https://orcid.org/0000-0003-1515-3440
Figures
Abstract
The movement behavior of open ocean fishes is challenged by metabolic demands resulting from sustained swimming and the availability of resources in a dynamic, ephemeral environment. Advances in electronic tagging and tracking technologies have permitted unprecedented opportunities to describe the movements of open ocean fishes in these environments, however, our understanding of the mechanistic drivers of individual variation in movement performance is limited. In this study, we tested the hypothesis that movement capacities of open ocean sharks would be related to physiology and body condition. We measured the physiological status (e.g., energy stores and body condition) of blue sharks (Prionace glauca) captured in the open ocean, and then tracked their movements over 45 days. We then explored for relationships between physiological metrics and individual differences in metrics of movement behavior (distance traveled, activity space, behavioral state, and tortuosity) of the tracked blue sharks. Analyses detected consistent positive relationships between individual plasma triglyceride concentrations and body condition (sampled at time of capture) on distance traveled, activity space, behavioral state, and tortuosity - up to 45 days post tagging, with models explaining up to 79% of individual variation in movement. These findings highlight the potential role of metabolic lipid reserves in shaping movement behavior of open ocean, predatory sharks in patchy, ephemeral environments. More broadly, they offer new insight into the factors which may influence individual variation in the timing and scale of movement in oceanic fishes.
Citation: Gallagher AJ, Byrnes EB, Skubel RA, Raymond B, Romeiro J, Cooke SJ, et al. (2026) Linking physiological state to movement dynamics in an open ocean predator, the blue shark (Prionace glauca). PLoS One 21(1): e0337589. https://doi.org/10.1371/journal.pone.0337589
Editor: Joel Harrison Gayford, James Cook University, AUSTRALIA
Received: May 23, 2025; Accepted: November 10, 2025; Published: January 7, 2026
Copyright: © 2026 Gallagher et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets used and/or analyzed for this study are available as Supplemental Information alongside the paper.
Funding: Oceana. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: Author BR was employed by the company Pelagic Expeditions, and Author JR was employed by the company 333 Productions. These both have declared that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. All remaining authors have declared that no competing interests exist.
Introduction
Electronic tagging has transformed the research community’s ability to track and describe the movements of highly mobile oceanic predators [1,2,3], capturing large-scale displacements, fine-scale movement patterns, and variation among individuals in space use and performance [e.g., 4,5,6]. These datasets have also been collated, synthesized, and integrated with big data approaches to inform conservation and management at ocean-basin scales [7,8]. While these advances have revealed remarkable complexity in movement behavior, the mechanisms underlying such individual differences remain poorly understood [9].
A comparative study of movements across different species of marine megafauna found that displacement, a daily measure of performance, was generally not affected by body length or mass [10], but instead, that complex movements were more related to microhabitat complexities which were hypothesized to be linked to nutritional or foraging-related benefits. Interestingly, work utilizing high-resolution tracking demonstrated that large predatory shark species such as the white shark (Carcharodon carcharias) and blue shark (Prionace glauca) seek out mesoscale oceanographic features known as eddies, where temperature anomalies create slightly warmed conditions and presumed access to energetically profitable prey [11–13]. These studies provided new insights into the importance of biological and physical features, which may influence dynamic animal movement in the open ocean.
Energy acquisition plays a major role in the movement ecology of fishes (reviewed in [14], and a horizon scan of the top 100 most-pressing questions for understanding the future of fish migration/movement, conservation, and policy identified the concept of internal state as the very first item on the list: “What are the internal physiological drivers of, and triggers for, migration?” [15]. While the theoretical link between animal movement and energy acquisition is well understood, studies integrating measures of internal physiology with movement are surprisingly lacking. However, there are a suite of non-invasive tools available to researchers to obtain small tissue biopsies (e.g., blood or muscle sample) from tagged animals, providing an opportunity to explore relationships between aspects of organismal condition (e.g., physiological status, health, gene expression) and their subsequent behavior and fate [1]. Furthermore, certain physiological endpoints assayed from blood plasma can be used as biomarkers for relevant questions related to energy use, metabolism, stress, and even prey quality [16–19]. One of the first studies to adopt this approach was on adult Pacific salmon (Oncorhynchus spp.), demonstrating physiological correlates of extensive movement behavior and premature mortality, thus revealing a mechanistic basis between movement behavior and fate [20].
Sharks generally store large quantities of fats in their liver, which can constitute a significant proportion of their body mass [21]. These reserves play a central role in lipid metabolism during extended movements and throughout parturition [19,22,23]. Triglycerides (TAG, as abbreviated in the present study) from the diet are conservatively transferred to heptic storage [24,25], and the low density of the liver relative to seawater also provides lift, thereby reducing locomotory costs [26]. Ketone bodies are recognized as a primary fuel during activity, but plasma triglycerides (TAG) remain a widely used and accessible proxy for long-term energy stores because they can be readily quantified from small blood samples in sharks [17] and have been linked to overall energetic condition in several species [16,18,27,28]. While the lipid metabolism of sharks is a complex and a somewhat limited area of study [16,17,29], these dynamics highlight the importance of the liver in both energy storage and movement ecology, and thus the value of examining lipid metabolites such as TAG in migratory sharks.
