients for distance to riverbank and home range size (AKDE, autocorrelated kernel density estimate) and b selection coefficients for number of focal vegetation species present and social rank among nine individuals in a Eurasian beaver population in southeastern Norway M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 11 of 19 Table 6 Eec ff t size (β), standard error (SE), lower (LCI) and upper [101, 129–131]. Beavers may preferably dive to bring (UCI) 95% confidence interval of explanatory variables for the large quantities of vegetation to the surface rather than analyses of distance to riverbank and distance to beaver lodge consuming small amounts of vegetation underwater, sav- among dives of nine individuals in a Eurasian beaver population ing energy by not diving repeatedly to exploit resources in southeastern Norway at depth, thereby also minimizing heat loss [132, 133]. In Variable Estimate β SE LCI UCI fact, an extensive study on diving in beavers by Graf et al. [99] found that animals had high dynamic body accelera- Distance to riverbank tion (DBA) (a good proxy for movement-based energy Intercept 2.226 0.071 2.087 2.365 expenditure [134]) for the descent, indicating work Log (water depth) 0.603 0.053 0.499 0.708 done against appreciable buoyancy, as has been noted Vegetation cover of 0.010 0.003 0.005 0.015 for many birds with their air-filled plumage [133]. Curi - focal species 2 2 ously though, in stark contrast to birds, which use this Marginal R : 0.35 Conditional R : 0.35 buoyancy for passive ascents [135], beavers also had high Distance to beaver lodge DBA values during their return to the surface, which was Intercept 5.801 0.410 − 0.375 11.976 suggested to be due to animals having to transport veg- Vegetation cover 0.006 0.002 0.002 0.009 etation from the bottom to the surface for consumption Hour − 0.071 0.023 − 0.116 − 0.025 [99]. Similar behaviour is commonly observed in water Julian day − 0.019 0.003 − 0.025 − 0.012 birds such as Eurasian coots (Fulica atra) when foraging Tail fat index 1.059 0.023 0.635 1.483 on aquatic vegetation [136, 137], but also in semi-aquatic 2 2 Marginal R : 0.18 Conditional R : 0.19 carnivores such as American mink (Neovison vison) that Distance to beaver lodge (dives within 150 m) occasionally consume aquatic food items at the surface or Intercept 4.227 0.091 4.049 4.404 on the riverbank [128, 138]. Critically, this behaviour is Sediment − 0.098 0.152 − 0.395 0.199 Clay most advantageous when large amounts of food can be Sediment 0.670 0.240 0.201 1.140 Sand brought to the surface during one dive, which can then be Sediment 0.309 0.147 0.021 0.597 Rocks consumed at leisure without the need for multiple, ener- 2 2 Marginal R : 0.25 Conditional R : 0.25 getically onerous dives in repetitive feeding bouts such as Effects were modelled using a GLMMs with gaussian distribution. Beaver ID was those performed by carnivorous species [138, 139]. included as random effect. Informative parameters are given in bold We found a decreased hourly diving probability in the Reference level for sediment = Mud early morning which is contrary to the findings of Graf et al. [99], although they had appreciable variation. Other studies have found a peak in general activity (meas- physical and vegetation characteristics in their diving ured via overall body dynamic acceleration) in the mid- locations, highlighting the degree of choice they exercise dle of the beavers’ principal activity period, suggesting for foraging behaviour although selection strength var- increased activity in the middle of the night [48]. Diving ied between individuals. Furthermore, spatial variations generally has a high overall body dynamic acceleration among dives indicate the energetic variability in aquatic compared to other behavioural activities and therefore habitat use. Often studies in freshwater-inhabiting semi- a higher movement-based energy cost [47, 134]. Div- aquatic mammals focus on the use of terrestrial habitat ing patterns in beavers may be implicitly represented by components, but we show how components of aquatic this general activity pattern, peaking in the middle of the habitats similarly may be an important resource which principal activity period, as aquatic habitat use may be can potentially have considerable fitness consequences costly for a semi-aquatic mammal [101, 140]. Similar div- for a semi-aquatic mammal like the beaver. ing patterns peaking in the middle of the activity period have been found in other semi-aquatic mammals, such Diving patterns as American mink, that perform temporal niche shifts The majority of our identified diving events were short, to avoid interspecific aggression from competitors [128, which matches previous findings for beavers [99, 101] 141]. and other semi-aquatic mammals [127–129]. Beavers Diving patterns may be structured according to the have previously been reported to spend less than 3% of activity peaks of potential terrestrial predators [142]. their nightly activity budget on diving activities which When beavers bring aquatic resources onto the riverbank corresponds well with their role as generalist herbivores to be handled, their risk of predation increases [93, 95]. that do not rely solely on aquatic foraging [99]. Similar Beavers are at a particular risk when on land because diving patterns have been found among semi-aquatic of their poor eyesight under low light conditions (i.e. generalist carnivores and may relate to semi-aquatic ani- they lack tapetum lucidum) [143] and dependence on mals being less specialized for the aquatic environment Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 12 of 19 Fig. 7 The predicted relationship ± 95% confidence interval between distance to riverbank and beaver lodge and a water depth, b vegetation cover of focal species, c vegetation cover, d hour of the night, e Julian day, and f tail fat index among dives of nine individuals in a Eurasian beaver population in southeastern Norway. Points represent actual distances olfaction to detect potential risks [144]. Consequently, active (including diving) in the middle of the night when they may not detect potential predators, such as wolves it is darkest. In North America, beavers make up a large (Canis lupus) and Eurasian lynx (Lynx lynx) that have proportion of wolf diet, especially during summer when advanced night vision [145, 146], before being detected wolves have been observed to ambush beavers at fre- themselves. However, wolves have been observed to quently used locations [95]. Natural predators are absent predate more at dawn, dusk, and during moonlit nights in our study area, but behavioural activities are known to [142], which could increase the benefits of being most be influenced by historical threats [84, 147]. In addition, M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 13 of 19 be treated as a central place [152], which makes clear the energetic costs of leaving the central place to acquire food. In that sense, the decreasing tendency to dive with increasing distance from the riverbank ties in with opti- mal foraging theory, which has animals maximizing reward and minimizing transit time and energy [153, 154]. Distance to the riverbank may also reflect decreas - ing aquatic foraging options in terms of decreased macro- phyte growth which is highly dependent on water depth and light penetration [114]. But short shallow dives may also be preferred as they are energetically cheaper for semi-aquatic animals like beavers that have high buoy- ancy [140]. In general, diving locations in semi-aquatic carnivores have been shown to follow distributions of aquatic food resources [127, 155]. We noted that selec- tion for diving locations near the riverbank was statisti- cally stronger among individuals that exploited a larger home range: individuals that exploit a larger home range may have a greater need to stay closer to the riverbank Fig. 8 The predicted relationship ± 95% confidence interval due to higher territory patrolling efforts whereas indi - between distance to beaver lodge and sediment type among dives viduals exploiting smaller home ranges may be able to within 150 m of the beaver lodge of nine individuals in a Eurasian beaver population in southeastern Norway. Points represent actual forage further away from the riverbank [86]. Individuals distances using larger areas or inhabiting larger territories may also have reduced resource depletion so that, conversely, bea- vers restricted to smaller areas may be forced to exploit human activities both on land and on water, which are foraging areas further away from the riverbank [33, 156]. naturally reduced at night, may also influence the activ - Larger home ranges may also have a greater area of shal- ity levels of beavers [147]. Consequently, diving for food low water, which, depending on the time spent diving, resources at dusk and dawn may be perceived to be too can be energetically easier to exploit for a semi-aquatic risky for beavers. animal [140]. Conversely, individuals with smaller home Diving is presumably not just shaped by risks, but also ranges may have to exploit all habitats to fulfil their ener - allows the animals to have access to important aquatic getic requirements [33]. We saw that water depth at the resources [4, 81]. Hourly diving probability decreased diving locations generally increased with increasing dis- with increasing distance from the territory borders which tance from the riverbank, which presumably represent may indicate a possible depletion of aquatic resources higher energetic costs of diving away from the riverbank, near beaver lodges that are often located in central but dives at longer distances from the riverbank also had parts of the territory [cf. 148,149]. Although this may be higher amounts of quillwort, shoreweed, and stonewort expected to be related to territory size [33], we did not present, possibly due to the Ashmole’s halo effect [148, find this, possibly because our sample size was too small. 149] operating on animals preferentially associated with Diving near the borders may also relate to territorial the shoreline. This indicates the interplay of depth and defence activities, which may help reduce aggressive ter- distance energetics in a semi-aquatic animal, which have ritorial encounters when territorial intruders swim away to be balanced with the calorific value of the food plants unseen [89, 150, 151]. However, we were unable to quan- and their location [132, 133, 140]. tify territorial behaviour. Beavers had a higher selection for diving at locations with clay sediment, which may be an important building Diving selection material for lodges and dams [96], although dams are not We found a high selection for diving closer to the river- present in our study site, and beavers may additionally bank. Similar short and shallow dives have been found make use of burrows dug into the riverbank whereby the in other studies of semi-aquatic mammals which prob- clay sediment aids in enhancing the structural integrity ably reflect the energetic constraints of bringing food of those burrows [157, 158]. Lodges and dams are mostly resources to the riverbank for consumption and a pref- repaired in the autumn [159, 160], but may be repaired erence for travelling along the riverbank [86, 99, 127, after flooding events too [161]. Mud is also widely used 128]. Association with a riverbank can, in some senses, for beaver constructions [97], although mud substrates, Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 14 of 19 despite being highly abundant in the study area, seemed their foraging strategy accordingly [4, 168]. Similarly, we not to be selected. Beavers continuously need to apply found that beavers with lower tail fat index tended to dive fresh mud and other fine debris to seal their construc - closer to their lodge, which may be a consequence of ter- tions as it is continuously washed away [96, 97]. We ritorial constraints as individuals with higher body condi- found that dives on mud and clay sediment within 150 m tion may be better able to cope with the increased cost of of the beaver lodge were generally located closer to the patrolling and protecting territory borders [18, 86]. Div- beaver lodges than dives located on sand and rock sedi- ing locations located further from the beaver lodge also ment which, we believe, indicates the importance of fine had higher amounts of vegetation cover which indicate sediment for building constructions. However, we cannot the energetic trade-offs between costs and benefits that rule out that mud and clay sediment may contain advan- a central place foraging individual experiences [4, 169]. tageous foraging options, although our ordination did Diving distance to the lodge decreased through the night not show strong correlation between mud and clay sedi- which may indicate a functional change in the purpose ment with aquatic species richness or vegetation cover. of the dive. As dives further from the lodge occurred at We found a higher selection for diving locations with locations with more aquatic vegetation, these dives may presence of species of quillwort, shoreweed, and stone- be intended for foraging whereas dives later in the night wort, which may represent important food resources may be used for building activities. We also saw that div- for the beaver. Other studies have similarly found a high ing distance to the lodge decreased from early spring to preference for quillwort in the early summer [73], and summer, which may relate to increased parenting activi- algae like stonewort may be selected for its protein con- ties that require the beaver to stay closer to the lodge tent and other nutrients [65]. Although beavers mainly when the kits are born in mid-May [58]. Diving for food forage on woody vegetation, aquatic vegetation seems resources may be perceived as less risk-taking than going to be seasonally important, with studies additionally on land [87] and could be preferred by parenting individ- reporting beavers foraging on, among others, water lilies uals to ensure the growth and survival of their offspring (Nymphaea spp.), pondweed (Potamogeton spp.), water [170]. horsetail (Equisetum spp.), waterweed (Elodea spp.), and water lobelia (Lobelia dortmanna) [64–66, 68, 69, Methodological limitations 71, 98]. Incorporating aquatic vegetation into a varied Using dead-reckoning together with an acceleration- diet could be a strategy to minimize risk of nutrient defi - based behavioural classification model to identify the ciency [162], which may be seasonally beneficial to some temporal and spatial distribution of clear diving events, individuals (e.g. females during lactation when energetic we identified considerably fewer diving events than a pre - requirements increase) [163, 164]. However, we did not vious study in beavers that identified typically 40 dives find any selection differences between males and females per night [99]. This may be related to seasonal variations which may be because of our limited number of individu- as we only tracked beavers from April to June, whereas als. Other studies on semi-aquatic mammals show how Graf et  al. [99] also included observations from the males and females in two species of shrew (Neomys fodi- autumn (September to October) where diving conditions ens and Sorex coronatus) use separate foraging habitats may be more favourable, because of increased water tem- during the breeding season [165]. Males and females in perature and life history patterns [15, 57, 99]. But dives our population have been shown to differ seasonally in may also be masked by our method. Despite a high classi- aquatic foraging with peaks in the spring and late sum- fication accuracy of diving events from the model by Graf mer for females, whereas males only foraged on aquatic et  al. [47], some events may not be identified because vegetation in the spring [166]. We found that selection for some behavioural activities, including behaviours not the focal plant species were statistically stronger among described by the model, can have similar patterns and subordinate individuals, which may be linked to their mask each other [47]. Therefore, it is important to clearly higher energetic requirements resulting from their activi- define each behavioural activity in an acceleration-based ties related to attempts to become dominant in a territory classification model, but also to include enough varia - [57] (e.g. performing more extra territorial movements tions of each activity from several individuals to improve [15]). A higher use of aquatic vegetation may also be a the precision of the model. Different diving styles may risk-avoiding strategy, minimizing predation risk on land be misclassified as other behavioural categories if the [92, 95]. In other studies, adult beavers have been found behavioural classification model is not trained on several to forage less on aquatic vegetation than subadult indi- variations of each behavioural category but only includes viduals (i.e. 2-year-old) [65] which may be less risk-will- typical acceleration patterns for each behavioural cat- ing as they potentially face higher fitness costs in terms egory. For example, the ecological difference between of future reproductive success [167] and therefore adjust a ‘dive’ and an ‘almost dive’ may be minimal when a M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 15 of 19 beaver can access aquatic resources by just sticking its beavers in a Eurasian beaver population in south-eastern Norway. S2. The head underwater, but they may fall within two different model selection results for the best candidate models investigating the diving probability among nine beavers in a Eurasian beaver population behavioural categories because of variable acceleration in south-eastern Norway. S3. List of aquatic species abundance in Saua patterns. The acceleration can also be affected by external river in south-eastern Norway. S4. The model selection results for the best environmental forces such as wave actions [134], which candidate models investigating the diving habitat selection among nine beavers in a Eurasian beaver population in south-eastern Norway. S5. The can be corrected by incorporating magnetism in the clas- model selection results for the best candidate models investigating the sification [50]. We also only gathered acceleration in 10 s individual context-dependent diving selection for distance to riverbank bursts for the classification model, which may mask some among nine beavers in a Eurasian beaver