Animal movement across different spatio-temporal scales enables individuals to fulfill life history requirements by responding to seasonal changes in ambient environmental conditions and resource availability, reproduce and avoid predation (Shaw, 2016). Mobile species often use distinct habitats for different life history requirements, highlighting the importance of disparate, essential habitats (Webster et al., 2002; Bauer and Hoye, 2014; Barnett et al., 2019). Linkages between these habitats have implications for connectivity, population dynamics, nutrient and energy transfer, and predator-driven effects on prey (Bauer and Hoye, 2014).

The drivers of movement shape ecological processes from individuals, to populations, and communities (Chapman et al., 2012; Shaw, 2016; Teitelbaum and Mueller, 2019). For example, movements driven by foraging behaviour and resource availability can reverberate through all trophic levels (Coughenour, 1991; Lima, 2002). Seasonal grazing by herbivores such as marine turtles, dugongs and ungulates can impact the cycling of nutrients, plant biomass and community structure as well as primary production (Zieman et al., 1984; Holdo et al., 2009; Lal et al., 2010; Bauer and Hoye, 2014). In turn, mobile predators can seasonally affect herbivore and mesopredator communities through direct predation pressure and indirect risk-effects (Heithaus et al., 2012; Barnett et al., 2017). For instance, wolves (Canis lupis) and tiger sharks (Galeocerdo cuvier) elicit similar anti-predator behaviours in their respective prey, elk (Cervus canadensis) and dugong (Dugong dugon), with wider ecosystem effects on primary producers (Wirsing and Ripple, 2011). Thus, information on how the trophic ecology of predators shapes their movement patterns and space-use is fundamental to understanding ecological processes across organisational levels.

Despite this importance, studies combining spatial and trophic ecology to determine how prey preference and/or resource availability shape predator space use and home ranges are lacking in marine predators (Barnett and Semmens, 2012; Heithaus et al., 2012). This is in part due to the inherent difficulty in observing predator-prey relationships in aquatic systems meaning they are often inferred solely from spatial overlap, without establishing direct trophic links (Suraci et al., 2022). Furthermore, many species such as elasmobranchs (sharks and rays), seabirds and cetaceans commonly exhibit behavioural variability in movement and resource use within species, as well as among closely related, sympatric species (Barnett et al., 2011; Geijer et al., 2016; Watts et al., 2018; Renshaw et al., 2023). However, this variability is frequently overlooked to infer general population or functional group patterns although growing evidence suggests that individual movement behaviour can indeed be context-dependent based on a variety of biotic and abiotic factors, questioning the usefulness of generalised patterns (Shaw, 2020; Catford et al., 2022; Lubitz et al., 2022).

Recently, a context-focused approach has been outlined to address this complexity and help close knowledge gaps regarding the drivers of animal movement (Lubitz et al., 2022). The framework suggests the application of multi-method approaches to investigate drivers of individual movement patterns while incorporating intra- or interspecific variability. This is achieved by comparing study systems that differ in their movement behaviour (e.g., males and females of the same species, conspecifics across geographical locations or individuals from sympatric species) by linking those differences to variation in the biotic/abiotic factors each study system is subjected to (Lubitz et al., 2022).

Here we apply a context-focused, multi-method approach to the great hammerhead shark (Sphyrna mokarran), a Critically Endangered, mobile marine predator, that primarily occupies coastal areas in tropical and subtropical regions (Gallagher and Klimley, 2018; Rigby et al., 2019). The overall ecology of this species continues to have critical knowledge gaps. For example, spatial ecology studies are limited to the southern United States and the Bahamas which suggest more localised movements compared to other large-bodied elasmobranchs, but with notable large-scale excursions (Graham et al., 2016; Guttridge et al., 2017). Trophic knowledge to date indicates that S. mokarran are benthic feeders, that can hunt in shallow waters, primarily targeting other elasmobranchs, in particular batoids (Cliff, 1995; Roemer et al., 2016; Raoult et al., 2019).

To link trophic and movement ecology (aspects of the ecological context, following Lubitz et al., 2022) in S. mokarran we explored spatio-temporal patterns of movement across northern Australia using acoustic and satellite telemetry. Furthermore, we tested whether space-use in S. mokarran is linked to potentially high, year-round abundance of benthic, batoid prey using drone surveys/video observations. Additionally, we used stable isotope analysis to explore the trophic link between S. mokarran and batoids. Lastly, we interpret the results on S. mokarran in the context of the movement and trophic ecology of the sympatric bull shark (Carcharhinus leucas), following the comparative approach outlined in Lubitz et al. (2022). This species is relatively well studied across the globe and can exhibit significant behavioural variability as residents and large-scale migrants co-occur across the distribution of this generalist predator (Heupel et al., 2015; Espinoza et al., 2016; Lubitz et al., 2023). Movement behaviour of C. leucas is in part driven by complex interactions of seasonal environmental change across latitudes, natal philopatry to pupping grounds and availability of their main prey, pelagic teleosts (Cliff and Dudley, 1991; Heupel et al., 2015; Niella et al., 2022; Lubitz et al., 2023).