In the present study, we investigated the potential physiological and behavioral correlates of intra-specific differences in movement of an oceanic predator, the blue shark, over short time sales (up to 45 days). Operating under the hypothesis that variability in movement may be related to differences in physiological state [30], we explored whether individual differences in distance traveled, activity space, behavioral state, and tortuosity of tagged blue sharks were correlated with plasma triglyceride levels (a measure of endogenous energy reserves), body condition (a proxy of overall health), and shark body length. Blue sharks are a useful oceanic species to examine these patterns as they are known to target open-ocean ephemeral features, like mesoscale eddies, along ocean frontal zones, where they are believed to feed [12,13].
Materials and methods
Capture, body measurements and physiological sampling
Blue sharks were captured using baited circle-hook handlines in federal waters off Point Judith, Rhode Island, USA (41.3615 N, −71.48139 W) from June 1 – September 1, 2016 (Table 1, Fig 1], overlapping with their seasonal offshore aggregation in the region [31]. All animal tagging and capture was ethically approved by the Carleton University Animal Care Committee. Sharks were immediately secured on a platform inside the boat and a saltwater hose and pump was inserted in the shark’s mouth. Each shark was measured for total length (TL in cm) and an additional five measurements were taken to estimate body condition based on a Span Condition Analysis (SCA) following Irschick and Hammerschlag [32]. These measures included: (1) pre-caudal length [PCL], the distance from the tip of the shark’s snout to the caudal peduncle; (2) lateral span [LS], the distance spanning across the surface of the shark between the anterior insertion points of the pectoral fins; (3) frontal span [FS], measured at the anterior insertion point of the first dorsal fin, the distance spanning the surface of the shark between lines parallel to the frontal plane that extend from the pectoral fins to the caudal fin; (4) proximal span [PS], measured at the posterior insertion point of the first dorsal fin, the distance spanning the surface of the shark between lines parallel to the frontal plane that extend from the pectoral fins to the caudal fin; and (5) caudal keel circumference [CKC], measured at the caudal peduncle, the circumference at the base of the caudal fin. From these measurements, the body condition of each shark was calculated, using the SCA equation: SCA = Σ (LS + FS + PS + CKC)/(PCL). Larger SCA values are suggestive of higher body condition, whereas lower SCA values confer a poorer condition [32].
The IDs of tracks are ordered and color coded by track length as follows: track length increases from IDs ordered along the blue to red color range as identified in the figure legend. Similarly, IDs listed earlier in the legend have relatively shorter tracking durations (colder color palate) and those ordered towards the end of the list have longer tracking durations (warmer color palate). Left display shows broader latitudinal range, whereas the right panel is zoomed in on those tracks from the core area off New England. Map developed in ESRI ArcGIS 10.3.3. Reprinted from Esri under a CC BY license with permission from Esri, original copyright 2016.
A 7mL blood sample was taken via caudal venipuncture on un-anesthetized sharks [33]. Whole blood was immediately spun down on the boat (5000 rpm for 4 minutes) and the plasma was then separated into three 2 mL vials and kept in a water-ice slurry for up to 8 hours before being transferred to a −20°C freezer back on land [34]. We ran metabolic assays on the frozen plasma samples to quantify levels of total plasma triglycerides [free and bound to lipoproteins] (TAG; mmol l–1, EnzyChrom Triglyceride Assay Kit, BioAssay Systems, Haywood, CA, USA) on a 96-well microplate absorbance reader at 570 nm (Tecan Sunrise, Tecan, Austria). We chose TAG because fatty acids have been suggested to serve as a useful metric for evaluating shark condition and energy over time [16,18,27,28], and the nature of the metabolite in sharks is such that its presence in the plasma is derivative from stored endogenous reserves [29]. While the exact turnover rate from feeding to TAG storage is unknown in sharks, the short-term time period of our study (i.e., + /- 1 month) was chosen such that the internal TAG values were likely reflective of recent feeding events along the same temporal scale (e.g., the time frame from when exogenous energy will become metabolically visible in blood chemistry; [25]). Concentrations of TAG values were determined using the appropriate standard curves. Inter-assay variation was 2.0%.
Satellite tagging and spatial data treatment
Prior to release, each shark was tagged with a Smart Position and Temperature Transmitting tag (SPOT6, Wildlife Computers). Tags were mounted to the shark’s first dorsal fin using titanium bolts, steel washers and high carbon steel nuts following Hammerschlag et al. [35]). Geographic locations of SPOT tagged sharks were determined by Doppler-shift calculations made by the Argos Data Collection and Location Service (Argos CLS; www.argos-system.org). Locations were acquired when the tag broke the water surface and transmitted a signal to passing Argos satellite. To improve location accuracy, all Doppler derived data were processed automatically by Argos CLS using Kalman filtering (KF). Argos provides location accuracy using location classes (LC) 3, 2, 1, 0, A, B, and Z (in decreasing accuracy), corresponding with the following error estimates: LC3 < 250 m, 250 m < LC2 < 500 m, and 500 m < LC1 < 1500 m. The error estimates associated with LC A and B are reported to be > 1 km and >5 km, respectively [36]. LC Z estimates are inaccurate or unreliable and were removed from the dataset prior to any analysis.
Due to irregular surfacing of sharks (and thus irregular transmission rates) and variation in satellite coverage at any given time, raw SPOT-derived data are often subject to data gaps, autocorrelation and spatial biases. Therefore, prior to any spatial analyses, several steps were taken to alleviate these potential issues, including track interpolation and regularization [see 6]. To prevent interpolating between positions with large temporal gaps, individual tracks with data gaps of more than five days between subsequent positions and/or individual tracks with less than 10 positions were split into separate track segments. Next, a hierarchical, first-difference, correlated, random-walk, state-space switching model [SSM; 37,38] was applied to these data following [39]. Specifically, position data were interpolated and regularized to daily estimates in the R statistical software package ‘bsam.’ This package also applied a speed filter of 5 m/s. All subsequent movement analyses were conducted on the interpolated, regularized positions estimates from the SSM, hereafter referred as position relocations.