population in south-eastern Norway. S6. The model selection results for the best candidate models of the diving events of shorter duration. A more detailed investigating the individual context-dependent diving selection for clay inspection of the fine-scale acceleration and body pos - sediment among nine beavers in a Eurasian beaver population in south- tures may improve the classification and provide more eastern Norway. S7. The model selection results for the best candidate models investigating the individual context-dependent diving selection information on the actual behavioural activity [171, 172]. for mud sediment among nine beavers in a Eurasian beaver population Placing the behavioural activities into a larger context (i.e. in south-eastern Norway. S8. The model selection results for the best what the animal did before and after an activity) would candidate models investigating the individual context-dependent diving selection for sand sediment among nine beavers in a Eurasian beaver furthermore help understand the ecological significance population in south-eastern Norway. S9. The model selection results for of each activity. In addition, the ability of dead-reckon- the best candidate models investigating the individual context-depend- ing procedures is spatially limited by the precision of the ent diving selection for rock sediment among nine beavers in a Eurasian beaver population in south-eastern Norway. S10. The model selection GPS positions that are used to ground-truth the dead- results for the best candidate models investigating the individual context- reckoned movement tracks [49, 51], which means that we dependent diving selection for species richness of focal species among will have introduced some spatial error in the locations of nine beavers in a Eurasian beaver population in south-eastern Norway. S11. The model selection results for the best candidate models investigat- the diving events [112, 113]. However, this potential error ing the diving distance to riverbank among nine beavers in a Eurasian will be consistent along the tracking period making it less beaver population in south-eastern Norway. S12. The model selection likely to bias our results. The high classification accuracy results for the best candidate models investigating the diving distance to beaver lodge among nine beavers in a Eurasian beaver population in of the model together with the filtering of less likely div - south-eastern Norway. S13. The model selection results for the best can- ing locations (e.g. on land or not in combination with didate models investigating the diving distance to beaver lodge among swimming) improve our confidence in our ability to clas - dives within 150 m of the lodge of nine beavers in a Eurasian beaver population in south-eastern Norway. sify relevant diving locations in our beaver population. Conclusion Acknowledgements This study was conducted under the Norwegian Beaver Project at University of By coupling fine-scaled information on individual bea - South-Eastern Norway. We thank every member within the NBP who contrib- vers’ movement and diving with comprehensive qualita- uted to the field work and especially H. K. Lodberg-Holm for also participating tive assessments of aquatic habitat characteristics within in early discussions of the project. We thank C. Catoni from Techno Smart and M. Holton from Wildbyte technologies for technical support. beaver territories, we provided new knowledge on the aquatic habitat use by a freshwater semi-aquatic mam- Authors’ contributions mal. We showed how energetic constraints may shape FR founded the NBP and RMM obtained additional supporting funding for this subproject. RMM, SR, MEH, and FR developed the study design. MEH led the beavers’ spatial use of the aquatic environment, and how field work of the study with support from RMM, SR and FR. RMM performed aquatic habitats may have great importance for both for- the statistical analyses with support from SR and MEH. RMM wrote the manu- aging, building materials and safety, even in absence of script with support from SR, RPW and FR. All authors read and approved the final manuscript. natural predators. However, future studies should inves- tigate the importance of aquatic habitat use relative to Funding terrestrial habitat use. Several groups of individuals expe- This study was funded by the University of South-Eastern Norway and partially supported by the Royal Norwegian Society of Sciences and Letters. riencing various ecological conditions may benefit greatly from the use of aquatic resources, consequently affecting Availability of data and materials their body condition, reproduction, and survival [57, 90], The datasets used and analysed during the current study are available from the corresponding author on reasonable request. which should be investigated further in future studies including more individuals and populations. Declarations Supplementary Information Ethics approval and consent to participate The online version contains supplementary material available at https:// doi. All capture and handling procedures were approved by the Norwegian org/ 10. 1186/ s40317- 021- 00259-7. Experimental Animal Board (FOTS ID15947) and by the Norwegian Direc- torate for Nature Management (2014/14415). Our study met the ASAB/ABS Guidelines for the treatment of animals in behavioural research and teaching Additional file 1: S1. The model selection results for the best candi- [173]. No individuals were injured during capture and handling, and they date models investigating the number of dives per night among nine were all successfully released. All methods were performed in accordance Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 16 of 19 with the relevant guidelines and regulations [174]. No short-term effects have 16. Morales JM, Moorcroft PR, Matthiopoulos J, Frair JL, Kie JG, Powell been observed on the movement after tagging [48]. Body mass of dominant RA, Merrill EH, Haydon DT. Building the bridge between animal individuals have been observed to decrease with number of capture and han- movement and population dynamics. Philos Trans R Soc B Biol Sci. dling events, but no statistically clear effects have been observed on survival 2010;365(1550):2289–301. or other body condition indices [57]. Number of capture and handling events 17. Larsen KW, Boutin S. Movements, survival, and settlement was also observed to affect reproduction, but the population seemed habitu- of red squirrel (Tamiasciurus hudsonicus) offspring. Ecology. ated to repeated capture and handling in the long-term [57]. To minimize 1994;75(1):214–23. potential risks, the NBP prioritized capture and handling individuals that were 18. Amsler SJ. Energetic costs of territorial boundary patrols by wild necessary for the monitoring and experiments with clear objectives. Chimpanzees. Am J Primatol. 2010;72(2):93–103. 19. Shier DM, Swaisgood RR. Fitness costs of neighborhood disruption in Consent for publication translocations of a solitary mammal. Conserv Biol. 2012;26(1):116–23. Not applicable. 20. Nathan R, Getz WM, Revilla E, Holyoak M, Kadmon R, Saltz D, Smouse PE. A movement ecology paradigm for unifying organismal move- Competing interests ment research. Proc Natl Acad Sci USA. 2008;105(49):19052–9. The authors declare that they have no competing interests. 21. Wheatley R, Buettel JC, Brook BW, Johnson CN, Wilson RP. Accidents alter animal fitness landscapes. Ecol Lett. 2021;24:920–34. Author details 22. Maher CR, Lott DF. Definitions of territoriality used in the Faculty of Technology, Natural Sciences, and Maritime Sciences, Department study of variation in vertebrate spacing systems. Anim Behav. of Natural Sciences and Environmental Health, University of South-Eastern 1995;49(6):1581–97. Norway, Bø i Telemark, Norway. Biosciences, College of Science, Swansea 23. Dunham ML, Warner RR, Lawson JW. The dynamics of territory University, Swansea, UK. acquisition: a model of two coexisting strategies. Theor Popul Biol. 1995;47(3):347–64. Received: 15 June 2021 Accepted: 11 August 2021 24. Orians GH, Wittenberger JF. Spatial and temporal scales in habitat selec- tion. Am Nat. 1991;137:S29–49. 25. Mysterud A, Ims RA. Functional responses in habitat use: avail- ability influences relative use in trade-off situations. Ecology. 1998;79(4):1435–41. 26. Creel S, Winnie J, Maxwell B, Hamlin K, Creel M. Elk alter habi- References tat selection as an antipredator response to wolves. Ecology. 1. Wilson RP, Quintana F, Hobson VJ. Construction of energy landscapes 2005;86(12):3387–97. can clarify the movement and distribution of foraging animals. Philos 27. Abrams PA. Adaptive change in the resource-exploitation traits of a Trans R Soc B Biol Sci. 2012;279(1730):975–80. generalist consumer: the evolution and coexistence of generalists and 2. Wall J, Douglas-Hamilton I, Vollrath F. Elephants avoid costly mountain- specialists. Evolution. 2006;60(3):427–39. eering. Curr Biol. 2006;16(14):R527–9. 28. Mangel M, Clark CW. Towards a unifield foraging theory. Ecology. 3. Wilson RP, Griffiths IW, Legg PA, Friswell MI, Bidder OR, Halsey LG, Lam- 1986;67(5):1127–38. bertucci SA, Shepard EL. Turn costs change the value of animal search 29. van Beest FM, Uzal A, Vander Wal E, Laforge MP, Contasti AL, Colville paths. Ecol Lett. 2013;16(9):1145–50. D, McLoughlin PD. Increasing density leads to generalization in both 4. Gallagher AJ, Creel S, Wilson RP, Cooke SJ. Energy landscapes and the coarse-grained habitat selection and fine-grained resource selection in landscape of fear. Trends Ecol Evol. 2017;32(2):88–96. a large mammal. J Anim Ecol. 2014;83(1):147–56. 5. Halsey LG. Terrestrial movement energetics: current knowledge 30. Fortin D, Morris DW, McLoughlin PD. Habitat selection and the evolu- and its application to the optimising animal. J Exp Biol. 2016;219(Pt tion of specialists in heterogeneous environments. Isr J Ecol Evol. 10):1424–31. 2008;54(3–4):311–28. 6. Shepard EL, Wilson RP, Rees WG, Grundy E, Lambertucci SA, Vosper 31. Veron G, Patterson BD, Reeves R. Global diversity of mammals (Mam- SB. Energy landscapes shape animal movement ecology. Am Nat. malia) in freshwater. Freshwater Animal Diversity Assessment. Berlin: 2013;182(3):298–312. Springer; 2007. p. 607–17. 7. Laundré JW, Hernández L, Ripple WJ. The landscape of fear: ecological 32. Dunstone N. Adaptations to the semi-aquatic habit and habitat. Behav implications of being afraid. Open Ecol J. 2010;3:1–7. Ecol Riparian Mamm. 1998;71:1–16. 8. Bleicher SS. The landscape of fear conceptual framework: definition and 33. Campbell RD, Rosell F, Nolet BA, Dijkstra VAA. Territory and group sizes review of current applications and misuses. PeerJ. 2017;5:e3772. in Eurasian beavers (Castor fiber): echoes of settlement and reproduc- 9. Gaynor KM, Brown JS, Middleton AD, Power ME, Brashares JS. Land- tion? Behav Ecol Sociobiol. 2005;58(6):597–607. scapes of fear: spatial patterns of risk perception and response. Trends 34. Johnson DH. The comparison of usage and availability measurements Ecol Evol. 2019;34:355–68. for evaluating resource preference. Ecology. 1980;61(1):65–71. 10. Dickie M, McNay SR, Sutherland GD, Cody M, Avgar T. Corridors or 35. Manly B, McDonald L, Thomas DL, McDonald TL, Erickson WP. Resource risk? Movement along, and use of, linear features varies predict- selection by animals: statistical design and analysis for field studies. ably among large mammal predator and prey species. J Anim Ecol. New York: Springer Science & Business Media; 2007. 2020;89(2):623–34. 36. McLoughlin PD, Coulson T, Clutton-Brock T. Cross-generational 11. Fortin D, Beyer HL, Boyce MS, Smith DW, Duchesne T, Mao JS. Wolves effects of habitat and density on life history in red deer. Ecology. influence elk movements: behavior shapes a trophic cascade in Yel- 2008;89(12):3317–26. lowstone National Park. Ecology. 2005;86(5):1320–30. 37. McLoughlin PD, Morris DW, Fortin D, Vander Wal E, Contasti AL. Consid- 12. Shaw AK, Couzin ID. Migration or residency? The evolution of move- ering ecological dynamics in resource selection functions. J Anim Ecol. ment behavior and information usage in seasonal environments. Am 2010;79(1):4–12. Nat. 2013;181(1):114–24. 38. Steyaert SMJG, Zedrosser A, Rosell F. Socio-ecological features other 13. Ariano-Sánchez D, Mortensen RM, Reinhardt S, Rosell F. Escaping than sex affect habitat selection in the socially obligate monogamous drought: seasonality effects on home range, movement patterns and Eurasian beaver. Oecologia. 2015;179(4):1023–32. habitat selection of the Guatemalan Beaded Lizard. Glob Ecol Conserv. 39. Bleicher SS, Rosenzweig ML. Too much of a good thing? A landscape- 2020;23:e01178. of-fear analysis for collared peccaries (Pecari tajacu) reveals hikers 14. Marshall BM, Crane M, Silva I, Strine CT, Jones MD, Hodges CW, Suwan- act as a greater deterrent than thorny or bitter food. Can J Zool. waree P, Artchawakom T, Waengsothorn S, Goode M. No room to roam: 2018;96(4):317–24. King Cobras reduce movement in agriculture. Mov Ecol. 2020;8(1):1–14. 40. Hebblewhite M, Haydon DT. Distinguishing technology from biology: a 15. Mayer M, Zedrosser A, Rosell F. Extra-territorial movements differ critical review of the use of GPS telemetry data in ecology. Philos Trans between territory holders and subordinates in a large, monogamous R Soc B Biol Sci. 2010;365(1550):2303–12. rodent. Sci Rep. 2017;7(1):15261. M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 17 of 19 41. Elliott KH. Measurement of flying and diving metabolic rate in wild 62. Haarberg O, Rosell F. Selective foraging on woody plant species animals: review and recommendations. Comp Biochem Physiol Part A. by the Eurasian beaver (Castor fiber) in Telemark, Norway. J Zool. 2016;202:63–77. 2006;270(2):201–8. 42. Wilson ADM, Wikelski M, Wilson RP, Cooke SJ. Utility of biological sensor 63. Fryxell JM, Doucet CM. Provisioning time and central-place foraging in tags in animal conservation. Conserv Biol. 2015;29(4):1065–75. beavers. Can J Zool. 1991;69(5):1308–13. 43. Cagnacci F, Boitani L, Powell RA, Boyce MS. Animal ecology meets GPS- 64. Milligan HE, Humphries MM. The importance of aquatic vegetation based radiotelemetry: a perfect storm of opportunities and challenges. in beaver diets and the seasonal and habitat specificity of aquatic– Philos Trans R Soc B. 2010;365(1550):2157–62. terrestrial ecosystem linkages in a subarctic environment. Oikos. 44. Williams HJ, Taylor LA, Benhamou S, Bijleveld AI, Clay TA, de Grissac S, 2010;119(12):1877–86. Demsar U, English HM, Franconi N, Gomez-Laich A, Griffiths RC, Kay WP, 65. Svendsen GE. Seasonal change in feeding patterns of beaver in south- Morales JM, Potts JR, Rogerson KF, Rutz C, Spelt A, Trevail AM, Wilson eastern Ohio. J Wildl Manag. 1980;44(1):285–90. RP, Borger L. Optimizing the use of biologgers for movement ecology 66. Doucet CM, Fryxell JM. The effect of nutritional quality on forage prefer - research. J Anim Ecol. 2019;89:186–206. ence by beavers. Oikos. 1993;67:201–8. 45. Hughey LF, Hein AM, Strandburg-Peshkin A, Jensen FH. Challenges and 67. Gerwing TG, Johnson CJ, Alström-Rapaport C. Factors influencing solutions for studying collective animal behaviour in the wild. Philos forage selection by the North American beaver (Castor canadensis). Trans R Soc B Biol Sci. 2018;373(1746):20170005. Mamm Biol. 2013;78(2):79–86. 46. Brown DD, Kays R, Wikelski M, Wilson R, Klimley AP. Observing the 68. Roberts TH, Arner DH. Food habits of beaver in east-central Mississippi. unwatchable through acceleration logging of animal behavior. Anim J Wildl Manag. 1984;48(4):1414–9. Biotelemetry. 2013;1(1):20. 69. Bełżecki G, Miltko R, Kowalik B, Demiaszkiewicz A, Lachowicz J, 47. Graf PM, Wilson RP, Qasem L, Hacklander K, Rosell F. The use of accel- Giżejewski Z, Obidziński A, McEwan N. Seasonal variations of the eration to code for animal behaviours; a case study in free-ranging digestive tract of the Eurasian beaver Castor fiber. Mamm Res. Eurasian beavers Castor fiber. PLoS ONE. 2015;10(8):e0136751. 2018;63(1):21–31. 48. Graf PM, Hochreiter J, Hacklander K, Wilson RP, Rosell F. Short-term 70. Law A, Jones KC, Willby NJ. Medium vs. short-term effects of herbivory effects of tagging on activity and movement patterns of Eurasian by Eurasian beaver on aquatic vegetation. Aquat Bot. 2014;116:27–34. beavers (Castor fiber). Eur J Wildl Res. 2016;62(6):725–36. 71. Law A, Bunnefeld N, Willby N. Beavers and lilies: selective herbivory and 49. Bidder OR, Walker JS, Jones MW, Holton MD, Urge P, Scantlebury DM, adaptive foraging behaviour. Freshw Biol. 2014;59(2):224–32. Marks NJ, Magowan EA, Maguire IE, Wilson RP. Step by step: reconstruc- 72. Vorel A, Válková L, Hamšíková L, Maloň J, Korbelová J. Beaver foraging tion of terrestrial animal movement paths by dead-reckoning. Mov Ecol. behaviour: seasonal foraging specialization by a choosy generalist 2015;3(1):23. herbivore. Behav Ecol Sociobiol. 2015;69(7):1221–35. 50. Williams HJ, Holton MD, Shepard ELC, Largey N, Norman B, Ryan PG, 73. Simonsen TA. Beverens næringsøkologi i vest-agder, Meddelelser fra Duriez O, Scantlebury M, Quintana F, Magowan EA, Marks NJ, Alagaili Statens viltundersøkelser, 2. serie, nr. 39; 1973. AN, Bennett NC, Wilson RP. Identification of animal movement patterns 74. Fraser D, Chavez E, Palohelmo J. Aquatic feeding by moose: selection of using tri-axial magnetometry. Mov Ecol. 2017;5:6. plant species and feeding areas in relation to plant chemical composi- 51. Gunner R, Holton M, Scantlebury M, Lv S, English H, Williams H, Hopkins tion and characteristics of lakes. Can J Zool. 1984;62(1):80–7. P, Quintana F, Gómez-Laich A, Börger L. Dead-reckoning animal move- 75. Bergman BG, Bump JK. Revisiting the role of aquatic plants in beaver ments in R-A reappraisal using Gundog.Tracks. Anim Biotelemetry. habitat selection. Am Midl Nat. 2018;179(2):222–46. 2021;9:23. 76. Nolet BA, Van Der Veer P, Evers E, Ottenheim MM. A linear program- 52. Wilson RP, Shepard ELC, Liebsch N. Prying into the intimate details of ming model of diet choice of free-living beavers. Netherlands J Zool. animal lives: use of a daily diary on animals. Endangered Species Res. 1995;45(3–4):315–37. 2008;4:123–37. 77. Harrison P, Mann K. Detritus formation from eelgrass (Zostera marina 53. Wright BM, Ford JK, Ellis GM, Deecke VB, Shapiro AD, Battaile BC, Trites L.): the relative effects of fragmentation, leaching, and decay. Limnol AW. Fine-scale foraging movements by fish-eating killer whales (Orcinus Oceanogr. 