The influence of dietary breadth and the spatio-temporal dynamics of preferred prey on predator movements is a fundamental component to understanding the functioning of marine and terrestrial systems (Madsen and Shine, 1996; Williams et al., 2004; Hayward and Kerley, 2008; Wirsing and Ripple, 2011). Yet, due to difficulties in observing predator-prey interactions in aquatic environments, how trophic ecology shapes space-use in marine predators is often inferred from spatial overlap of predator and suspected prey (e.g. Fitzpatrick et al., 2012; Hammerschlag et al., 2012; Griffin et al., 2022; Suraci et al., 2022). By utilising a context-focused, multi-method approach, combining telemetry, drone and video surveys and stable isotope analysis, we show a link between great hammerhead shark (Sphyrna mokarran) movements and a possible specialisation on year-round abundant batoid prey. This contrasts with the high intra-specific variability in movement and trophic ecology of bull sharks (Carcharhinus leucas).

Sphyrna mokarran exhibited comparatively small core use areas which consisted mainly of shallow inshore areas, and limited intra-specific variability. The use of shallow inshore shelf habitats by S. mokarran has been reported from our study region and other localities (Harry et al., 2011; Gallagher and Klimley, 2018). This limited dispersal is unusual, as large-bodied marine predators generally move over long-distances, often tracking seasonally changing resource availabilities (Block et al., 2011; Abrahms et al., 2019). Meanwhile, S. mokarran in other areas have shown large-scale movements that contrast with the patterns seen here (Hammerschlag et al., 2011; Guttridge et al., 2017). This discrepancy could be explained by different trophic dynamics and prey availability, as northern Australia is characterised by a large, productive coastal shelf with abundant estuaries and shallow habitats supporting batoid abundance and diversity, compared to the potentially less productive North-West Atlantic coastal habitats (Nagelkerken et al., 2015; Sheaves et al., 2015). Nonetheless, as demonstrated by the individual tagged at North-West Island, some S. mokarran may also occasionally move over larger spatial scales in northern Australia, although such movements away from core use areas appear short in duration (e.g., 6 weeks) in comparison to seasonal migrations in C. leucas. Terrestrial predators such as lions and wolves adjust home range size with prey availability, while theoretical models demonstrate a strong link between the distribution of resources and predator space-use (Walton et al., 2001; Mitchell and Powell, 2004; Tumenta et al., 2013). Although our results arise from just one batoid nursery, studies from other parts of northern Australia indicate similar patterns of high, year-round abundance of batoids in shallow nurseries (Cerutti-Pereyra et al., 2014; Martins et al., 2020). Thus, year-round availability of batoid prey across northern Australia may allow for smaller home ranges in S. mokarran.

Many species of batoids use inshore reef and sand flat habitats as nurseries (Cerutti-Pereyra et al., 2014; Crook, 2020; Martins et al., 2020) and indeed, our drone surveys confirm high, year-round occurrence of batoids in these habitats. Across inshore nurseries in northern Australia, batoid abundance is high and stable throughout the year, providing a consistent, year-round, food source to S. mokarran (Cerutti-Pereyra et al., 2014; Crook, 2020; Martins et al., 2020). In the present study, stable isotope results suggests that S. mokarran utilise the same relatively small areas through time (in agreement with tracking), consistently feeding on benthic organisms that are part of a localised food web. Indeed, S. mokarran δ¹³C differed between individuals sampled from coastal sand flat- vs. reef-dominated areas, and corrected S. mokarran δ¹³C fell close to rays sampled from equivalent habitats, with both prey and predator δ¹³C values agreeing well with the predominant benthic sources of nutrition at each habitat (microphytobenthos in coastal sand flat habitats and coral in coral reef habitats). The presence of these two groups led to the larger variability in overall S. mokarran δ¹³C when compared to C. leucas, despite the migratory behaviour of some C. leucas. In the seagrass-dominated Shark Bay, WA, S. mokarran also had stable isotope values that indicate feeding on rays (Heithaus et al., 2013). Although blood and plasma samples were only available for two mature S. mokarran, these two individuals had low inter-tissue differences in δ¹³C, suggesting a consistent diet over time. This was particularly the case for the largest individual sampled, which had similar δ¹³C for all three tissues. Combined, these results suggest that, across northern Australia, S. mokarran specialise on batoid prey, which appears to be in high abundance year-round, allowing for the majority of S. mokarran to move over smaller spatial scales.