Movement performance metrics
We calculated four metrics of movement behavior for each shark: distance travelled (km), behavioral state, tortuosity, and activity space (km2). Behavioral state and tortuosity were chosen because they have previously been linked to foraging activities. More spatially restricted, convoluted movements are associated with exploiting resources, whereas more directed, transient movements are associated with moving between resource patches. Moreover, switching behavior states may be part of an overall oceanic foraging strategy for species like blue sharks. To allow comparisons among individuals in movement performance over standardized tracking periods and to also permit evaluating potential relationships between biological variables and movement metrics for different tracking durations, all movement performance metrics (including tortuosity and behavioral state) were calculated for cumulative five-day tracking intervals (i.e., 0–5 days, 0–10 days, 0–15 days etc., inclusive). Analysis of movement data over cumulative five-day intervals for the 45-day period was chosen as it aligns with the relative window in which some lipid metabolites are known to continue to increase in sharks following feeding [25]. Movement metrics were not generated beyond 45 days because the satellite tag batteries of most tracked sharks were depleted due to unexpected high rate of transmissions (average 1000 transmissions per day) from blue shark surfacing [40,41]. Despite this technological limitation, this surface behavior likely benefitted the spatial accuracy of onboard tags as they had more time to uplink to Earth-orbiting satellites [42].
Distance traveled was calculated as the total Euclidian distance (km) moved for each 24-hour period, summed over each cumulative five-day tracking interval. Estimates of activity space (km2) were calculated for each cumulative five-day tracking interval by determining the area of the smallest polygon that encompassed all position relocations. Metrics of activity space were computed using the Minimum Bounding Geometry tool in ArcGIS 10.3.3. [ESRI 2016]. This metric of activity space provides a simple metric of evaluating space use, without any assumption about habitat preferences. While other approaches for evaluating space use are available, such as kernel density estimates [43], such metrics are used to investigate home range and spatial preferences, which was not the goal of this work. Tortuosity, a measure of complexity in an animal’s track, was calculated here using VFractal [44] to give an estimate of the interpolated track’s turning angle across each discrete tracking interval. Measured between 1 and 2, lower values near 1 suggest more direct path movement, whereas values approaching 2 signify tortuous and complex path movements. For each independent tracking interval, behavioral state – the classification of movement patterns into categories that approximate underlying behaviors – was calculated from data generated from the Bayesian state-spaced model [39], which allows for individuals to switch between two behavioral states: a “resident” state (area-restricted searching, as values approach ‘2’), and a “transient” state (ranging behavior, as values approach ‘1’). The state-spaced model output produced a behavioral state metric for each 24-hour period, with a value ranging on a continuum between 1 (transient) and 2 (resident). Uncertainty in the value of behavioral state at time T is quantified with a Markov Chain Monte Carlo (MCMC) algorithm providing values between 1 and 2. For movement analyses, we calculated the mean behavioral state value and tortuosity for each cumulative five-day tracking interval for each shark.
All research activities and were acknowledged and approved under formal Letters of Acknowledgement (LOA) issued by the National Marine Fisheries Service to AJG and NH (SHK-LOA-16). All activities took place in US Federal waters; none of these areas were privately owned nor were they protected under any specific management plans. All research was non-take, and no protected species were sampled. The research and protocols conducted in this study complied with relevant ethical regulations as approved by the Carleton University Animal Care Committee.
Statistical analyses
We evaluated the relationships between internal state (TAG, body condition, shark size) and movement (distance traveled, rate of movement, activity space, tortuosity, behavioral state) over the entire tracking 45-day tracking duration using a combination of linear (LM) and generalized linear models (GLM). For distance traveled and activity space, we observed that these metrics of animal movement often saturate toward some maximum, typically in a log-linear fashion [e.g., due to home-ranging or otherwise spatially restricted movements; [45,46]. Thus, we structured these LMs such that each log-transformed response variable (distance traveled, activity space) was expected to increase with tracking interval but at diminishing rates following a simple quadratic function (i.e., , where t is the time interval since capture). However, our hypotheses posited that the rate of change in the response with time is determined by the physiological and morphological traits of the animals at capture. We thus included first-order interactions between the main linear effect of time and TAG, body condition, and shark size (TL). This parameterization allowed us to specifically ask, after accommodating the known and strong effect of time on our movement metrics: do our physiological and morphological measurements of the animals meaningfully explain the remaining variation in movement metrics? If our hypotheses were correct, we expected to see strong interactions between t and TAG, body condition, and TL. We also centered-and-scaled TAG, body condition, and TL such that their respective effects on each response would be directly comparable. This model structure allowed us to evaluate whether TAG, body condition, and TL had any effect on the response variables after controlling for the well-established log-linear relationships between the response variables and time [45]. Additionally, the center-and-scaling allows us to compare the predicted effects on our movement metrics at low (−2 standard deviations from the mean), medium (the mean), and high (+2 standard deviations from the mean) values of each physiological or morphological trait.