1975;20:924–34. orca) relate to the vertical distributions and escape responses of salmo- 78. Haraguchi A. Seasonal changes in redox properties of peat, nutrition nid prey (Oncorhynchus spp.). Mov Ecol. 2017;5(1):1–18. and phenology of Menyanthes trifoliata L. in a floating peat mat in Mizo - 54. Studd EK, Boudreau MR, Majchrzak YN, Menzies AK, Peers MJL, Seguin rogaike Pond, central Japan. Aquat Ecol. 2004;38(3):351–7. JL, Lavergne SG, Boonstra R, Murray DL, Boutin S, Humphries MM. Use 79. Belovsky GE. Summer diet optimization by beaver. Am Midland Nat. of acceleration and acoustics to classify behavior, generate time budg- 1984;111:209–22. ets, and evaluate responses to moonlight in free-ranging snowshoe 80. Fryxell JM, Vamosi SM, Walton RA, Doucet CM. Retention time and the hares. Front Ecol Evol. 2019;7:154. functional response of beavers. Oikos. 1994;71:207–14. 55. Jeantet L, Planas-Bielsa V, Benhamou S, Geiger S, Martin J, Siegwalt F, 81. Gallant D, Leger L, Tremblay E, Berteaux D, Lecomte N, Vasseur L. Lelong P, Gresser J, Etienne D, Hiélard G. Behavioural inference from Linking time budgets to habitat quality suggests that beavers (Castor signal processing using animal-borne multi-sensor loggers: a novel canadensis) are energy maximizers. Can J Zool. 2016;94(10):671–6. solution to extend the knowledge of sea turtle ecology. R Soc Open Sci. 82. Rosell F, Bozser O, Collen P, Parker H. Ecological impact of beavers Castor 2020;7(5):200139. fiber and Castor canadensis and their ability to modify ecosystems. 56. Chimienti M, van Beest FM, Beumer LT, Desforges JP, Hansen LH, Stelvig Mamm Rev. 2005;35(3–4):248–76. M, Schmidt NM. Quantifying behavior and life-history events of an Arc- 83. Basey JM, Jenkins SH. Influences of predation risk and energy maximiza- tic ungulate from year-long continuous accelerometer data. Ecosphere. tion on food selection by beavers (Castor canadensis). Can J Zool. 2021;12(6):e03565. 1995;73(12):2197–208. 57. Mortensen RM, Rosell F. Long-term capture and handling effects on 84. Rosell F, Sanda J. Potential risks of olfactory signaling: the effect of body condition, reproduction and survival in a semi-aquatic mammal. predators on scent marking by beavers. Behav Ecol. 2006;17(6):897–904. Sci Rep. 2020;10(1):1–16. 85. Krojerová-Prokešová J, Barančeková M, Hamšíková L, Vorel A. Feeding 58. Parker H, Rosell F. Parturition dates for Eurasian beaver Castor fiber: habits of reintroduced Eurasian beaver: spatial and seasonal variation in when should spring hunting cease? Wildl Biol. 2001;7(3):237–41. the use of food resources. J Zool. 2010;281(3):183–93. 59. Parker H, Zedrosser A, Rosell F. Age-specific reproduction in rela- 86. Graf PM, Mayer M, Zedrosser A, Hacklander K, Rosell F. Territory size and tion to body size and condition in female Eurasian beavers. J Zool. age explain movement patterns in the Eurasian beaver. Mamm Biol. 2017;302(4):236–43. 2016;81(6):587–94. 60. Sun L, Müller-Schwarze D, Schulte BA. Dispersal pattern and effective 87. Bartra Cabré L, Mayer M, Steyaert S, Rosell F. Beaver (Castor fiber) activity population size of the beaver. Can J Zool. 2000;78(3):393–8. and spatial movement in response to light and weather conditions. 61. Gallant D, Bérubé CH, Tremblay E, Vasseur L. An extensive study of the Mamm Biol. 2020;100:1–11. foraging ecology of beavers (Castor canadensis) in relation to habitat 88. Sharpe F, Rosell F. Time budgets and sex differences in the Eurasian bea- quality. Can J Zool. 2004;82(6):922–33. ver. Anim Behav. 2003;66(6):1059–67. Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 18 of 19 89. Mayer M, Frank SC, Zedrosser A, Rosell F. Causes and consequences of 116. Kent M. Vegetation description and data analysis: a practical inverse density-dependent territorial behaviour and aggression in a approach. Chichester: John Wiley & Sons; 2012. monogamous mammal. J Anim Ecol. 2019;89(2):577–88. 117. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’hara R, 90. Severud WJ, Windels SK, Belant JL, Bruggink JG. The role of forage avail- Simpson GL, Solymos P, Stevens MHH, Wagner H. Package ‘vegan’. ability on diet choice and body condition in American beavers (Castor Community ecology package, version 2.5-7; 2013. canadensis). Mamm Biol. 2013;78(2):87–93. 118. Lele SR, Merrill EH, Keim J, Boyce MS. Selection, use, choice and 91. Gable TD, Windels SK, Romanski MC, Rosell F. The forgotten prey of an occupancy: clarifying concepts in resource selection studies. J Anim iconic predator: a review of interactions between grey wolves Canis Ecol. 2013;82(6):1183–91. lupus and beavers Castor spp. Mammal Rev. 2018;48(2):123–38. 119. Gillies CS, Hebblewhite M, Nielsen SE, Krawchuk MA, Aldridge CL, 92. Fryxell JM, Doucet CM. Diet choice and the functional response of Frair JL, Saher DJ, Stevens CE, Jerde CL. Application of random beavers. Ecology. 1993;74(5):1297–306. effects to the study of resource selection by animals. J Anim Ecol. 93. Gable TD, Windels SK, Bruggink JG, Homkes AT. Where and How 2006;75(4):887–98. Wolves (Canis lupus) Kill Beavers (Castor canadensis). PLoS ONE. 120. Zuur AF, Ieno EN, Walker NJ, Saveliev AA, Smith GM. Mixed effects 2016;11(12):e0165537. models and extensions in ecology with R. New York: Spring Science 94. Law A, Gaywood MJ, Jones KC, Ramsay P, Willby NJ. Using ecosystem and Business Media; 2009. engineers as tools in habitat restoration and rewilding: beaver and 121. Burnham KP, Anderson DR. Model selection and multimodel infer- wetlands. Sci Total Environ. 2017;605–606:1021–30. ence: a practical information-theoretic approach. New York: Springer 95. Gable TD, Homkes AT, Johnson-Bice SM, Windels SK, Bump JK. Wolves Science & Business Media; 2002. choose ambushing locations to counter and capitalize on the sensory 122. Magnusson A, Skaug H, Nielsen A, Berg C, Kristensen K, Maechler M, abilities of their prey. Behav Ecol. 2021;32:339–48. van Bentham K, Bolker B, Brooks M, Brooks MM. Package ‘glmmTMB’. R 96. Gurnell AM. The hydrogeomorphological effects of beaver dam-build- package version 1.0.2.1; 2017. ing activity. Prog Phys Geogr. 1998;22(2):167–89. 123. Barton K. MuMIn: Multi-Model Inference. R package version 1.43.17; 97. Woo M-K, Waddington JM. Eec ff ts of beaver dams on subarctic wetland 2018. hydrology. Arctic. 1990;43:223–30. 124. Arnold T W. Uninformative parameters and model selection using 98. Severud WJ, Belant JL, Windels SK, Bruggink JG. Seasonal varia- Akaike’s Information Criterion. J Wildl Manag. 2010;74(6):1175–8. tion in assimilated diets of American beavers. Am Midland Nat. 125. Hartig F. DHARMa: residual diagnostics for hierarchical (multi-level/ 2013;169:30–42. mixed) regression models. R package version 0.4.1; 2017. 99. Graf PM, Wilson RP, Sanchez LC, Hackländer K, Rosell F. Diving behavior 126. Team RC. R: a language and environment for statistical computing, r in a free-living, semi-aquatic herbivore, the Eurasian beaver Castor fiber. foundation for statistical computing, Austria; 2015, 2018. Ecol Evol. 2018;8(2):997–1008. 127. Lardet J-P. Spatial behaviour and activity patterns of the water shrew 100. Fish FE. Biomechanics and energetics in aquatic and semiaquatic mam- Neomys fodiens in the field. Acta Theriol. 1988;33(21):293–303. mals: platypus to whale. Physiol Biochem Zool. 2000;73(6):683–98. 128. Hays GC, Forman DW, Harrington LA, Harrington AL, MacDonald DW, 101. Allers D, Culik BM. Energy requirements of beavers (Castor canadensis) Righton D. Recording the free-living behaviour of small-bodied, shal- swimming underwater. Physiol Zool. 1997;70(4):456–63. low-diving animals with data loggers. J Anim Ecol. 2007;76(1):183–90. 102. Fish FE. Comparison of swimming kinematics between terrestrial and 129. Harrington LA, Hays GC, Fasola L, Harrington AL, Righton D, Macdon- semiaquatic opossums. J Mamm. 1993;74(2):275–84. ald DW. Dive performance in a small-bodied, semi-aquatic mammal 103. Pinto B, Santos MJ, Rosell F. Habitat selection of the Eurasian beaver in the wild. J Mamm. 2012;93(1):198–210. (Castor fiber) near its carrying capacity: an example from Norway. Can J 130. Bagniewska JM, Harrington LA, Hart T, Harrington AL, Fasola L, Mac- Zool. 2009;87(4):317–25. donald DW. Persistence in diving American mink. Anim Biotelemetry. 104. Olstad O. Beverens (Castor fiber) utbredelse i Norge. Statens vil- 2015;3(1):1–10. tundersøkelser. 1937;77:217–73. 131. Vogel P, Bodmer C, Spreng M, Aeschimann J, Dunstone N, Gorman 105. Herfindal I, Linnell JD, Odden J, Nilsen EB, Andersen R. Prey density, M. Diving capacity and foraging behaviour of the water shrew. Cam- environmental productivity and home-range size in the Eurasian lynx bridge: Cambridge University Press; 1998. (Lynx lynx). J Zool. 2005;265(1):63–71. 132. Ciancio JE, Quintana F, Sala JE, Wilson RP. Cold birds under pressure: 106. Rosell F, Hovde B. Methods of aquatic and terrestrial netting to capture can thermal substitution ease heat loss in diving penguins? Mar Biol. Eurasian beavers. Wildl Soc Bull. 2001;29(1):269–74. 2016;163(2):43. 107. Rosell F, Sun L. Use of anal gland secretion to distinguish the two bea- 133. Wilson RP, Hustler K, Ryan PG, Burger AE, Noldeke EC. Diving birds in ver species Castor canadensis and C. fiber. Wildl Biol. 1999;5(2):119–23. cold water: do Archimedes and Boyle determine energetic costs? Am 108. Rosell F, Zedrosser A, Parker H. Correlates of body measurements and Nat. 1992;140(2):179–200. age in Eurasian beaver from Norway. Eur J Wildl Res. 2010;56(1):43–8. 134. Wilson RP, Borger L, Holton MD, Scantlebury DM, Gomez-Laich A, 109. Mayer M, Künzel F, Zedrosser A, Rosell F. The 7-year itch: non- Quintana F, Rosell F, Graf PM, Williams H, Gunner R, Hopkins L, Marks adaptive mate change in the Eurasian beaver. Behav Ecol Sociobiol. N, Geraldi NR, Duarte CM, Scott R, Strano MS, Robotka H, Eizaguirre 2017;71(2):32. C, Fahlman A, Shepard ELC. Estimates for energy expenditure in free- 110. Fleming CH, Fagan WF, Mueller T, Olson KA, Leimgruber P, Calabrese JM. living animals using acceleration proxies: a reappraisal. J Anim Ecol. Rigorous home range estimation with movement data: a new autocor- 2020;89(1):161–72. related kernel density estimator. Ecology. 2015;96(5):1182–8. 135. Wilson RP, Shepard EL, Laich AG, Frere E, Quintana F. Pedalling down- 111. Walker JS, Jones MW, Laramee RS, Holton MD, Shepard EL, Williams hill and freewheeling up; a penguin perspective on foraging. Aquat HJ, Scantlebury DM, Marks NJ, Magowan EA, Maguire IE, Bidder OR, Di Biol. 2010;8(3):193–202. Virgilio A, Wilson RP. Prying into the intimate secrets of animal lives; soft- 136. Perrow MR, Schutten JH, Howes JR, Holzer T, Madgwick FJ, Jowitt AJ. ware beyond hardware for comprehensive annotation in “Daily Diary” Interactions between coot (Fulica atra) and submerged macrophytes: tags. Mov Ecol. 2015;3(1):29. the role of birds in the restoration process. Shallow Lakes’ 95. Berlin: 112. Lewis JS, Rachlow JL, Garton EO, Vierling LA. Eec ff ts of habitat on GPS Springer; 1997. p. 241–55. collar performance: using data screening to reduce location error. J 137. Irwin S, O’Halloran J, editors. The wintering behaviour of coot Fulica Appl Ecol. 2007;44(3):663–71. atra L. at Cork Lough, south-west Ireland. In: Biology and environ- 113. Justicia LS, Rosell F, Mayer M. Performance of GPS units for deployment ment: proceedings of the royal Irish academy. JSTOR; 1997. on semiaquatic animals. PLoS ONE. 2018;13(12):e0207938. 138. Bagniewska JM, Hart T, Harrington LA, Macdonald DW. Hidden 114. Middelboe AL, Markager S. Depth limits and minimum light require- Markov analysis describes dive patterns in semiaquatic animals. ments of freshwater macrophytes. Freshw Biol. 1997;37(3):553–68. Behav Ecol. 2013;24(3):659–67. 115. Artsdatabanken. Artsnavnebase. Norsk taksonomisk database; 2015. 139. De Leeuw JJ. Food intake rates and habitat segregation of tufted http:// www2. artsd ataba nken. no/ artsn avn/ Conte ntpag es/ Sok. aspx. duck Aythya fuligula and scaup Aythya marila exploiting zebra mus- Accessed 4 Feb 2020. sels Dreissena polymorpha. Ardea. 1999;87(1):15–31. M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 19 of 19 140. Fish FE, Smelstoys J, Baudinette RV, Reynolds PS. Fur doesn’t fly, it 159. Nolet BA, Rosell F. Comeback of the beaver Castor fiber: An overview of floats: buoyancy of pelage in semi-aquatic mammals. Aquat Mamm. old and new conservation problems. Biol Cons. 1998;83(2):165–73. 2002;28(2):103–12. 160. Macdonald DW, Tattersall F, Brown E, Balharry D. Reintroducing the 141. Harrington LA, Harrington AL, Yamaguchi N, Thom MD, Ferreras P, European beaver to Britain: nostalgic meddling or restoring biodiver- Windham TR, Macdonald DW. The impact of native competitors on an sity? Mamm Rev. 1995;25(4):161–200. alien invasive: temporal niche shifts to avoid interspecific aggression. 161. Andersen DC, Shafroth PB. Beaver dams, hydrological thresholds, and Ecology. 2009;90(5):1207–16. controlled floods as a management tool in a desert riverine ecosystem, 142. Theuerkauf J, Jȩdrzejewski W, Schmidt K, Okarma H, Ruczyński I, Śniezko Bill Williams River. Arizona Ecohydrol. 2010;3(3):325–38. S, Gula R. Daily patterns and duration of wolf activity in the Białowieza 162. Nolet BA, Hoekstra A, Ottenheim MM. Selective foraging on woody spe- Forest. Poland J Mammal. 2003;84(1):243–53. cies by the beaver Castor fiber, and its impact on a riparian willow forest. 143. Rodriguez-Ramos Fernandez J, Dubielzig RR. Ocular comparative Biol Cons. 1994;70(2):117–28. anatomy of the family Rodentia. Vet Ophthalmol. 2013;16:94–9. 163. Logan M, Sanson GD. The effects of lactation on the feeding behaviour 144. Campbell-Palmer R, Rosell F. Conservation of the Eurasian beaver Castor and activity patterns of free-ranging female koalas (Phascolarctos fiber: an olfactory perspective. Mamm Rev. 2010;40(4):293–312. cinereus Goldfuss). Aust J Zool. 2003;51(4):415–28. 145. Ollivier F, Samuelson D, Brooks D, Lewis P, Kallberg M, Komáromy A. 164. Zoller H, Drygala F. Activity patterns of the invasive raccoon dog Comparative morphology of the Tapetum lucidum (among selected (Nyctereutes procyonoides) in North East Germany. J Vertebr Biol. species). Vet Ophthalmol. 2004;7(1):11–22. 2013;62(4):290–6. 146. Maffei L, Fiorentini A, Bisti S. The visual acuity of the lynx. Vision Res. 165. Cantoni D. Social and spatial organization of free-ranging shrews, Sorex 1990;30(4):527–8. coronatus and Neomys fodiens (Insectivora, Mammalia). Anim Behav. 147. Swinnen KRR, Hughes NK, Leirs H. Beaver (Castor fiber) activity patterns 1993;45(5):975–95. in a predator-free landscape. What is keeping them in the dark? Mamm 166. Lodberg-Holm HK, Steyaert S, Reinhardt S, Rosell F. Size is not every- Biol. 2015;80(6):477–83. thing: differing activity and foraging patterns between the sexes in a 148. Birt V, Birt T, Goulet D, Cairns D, Montevecchi W. Ashmole’s halo: direct monomorphic mammal. Behav Ecol Sociobiol. 2021;75(4):1–14. evidence for prey depletion by a seabird. Mar Ecol Progr Ser Oldendorf. 167. Mayer M, Zedrosser A, Rosell F. Couch potatoes do better: Delayed 1987;40(3):205–8. dispersal and territory size affect the duration of territory occupancy in 149. Elliott KH, Woo KJ, Gaston AJ, Benvenuti S, Dall’Antonia L, Davoren GK. a monogamous mammal. Ecol Evol. 2017;7(12):4347–56. Central-place foraging in an Arctic seabird provides evidence for Storer- 168. Mayer M, Fog Bjerre DH, Sunde P. Better safe than sorry: The response to Ashmole’s halo. Auk. 2009;126(3):613–25. a simulated predator and unfamiliar scent by the European hare. Ethol- 150. Nolet BA, Rosell F. Territoriality and time budgets in beavers during ogy. 2020;126(7):704–15. sequential settlement. Can J Zool. 1994;72(7):1227–37. 169. Benkwitt CE. Central-place foraging and ecological effects of an inva- 151. Lima SL. Ecological and evolutionary perspectives on escape from sive predator across multiple habitats. Ecology. 2016;97(10):2729–39. predatory attack: a survey of North American birds. The Wilson Bulletin. 170. Dale S, Gustavsen R, Slagsvold T. Risk taking during parental care: a test 1993;105(1):1–47. of three hypotheses applied to the pied flycatcher. Behav Ecol Socio - 152. Orians GH, Pearson NE. On the theory of central place foraging. In: biol. 1996;39(1):31–42. Horn DJ, Mitchell RD, Stairs GR, editors. Analysis of ecological systems. 171. Chimienti M, Cornulier T, Owen E, Bolton M, Davies IM, Travis JMJ, Scott Columbus: The Ohio State University Press; 1979. p. 154–77. BE. The use of an unsupervised learning approach for characterizing 153. Wetterer JK. Central place foraging theory: when load size affects travel latent behaviors in accelerometer data. Ecol Evol. 2016;6(3):727–41. time. Theor Popul Biol. 1989;36(3):267–80. 172. Gunner RM, Wilson RP, Holton MD, Scott R, Hopkins P, Duarte CM. A new 154. Houston AI, McNamara JM. A general theory of central place foraging direction for differentiating animal activity based on measuring angular for single-prey loaders. Theor Popul Biol. 1985;28(3):233–62. velocity about the yaw axis. Ecol Evol. 2020;10(14):7872–86. 155. DuPasquier A, Cantoni D. Shifts in benthic macroinvertebrate com- 173. Buchanan K, Burt de Perera T, Carere C, Carter T, Hailey A, Hubrecht R, munity and food habits of the water shrew, Neomys fodiens (Soricidae, Jennings D, Metcalfe N, Pitcher T, Peron F. Guidelines for the treat- Insectivora). Acta Oecol. 1992;13(1):81–99. ment of animals in behavioural research and teaching. Anim Behav. 156. Goryainova Z, Katsman E, Zavyalov N, Khlyap L, Petrosyan V. Evaluation 2012;83(1):301–9. of tree and shrub resources of the Eurasian beaver (Castor fiber L.) and 174. Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improving changes in beaver foraging strategy after resources depletion. Russ J bioscience research reporting: the ARRIVE guidelines for reporting Biol Invas. 2014;5(4):242–54. animal research. PLoS Biol. 2010;8(6):e1000412. 157. Danilov P, Fyodorov F. Comparative characterization of the building activity of Canadian and European beavers in northern European Rus- Publisher’s Note sia. Russ J Ecol. 2015;46(3):272–8. Springer Nature remains neutral with regard to jurisdictional claims in pub- 158. Collen P, Gibson R. The general ecology of beavers (Castor spp.), as lished maps and institutional affiliations. related to their influence on stream ecosystems and riparian habitats, and the subsequent effects on fish—a review. Rev Fish Biol Fish. 2000;10(4):439–61. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Animal Biotelemetry Springer Journals

Aquatic habitat use in a semi-aquatic mammal: the Eurasian beaver

Loading next page...