Tortuosity values from VFractal and behavioral state estimates from the state-space model are both naturally bounded between 1 and 2. For consistency in modeling, we rescaled both variables to the interval (0,1) by subtracting 1 from all values. We then constructed GLMs for these response variables assuming a beta distribution with a logit link, following the same model structure as for the LMs described above (i.e., logit Ŷ = β₀ + β₁t + β₂t², with first-order interactions between t and centered-and-scaled TAG, body condition, and TL).
Diagnostic plots for all LMs and GLMs revealed no violations of model assumptions or overdispersion. VIF scores indicated multicollinearity only for time interval and its square (as expected since one is derived from the other). Models were evaluated with respect to R2 (LM) or pseudo-R2 (GLM), with statistical significance declared at . All analyses were performed in R v. 4.0.4 (R Core Team 2021).
Results
A total of 13 male blue sharks were tagged with fin-mounted SPOT satellite tags and tracked between of 12–82 days (Table 1, Fig 1). Only male blue sharks were encountered during sampling (sex-specific aggregations are not uncommon in the region), and all sharks were of a similar size class suggesting maturity; the average body length was 264.0 ± 21.0 cm (mean ± SD; cm, total length). TAG ranged from 0.02–0.95 mmol/L, with a mean of 0.30 ± 0.30 mmol l–1. Body condition ranged from 0.89–1.09 (0.98 ± 0.06). The total distance traveled for all sharks combined was 9,511 km across a total of 455 tracking days (Fig 1a), with an average rate of movement of 18.49 ± 11.99 km/day across this period. The total activity space of all sharks tracked across the analysis period was 592,624 km2 (Fig 2a). The mean behavioral state exhibited by all tracked sharks throughout the sampling period was 1.74 ± 0.23 (Fig 3a), suggesting an overall resident state, whereas calculated tortuosity values averaged 0.70 ± 0.10, suggesting mostly linear movement.
The individual IDs for activity space are ordered and color coded by track length as follows: activity space increases from IDs ordered along the blue to red color range as identified in the figure legend. Similarly, IDs listed earlier in the legend have relatively smaller activity spaces (colder color palate) and those ordered towards the end of the list have relatively larger activity spaces (warmer color palate). Left display shows broader latitudinal range, whereas the right panel is zoomed in on those tracks from the core area off New England. Map developed in ESRI ArcGIS 10.3.3. Reprinted from Esri under a CC BY license with permission from Esri, original copyright 2016.
Darker color shades represent values closer to 2, suggesting a “resident” state; whereas lighter color shades approaching 1 suggest a “transient” state. Left display shows broader latitudinal range, whereas the right panel is zoomed in on those tracks from the core area off New England. Map developed in ESRI ArcGIS 10.3.3. Reprinted from Esri under a CC BY license with permission from Esri, original copyright 2016.
When controlling for elapsed track time, distance traveled was positively and significantly influenced by TAG (t = 5.38, p < 0.001) and condition (t = 6.02, p < 0.001) while keeping constant the other variable, with the overall model explaining 79% of the variation (F5,84 = 66.17, p < 0.001; Table 2, Fig 4). Using the same modeling approach, activity space was positively and significantly influenced by TAG (t = 5.37, p < 0.001) and condition (t = 5.61, p < 0.001) while keeping constant the other variable, with the overall model explaining 65% of the variation (F5,84 = 30.72, p < 0.001; Table 2, Fig 4). Across all 3 of these models, we consistently observed that both TAG and body condition had strong positive interacting effects with time elapsed since capture (Table 2, Fig 4). Total length (TL) also had a positive interaction effect on these responses, but its effect was substantially smaller than either TAG or body condition, and consistently insignificant across all LMs (Table 2).
Lines are drawn following the regression equation as described in the Methods, corresponding to “low, medium, or high” levels of TAG, body condition, or TL (while keeping the other 2 constant). A “low” value is 2 standard deviations below the mean, a “high” value is 2 standard deviations above the mean, and a “medium” value is the mean value for TAG, condition, or TL.
The beta regression model for tortuosity over the elapsed track time showed a significant (positive) effect of only body condition (z = 2.63, p < 0.001), suggesting that after controlling for all independent variables, every one-unit increase in condition was associated with a 0.007 unit increase in tortuosity (i.e., higher path complexity, Table 2, Fig 4). The beta regression model for behavioral state showed significant effects of TAG (z = −2.71, p < 0.01) and condition (z = −3.12, p < 0.01), suggesting that when TAG and condition were each interpedently increased by one-unit, there were decreases in the logit of behavioral state (i.e., a trend towards ‘transient’ behavior) by 0.007 and 0.009, respectively, after controlling for all independent variables (Table 2). Neither tortuosity nor behavioral state exhibited statistically significant relationships with time internal or its square, implying that these are temporally scale-invariant phenomena (Table 2). Animal size did not show a significant effect in any of the models across each of the movement metrics (Table 2).
Discussion
Understanding the physiological drivers of behavioral plasticity in open ocean fishes offers a pathway to link individual condition with variable patterns of movement and distribution. A recent meta-analysis explored whether physiological traits predicted activity, exploration, and dispersal, revealing effects of metabolism and locomotory performance, which were especially strong in birds and amphibians [30]. Our results show that physiological metrics were associated with variation in short-term (up to 45 days) movement dynamics of blue sharks in the open ocean. This conclusion was supported by the observed correlative effects between stored energy and body condition on both distance traveled and activity space, as well as behavioral state (which were independent of tracking duration).
The total distance traveled, and activity space covered, by blue sharks in our study showed positive relationships with plasma triglycerides when controlling for total elapsed track time on an individual basis (Fig 4), with these models explaining a relatively high (64–79%) proportion of the variation. Taken together, these results suggest that higher levels of liver-derived lipid stores may increase swimming distance and activity space in open-ocean sharks for periods of up to a month and a half. Given that animals will expand their ranges or move between distant patches when local resources are depleted or if resources are improved in other areas, it is possible that blue sharks which have recently exploited a feeding patch, thus exhibiting higher triglyceride levels, will subsequently be able to undertake more dispersive movements and or exploit distant resource patches (see Fig 5 for a conceptual diagram of these relationships). Indeed, plasma TAG have been correlated with endurance-flight in migratory bird species [47], whereby individuals will ‘fatten-up’ before long periods of flight or re-fuel at known ‘stop-over’ sites [48]. While TAG could, in theory, reflect prior foraging success, this is unlikely in our case as sharks were encountered at the start of the seasonal aggregation, following their northward migration to the shelf where feeding is expected to occur. We therefore interpret elevated TAG as indicative of capacity for future movement, while acknowledging that it may also integrate aspects of recent energetic history.
Triglyceride (as seen in the molecular structures) energy reserves are stored at high levels ( larger molecules suggesting higher levels) and should reflect elevated condition (sharks on top have greater ‘girth’) and should be utilized and depleted over a period of weeks. Sharks with higher energy may remain in transient states at longer periods, whereas those with lower energy may switch to a more ‘resident’ state, whereby individuals may search to acquire prey resources to rebuild energy stores, thus preparing for the next phase of their movement.
When fed lipid-rich diets, large open ocean sharks also increase their body condition (e.g., girth relative to length) as these metabolites are deposited in liver tissue [26]. However, body condition is more of a whole-body proxy of overall health accumulated over longer periods than circulating lipid metabolites [49,50,23]. Indeed, body condition has been linked to health and internal energy stores in some species of shark [27]. Body condition was also positively associated with distance traveled and activity space in tagged blue sharks, perhaps reflecting an important dynamic between stored energy and the capacity to move or the ability to take advantage of distant resource patches. Alternatively, but not mutually exclusive, individuals exhibiting higher body condition need not be constrained by local food resources and may search for more profitable resource patches.
Behavioral state analyses of tagged blue sharks suggested that animals with higher energy reserves at the time of capture exhibited more ‘transient’ behavior and more directed movements over the tracking period. We interpret this result to be reflective of a metabolic benefit of stored energy, whereby endogenous energy stores were metabolized to power directed movement to take advantage of (or to search for) more distant areas/resources (Fig 5). Conversely, those individuals with lower energy at the time of capture may not have sufficient energy reserves to exploit distant resource patches and thus show more restricted, convoluted movements, in search of local resources. Yet, behavioral switching may indeed be part of the strategy for these species in the open ocean, as previous data suggest that pelagic sharks periodically adjust their behavioral state during extensive oceanic movements. Indeed, previous research using satellite tags to examine the movements of blue sharks demonstrated behavioral switching from ‘transiency’ to ‘residency’ within the time frame of 20–30 days [40,51]. Moreover, other research demonstrated that blue sharks exhibited frequent ‘knifing’ surfacing behavior, thought to be linked to prey searching [off Ireland, 42]. Work has shown that blue sharks seek out mesoscale oceanographic features (e.g., anticyclonic eddies), possibly to take advantage of deep, mesopelagic prey [off northastern USA, 12,13], with a similar marked switching of modeled behavioral states. Variation in biological oceanography and associated prey fields clearly play a role in determining the decision-making of oceanic species, and these features (e.g., mesopelagic prey, deep scattering layers), if exploited, may likely interact with internal energetic state. It is noteworthy that neither time interval nor its square were statistically significant in the tortuosity and behavioral state models (Table 2). This is unsurprising for behavioral state, which should not necessarily be expected to vary with time, and tortuosity, which was estimated using fractal dimension and is thus by definition scale-invariant [44,46].
Due to the correlative nature of our study and the somewhat limited information on the dyanmics of energy metabolism in large, free-ranging sharks, there are a few important assumptions which deserve transparency. First, plasma triglyceride concentrations depend on a number of factors such as season, maturity status, and gender [29]. However, all tagged sharks were males of similar length and sampled at the same location during the same time of year (summer). It should be noted that although we did not test for the effects of temperature on movement in this study, all sharks were tagged and tracked during the summer season, and the variation in sea-surface temperature was less than 4°C for the temporal period where the majority of individuals (11 out of the 13 blue sharks) were tracked (National Oceanic and Atmospheric Administration 2016). All sharks were also captured using handlines with negligible hook times (< 3 minutes), and were landed instantly, thus reducing potential stress-related effects on energy mobilization. Additional replication and increased satellite track lengths will help to further test the relationships we explored here. We also attempted to mitigate differences of individual tracking duration on our study by standardizing each analysis to a discrete temporal widow. Yet, despite our somewhat limited sample of sharks, we were able to quantitatively ascribe a large proportion of the variation (>75% in many cases) in movement metrics and behavior to a small set of three metabolic variables. Ketone bodies, which were not sampled here, are preferentially oxidized in muscle tissue relative to lipids [24], and as such they may be an important source of metabolic fuel for sharks which spend a significant portion of their lives in the open ocean [17]. Therefore, additional metabolite assays on parameters such as ketone bodies should be considered in future studies to look at differences in metabolic strategies in relation to triglyceride usage across a greater sample of sizes and species. We recognize that snapshot measures of physiology are challenging for pinpointing a specific element of an animal’s life history or any associated physiological mechanisms, and innovations to allow for repeated sampling of blood or metabolic state while an animal is being tracked (using bio-logging devices fitted with probes and sensors) are an important consideration for future work. Furthermore, as the omics toolbox expands, assessing relationships between gene expression and movement behavior should become increasingly common [e.g., 52] and will represent a new frontier in animal performance studies in the wild. As with such studies, including our own here, we caution that the relationships revealed are indeed correlative and we were unable to experimentally test the physiological mechanisms driving variability in movement behaviors.