 
/lp/springer-journals/aquatic-habitat-use-in-a-semi-aquatic-mammal-the-eurasian-beaver-zqP00r8cO4
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2050-3385
DOI
10.1186/s40317-021-00259-7
Publisher site
See Article on Publisher Site

Abstract

Background: Semi-aquatic mammals exploit resources both on land and in water and may require both to meet their habitat requirements including food- and building resources, refuges, and for social interactions with conspe- cifics. Within this, the specific availability of both terrestrial and aquatic resources is expected to impact individual fitness. Beavers are highly dependent on water for movement and protection from predators. They are central place foragers and mostly forage on woody vegetation near water although aquatic vegetation may also be an important food resource. However, little is known about their use of aquatic habitats. We aimed to address this knowledge gap by dead-reckoning fine-scale movement tracks and classifying fine-scale diving events, which we then related to the spatial distribution of aquatic vegetation and habitat components within the territory. Results: Overall, there was a statistically clear decrease in probability that diving would occur at dawn and with increasing distance from territory borders. In addition, the distance from the lodge at which animals dived decreased through the night and during the spring/early summer. There was strong selection for diving habitats located closer to the riverbank, with stronger selection for these areas being observed in individuals with larger home ranges. We saw a higher selection for diving above clay sediment, and within 150 m from the lodge, presumably because mud and clay sediment tended to be located closer to the lodge than sand and rock sediment. Furthermore, we found a clear selection for diving in the presence of quillwort (Isoetes spp.), shoreweed (Littorella uniflora), and stonewort (Nitella spp.). Selection for these focal species was stronger among subordinate individuals. Individuals with lower body condition dived closer to the beaver lodge, and dives located further from the lodge were associated with high densities of aquatic vegetation. Conclusion: We provide new knowledge on the aquatic habitat use in a semi-aquatic mammal and show how energetic constraints may shape how beavers spatially use the aquatic environment, whereby short and shallow dives appear most beneficial. We show how aquatic habitats may have great importance for both foraging, building materi- als and safety, and discuss to how they may affect the fitness of individuals. Keywords: Aquatic foraging, Behavioural ecology, Castor fibre, Dead-reckoning, Habitat selection, Movement ecology, Resource selection functions Background Animal movements are affected by a suite of factors, including the cost of movement [1–6], likelihood of pre- *Correspondence: rmo@usn.no dation [7–11], resource distribution [12–14], reproduc- Faculty of Technology, Natural Sciences, and Maritime Sciences, tion [15, 16], and social interactions [12, 17–19]. This, Department of Natural Sciences and Environmental Health, University in part, explains why there is so much interest in animal of South-Eastern Norway, Bø i Telemark, Norway Full list of author information is available at the end of the article movement ecology, but elucidating causality to explain © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 2 of 19 animal movement is challenging as behavioural ecolo- be combined with fine-scale animal movement deter - gists have to juggle with multiple interacting factors [20, mined by dead-reckoning [49–51] to provide informa- 21]. tion on what animals do in the spaces they inhabit [52]. A major movement delimitator results from competi- Altogether, fine-scale information on behaviour and tion, when animals may exclude potential competitors movement, obtained from the use of acceleration and from an area by being territorial [22, 23]. As a result, dead-reckoning, respectively, are being used increas- some individuals end up defending habitats that do not ingly to study wild animals that are hard to observe have favourable combinations of essential resources, directly and/or without bias [44, 49, 52–56]. and thus have to trade-off resources within their ulti - Beavers are socially monogamous, monomorphic, mate habitat acquisition [24–26]. Consequently, some nocturnal mammals that inhabit various freshwater species benefit by being generalists rather than special - bodies [38, 57]. They live in family groups consist - ists, because the consequences of losing a resource are ing of the dominant breeding pair, kits of the year, and less severe since animals can exploit a greater variety of older non-breeding offspring [33, 58, 59]. Beavers reach resources [27–30]. For example, semi-aquatic mammals sexual maturity during their second winter [59], and have adapted to exploit resources on land and in water, give birth to one to five kits in mid-May at northern and individuals of this group may express considerable latitudes [57]. The kits emerge from the lodge in July plasticity to meet their habitat requirements according to when they start feeding on their own [33]. At around the available food resources, shelter, and social interac- 2–3.5  years old, beavers tend to disperse from their tions [31–33]. natal territory to establish their own territory [15, 60]. Habitat selection happens at various spatial scales [34] Beavers are central place foragers, mostly foraging near and is described as the use of resources (habitat) in a water and their lodges [61–63]. Their diet consists mainly manner that is disproportionate to their availability [35]. of woody vegetation but varies seasonally, and comprised Critically, resource-use may not necessarily be directly primarily bark from deciduous trees during winter to proportional to resource availability, but it may also be more nutritiously rich deciduous leaves, aquatic vegeta- modulated by other ecological factors such as competi- tion, and herbaceous plants in spring and summer [64– tion and predation [25, 36–38]. Although it has always 72]. In some areas, aquatic plants may seasonally account been a challenge to quantify which habitats animals have for up to 90% of the diet [65, 66, 73]. Aquatic vegetation available to them, and how much they use them, today, may offer some nutritional benefits over terrestrial veg - this information is important to inform space-value dis- etation, including better digestibility, higher crude pro- cussions so that confounding ecological variables affect - tein, and higher sodium and iron content [64, 66, 74, 75]. ing habitat value for the animal can be put into context Low concentrations of secondary compounds might also [39–41]. make aquatic vegetation more palatable [66], but this may In this study, we examine aquatic habitat use in Eura- vary with species [76]. Seasonally, rhizomes of aquatic sian beavers (Castor fibre) to identify important char - plants can provide great nutritional value in winter and acteristics of aquatic habitats within beaver territories spring when plants store nutrients in the rhizomes in and investigate potential differences in aquatic habitat preparation for spring growth [77, 78]. Diet variation may use among individuals. To achieve our aim, we combine depend on nutrient content and digestibility of available sophisticated animal-attached tags (GPS loggers and forage as individual beavers attempt to maximize energy Daily Diary units), that allow determination of animal intake over time [66, 76, 79–81]. Beaver foraging behav- behaviour with locations, with a comprehensive assess- iour varies according to environmental factors that affect ment of aquatic habitat characteristics. the distribution of food items [82], but ecological factors Technological developments have hugely enhanced such as food plant density, human disturbance, presence what animal biotelemetry can do for us, elucidating, of conspecifics, and predator activity may also affect their for example, fine-scale spatiotemporal location data on foraging choices and foraging locations [67, 83, 84]. No an increasing range of animals across various environ- clear dietary differences have been found between sexes, ments [40–43]. In particular, tri-axial accelerometers ages, or social ranks in beavers [65, 69, 85], but several are increasingly being used to study wild animals [44, studies indicate that foraging behaviour may differ sea - 45], because they allow determination of an individual’s sonally as territorial movements vary among individu- behaviour [46]. They have been used to classify behav - als [86–88]. Furthermore, individuals may be affected by iour and activity level patterns in beavers, distinguish- various ecological conditions during their lifetime, such ing seven behaviours with high precision, including as loss and acquisition of territories [15, 89], that can swimming and diving [47, 48]. These behaviours can affect their behavioural time-budgets and consequently their body condition, reproduction, and survival [57, 90]. M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 3 of 19 Water acts as a refuge for beavers [91] and is specifi - Using fine-scaled dead-reckoned animal tracks to cally used to minimize predation risk when foraging on determine spatial and temporal locations of aquatic dives, land [92]. Even though beavers depend on the aquatic we aim to examine important characteristics of aquatic environment for movement and safety [83, 86, 87, 93] habitat use by Eurasian beavers and investigate poten- and may even manipulate the environment to optimize tial individual differences. Assuming that dives indicate aquatic conditions [82, 94], aquatic behaviour and habitat aquatic habitat use in beavers, we hypothesize that habi- use have not been well studied in these animals. Although tat use vary temporally and spatially between individuals beavers use aquatic habitats for foraging [64–66, 71], pro- of various ages, sexes, social ranks, and by the composi- tection from predators [92, 95], and collecting resources tion of biotic and abiotic factors within their individual for lodge- and dam-building [96, 97], aquatic habitats territories. may be particularly important when terrestrial vegetation is difficult to access, or is of low nutritional quality [38, Methods 64, 66, 75, 98]. However, their spatial exploitation of the Study site aquatic components of their territory is poorly under- Our study site was located at the lower reaches of the stood. Research has found, however, that habitat use river Sauar in Vestfold and Telemark County, southeast- may differ between age groups and according to risk lev - ern Norway (Fig. 1). The river drains the lake Heddalsvat - els [87]. Diving behaviour has been studied using accel- net in the north and forms part of the catchment of the erometers, which has highlighted a preference for short lake Norsjø in the south, stretching over approximately (< 30  s) and shallow (up to 4  m, but most < 1  m) dives, 13  km with a width of 45–250  m. The river sections are which indicate some form of aquatic resource selection, generally slow-flowing with stable water levels because of although the link between diving and space use is vague natural lakes and man-made impoundments along part [99]. Being only semi-aquatic [cf. 100], beavers may expe- of its length [103], although flooding events frequently rience a higher cost exploiting aquatic resources [101, occur. The river flows through small villages, farmlands 102] than a fully aquatic equivalent may do. This may and fields interspersed with riparian woodland that com - explain why studies have found them to be diving for less prised mostly Norway spruce (Picea abies), Scots pine than 3% of their nightly activity budget [99]. However, (Pinus sylvestris), birch (Betula spp.), grey alder (Alnus energy requirements have been reported to compare well incana), aspen (Populus tremula) and mountain ash (Sor- with more fully aquatic mammals and birds [101]. bus aucuparia) [62, 103]. Fig. 1 a The location of the study site (red square) in Telemark and Vestfold County, Norway. b Overview of study river with random available sites (yellow circles) and identified beaver diving locations (red triangles) within beaver territories Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 4 of 19 Beavers have inhabited the area since the 1920s when maintain their social rank until they died or disappeared they recolonized the rivers [104]. The population is at from the territory [109]. carrying capacity, as territories of various sizes directly We captured and equipped nine beavers (five males, border each other [33]. Territory borders are identified four females, Table 1) with GPS loggers (Gipsy-5, Techno based on scent mound concentrations, sight observations Smart) and daily diary units (including accelerometer, of known patrolling beavers backed up by GPS data. The magnetometer, thermometer, Wildbyte technologies main river contains ten distinguishable territories inhab- [52]) in the spring and early summer of 2018 (one beaver) ited by approximately 32 individuals [33, 57]. Predation and 2019 (eight beavers). The data loggers were glued pressure is low as wolves (Canis lupus) and bears (Ursus onto the fur on the lower back of the beavers, approxi- arctos) are functionally extinct in the area, and lynx (Lynx mately 15 cm above the tail following the spine, and were lynx) only occur at low densities [84, 105]. removed again after 2 to 3  weeks if they had not fallen The study river is part of a larger monitoring project off by themselves [47]. To extend battery life, GPS log - where beavers in the area have been monitored through gers were programmed to take a fix position every 15 min an extensive capture programme, the Norwegian Beaver between 7  p.m. and 7  a.m. to reduce numbers of unsuc- Project (NBP), since 1997 [57]. The long-term monitor - cessful GPS fix attempts from within the beaver lodges. ing project aims to capture all newcomers (kits and dis- Daily diary units logged continuously at 40 Hz. From the persers from outside the study site) annually, enabling GPS positions, we identified territory borders to esti - identification of individuals at later encounters and fam - mate territory size, expressed in bank length (km) and ily group sizes. calculated 95% autocorrelated kernel density estimates (AKDE) to estimate the overall space use (home range) of Capture and tracking protocol each beaver [110]. Captured beavers were released near Individuals were detected from a motorboat using the capture site within their territory after approximately searchlights and captured at night with large diving- 40 min of handling time [106]. nets in shallow water or with land-nets [106]. Captured individuals were immobilized in cloth sacks, enabling Identification of dives easy handling without anaesthesia, and identified via Accelerometers in tandem with magnetometers can be microchips (PIT tag) and unique combinations of plastic used in dead-reckoning to accurately predict and recon- and metal ear-tags. Beavers were weighed to the near- struct animals’ fine-scale three-dimensional movement est 100  g. Body length was measured following the cur- paths in space and time by sequentially integrating cal- vature of the spine from nose tip to the base of the tail. culated travel vectors [49, 51]. However, the estimated Tail length was measured from the base to the tip of the movement track accumulates error and therefore drifts tail, and tail width was measured from edge to edge of over time, so it needs to be corrected through ground- the dorsal surface at the midpoint between tail base and truthing, e.g. correcting the track according to GPS fixes tip. Measurements of body length and tail proportions in [49]. cm were used to calculate tail fat index ((tail length × tail We calibrated the daily diary data in the software width)/body length), representing the body condition of DDMT (Daily Diary Multiple Trace, Wildbyte Technolo- beavers [57, 59]. gies). Using the acceleration and magnetism data, we Individuals were sexed based on the colour and vis- dead-reckoned the movement track of each beaver in the cosity of their anal gland secretion [107] and assigned a software Framework4 [111]. The dead-reckoned move - minimum age based on body mass at first capture [57, ment tracks were hereafter corrected using the GPS posi- 108]; minimum 2  years (subadult) when body mass was tions as ground-truthing [49]. GPS positions were filtered between 17 to 19.5  kg inclusive, and minimum 3  years to remove positions with horizontal dilution of precision (adult) when body mass was above 19.5  kg. Territorial (HDOP) values above five and with less than four avail - dominance was in most cases attributed to adult terri- able satellites to reduce the effects of imprecise GPS posi - torial residents of each sex. Territorial dominance was tions [112, 113]. verified by eventual dispersal of the alternative candidate, To identify diving locations, we divided the dead-reck- greatest body weight among same-sex group members oned movement tracks into ten second bursts and used or lactation in females (large nipples, i.e. > 0.5 cm). Indi- the acceleration to assign behavioural activities to each viduals dispersing into a territory were posited to have burst based on the acceleration-based behavioural classi- achieved the dominant breeding position when the pre- fication model by Graf et al. [47]. The classification model vious dominant same-sex individual had disappeared, or can clearly differentiate the acceleration between seven evidence outlined above was applicable. Unless proven behaviours: swimming, diving, sleeping, feeding, stand- otherwise, dominant individuals were assumed to ing, walking, and grooming. To furthermore filter out M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 5 of 19 Table 1 Overview of tracked individuals with daily diaries and GPS loggers in a population of Eurasian beavers in southeastern Norway Beaver Tracking period Nights Dives Territory Home range, ha Sex Age, years Social rank Mass, kg Body Tail fat index size, km length, cm Anna April 2–April 15 2019 13 92 4.9 1.3 F 7 Dominant 23.3 85.5 4.3 Ceasar April 2–April 9 2019 7 143 4.9 1.1 M 9 Dominant 17.7 81.0 3.9 Dylan May 9–May 11 2018 2 110 6.6 10.3 M 6 Dominant 21.2 82.0 4.0 Laurits