Conclusions
Assessments of the individual physiological state have been surprisingly underrepresented in movement studies of marine fishes despite their potential in explaining patterns of variation. A primary result from this study stems from the inferred influence of triglycerides and body condition on the short-term movements of migratory, oceanic sharks obtained by satellite tag telemetry [28,51,53–62]. Our results suggest that variation in these metrics may be important for determining the scale and type of movement of large pelagic consumers, and as such future studies integrating physiology and behavior may help explain variable dynamics resulting from short-term, partial, and long-term movements in tracked animals.
Acknowledgments
Research efforts were permitted under National Marine Fisheries Service LOA’s for federal fisheries research provided to AJG and NH. This project would not have been possible without the dedicated support of Rhode Island Shark Diving, the University of Miami, and Beneath The Waves.
References
- 1. Hussey NE, Kessel ST, Aarestrup K, Cooke SJ, Cowley PD, Fisk AT, et al. Aquatic animal telemetry: a panoramic window into the underwater world. Science. 2015;348(6240):1255642.
- 2. Calich HJ, Rodríguez JP, Eguíluz VM, Hammerschlag N, Pattiaratchi C, Duarte CM, et al. Comprehensive analytical approaches reveal species‐specific search strategies in sympatric apex predatory sharks. Ecography. 2021;44(10):1544–56.
- 3. Renshaw S, Hammerschlag N, Gallagher AJ, Lubitz N, Sims DW. Global tracking of shark movements, behaviour and ecology: a review of the renaissance years of satellite tagging studies, 2010–2020. Journal of Experimental Marine Biology and Ecology. 2023;560:151841.
- 4. Lea JSE, Wetherbee BM, Queiroz N, Burnie N, Aming C, Sousa LL, et al. Repeated, long-distance migrations by a philopatric predator targeting highly contrasting ecosystems. Sci Rep. 2015;5:11202. pmid:26057337
- 5. Rooker JR, Dance MA, Wells RJD, Ajemian MJ, Block BA, Castleton MR, et al. Population connectivity of pelagic megafauna in the Cuba-Mexico-United States triangle. Sci Rep. 2019;9(1):1663. pmid:30733508
- 6. Hammerschlag N, McDonnell LH, Rider MJ, Street GM, Hazen EL, Natanson LJ, et al. Ocean warming alters the distributional range, migratory timing, and spatial protections of an apex predator, the tiger shark (Galeocerdo cuvier). Glob Chang Biol. 2022;28(6):1990–2005. pmid:35023247
- 7. Queiroz N, Humphries NE, Couto A, Vedor M, da Costa I, Sequeira AMM, et al. Global spatial risk assessment of sharks under the footprint of fisheries. Nature. 2019;572(7770):461–6. pmid:31340216
- 8. Sequeira AM, Rodríguez JP, Marley SA, Calich HJ, van der Mheen M, VanCompernolle M, et al. Global tracking of marine megafauna space use reveals how to achieve conservation targets. Science. 2025;388(6751):1086–97.
- 9. Hays GC, Ferreira LC, Sequeira AM, Meekan MG, Duarte CM, Bailey H, et al. Key questions in marine megafauna movement ecology. Trends Ecol Evol. 2016;31(6):463–75. https://doi.org/10.1016/j.tree.2016.02.015
- 10. Sequeira AMM, Rodríguez JP, Eguíluz VM, Harcourt R, Hindell M, Sims DW, et al. Convergence of marine megafauna movement patterns in coastal and open oceans. Proc Natl Acad Sci. 2018;115(12):3072–7. https://doi.org/10.1073/pnas.1716137115
- 11. Gaube P, Braun CD, Lawson GL, McGillicuddy Jr DJ, Penna AD, Skomal GB, Fischer C, Thorrold SR. Mesoscale eddies influence the movements of mature female white sharks in the Gulf Stream and Sargasso Sea. Scientific reports. 2018 May 9;8(1):7363.
- 12. Braun CD, Gaube P, Sinclair-Taylor TH, Skomal GB, Thorrold SR. Mesoscale eddies release pelagic sharks from thermal constraints to foraging in the ocean twilight zone. Proc Natl Acad Sci U S A. 2019;116(35):17187–92. pmid:31387979
- 13. Braun CD, Della Penna A, Arostegui MC, Afonso P, Berumen ML, Block BA, et al. Linking vertical movements of large pelagic predators with distribution patterns of biomass in the open ocean. Proc Natl Acad Sci U S A. 2023;120(47):e2306357120. pmid:38150462
- 14. Cooke SJ, Bergman JN, Twardek WM, Piczak ML, Casselberry GA, Lutek K, et al. The movement ecology of fishes. J Fish Biol. 2022;101(4):756–79. pmid:35788929
- 15. Lennox RJ, Engler-Palma C, Kowarski K, Filous A, Whitlock R, Cooke SJ, et al. Optimizing marine spatial plans with animal tracking data. Can J Fish Aquat Sci. 2019;76(3):497–509.