April 26–May 7 2019 12 37 2.2 0.4 M 13 Dominant 19.0 77.5 3.8 Mason April 2–April 10 2019 7 49 3.4 2.2 M 6 Subordiante 17.9 72.0 4.2 Mattanja April 25–May 4 2019 9 89 2.2 4.0 M 3 Subordinate 18.0 72.0 3.9 Maximus April 25–April 29 2019 4 9 6.6 6.2 M 2 Subordinate 17.1 79.0 3.6 Tanja May 28–June 13 2019 17 269 4.4 1.1 F 16 Dominant 18.6 75.0 3.1 Tatjana April 2–April 6 2019 4 47 3.4 0.7 F 3 Subordinate 16.4 70.0 4.0 Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 6 of 19 potentially falsely classified dives, we focused on diving diving event was identified within a given hour, 0 = no sections where the previous 10-s bursts were additionally diving events were identified within a given hour) varied predicted to be swimming. through the night and between individuals. We analysed the effects of spatiotemporal variables (date, hour of the night, distance to riverbank, lodge, and territory border), Assessment of aquatic habitats individual differences (sex, social rank, age, body sizes, We visited and assessed the aquatic habitat of all div- and home range size), and between components of the ing locations as well as random aquatic locations avail- territories (territory size, mean water depth, mean veg- able within the territories of each beaver between June etation cover, and mean species richness). and October 2019 (Fig.  1b). We only sampled sites with We used a detrended correspondence analysis (DCA) a water depth of less than 10  m since vegetation growth from the R package ‘vegan’ v. 2.5–7 [117] to assess the here is more abundant due to light conditions [114], general distribution pattern of the aquatic vegetation in whilst also accounting for known diving depths in bea- the study river in relation to physical (water depth and vers [99]. sediment type), spatial (distance to riverbank, beaver For vegetation sampling at each location, sites were lodge, and territory border) and vegetation characteris- sampled using a 1 × 1  m quadrat with an aluminium tics (vegetation cover and species richness) of the sites. A frame. The quadrat was placed as close to the location matrix including aggregated abundance per species was as wind and currents allowed. The frame construction used for the DCA. However, species that only occurred had a pyramidal shape, enabling a GoPro camera (GoPro in five sites or less were removed from the matrix. The Hero5) to be attached to the top, 0.8  m above the sur- correlation between the aquatic species compositions face, which would keep the quadrat within the camera and physical, spatial, and vegetation characteristics of view. The quadrat was left at the bottom of the site after the sites were assessed by passively fitting them to the sediment settled, and pictures and films were recorded. ordination (permutations = 999). From the ordination, When water depth allowed it, aquatic plants and physi- comparing species with use, we could furthermore iden- cal characteristics were recorded in  situ using an aqua tify species that appeared to be of potential importance scope. Aquatic plants were collected with a rake when to the beavers when diving. For the subsequent resource identification required closer inspection. Plant species selection functions, we included variables for the number identification followed the database of Artsdatabanken of focal species present at a site and vegetation cover of [115]. Species abundance was quantified as coverage in these focal species. percentage, rounding to the nearest 5% [116]. Plants with We investigated the aquatic habitat selection within less than 5% cover were registered as 1% per species. territories using GLMMs with Bernoulli distribution, We categorized each site according to physical charac- logit link, and beaver ID as a random effect (1 = diving teristics [water depth and sediment type (clay, mud, sand, site, 0 = random available site within the territory) [34, and rock)], spatial characteristics (distance to riverbank, 35]. We analysed whether aquatic habitats located at beaver lodge, and territory border), and characteristics varying water depths, sediment types, distances to river- of the aquatic vegetation. We characterized the aquatic bank, beaver lodge, and territory border, or with varying vegetation by cover and species richness (number of spe- vegetation cover (overall and focal species), and spe- cies) to evaluate importance of quantity and diversity, cies richness (overall and focal species) were used more respectively. than was generally available in the territories [35, 118]. Additionally, we analysed the variations in diving selec- Statistical analysis tion among individuals in univariate models weighted by We used generalized linear mixed-effects models number of identified dives by fitting the resource selec - (GLMMs) with Poisson distribution, log link, and beaver tion function to each individual [119]. This enabled us to ID as a random effect to investigate how the number of analyse how the individual diving selection coefficients identified dives per night varied between individuals and varied between beavers of different age, sex, body size, components of the territories. We analysed the effects of and social rank that furthermore inhabit territories of sex (male, female), social rank (dominant, subordinate), different size and with varying amount of available water age (years), body size (body mass, body length, and tail fat depth, vegetation cover (overall and focal species), and index), and home range size (95% AKDE), and the effects species richness (overall and focal species) [13, 25, 119]. of territory size (bank length in km), mean water depth As dives may have different purposes according to their (m) and mean vegetation cover and species richness. spatial location, we furthermore investigated how log Using GLMMs with Bernoulli distribution, logit link, transformed distance to the lodge and riverbank varied with beaver ID and tracking night as random effects, we temporally (date and hour of the night), by environmental also investigated how hourly diving probability (1 = a M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 7 of 19 characteristics (depth, sediment, vegetation cover, spe- out using the R packages ‘glmmTMB’ v. 1.0.2.1 [122] and cies richness), and among individuals of different age, ‘MuMIn’ v. 1.43.17 [123]. The most parsimonious mod - sex, body size, social rank, territory size, and home range els within ΔAICc < 2 were chosen as the best models to by using GLMMs with gaussian distribution and beaver describe the variation [121, 124]. In each model, variables ID as a random effect. To examine characteristics associ - that included zero within their 95% confidence inter - ated to the proximate area of the beaver lodge, we fur- val (CI) were considered uninformative and reported as thermore analysed how dives within 150 m of the beaver unclear effects [124]. The best models were visually vali - lodge varied according to the above variables using uni- dated using the R package ‘DHARMa’ v. 0.4.1 [125] to variate GLMMs with Gaussian distribution and beaver plot standardized model residuals against the fitted val - ID as a random effect. ues [120] and, when relevant, furthermore checked for In all analyses, a list of candidate models was created zero-inflation. Top candidate models for all analyses can using ecologically relevant combinations of fixed effects be found in the supplemental material (Additional file  1). to account for variability in endogenous (such as sex, age, All analyses were conducted in R 4.0.3 [126]. and social rank) and exogenous factors (such as territory size, vegetation composition) that may be important in describing the ecology of beavers (Fig. 2). Because of the Results sample size, individual effects (sex, age, social rank, ter - Nine beavers were tracked with data loggers (Wild- ritory size, and home range size) should be interpreted byte technologies, Daily Diaries [52]) and GPS loggers with care as they only imply possible ecological effects (Techno Smart) affixed to the lower back for a total of that should be investigated with more individuals in 77 nights. Identified diving events lasted between 10 future studies. We included spatiotemporal interactions and 110  s, with the majority (80%) lasting 10  s or less. (between hour and distance to riverbank, lodge, and We identified on average (mean ± SD) 9.5 ± 3.1 dives per territory border, respectively) in the analysis for diving night for each beaver. We found no clear differences in probability, but excluded interactions in all other analyses the number of dives per night between males and females because of the limited sample size. Individual selection (10.9 ± 3.3 and 8.8 ± 3.0, respectively), among dominants coefficients were similarly analysed in univariate models and subordinates (12.9 ± 3.6 and 6.4 ± 2.5, respectively) because of the limited sample size. The fixed effects used or as a function of age, body size and tail fat index (Addi- in all analyses were not correlated (Pearson r coefficients tional file  1). Furthermore, territory size, home range less than 0.5) and variance inflation factor values were size, mean water depth in territory, and mean vegeta- less than 3 [120]. tion cover and species richness in territory did not have Model selection was based on Akaike’s information cri- a clear effect on the number of identified dives per night terion corrected for small sample size [121], and carried (Additional file 1). Fig. 2 Alluvial diagram showing how covariates included in the analyses (left) may relate to various ecological mechanisms (middle) which ecologically may have physical, behavioural, and social consequences (right) for the fitness of beavers Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 8 of 19 Table 3 Detrended correspondence analysis results for aquatic vegetation sites in a population of Eurasian beaver in southeastern Norway DCA1 DCA2 DCA3 DCA4 Eigenvalues 0.62 0.60 0.43 0.42 Axis lengths 6.15 4.94 4.30 3.51 Proportion explained % 9.63 9.39 6.78 6.52 Cumulative prop. explained % 9.63 19.02 25.80 32.32 Fig. 3 The predicted relationship ± 95% confidence interval between hourly diving probability, time of night, and distance from territory border among nine individuals in a Eurasian beaver population in southeastern Norway. Yellow boxes represent time of sunset and sunrise through the tracking period. Red boxes represent time of dusk and dawn through the tracking period Table 2 Eec ff t size (β), standard error (SE), lower (LCI) and upper (UCI) 95% confidence interval of explanatory variables for the analysis of nightly diving probability among nine individuals in a Eurasian beaver population in southeastern Norway Fig. 4 Detrended correspondence analysis (DCA) of aquatic Variable Estimate β SE LCI UCI vegetation sites (points) within territories of nine individuals in a Eurasian beaver population in southeastern Norway. Arrows represent Intercept − 2.970 0.652 − 4.247 − 1.692 passively fitted environmental gradients and red labels represent Hour − 0.114 0.002 − 0.119 − 0.110 environmental centroids. The green ellipse encircles species that may Log (distance to − 0.184 0.007 − 0.197 − 0.171 represent important resources at diving locations of the beavers territory border) Hour × log − 0.064 0.002 − 0.069 − 0.059 (distance to ter- ritory border) depth in territory and mean vegetation diversity within 2 2 Marginal R : 0.01 Conditional R : 0.84 the territory on nightly diving probability (Table 2, Addi- Effects were modelled using a GLMM with Bernoulli distribution. Beaver ID and tional file 1). tracking night were included as random effects. Informative parameters are given in bold Characterization of aquatic vegetation The DCA ordination described up to 32.3% of the vari - Diving probability ation in the aquatic vegetation composition within the Hourly diving probability varied through the night, beaver territories with DCA1 and DCA2 describing 9.6 decreasing over the final hours of the night (Fig.  3) and and 9.4% of the variation, respectively (Table  3). The with increasing distance from the territory borders aquatic vegetation within the territories showed great (Table 2, Fig. 3). We found no differences in hourly diving variation and differed clearly with increasing water depth, probability between sexes, social ranks, age, or relative to vegetation cover, species richness, and sediment type body size and tail fat index. Furthermore, we found no (Fig.  4). Diving sites and random available sites within clear effect of date, distance to riverbank, distance from the beaver territories clearly differed in species composi the lodge, territory size, home range size, mean water tion. Diving sites were especially associated with varying M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 9 of 19 amounts of quillwort (Isoetes spp.), shoreweed (Littorella with decreasing water depths, increasing species rich- uniflora), and stonewort (Nitella spp.) (Fig.  4). Further- ness, and decreasing vegetation cover. A list of all aquatic more, sites were predominantly more used for diving species can be found in Additional file 1. Resource selection of aquatic habitats within the territory Table 4 Eec ff t size (β), standard error (SE), lower (LCI) and upper We found a clear diving selection for aquatic locations (UCI) 95% confidence interval of explanatory variables for the closer to the riverbank but found no clear diving selection analysis of diving location selection among nine individuals in a as a function of distance from lodge, distance from terri- Eurasian beaver population in southeastern Norway tory border, or water depth (Table  4, Fig.  5a, Additional Variable Estimate β SE LCI UCI file  1). Diving selection probability varied as a function of sediment type. Locations having either mud, sand, or Intercept − 0.934 0.216 − 1.358 − 0.510 rock sediment were less selected than locations with clay Log (distance to riverbank) − 0.686 0.086 − 0.855 − 0.517 sediment (Table 4, Fig. 5b). Furthermore, diving selection Sediment 0.857 0.265 0.337 1.376 Clay probability increased when several of either quillwort, Sediment − 0.349 0.308 − 0.952 0.254 Sand shoreweed, and stonewort were present, but we found no Sediment − 0.461 0.286 − 1.022 0.100 Rocks effect of vegetation cover, overall species richness, or veg - Number of focal species 0.340 0.141 0.065 0.616 etation cover of the focal species on the diving selection present 2 2 probability (Table 4, Fig. 5c, Additional file 1). Marginal R : 0.19 Conditional R : 0.26 Individual diving selection coefficients varied among Effects were modelled using a GLMM with Bernoulli distribution. Beaver ID was individuals and territories. Individuals exploiting larger included as random effect. Informative parameters are given in bold home ranges (i.e. 95% AKDE) had a weaker selection for Reference level for sediment = Mud Fig. 5 The predicted relationship ± 95% confidence interval between diving selection probability and a distance to riverbank, b sediment type and c number of focal vegetation species present among nine individuals in a Eurasian beaver population in southeastern Norway Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 10 of 19 diving locations closer to the riverbank than individuals individuals than dominant individuals (Table  5, Fig.  6b). with smaller home ranges (Table 5, Fig. 6a). We found no We found no conditional effects of other variables on the clear context dependent effects on selection coefficients selection coefficients (Additional file 1). for the various sediment types (Additional file  1). Social We found that diving locations further from the riv- rank had an effect on diving selection for number of focal erbank had a deeper water depth and higher vegeta- species present which were stronger among subordinate tion cover of focal species compared to diving locations closer to the riverbank (Table  6, Fig.  7a, b). Diving loca- tions located further from the beaver lodge had a higher vegetation cover than diving locations closer to the bea- Table 5 Eec ff t size (β), standard error (SE), lower (LCI) and ver lodge (Table  6, Fig.  7c). Diving distance to the bea- upper (UCI) 95% confidence interval of explanatory variables ver lodge also increased with increasing tail fat index and for the analysis of individual selection coefficient for distance decreased during the night and during the spring/early to riverbank and number of focal species present among nine individuals in a Eurasian beaver population in southeastern summer (Table  6, Fig.  7d–f ). Focusing on dives within Norway 150 m of the beaver lodge, we found a difference among sediment types with dives located on clay and mud sedi- Variable Estimate β SE LCI UCI ment being closer to the lodge than dives located on sand Selection for distance to riverbank and rock (Table 6, Fig. 8). Intercept − 0.465 0.169 − 0.864 − 0.067 Log (home range) − 0.420 0.142 − 0.757 − 0.084 Discussion 2 2 R : 0.56 R : 0.49 adjusted We provide new knowledge on aquatic habitat use in a Selection for number of focal species present semi-aquatic mammal, the Eurasian beaver, by examin- Intercept 0.175 0.131 − 0.135 0.485 ing finely resolved information on beaver movement and Social rank 0.922 0.228 0.384 1.460 subordinate diving in relation to fine-scaled qualitative assessments of 2 2 R : 0.70 R : 0.66 adjusted aquatic habitat characteristics within beaver territories. Effects were modelled using a GLM with gaussian distribution. Analyses were We observed clear spatiotemporal variations in hourly weighed by number of diving sites for each individual. Informative parameters diving probability but found no differences among indi - are given in bold viduals or territories. Beavers selected for both spatial, Reference level for social rank = Dominant Fig. 6 The predicted relationship ± 95% confidence interval between a selection coefficients for distance to riverbank and home range size (AKDE, autocorrelated kernel density estimate) and b selection coefficients for number of focal vegetation species present and social rank among nine individuals in a Eurasian beaver population in southeastern Norway M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 11 of 19 Table 6 Eec ff t size (β), standard error (SE), lower (LCI) and upper [101, 129–131]. Beavers may preferably dive to bring (UCI) 95% confidence interval of explanatory variables for the large