- 16. Gallagher AJ, Skubel RA, Pethybridge HR, Hammerschlag N. Energy metabolism in mobile, wild-sampled sharks inferred by plasma lipids. Conserv Physiol. 2017;5(1):cox002. pmid:28852506
- 17. Gallagher A, Meyer L, Pethybridge H, Huveneers C, Butcher P. Effects of short-term capture on the physiology of white sharks Carcharodon carcharias: amino acids and fatty acids. Endang Species Res. 2019;40:297–308.
- 18. Meyer L, Pethybridge H, Nichols PD, Beckmann C, Huveneers C. Abiotic and biotic drivers of fatty acid tracers in ecology: A global analysis of chondrichthyan profiles. Functional Ecology. 2019;33(7):1243–55.
- 19. S. Rangel B, Hammerschlag N, Sulikowski JA, Moreira RG. Physiological markers suggest energetic and nutritional adjustments in male sharks linked to reproduction. Oecologia. 2021;196(4):989–1004. https://doi.org/10.1007/s00442-021-04999-4
- 20. Cooke SJ, Hinch SG, Crossin GT, Patterson DA, English KK, Healey MC, et al. Mechanistic basis of individual mortality in Pacific salmon during spawning migrations. Ecology. 2006;87(6):1575–86. pmid:16869433
- 21. Hussey NE, Cocks DT, Dudley SF, McCarthy ID, Wintner SP. The condition conundrum: application of multiple condition indices to the dusky shark Carcharhinus obscurus. Marine Ecology Progress Series. 2009;380:199–212.
- 22. Meyer L, Pethybridge H, Nichols PD, Beckmann C, Bruce BD, Werry JM, et al. Assessing the Functional Limitations of Lipids and Fatty Acids for Diet Determination: The Importance of Tissue Type, Quantity, and Quality. Front Mar Sci. 2017;4.
- 23. Hammerschlag N, Skubel RA, Sulikowski J, Irschick DJ, Gallagher AJ. A Comparison of Reproductive and Energetic States in a Marine Apex Predator (the Tiger Shark, Galeocerdo cuvier). Physiol Biochem Zool. 2018;91(4):933–42. pmid:29782225
- 24. Speers-Roesch B, Treberg JR. The unusual energy metabolism of elasmobranch fishes. Comp Biochem Physiol A Mol Integr Physiol. 2010;155(4):417–34. pmid:19822221
- 25. Beckmann CL, Mitchell JG, Seuront L, Stone DAJ, Huveneers C. Experimental evaluation of fatty acid profiles as a technique to determine dietary composition in benthic elasmobranchs. Physiol Biochem Zool. 2013;86(2):266–78. pmid:23434786
- 26. Del Raye G, Jorgensen SJ, Krumhansl K, Ezcurra JM, Block BA. Travelling light: white sharks (Carcharodon carcharias) rely on body lipid stores to power ocean-basin scale migration. Proc Biol Sci. 2013;280(1766):20130836. pmid:23864595
- 27. Gallagher AJ, Wagner DN, Irschick DJ, Hammerschlag N. Body condition predicts energy stores in apex predatory sharks. Conserv Physiol. 2014;2(1):cou022. pmid:27293643
- 28. Moorhead SG, Gallagher AJ, Merly L, Hammerschlag N. Variation of body condition and plasma energy substrates with life stage, sex, and season in wild-sampled nurse sharks Ginglymostoma cirratum. J Fish Biol. 2021;98(3):680–93. pmid:33161578
- 29. Ballantyne JS. Jaws: The Inside Story. The Metabolism of Elasmobranch Fishes. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology. 1997;118(4):703–42.
- 30. Wu NC, Seebacher F. Physiology can predict animal activity, exploration, and dispersal. Commun Biol. 2022;5(1):109. pmid:35115649
- 31. Shea BD, Coulter SK, Dooling KE, Isihara HL, Roth JC, Sudal E, et al. Recreational fishing fight times are not correlated with physiological status of blue sharks (Prionace glauca) in the Northwestern Atlantic. Fisheries Research. 2022;248:106220.
- 32. Irschick DJ, Hammerschlag N. A new metric for measuring condition in large predatory sharks. J Fish Biol. 2014;85(3):917–26. pmid:25130454
- 33. Lawrence MJ, Jain-Schlaepfer S, Zolderdo AJ, Algera DA, Gilmour KM, Gallagher AJ, et al. Are 3 minutes good enough for obtaining baseline physiological samples from teleost fish?. Can J Zool. 2018;96(7):774–86. https://doi.org/10.1139/cjz-2017-0093
- 34. Campbell BA, Shipley ON, Jones TR, Gallagher AJ, Sulikowski JA. Observations of biennial reproduction in Caribbean reef sharks ‘Carcharhinus perezi’. Front Mar Sci. 2024;10.
- 35. Hammerschlag N, Gallagher AJ, Wester J, Luo J, Ault JS. Don’t bite the hand that feeds: assessing ecological impacts of provisioning ecotourism on an apex marine predator. Functional Ecology. 2012;26(3):567–76.
- 36. Tougaard J, Teilmann J, Tougaard S. Harbour seal spatial distribution estimated from Argos satellite telemetry: overcoming positioning errors. Endangered Species Research. 2008;4(1–2):113–22.
- 37. Jonsen ID, Myers RA, Flemming JM. Meta-analysis of Animal Movement using State-space Models. Ecology. 2003;84(11):3055–63.
- 38. Jonsen ID, Flemming JM, Myers RA. Robust State–space Modeling of Animal Movement Data. Ecology. 2005;86(11):2874–80.
- 39. Acuña-Marrero D, Smith ANH, Hammerschlag N, Hearn A, Anderson MJ, Calich H, et al. Residency and movement patterns of an apex predatory shark (Galeocerdo cuvier) at the Galapagos Marine Reserve. PLoS One. 2017;12(8):e0183669. pmid:28829820
- 40. Stevens JD, Bradford RW, West GJ. Satellite tagging of blue sharks (Prionace glauca) and other pelagic sharks off eastern Australia: depth behaviour, temperature experience and movements. Mar Biol. 2009;157(3):575–91.
- 41. Vandeperre F, Aires-da-Silva A, Fontes J, Santos M, Serrão Santos R, Afonso P, 2014. Movements of blue sharks (Prionace glauca) across their life history. PloS one;9(8):e103538.
- 42. Doyle TK, Bennison A, Jessopp M, Haberlin D, Harman LA. A dawn peak in the occurrence of ‘knifing behaviour’ in blue sharks. Anim Biotelemetry. 2015;3(1).
- 43. Worton BJ. Kernel Methods for Estimating the Utilization Distribution in Home‐Range Studies. Ecology. 1989;70(1):164–8.
- 44. Nams VO. The VFractal: a new estimator for fractal dimension of animal movement paths. Landscape Ecol. 1996;11(5):289–97.
- 45.
Turchin P. Quantitative analysis of movement: measuring and modeling population redistribution in animals and plants. Mansfield, CT: Beresta Books. 2015.
- 46. Street GM, Avgar T, Börger L. Net displacement and temporal scaling: Model fitting, interpretation and implementation. Methods Ecol Evol. 2018;9(6):1503–17.
- 47. Jenni-Eiermann S, Jenni L. High Plasma Triglyceride Levels in Small Birds during Migratory Flight: A New Pathway for Fuel Supply during Endurance Locomotion at Very High Mass-Specific Metabolic Rates? Physiological Zoology. 1992;65(1):112–23.
- 48. Schaub M, Jenni L. Variation of fuelling rates among sites, days and individuals in migrating passerine birds. Functional Ecology. 2001;15(5):584–94.
- 49. Jakob EM, Marshall SD, Uetz GW. Estimating fitness: a comparison of body condition indices. Oikos. 1996;:61–7.
- 50. Green AJ. Mass/length residuals: measures of body condition or generators of spurious results? Ecology. 2001;82(5):1473–83.
- 51. Queiroz N, Humphries NE, Noble LR, Santos AM, Sims DW. Spatial dynamics and expanded vertical niche of blue sharks in oceanographic fronts reveal habitat targets for conservation. PLoS One. 2012;7(2):e32374. pmid:22393403
- 52. Miller KM, Li S, Kaukinen KH, Ginther N, Hammill E, Curtis JMR, et al. Genomic signatures predict migration and spawning failure in wild Canadian salmon. Science. 2011;331(6014):214–7. pmid:21233388
- 53. Bonfil R, Meÿer M, Scholl MC, Johnson R, O’Brien S, Oosthuizen H, et al. Transoceanic migration, spatial dynamics, and population linkages of white sharks. Science. 2005;310(5745):100–3.
- 54. Calich H, Estevanez M, Hammerschlag N. Overlap between highly suitable habitats and longline gear management areas reveals vulnerable and protected regions for highly migratory sharks. Marine Ecology Progress Series. 2018;602:183–95.
- 55. Hein AM, Hou C, Gillooly JF. Energetic and biomechanical constraints on animal migration distance. Ecol Lett. 2012;15(2):104–10. pmid:22093885
- 56. Johnson CJ, Parker KL, Heard DC, Gillingham MP. Movement parameters of ungulates and scale‐specific responses to the environment. Journal of Animal Ecology. 2002;71(2):225–35.
- 57. Jorgensen SJ, Arnoldi NS, Estess EE, Chapple TK, Rückert M, Anderson SD, et al. Eating or meeting? Cluster analysis reveals intricacies of white shark (Carcharodon carcharias) migration and offshore behavior. PLoS One. 2012;7(10):e47819. pmid:23144707
- 58. Papastamatiou YP, Meyer CG, Carvalho F, Dale JJ, Hutchinson MR, Holland KN. Telemetry and random-walk models reveal complex patterns of partial migration in a large marine predator. Ecology. 2013;94(11):2595–606. pmid:24400511
- 59. de Sousa Rangel B, Hammerschlag N, Sulikowski JA, Moreira RG. Dietary and reproductive biomarkers in a generalist apex predator reveal differences in nutritional ecology across life stages. Marine Ecology Progress Series. 2021;664:149–63.
- 60. Riekkola L, Andrews‐Goff V, Friedlaender A, Zerbini AN, Constantine R. Longer migration not necessarily the costliest strategy for migrating humpback whales. Aquatic Conservation. 2020;30(5):937–48.
- 61. Sequeira AM, Rodríguez JP, Eguíluz VM, Harcourt R, Hindell M, Sims DW, et al. Convergence of marine megafauna movement patterns in coastal and open oceans. Proceedings of the National Academy of Sciences. 2018;115(12):3072–7.
- 62. Watson RR, Dickson KA. Enzyme activities support the use of liver lipid-derived ketone bodies as aerobic fuels in muscle tissues of active sharks. Physiol Biochem Zool. 2001;74(2):273–82. pmid:11247746