quantities of vegetation to the surface rather than analyses of distance to riverbank and distance to beaver lodge consuming small amounts of vegetation underwater, sav- among dives of nine individuals in a Eurasian beaver population ing energy by not diving repeatedly to exploit resources in southeastern Norway at depth, thereby also minimizing heat loss [132, 133]. In Variable Estimate β SE LCI UCI fact, an extensive study on diving in beavers by Graf et al. [99] found that animals had high dynamic body accelera- Distance to riverbank tion (DBA) (a good proxy for movement-based energy Intercept 2.226 0.071 2.087 2.365 expenditure [134]) for the descent, indicating work Log (water depth) 0.603 0.053 0.499 0.708 done against appreciable buoyancy, as has been noted Vegetation cover of 0.010 0.003 0.005 0.015 for many birds with their air-filled plumage [133]. Curi - focal species 2 2 ously though, in stark contrast to birds, which use this Marginal R : 0.35 Conditional R : 0.35 buoyancy for passive ascents [135], beavers also had high Distance to beaver lodge DBA values during their return to the surface, which was Intercept 5.801 0.410 − 0.375 11.976 suggested to be due to animals having to transport veg- Vegetation cover 0.006 0.002 0.002 0.009 etation from the bottom to the surface for consumption Hour − 0.071 0.023 − 0.116 − 0.025 [99]. Similar behaviour is commonly observed in water Julian day − 0.019 0.003 − 0.025 − 0.012 birds such as Eurasian coots (Fulica atra) when foraging Tail fat index 1.059 0.023 0.635 1.483 on aquatic vegetation [136, 137], but also in semi-aquatic 2 2 Marginal R : 0.18 Conditional R : 0.19 carnivores such as American mink (Neovison vison) that Distance to beaver lodge (dives within 150 m) occasionally consume aquatic food items at the surface or Intercept 4.227 0.091 4.049 4.404 on the riverbank [128, 138]. Critically, this behaviour is Sediment − 0.098 0.152 − 0.395 0.199 Clay most advantageous when large amounts of food can be Sediment 0.670 0.240 0.201 1.140 Sand brought to the surface during one dive, which can then be Sediment 0.309 0.147 0.021 0.597 Rocks consumed at leisure without the need for multiple, ener- 2 2 Marginal R : 0.25 Conditional R : 0.25 getically onerous dives in repetitive feeding bouts such as Effects were modelled using a GLMMs with gaussian distribution. Beaver ID was those performed by carnivorous species [138, 139]. included as random effect. Informative parameters are given in bold We found a decreased hourly diving probability in the Reference level for sediment = Mud early morning which is contrary to the findings of Graf et al. [99], although they had appreciable variation. Other studies have found a peak in general activity (meas- physical and vegetation characteristics in their diving ured via overall body dynamic acceleration) in the mid- locations, highlighting the degree of choice they exercise dle of the beavers’ principal activity period, suggesting for foraging behaviour although selection strength var- increased activity in the middle of the night [48]. Diving ied between individuals. Furthermore, spatial variations generally has a high overall body dynamic acceleration among dives indicate the energetic variability in aquatic compared to other behavioural activities and therefore habitat use. Often studies in freshwater-inhabiting semi- a higher movement-based energy cost [47, 134]. Div- aquatic mammals focus on the use of terrestrial habitat ing patterns in beavers may be implicitly represented by components, but we show how components of aquatic this general activity pattern, peaking in the middle of the habitats similarly may be an important resource which principal activity period, as aquatic habitat use may be can potentially have considerable fitness consequences costly for a semi-aquatic mammal [101, 140]. Similar div- for a semi-aquatic mammal like the beaver. ing patterns peaking in the middle of the activity period have been found in other semi-aquatic mammals, such Diving patterns as American mink, that perform temporal niche shifts The majority of our identified diving events were short, to avoid interspecific aggression from competitors [128, which matches previous findings for beavers [99, 101] 141]. and other semi-aquatic mammals [127–129]. Beavers Diving patterns may be structured according to the have previously been reported to spend less than 3% of activity peaks of potential terrestrial predators [142]. their nightly activity budget on diving activities which When beavers bring aquatic resources onto the riverbank corresponds well with their role as generalist herbivores to be handled, their risk of predation increases [93, 95]. that do not rely solely on aquatic foraging [99]. Similar Beavers are at a particular risk when on land because diving patterns have been found among semi-aquatic of their poor eyesight under low light conditions (i.e. generalist carnivores and may relate to semi-aquatic ani- they lack tapetum lucidum) [143] and dependence on mals being less specialized for the aquatic environment Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 12 of 19 Fig. 7 The predicted relationship ± 95% confidence interval between distance to riverbank and beaver lodge and a water depth, b vegetation cover of focal species, c vegetation cover, d hour of the night, e Julian day, and f tail fat index among dives of nine individuals in a Eurasian beaver population in southeastern Norway. Points represent actual distances olfaction to detect potential risks [144]. Consequently, active (including diving) in the middle of the night when they may not detect potential predators, such as wolves it is darkest. In North America, beavers make up a large (Canis lupus) and Eurasian lynx (Lynx lynx) that have proportion of wolf diet, especially during summer when advanced night vision [145, 146], before being detected wolves have been observed to ambush beavers at fre- themselves. However, wolves have been observed to quently used locations [95]. Natural predators are absent predate more at dawn, dusk, and during moonlit nights in our study area, but behavioural activities are known to [142], which could increase the benefits of being most be influenced by historical threats [84, 147]. In addition, M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 13 of 19 be treated as a central place [152], which makes clear the energetic costs of leaving the central place to acquire food. In that sense, the decreasing tendency to dive with increasing distance from the riverbank ties in with opti- mal foraging theory, which has animals maximizing reward and minimizing transit time and energy [153, 154]. Distance to the riverbank may also reflect decreas - ing aquatic foraging options in terms of decreased macro- phyte growth which is highly dependent on water depth and light penetration [114]. But short shallow dives may also be preferred as they are energetically cheaper for semi-aquatic animals like beavers that have high buoy- ancy [140]. In general, diving locations in semi-aquatic carnivores have been shown to follow distributions of aquatic food resources [127, 155]. We noted that selec- tion for diving locations near the riverbank was statisti- cally stronger among individuals that exploited a larger home range: individuals that exploit a larger home range may have a greater need to stay closer to the riverbank Fig. 8 The predicted relationship ± 95% confidence interval due to higher territory patrolling efforts whereas indi - between distance to beaver lodge and sediment type among dives viduals exploiting smaller home ranges may be able to within 150 m of the beaver lodge of nine individuals in a Eurasian beaver population in southeastern Norway. Points represent actual forage further away from the riverbank [86]. Individuals distances using larger areas or inhabiting larger territories may also have reduced resource depletion so that, conversely, bea- vers restricted to smaller areas may be forced to exploit human activities both on land and on water, which are foraging areas further away from the riverbank [33, 156]. naturally reduced at night, may also influence the activ - Larger home ranges may also have a greater area of shal- ity levels of beavers [147]. Consequently, diving for food low water, which, depending on the time spent diving, resources at dusk and dawn may be perceived to be too can be energetically easier to exploit for a semi-aquatic risky for beavers. animal [140]. Conversely, individuals with smaller home Diving is presumably not just shaped by risks, but also ranges may have to exploit all habitats to fulfil their ener - allows the animals to have access to important aquatic getic requirements [33]. We saw that water depth at the resources [4, 81]. Hourly diving probability decreased diving locations generally increased with increasing dis- with increasing distance from the territory borders which tance from the riverbank, which presumably represent may indicate a possible depletion of aquatic resources higher energetic costs of diving away from the riverbank, near beaver lodges that are often located in central but dives at longer distances from the riverbank also had parts of the territory [cf. 148,149]. Although this may be higher amounts of quillwort, shoreweed, and stonewort expected to be related to territory size [33], we did not present, possibly due to the Ashmole’s halo effect [148, find this, possibly because our sample size was too small. 149] operating on animals preferentially associated with Diving near the borders may also relate to territorial the shoreline. This indicates the interplay of depth and defence activities, which may help reduce aggressive ter- distance energetics in a semi-aquatic animal, which have ritorial encounters when territorial intruders swim away to be balanced with the calorific value of the food plants unseen [89, 150, 151]. However, we were unable to quan- and their location [132, 133, 140]. tify territorial behaviour. Beavers had a higher selection for diving at locations with clay sediment, which may be an important building Diving selection material for lodges and dams [96], although dams are not We found a high selection for diving closer to the river- present in our study site, and beavers may additionally bank. Similar short and shallow dives have been found make use of burrows dug into the riverbank whereby the in other studies of semi-aquatic mammals which prob- clay sediment aids in enhancing the structural integrity ably reflect the energetic constraints of bringing food of those burrows [157, 158]. Lodges and dams are mostly resources to the riverbank for consumption and a pref- repaired in the autumn [159, 160], but may be repaired erence for travelling along the riverbank [86, 99, 127, after flooding events too [161]. Mud is also widely used 128]. Association with a riverbank can, in some senses, for beaver constructions [97], although mud substrates, Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 14 of 19 despite being highly abundant in the study area, seemed their foraging strategy accordingly [4, 168]. Similarly, we not to be selected. Beavers continuously need to apply found that beavers with lower tail fat index tended to dive fresh mud and other fine debris to seal their construc - closer to their lodge, which may be a consequence of ter- tions as it is continuously washed away [96, 97]. We ritorial constraints as individuals with higher body condi- found that dives on mud and clay sediment within 150 m tion may be better able to cope with the increased cost of of the beaver lodge were generally located closer to the patrolling and protecting territory borders [18, 86]. Div- beaver lodges than dives located on sand and rock sedi- ing locations located further from the beaver lodge also ment which, we believe, indicates the importance of fine had higher amounts of vegetation cover which indicate sediment for building constructions. However, we cannot the energetic trade-offs between costs and benefits that rule out that mud and clay sediment may contain advan- a central place foraging individual experiences [4, 169]. tageous foraging options, although our ordination did Diving distance to the lodge decreased through the night not show strong correlation between mud and clay sedi- which may indicate a functional change in the purpose ment with aquatic species richness or vegetation cover. of the dive. As dives further from the lodge occurred at We found a higher selection for diving locations with locations with more aquatic vegetation, these dives may presence of species of quillwort, shoreweed, and stone- be intended for foraging whereas dives later in the night wort, which may represent important food resources may be used for building activities. We also saw that div- for the beaver. Other studies have similarly found a high ing distance to the lodge decreased from early spring to preference for quillwort in the early summer [73], and summer, which may relate to increased parenting activi- algae like stonewort may be selected for its protein con- ties that require the beaver to stay closer to the lodge tent and other nutrients [65]. Although beavers mainly when the kits are born in mid-May [58]. Diving for food forage on woody vegetation, aquatic vegetation seems resources may be perceived as less risk-taking than going to be seasonally important, with studies additionally on land [87] and could be preferred by parenting individ- reporting beavers foraging on, among others, water lilies uals to ensure the growth and survival of their offspring (Nymphaea spp.), pondweed (Potamogeton spp.), water [170]. horsetail (Equisetum spp.), waterweed (Elodea spp.), and water lobelia (Lobelia dortmanna) [64–66, 68, 69, Methodological limitations 71, 98]. Incorporating aquatic vegetation into a varied Using dead-reckoning together with an acceleration- diet could be a strategy to minimize risk of nutrient defi - based behavioural classification model to identify the ciency [162], which may be seasonally beneficial to some temporal and spatial distribution of clear diving events, individuals (e.g. females during lactation when energetic we identified considerably fewer diving events than a pre - requirements increase) [163, 164]. However, we did not vious study in beavers that identified typically 40 dives find any selection differences between males and females per night [99]. This may be related to seasonal variations which may be because of our limited number of individu- as we only tracked beavers from April to June, whereas als. Other studies on semi-aquatic mammals show how Graf et  al. [99] also included observations from the males and females in two species of shrew (Neomys fodi- autumn (September to October) where diving conditions ens and Sorex coronatus) use separate foraging habitats may be more favourable, because of increased water tem- during the breeding season [165]. Males and females in perature and life history patterns [15, 57, 99]. But dives our population have been shown to differ seasonally in may also be masked by our method. Despite a high classi- aquatic foraging with peaks in the spring and late sum- fication accuracy of diving events from the model by Graf mer for females, whereas males only foraged on aquatic et  al. [47], some events may not be identified because vegetation in the spring [166]. We found that selection for some behavioural activities, including behaviours not the focal plant species were statistically stronger among described by the model, can have similar patterns and subordinate individuals, which may be linked to their mask each other [47]. Therefore, it is important to clearly higher energetic requirements resulting from their activi- define each behavioural activity in an acceleration-based ties related to attempts to become dominant in a territory classification model, but also to include enough varia - [57] (e.g. performing more extra territorial movements tions of each activity from several individuals to improve [15]). A higher use of aquatic vegetation may also be a the precision of the model. Different diving styles may risk-avoiding strategy, minimizing predation risk on land be misclassified as other behavioural categories if the [92, 95]. In other studies, adult beavers have been found behavioural classification model is not trained on several to forage less on aquatic vegetation than subadult indi- variations of each behavioural category but only includes viduals (i.e. 2-year-old) [65] which may be less risk-will- typical acceleration patterns for each behavioural cat- ing as they potentially face higher fitness costs in terms egory. For example, the ecological difference between of future reproductive success [167] and therefore adjust a ‘dive’ and an ‘almost dive’ may be minimal when a M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 15 of 19 beaver can access aquatic resources by just sticking its beavers in a Eurasian beaver population in south-eastern Norway. S2. The head underwater, but they may fall within two different model selection results for the best candidate models investigating the diving probability among nine beavers in a Eurasian beaver population behavioural categories because of variable acceleration in south-eastern Norway. S3. List of aquatic species abundance in Saua patterns. The acceleration can also be affected by external river in south-eastern Norway. S4. The model selection results for the best environmental forces such as wave actions [134], which candidate models investigating the diving habitat selection among nine beavers in a Eurasian beaver population in south-eastern Norway. S5. The can be corrected by incorporating magnetism in the clas- model selection results for the best candidate models investigating the sification [50]. We also only gathered acceleration in 10 s individual context-dependent diving selection for distance to riverbank bursts for the classification model, which may mask some among nine beavers in a Eurasian beaver population in south-eastern Norway. S6. The model selection results for the best candidate models of the diving events of shorter duration. A more detailed investigating the individual context-dependent diving selection for clay inspection of the fine-scale acceleration and body pos - sediment among nine beavers in a Eurasian beaver population in south- tures may improve the classification and provide more eastern Norway. S7. The model selection results for the best candidate models investigating the individual context-dependent diving selection information on the actual behavioural activity [171, 172]. for mud sediment among nine beavers in a Eurasian beaver population Placing the behavioural activities into a larger context (i.e. in south-eastern Norway. S8. The model selection results for the best what the animal did before and after an activity) would candidate models investigating the individual context-dependent diving selection for sand sediment among nine beavers in a Eurasian beaver furthermore help understand the ecological significance population in south-eastern Norway. S9. The model selection results for of each activity. In addition, the ability of dead-reckon- the best candidate models investigating the individual context-depend- ing procedures is spatially limited by the precision of the ent diving selection for rock sediment among nine beavers in a Eurasian beaver population in south-eastern Norway. S10. The model selection GPS positions that are used to ground-truth the dead- results for the best candidate models investigating the individual context- reckoned movement tracks [49, 51], which means that we dependent diving selection for species richness of focal species among will have introduced some spatial error in the locations of nine beavers in a Eurasian beaver population in south-eastern Norway. S11. The model selection results for the best candidate models investigat- the diving events [112, 113]. However, this potential error ing the diving distance to riverbank among nine beavers in a Eurasian will be consistent along the tracking period making it less beaver population in south-eastern Norway. S12. The model selection likely to bias our results. The high classification accuracy results for the best candidate models investigating the diving distance to beaver lodge among nine beavers in a Eurasian beaver population in of the model together with the filtering of less likely div - south-eastern Norway. S13. The model selection results for the best can- ing locations (e.g. on land or not in combination with didate models investigating the diving distance to beaver lodge among swimming) improve our confidence in our ability to clas - dives within 150 m of the lodge of nine beavers in a Eurasian beaver population in south-eastern Norway. sify relevant diving locations in our beaver population. Conclusion Acknowledgements This study was conducted under the Norwegian Beaver Project at University of By coupling fine-scaled information on individual bea - South-Eastern Norway. We thank every member within the NBP who contrib- vers’ movement and diving with comprehensive qualita- uted to the field work and especially H. K. Lodberg-Holm for also participating tive assessments of aquatic habitat characteristics within in early discussions of the project. We thank C. Catoni from Techno Smart and M. Holton from Wildbyte technologies for technical support. beaver territories, we provided new knowledge on the aquatic habitat use by a freshwater semi-aquatic mam- Authors’ contributions mal. We showed how energetic constraints may shape FR founded the NBP and RMM obtained additional supporting funding for this subproject. RMM, SR, MEH, and FR developed the study design. MEH led the beavers’ spatial use of the aquatic environment, and how field work of the study with support from RMM, SR and FR. RMM performed aquatic habitats may have great importance for both for- the statistical analyses with support from SR and MEH. RMM wrote the manu- aging, building materials and safety, even in absence of script with support from SR, RPW and FR. All authors read and approved the final manuscript. natural predators. However, future studies should inves- tigate the importance of aquatic habitat use relative to Funding terrestrial habitat use. Several groups of individuals expe- This study was funded by the University of South-Eastern Norway and partially supported by the Royal Norwegian Society of Sciences and Letters. riencing various ecological conditions may benefit greatly from the use of aquatic resources, consequently affecting Availability of data and materials their body condition, reproduction, and survival [57, 90], The datasets used and analysed during the current study are available from the corresponding author on reasonable request. which should be investigated further in future studies including more individuals and populations. Declarations Supplementary Information Ethics approval and consent to participate The online version contains supplementary material available at https:// doi. All capture and handling procedures were approved by the Norwegian org/ 10. 1186/ s40317- 021- 00259-7. Experimental Animal Board (FOTS ID15947) and by the Norwegian Direc- torate for Nature Management (2014/14415). Our study met the ASAB/ABS Guidelines for the treatment of animals in behavioural research and teaching Additional file 1: S1. The model selection results for the best candi- [173]. No individuals were injured during capture and handling, and they date models investigating the number of dives per night among nine were all successfully released. All methods were performed in accordance Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 16 of 19 with the relevant guidelines and regulations [174]. No short-term effects have 16. Morales JM, Moorcroft PR, Matthiopoulos J, Frair JL, Kie JG, Powell been observed on the movement after tagging [48]. Body mass of dominant RA, Merrill EH, Haydon DT. Building the bridge between animal individuals have been observed to decrease with number of capture and han- movement and population dynamics. Philos Trans R Soc B Biol Sci. dling events, but no statistically clear effects have been observed on survival 2010;365(1550):2289–301. or other body condition indices [57]. Number of capture and handling events 17. Larsen KW, Boutin S. Movements, survival, and settlement was also observed to affect reproduction, but the population seemed habitu- of red squirrel (Tamiasciurus hudsonicus) offspring. Ecology. ated to repeated capture and handling in the long-term [57]. To minimize 1994;75(1):214–23. potential risks, the NBP prioritized capture and handling individuals that were 18. Amsler SJ. Energetic costs of territorial boundary patrols by wild necessary for the monitoring and experiments with clear objectives. Chimpanzees. Am J Primatol. 2010;72(2):93–103. 19. Shier DM, Swaisgood RR. Fitness costs of neighborhood disruption in Consent for publication translocations of a solitary mammal. Conserv Biol. 2012;26(1):116–23. Not applicable. 20. Nathan R, Getz WM, Revilla E, Holyoak M, Kadmon R, Saltz D, Smouse PE. A movement ecology paradigm for unifying organismal move- Competing interests ment research. Proc Natl Acad Sci USA. 2008;105(49):19052–9. The authors declare that they have no competing interests. 21. Wheatley R, Buettel JC, Brook BW, Johnson CN, Wilson RP. Accidents alter animal fitness landscapes. Ecol Lett. 2021;24:920–34. Author details 22. Maher CR, Lott DF. Definitions of territoriality used in the Faculty of Technology, Natural Sciences, and Maritime Sciences, Department study of variation in vertebrate spacing systems. Anim Behav. of Natural Sciences and Environmental Health, University of South-Eastern 1995;49(6):1581–97. Norway, Bø i Telemark, Norway. Biosciences, College of Science, Swansea 23. Dunham ML, Warner RR, Lawson JW. The dynamics of territory University, Swansea, UK. acquisition: a model of two coexisting strategies. Theor Popul Biol. 1995;47(3):347–64. Received: 15 June 2021 Accepted: 11 August 2021 24. Orians GH, Wittenberger JF. Spatial and temporal scales in habitat selec- tion. Am Nat. 1991;137:S29–49. 25. Mysterud A, Ims RA. Functional responses in habitat use: avail- ability influences relative use in trade-off situations. Ecology. 1998;79(4):1435–41. 26. Creel S, Winnie J, Maxwell B, Hamlin K, Creel M. Elk alter habi- References tat selection as an antipredator response to wolves. Ecology. 1. Wilson RP, Quintana F, Hobson VJ. Construction of energy landscapes 2005;86(12):3387–97. can clarify the movement and distribution of foraging animals. Philos 27. Abrams PA. Adaptive change in the resource-exploitation traits of a Trans R Soc B Biol Sci. 2012;279(1730):975–80. generalist consumer: the evolution and coexistence of generalists and 2. Wall J, Douglas-Hamilton I, Vollrath F. Elephants avoid costly mountain- specialists. Evolution. 2006;60(3):427–39. eering. Curr Biol. 2006;16(14):R527–9. 28. Mangel M, Clark CW. Towards a unifield foraging theory. Ecology. 3. Wilson RP, Griffiths IW, Legg PA, Friswell MI, Bidder OR, Halsey LG, Lam- 1986;67(5):1127–38. bertucci SA, Shepard EL. Turn costs change the value of animal search 29. van Beest FM, Uzal A, Vander Wal E, Laforge MP, Contasti AL, Colville paths. Ecol Lett. 2013;16(9):1145–50. D, McLoughlin PD. Increasing density leads to generalization in both 4. Gallagher AJ, Creel S, Wilson RP, Cooke SJ. Energy landscapes and the coarse-grained habitat selection and fine-grained resource selection in landscape of fear. Trends Ecol Evol. 2017;32(2):88–96. a large mammal. J Anim Ecol. 2014;83(1):147–56. 5. Halsey LG. Terrestrial movement energetics: current knowledge 30. Fortin D, Morris DW, McLoughlin PD. Habitat selection and the evolu- and its application to the optimising animal. J Exp Biol. 2016;219(Pt tion of specialists in heterogeneous environments. Isr J Ecol Evol. 10):1424–31. 2008;54(3–4):311–28. 6. Shepard EL, Wilson RP, Rees WG, Grundy E, Lambertucci SA, Vosper 31. Veron G, Patterson BD, Reeves R. Global diversity of mammals (Mam- SB. Energy landscapes shape animal movement ecology. Am Nat. malia) in freshwater. Freshwater Animal Diversity Assessment. Berlin: 2013;182(3):298–312. Springer; 2007. p. 607–17. 7. Laundré JW, Hernández L, Ripple WJ. The landscape of fear: ecological 32. Dunstone N. Adaptations to the semi-aquatic habit and habitat. Behav implications of being afraid. Open Ecol J. 2010;3:1–7. Ecol Riparian Mamm. 1998;71:1–16. 8. Bleicher SS. The landscape of fear conceptual framework: definition and 33. Campbell RD, Rosell F, Nolet BA, Dijkstra VAA. Territory and group sizes review of current applications and misuses. PeerJ. 2017;5:e3772. in Eurasian beavers (Castor fiber): echoes of settlement and reproduc- 9. Gaynor KM, Brown JS, Middleton AD, Power ME, Brashares JS. Land- tion? Behav Ecol Sociobiol. 2005;58(6):597–607. scapes of fear: spatial patterns of risk perception and response. Trends 34. Johnson DH. The comparison of usage and availability measurements Ecol Evol. 2019;34:355–68. for evaluating resource preference. Ecology. 1980;61(1):65–71. 10. Dickie M, McNay SR, Sutherland GD, Cody M, Avgar T. Corridors or 35. Manly B, McDonald L, Thomas DL, McDonald TL, Erickson WP. Resource risk? Movement along, and use of, linear features varies predict- selection by animals: statistical design and analysis for field studies. ably among large mammal predator and prey species. J Anim Ecol. New York: Springer Science & Business Media; 2007. 2020;89(2):623–34. 36. McLoughlin PD, Coulson T, Clutton-Brock T. Cross-generational 11. Fortin D, Beyer HL, Boyce MS, Smith DW, Duchesne T, Mao JS. Wolves effects of habitat and density on life history in red deer. Ecology. influence elk movements: behavior shapes a trophic cascade in Yel- 2008;89(12):3317–26. lowstone National Park. Ecology. 2005;86(5):1320–30. 37. McLoughlin PD, Morris DW, Fortin D, Vander Wal E, Contasti AL. Consid- 12. Shaw AK, Couzin ID. Migration or residency? The evolution of move- ering ecological dynamics in resource selection functions. J Anim Ecol. ment behavior and information usage in seasonal environments. Am 2010;79(1):4–12. Nat. 2013;181(1):114–24. 38. Steyaert SMJG, Zedrosser A, Rosell F. Socio-ecological features other 13. Ariano-Sánchez D, Mortensen RM, Reinhardt S, Rosell F. Escaping than sex affect habitat selection in the socially obligate monogamous drought: seasonality effects on home range, movement patterns and Eurasian beaver. Oecologia. 2015;179(4):1023–32. habitat selection of the Guatemalan Beaded Lizard. Glob Ecol Conserv. 39. Bleicher SS, Rosenzweig ML. Too much of a good thing? A landscape- 2020;23:e01178. of-fear analysis for collared peccaries (Pecari tajacu) reveals hikers 14. Marshall BM, Crane M, Silva I, Strine CT, Jones MD, Hodges CW, Suwan- act as a greater deterrent than thorny or bitter food. Can J Zool. waree P, Artchawakom T, Waengsothorn S, Goode M. No room to roam: 2018;96(4):317–24. King Cobras reduce movement in agriculture. Mov Ecol. 2020;8(1):1–14. 40. Hebblewhite M, Haydon DT. Distinguishing technology from biology: a 15. Mayer M, Zedrosser A, Rosell F. Extra-territorial movements differ critical review of the use of GPS telemetry data in ecology. Philos Trans between territory holders and subordinates in a large, monogamous R Soc B Biol Sci. 2010;365(1550):2303–12. rodent. Sci Rep. 2017;7(1):15261. M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 17 of 19 41. Elliott KH. Measurement of flying and diving metabolic rate in wild 62. Haarberg O, Rosell F. Selective foraging on woody plant species animals: review and recommendations. Comp Biochem Physiol Part A. by the Eurasian beaver (Castor fiber) in Telemark, Norway. J Zool. 2016;202:63–77. 2006;270(2):201–8. 42. Wilson ADM, Wikelski M, Wilson RP, Cooke SJ. Utility of biological sensor 63. Fryxell JM, Doucet CM. Provisioning time and central-place foraging in tags in animal conservation. Conserv Biol. 2015;29(4):1065–75. beavers. Can J Zool. 1991;69(5):1308–13. 43. Cagnacci F, Boitani L, Powell RA, Boyce MS. Animal ecology meets GPS- 64. Milligan HE, Humphries MM. The importance of aquatic vegetation based radiotelemetry: a perfect storm of opportunities and challenges. in beaver diets and the seasonal and habitat specificity of aquatic– Philos Trans R Soc B. 2010;365(1550):2157–62. terrestrial ecosystem linkages in a subarctic environment. Oikos. 44. Williams HJ, Taylor LA, Benhamou S, Bijleveld AI, Clay TA, de Grissac S, 2010;119(12):1877–86. Demsar U, English HM, Franconi N, Gomez-Laich A, Griffiths RC, Kay WP, 65. Svendsen GE. Seasonal change in feeding patterns of beaver in south- Morales JM, Potts JR, Rogerson KF, Rutz C, Spelt A, Trevail AM, Wilson eastern Ohio. J Wildl Manag. 1980;44(1):285–90. RP, Borger L. Optimizing the use of biologgers for movement ecology 66. Doucet CM, Fryxell JM. The effect of nutritional quality on forage prefer - research. J Anim Ecol. 2019;89:186–206. ence by beavers. Oikos. 1993;67:201–8. 45. Hughey LF, Hein AM, Strandburg-Peshkin A, Jensen FH. Challenges and 67. Gerwing TG, Johnson CJ, Alström-Rapaport C. Factors influencing solutions for studying collective animal behaviour in the wild. Philos forage selection by the North American beaver (Castor canadensis). Trans R Soc B Biol Sci. 2018;373(1746):20170005. Mamm Biol. 2013;78(2):79–86. 46. Brown DD, Kays R, Wikelski M, Wilson R, Klimley AP. Observing the 68. Roberts TH, Arner DH. Food habits of beaver in east-central Mississippi. unwatchable through acceleration logging of animal behavior. Anim J Wildl Manag. 1984;48(4):1414–9. Biotelemetry. 2013;1(1):20. 69. Bełżecki G, Miltko R, Kowalik B, Demiaszkiewicz A, Lachowicz J, 47. Graf PM, Wilson RP, Qasem L, Hacklander K, Rosell F. The use of accel- Giżejewski Z, Obidziński A, McEwan N. Seasonal variations of the eration to code for animal behaviours; a case study in free-ranging digestive tract of the Eurasian beaver Castor fiber. Mamm Res. Eurasian beavers Castor fiber. PLoS ONE. 2015;10(8):e0136751. 2018;63(1):21–31. 48. Graf PM, Hochreiter J, Hacklander K, Wilson RP, Rosell F. Short-term 70. Law A, Jones KC, Willby NJ. Medium vs. short-term effects of herbivory effects of tagging on activity and movement patterns of Eurasian by Eurasian beaver on aquatic vegetation. Aquat Bot. 2014;116:27–34. beavers (Castor fiber). Eur J Wildl Res. 2016;62(6):725–36. 71. Law A, Bunnefeld N, Willby N. Beavers and lilies: selective herbivory and 49. Bidder OR, Walker JS, Jones MW, Holton MD, Urge P, Scantlebury DM, adaptive foraging behaviour. Freshw Biol. 2014;59(2):224–32. Marks NJ, Magowan EA, Maguire IE, Wilson RP. Step by step: reconstruc- 72. Vorel A, Válková L, Hamšíková L, Maloň J, Korbelová J. Beaver foraging tion of terrestrial animal movement paths by dead-reckoning. Mov Ecol. behaviour: seasonal foraging specialization by a choosy generalist 2015;3(1):23. herbivore. Behav Ecol Sociobiol. 2015;69(7):1221–35. 50. Williams HJ, Holton MD, Shepard ELC, Largey N, Norman B, Ryan PG, 73. Simonsen TA. Beverens næringsøkologi i vest-agder, Meddelelser fra Duriez O, Scantlebury M, Quintana F, Magowan EA, Marks NJ, Alagaili Statens viltundersøkelser, 2. serie, nr. 39; 1973. AN, Bennett NC, Wilson RP. Identification of animal movement patterns 74. Fraser D, Chavez E, Palohelmo J. Aquatic feeding by moose: selection of using tri-axial magnetometry. Mov Ecol. 2017;5:6. plant species and feeding areas in relation to plant chemical composi- 51. Gunner R, Holton M, Scantlebury M, Lv S, English H, Williams H, Hopkins tion and characteristics of lakes. Can J Zool. 1984;62(1):80–7. P, Quintana F, Gómez-Laich A, Börger L. Dead-reckoning animal move- 75. Bergman BG, Bump JK. Revisiting the role of aquatic plants in beaver ments in R-A reappraisal using Gundog.Tracks. Anim Biotelemetry. habitat selection. Am Midl Nat. 2018;179(2):222–46. 2021;9:23. 76. Nolet BA, Van Der Veer P, Evers E, Ottenheim MM. A linear program- 52. Wilson RP, Shepard ELC, Liebsch N. Prying into the intimate details of ming model of diet choice of free-living beavers. Netherlands J Zool. animal lives: use of a daily diary on animals. Endangered Species Res. 1995;45(3–4):315–37. 2008;4:123–37. 77. Harrison P, Mann K. Detritus formation from eelgrass (Zostera marina 53. Wright BM, Ford JK, Ellis GM, Deecke VB, Shapiro AD, Battaile BC, Trites L.): the relative effects of fragmentation, leaching, and decay. Limnol AW. Fine-scale foraging movements by fish-eating killer whales (Orcinus Oceanogr. 1975;20:924–34. orca) relate to the vertical distributions and escape responses of salmo- 78. Haraguchi A. Seasonal changes in redox properties of peat, nutrition nid prey (Oncorhynchus spp.). Mov Ecol. 2017;5(1):1–18. and phenology of Menyanthes trifoliata L. in a floating peat mat in Mizo - 54. Studd EK, Boudreau MR, Majchrzak YN, Menzies AK, Peers MJL, Seguin rogaike Pond, central Japan. Aquat Ecol. 2004;38(3):351–7. JL, Lavergne SG, Boonstra R, Murray DL, Boutin S, Humphries MM. Use 79. Belovsky GE. Summer diet optimization by beaver. Am Midland Nat. of acceleration and acoustics to classify behavior, generate time budg- 1984;111:209–22. ets, and evaluate responses to moonlight in free-ranging snowshoe 80. Fryxell JM, Vamosi SM, Walton RA, Doucet CM. Retention time and the hares. Front Ecol Evol. 2019;7:154. functional response of beavers. Oikos. 1994;71:207–14. 55. Jeantet L, Planas-Bielsa V, Benhamou S, Geiger S, Martin J, Siegwalt F, 81. Gallant D, Leger L, Tremblay E, Berteaux D, Lecomte N, Vasseur L. Lelong P, Gresser J, Etienne D, Hiélard G. Behavioural inference from Linking time budgets to habitat quality suggests that beavers (Castor signal processing using animal-borne multi-sensor loggers: a novel canadensis) are energy maximizers. Can J Zool. 2016;94(10):671–6. solution to extend the knowledge of sea turtle ecology. R Soc Open Sci. 82. Rosell F, Bozser O, Collen P, Parker H. Ecological impact of beavers Castor 2020;7(5):200139. fiber and Castor canadensis and their ability to modify ecosystems. 56. Chimienti M, van Beest FM, Beumer LT, Desforges JP, Hansen LH, Stelvig Mamm Rev. 2005;35(3–4):248–76. M, Schmidt NM. Quantifying behavior and life-history events of an Arc- 83. Basey JM, Jenkins SH. Influences of predation risk and energy maximiza- tic ungulate from year-long continuous accelerometer data. Ecosphere. tion on food selection by beavers (Castor canadensis). Can J Zool. 2021;12(6):e03565. 1995;73(12):2197–208. 57. Mortensen RM, Rosell F. Long-term capture and handling effects on 84. Rosell F, Sanda J. Potential risks of olfactory signaling: the effect of body condition, reproduction and survival in a semi-aquatic mammal. predators on scent marking by beavers. Behav Ecol. 2006;17(6):897–904. Sci Rep. 2020;10(1):1–16. 85. Krojerová-Prokešová J, Barančeková M, Hamšíková L, Vorel A. Feeding 58. Parker H, Rosell F. Parturition dates for Eurasian beaver Castor fiber: habits of reintroduced Eurasian beaver: spatial and seasonal variation in when should spring hunting cease? Wildl Biol. 2001;7(3):237–41. the use of food resources. J Zool. 2010;281(3):183–93. 59. Parker H, Zedrosser A, Rosell F. Age-specific reproduction in rela- 86. Graf PM, Mayer M, Zedrosser A, Hacklander K, Rosell F. Territory size and tion to body size and condition in female Eurasian beavers. J Zool. age explain movement patterns in the Eurasian beaver. Mamm Biol. 2017;302(4):236–43. 2016;81(6):587–94. 60. Sun L, Müller-Schwarze D, Schulte BA. Dispersal pattern and effective 87. Bartra Cabré L, Mayer M, Steyaert S, Rosell F. Beaver (Castor fiber) activity population size of the beaver. Can J Zool. 2000;78(3):393–8. and spatial movement in response to light and weather conditions. 61. Gallant D, Bérubé CH, Tremblay E, Vasseur L. An extensive study of the Mamm Biol. 2020;100:1–11. foraging ecology of beavers (Castor canadensis) in relation to habitat 88. Sharpe F, Rosell F. Time budgets and sex differences in the Eurasian bea- quality. Can J Zool. 2004;82(6):922–33. ver. Anim Behav. 2003;66(6):1059–67. Mortensen et al. Anim Biotelemetry (2021) 9:35 Page 18 of 19 89. Mayer M, Frank SC, Zedrosser A, Rosell F. Causes and consequences of 116. Kent M. Vegetation description and data analysis: a practical inverse density-dependent territorial behaviour and aggression in a approach. Chichester: John Wiley & Sons; 2012. monogamous mammal. J Anim Ecol. 2019;89(2):577–88. 117. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’hara R, 90. Severud WJ, Windels SK, Belant JL, Bruggink JG. The role of forage avail- Simpson GL, Solymos P, Stevens MHH, Wagner H. Package ‘vegan’. ability on diet choice and body condition in American beavers (Castor Community ecology package, version 2.5-7; 2013. canadensis). Mamm Biol. 2013;78(2):87–93. 118. Lele SR, Merrill EH, Keim J, Boyce MS. Selection, use, choice and 91. Gable TD, Windels SK, Romanski MC, Rosell F. The forgotten prey of an occupancy: clarifying concepts in resource selection studies. J Anim iconic predator: a review of interactions between grey wolves Canis Ecol. 2013;82(6):1183–91. lupus and beavers Castor spp. Mammal Rev. 2018;48(2):123–38. 119. Gillies CS, Hebblewhite M, Nielsen SE, Krawchuk MA, Aldridge CL, 92. Fryxell JM, Doucet CM. Diet choice and the functional response of Frair JL, Saher DJ, Stevens CE, Jerde CL. Application of random beavers. Ecology. 1993;74(5):1297–306. effects to the study of resource selection by animals. J Anim Ecol. 93. Gable TD, Windels SK, Bruggink JG, Homkes AT. Where and How 2006;75(4):887–98. Wolves (Canis lupus) Kill Beavers (Castor canadensis). PLoS ONE. 120. Zuur AF, Ieno EN, Walker NJ, Saveliev AA, Smith GM. Mixed effects 2016;11(12):e0165537. models and extensions in ecology with R. New York: Spring Science 94. Law A, Gaywood MJ, Jones KC, Ramsay P, Willby NJ. Using ecosystem and Business Media; 2009. engineers as tools in habitat restoration and rewilding: beaver and 121. Burnham KP, Anderson DR. Model selection and multimodel infer- wetlands. Sci Total Environ. 2017;605–606:1021–30. ence: a practical information-theoretic approach. New York: Springer 95. Gable TD, Homkes AT, Johnson-Bice SM, Windels SK, Bump JK. Wolves Science & Business Media; 2002. choose ambushing locations to counter and capitalize on the sensory 122. Magnusson A, Skaug H, Nielsen A, Berg C, Kristensen K, Maechler M, abilities of their prey. Behav Ecol. 2021;32:339–48. van Bentham K, Bolker B, Brooks M, Brooks MM. Package ‘glmmTMB’. R 96. Gurnell AM. The hydrogeomorphological effects of beaver dam-build- package version 1.0.2.1; 2017. ing activity. Prog Phys Geogr. 1998;22(2):167–89. 123. Barton K. MuMIn: Multi-Model Inference. R package version 1.43.17; 97. Woo M-K, Waddington JM. Eec ff ts of beaver dams on subarctic wetland 2018. hydrology. Arctic. 1990;43:223–30. 124. Arnold T W. Uninformative parameters and model selection using 98. Severud WJ, Belant JL, Windels SK, Bruggink JG. Seasonal varia- Akaike’s Information Criterion. J Wildl Manag. 2010;74(6):1175–8. tion in assimilated diets of American beavers. Am Midland Nat. 125. Hartig F. DHARMa: residual diagnostics for hierarchical (multi-level/ 2013;169:30–42. mixed) regression models. R package version 0.4.1; 2017. 99. Graf PM, Wilson RP, Sanchez LC, Hackländer K, Rosell F. Diving behavior 126. Team RC. R: a language and environment for statistical computing, r in a free-living, semi-aquatic herbivore, the Eurasian beaver Castor fiber. foundation for statistical computing, Austria; 2015, 2018. Ecol Evol. 2018;8(2):997–1008. 127. Lardet J-P. Spatial behaviour and activity patterns of the water shrew 100. Fish FE. Biomechanics and energetics in aquatic and semiaquatic mam- Neomys fodiens in the field. Acta Theriol. 1988;33(21):293–303. mals: platypus to whale. Physiol Biochem Zool. 2000;73(6):683–98. 128. Hays GC, Forman DW, Harrington LA, Harrington AL, MacDonald DW, 101. Allers D, Culik BM. Energy requirements of beavers (Castor canadensis) Righton D. Recording the free-living behaviour of small-bodied, shal- swimming underwater. Physiol Zool. 1997;70(4):456–63. low-diving animals with data loggers. J Anim Ecol. 2007;76(1):183–90. 102. Fish FE. Comparison of swimming kinematics between terrestrial and 129. Harrington LA, Hays GC, Fasola L, Harrington AL, Righton D, Macdon- semiaquatic opossums. J Mamm. 1993;74(2):275–84. ald DW. Dive performance in a small-bodied, semi-aquatic mammal 103. Pinto B, Santos MJ, Rosell F. Habitat selection of the Eurasian beaver in the wild. J Mamm. 2012;93(1):198–210. (Castor fiber) near its carrying capacity: an example from Norway. Can J 130. Bagniewska JM, Harrington LA, Hart T, Harrington AL, Fasola L, Mac- Zool. 2009;87(4):317–25. donald DW. Persistence in diving American mink. Anim Biotelemetry. 104. Olstad O. Beverens (Castor fiber) utbredelse i Norge. Statens vil- 2015;3(1):1–10. tundersøkelser. 1937;77:217–73. 131. Vogel P, Bodmer C, Spreng M, Aeschimann J, Dunstone N, Gorman 105. Herfindal I, Linnell JD, Odden J, Nilsen EB, Andersen R. Prey density, M. Diving capacity and foraging behaviour of the water shrew. Cam- environmental productivity and home-range size in the Eurasian lynx bridge: Cambridge University Press; 1998. (Lynx lynx). J Zool. 2005;265(1):63–71. 132. Ciancio JE, Quintana F, Sala JE, Wilson RP. Cold birds under pressure: 106. Rosell F, Hovde B. Methods of aquatic and terrestrial netting to capture can thermal substitution ease heat loss in diving penguins? Mar Biol. Eurasian beavers. Wildl Soc Bull. 2001;29(1):269–74. 2016;163(2):43. 107. Rosell F, Sun L. Use of anal gland secretion to distinguish the two bea- 133. Wilson RP, Hustler K, Ryan PG, Burger AE, Noldeke EC. Diving birds in ver species Castor canadensis and C. fiber. Wildl Biol. 1999;5(2):119–23. cold water: do Archimedes and Boyle determine energetic costs? Am 108. Rosell F, Zedrosser A, Parker H. Correlates of body measurements and Nat. 1992;140(2):179–200. age in Eurasian beaver from Norway. Eur J Wildl Res. 2010;56(1):43–8. 134. Wilson RP, Borger L, Holton MD, Scantlebury DM, Gomez-Laich A, 109. Mayer M, Künzel F, Zedrosser A, Rosell F. The 7-year itch: non- Quintana F, Rosell F, Graf PM, Williams H, Gunner R, Hopkins L, Marks adaptive mate change in the Eurasian beaver. Behav Ecol Sociobiol. N, Geraldi NR, Duarte CM, Scott R, Strano MS, Robotka H, Eizaguirre 2017;71(2):32. C, Fahlman A, Shepard ELC. Estimates for energy expenditure in free- 110. Fleming CH, Fagan WF, Mueller T, Olson KA, Leimgruber P, Calabrese JM. living animals using acceleration proxies: a reappraisal. J Anim Ecol. Rigorous home range estimation with movement data: a new autocor- 2020;89(1):161–72. related kernel density estimator. Ecology. 2015;96(5):1182–8. 135. Wilson RP, Shepard EL, Laich AG, Frere E, Quintana F. Pedalling down- 111. Walker JS, Jones MW, Laramee RS, Holton MD, Shepard EL, Williams hill and freewheeling up; a penguin perspective on foraging. Aquat HJ, Scantlebury DM, Marks NJ, Magowan EA, Maguire IE, Bidder OR, Di Biol. 2010;8(3):193–202. Virgilio A, Wilson RP. Prying into the intimate secrets of animal lives; soft- 136. Perrow MR, Schutten JH, Howes JR, Holzer T, Madgwick FJ, Jowitt AJ. ware beyond hardware for comprehensive annotation in “Daily Diary” Interactions between coot (Fulica atra) and submerged macrophytes: tags. Mov Ecol. 2015;3(1):29. the role of birds in the restoration process. Shallow Lakes’ 95. Berlin: 112. Lewis JS, Rachlow JL, Garton EO, Vierling LA. Eec ff ts of habitat on GPS Springer; 1997. p. 241–55. collar performance: using data screening to reduce location error. J 137. Irwin S, O’Halloran J, editors. The wintering behaviour of coot Fulica Appl Ecol. 2007;44(3):663–71. atra L. at Cork Lough, south-west Ireland. In: Biology and environ- 113. Justicia LS, Rosell F, Mayer M. Performance of GPS units for deployment ment: proceedings of the royal Irish academy. JSTOR; 1997. on semiaquatic animals. PLoS ONE. 2018;13(12):e0207938. 138. Bagniewska JM, Hart T, Harrington LA, Macdonald DW. Hidden 114. Middelboe AL, Markager S. Depth limits and minimum light require- Markov analysis describes dive patterns in semiaquatic animals. ments of freshwater macrophytes. Freshw Biol. 1997;37(3):553–68. Behav Ecol. 2013;24(3):659–67. 115. Artsdatabanken. Artsnavnebase. Norsk taksonomisk database; 2015. 139. De Leeuw JJ. Food intake rates and habitat segregation of tufted http:// www2. artsd ataba nken. no/ artsn avn/ Conte ntpag es/ Sok. aspx. duck Aythya fuligula and scaup Aythya marila exploiting zebra mus- Accessed 4 Feb 2020. sels Dreissena polymorpha. Ardea. 1999;87(1):15–31. M ortensen et al. Anim Biotelemetry (2021) 9:35 Page 19 of 19 140. Fish FE, Smelstoys J, Baudinette RV, Reynolds PS. Fur doesn’t fly, it 159. Nolet BA, Rosell F. Comeback of the beaver Castor fiber: An overview of floats: buoyancy of pelage in semi-aquatic mammals. Aquat Mamm. old and new conservation problems. Biol Cons. 1998;83(2):165–73. 2002;28(2):103–12. 160. Macdonald DW, Tattersall F, Brown E, Balharry D. Reintroducing the 141. Harrington LA, Harrington AL, Yamaguchi N, Thom MD, Ferreras P, European beaver to Britain: nostalgic meddling or restoring biodiver- Windham TR, Macdonald DW. The impact of native competitors on an sity? Mamm Rev. 1995;25(4):161–200. alien invasive: temporal niche shifts to avoid interspecific aggression. 161. Andersen DC, Shafroth PB. Beaver dams, hydrological thresholds, and Ecology. 2009;90(5):1207–16. controlled floods as a management tool in a desert riverine ecosystem, 142. Theuerkauf J, Jȩdrzejewski W, Schmidt K, Okarma H, Ruczyński I, Śniezko Bill Williams River. Arizona Ecohydrol. 2010;3(3):325–38. S, Gula R. Daily patterns and duration of wolf activity in the Białowieza 162. Nolet BA, Hoekstra A, Ottenheim MM. Selective foraging on woody spe- Forest. Poland J Mammal. 2003;84(1):243–53. cies by the beaver Castor fiber, and its impact on a riparian willow forest. 143. Rodriguez-Ramos Fernandez J, Dubielzig RR. Ocular comparative Biol Cons. 1994;70(2):117–28. anatomy of the family Rodentia. Vet Ophthalmol. 2013;16:94–9. 163. Logan M, Sanson GD. The effects of lactation on the feeding behaviour 144. Campbell-Palmer R, Rosell F. Conservation of the Eurasian beaver Castor and activity patterns of free-ranging female koalas (Phascolarctos fiber: an olfactory perspective. Mamm Rev. 2010;40(4):293–312. cinereus Goldfuss). Aust J Zool. 2003;51(4):415–28. 145. Ollivier F, Samuelson D, Brooks D, Lewis P, Kallberg M, Komáromy A. 164. Zoller H, Drygala F. Activity patterns of the invasive raccoon dog Comparative morphology of the Tapetum lucidum (among selected (Nyctereutes procyonoides) in North East Germany. J Vertebr Biol. species). Vet Ophthalmol. 2004;7(1):11–22. 2013;62(4):290–6. 146. Maffei L, Fiorentini A, Bisti S. The visual acuity of the lynx. Vision Res. 165. Cantoni D. Social and spatial organization of free-ranging shrews, Sorex 1990;30(4):527–8. coronatus and Neomys fodiens (Insectivora, Mammalia). Anim Behav. 147. Swinnen KRR, Hughes NK, Leirs H. Beaver (Castor fiber) activity patterns 1993;45(5):975–95. in a predator-free landscape. What is keeping them in the dark? Mamm 166. Lodberg-Holm HK, Steyaert S, Reinhardt S, Rosell F. Size is not every- Biol. 2015;80(6):477–83. thing: differing activity and foraging patterns between the sexes in a 148. Birt V, Birt T, Goulet D, Cairns D, Montevecchi W. Ashmole’s halo: direct monomorphic mammal. Behav Ecol Sociobiol. 2021;75(4):1–14. evidence for prey depletion by a seabird. Mar Ecol Progr Ser Oldendorf. 167. Mayer M, Zedrosser A, Rosell F. Couch potatoes do better: Delayed 1987;40(3):205–8. dispersal and territory size affect the duration of territory occupancy in 149. Elliott KH, Woo KJ, Gaston AJ, Benvenuti S, Dall’Antonia L, Davoren GK. a monogamous mammal. Ecol Evol. 2017;7(12):4347–56. Central-place foraging in an Arctic seabird provides evidence for Storer- 168. Mayer M, Fog Bjerre DH, Sunde P. Better safe than sorry: The response to Ashmole’s halo. Auk. 2009;126(3):613–25. a simulated predator and unfamiliar scent by the European hare. Ethol- 150. Nolet BA, Rosell F. Territoriality and time budgets in beavers during ogy. 2020;126(7):704–15. sequential settlement. Can J Zool. 1994;72(7):1227–37. 169. Benkwitt CE. Central-place foraging and ecological effects of an inva- 151. Lima SL. Ecological and evolutionary perspectives on escape from sive predator across multiple habitats. Ecology. 2016;97(10):2729–39. predatory attack: a survey of North American birds. The Wilson Bulletin. 170. Dale S, Gustavsen R, Slagsvold T. Risk taking during parental care: a test 1993;105(1):1–47. of three hypotheses applied to the pied flycatcher. Behav Ecol Socio - 152. Orians GH, Pearson NE. On the theory of central place foraging. In: biol. 1996;39(1):31–42. Horn DJ, Mitchell RD, Stairs GR, editors. Analysis of ecological systems. 171. Chimienti M, Cornulier T, Owen E, Bolton M, Davies IM, Travis JMJ, Scott Columbus: The Ohio State University Press; 1979. p. 154–77. BE. The use of an unsupervised learning approach for characterizing 153. Wetterer JK. Central place foraging theory: when load size affects travel latent behaviors in accelerometer data. Ecol Evol. 2016;6(3):727–41. time. Theor Popul Biol. 1989;36(3):267–80. 172. Gunner RM, Wilson RP, Holton MD, Scott R, Hopkins P, Duarte CM. A new 154. Houston AI, McNamara JM. A general theory of central place foraging direction for differentiating animal activity based on measuring angular for single-prey loaders. Theor Popul Biol. 1985;28(3):233–62. velocity about the yaw axis. Ecol Evol. 2020;10(14):7872–86. 155. DuPasquier A, Cantoni D. Shifts in benthic macroinvertebrate com- 173. Buchanan K, Burt de Perera T, Carere C, Carter T, Hailey A, Hubrecht R, munity and food habits of the water shrew, Neomys fodiens (Soricidae, Jennings D, Metcalfe N, Pitcher T, Peron F. Guidelines for the treat- Insectivora). Acta Oecol. 1992;13(1):81–99. ment of animals in behavioural research and teaching. Anim Behav. 156. Goryainova Z, Katsman E, Zavyalov N, Khlyap L, Petrosyan V. Evaluation 2012;83(1):301–9. of tree and shrub resources of the Eurasian beaver (Castor fiber L.) and 174. Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improving changes in beaver foraging strategy after resources depletion. Russ J bioscience research reporting: the ARRIVE guidelines for reporting Biol Invas. 2014;5(4):242–54. animal research. PLoS Biol. 2010;8(6):e1000412. 157. Danilov P, Fyodorov F. Comparative characterization of the building activity of Canadian and European beavers in northern European Rus- Publisher’s Note sia. Russ J Ecol. 2015;46(3):272–8. Springer Nature remains neutral with regard to jurisdictional claims in pub- 158. Collen P, Gibson R. The general ecology of beavers (Castor spp.), as lished maps and institutional affiliations. related to their influence on stream ecosystems and riparian habitats, and the subsequent effects on fish—a review. Rev Fish Biol Fish. 2000;10(4):439–61. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Animal BiotelemetrySpringer Journals

Published: Aug 18, 2021

Keywords: Aquatic foraging; Behavioural ecology; Castor fibre; Dead-reckoning; Habitat selection; Movement ecology; Resource selection functions

References