The two immature S. mokarran had muscle δ¹³C values closer to C. leucas, suggestive of feeding on teleosts, which have been recorded as prey for smaller S. mokarran (Stevens and Lyle, 1989; Cliff, 1995; Hsu et al., 2022). The larger of these juveniles (a 225 cm TL female) also had largest differences in δ¹³C between tissue types, particularly between muscle (a tissue a turnover rate of months to years) and blood/plasma (tissues with much faster turnover rates), suggesting a recent ontogenetic shift in diet from a teleost-based juvenile diet to a benthic adult diet. Note also that the δ¹³C values of muscle were the lowest and similar to those of C. leucas, whereas plasma had higher δ¹³C, close to muscle samples from adults from the same (sand flat) habitat (and also close to plasma of the two mature S. mokarran sampled). S. mokarran females mature at 210–220 cm TL (Last and Stevens, 2009), so this individual was likely transitioning to maturity when it was sampled, and this study suggest that this transition is accompanied by a shift in diet. In contrast to S. mokarran, both migratory and resident C. leucas derived carbon mostly from plankton-based, pelagic, food webs, and had limited isotopic niche overlap with S. mokarran. Pelagic food webs are often dynamic and dominated by pulse-prey events (Rooney et al., 2006; MacKenzie et al., 2019), which may drive seasonal large-scale migrations in some C. leucas (Espinoza et al., 2016; Lubitz et al., 2023). Similar to this study, some bull sharks remain resident in subtropical areas in southern Africa, even after seasonally aggregating teleost prey disperse, despite other sharks migrating away (Lubitz et al., 2023). As previously suggested for southern Africa, this could be due to the dynamic nature of pelagic food webs, where both residents and migrants are seasonally attracted to a high abundance of localised resources, but migrants leave after resources decrease, while some residents sustain themselves on remaining resources or switch to other prey (Lubitz et al., 2023). However, seasonal temperature change and reproductive philopatry, where females will return to estuaries to pup, also play a role in driving bull shark movements, indicating a potentially complex interaction between environmental factors, prey, and reproduction in driving variable movement strategies in C. leucas (Tillett et al., 2012; Heupel et al., 2015).

In this study, both migrant and resident C. leucas had relatively high inter-tissue variability in δ¹³C, suggesting a broader trophic niche. Migrants had higher differences compared to residents, in line with large-scale foraging excursions across distant habitats, such as tropical coral reefs and temperate estuaries. In contrast to this high intra-specific variability in C. leucas, high year-round resource availability in benthic food webs may cause increased residency in predators specialising in resident, benthic prey, such as in S. mokarran. These differences in trophic and spatial ecology of S. mokarran and C. leucas are similar to sympatric predators in other systems. Differences in stable isotope composition was driven by migration distance and foraging habitat in pinniped species (Burton and Koch, 1999), while dietary/niche specialists have smaller home ranges than sympatric dietary/niche generalists in some terrestrial carnivores and birds (Walker et al., 2007; Gómez et al., 2016; Huaranca et al., 2019; Lagos et al., 2021).

Despite their relatively small-scale movements within the tropical study sites, where water temperatures likely remain favourable year-round, S. mokarran repeatedly moved between inshore habitats and mid-shelf coral reefs such as Batt Reef and reefs off the Palm Islands/Whitsunday Islands/Magnetic Island. Such habitat connectivity may constitute important trophic linkages, as both inshore habitats and reef flats are rich in batoids (Crook, 2020; Martins et al., 2020; Barnett et al., 2022a). This may also explain the almost tenfold difference between the 50% UD and 95% in S. mokarran where small core use areas (hunting grounds) seem connected via movement. Hunting in water, often as shallow as half a body height, has been observed numerous times in hammerhead sharks in the north-west Atlantic (Roemer et al., 2016). This is also evident in our collected videos of successful and unsuccessful predation events on batoids in shallow inter-tidal sand/reef habitats (Supplemental Material). The reduced three-dimensionality in these habitats combined with morphological and physiological adaptions likely offer a distinct advantage to S. mokarran. For instance, increased electro-reception in the cephalofoil and higher maneuverability compared to other carcharhiniforms appears ideal for locating buried batoids and pursuing them rapidly in shallow water (Kajiura, 2001; Kajiura et al., 2003). The use of the cephalofoil to pin batoids to the substrate, pivoting sideways to avoid the barb, and immobilising rays by tearing a wing off are also specialised behaviours for hunting batoids, providing further evidence for a trophic specialisation in S. mokarran (Strong et al., 1990; Chapman and Gruber, 2002). This active, shallow water hunting mechanism contrasts with the ambush tactics of the less maneuverable tiger shark (Galeocerdo cuvier) when hunting air-breathing prey in shallow two-dimensional seagrass habitats (Heithaus et al., 2002).

Direct predation pressure and non-consumptive (risk) effects induced by S. mokarran could have consequences for batoids. Indeed, juvenile batoids select micro-habitats on sand and reef flats with often suboptimal temperatures in favour of predator avoidance (Vaudo and Heithaus, 2013; Crook, 2020; Martins et al., 2020). Drone surveys observed P. ater and G. typus sheltering against the shoreline at high tide, with the smallest individuals situated closest to shore (Supplemental Material). Stingrays are often described as keystone species, influencing invertebrate communities, and physical processes through bioturbation (O'Shea et al., 2011; Flowers et al., 2021; Crook et al., 2022). Therefore, S. mokarran predation may influence the spatio-temporal distribution of batoids and their feeding and bioturbation rates, shaping ecosystem processes in inshore habitats (Rupp and Bornatowski, 2021).

Drivers of movement and habitat use can also be linked to reproduction (Shaw, 2016). For example, some sea birds and tuna migrate between breeding and feeding habitats, suggesting that both life history requirements need to be fulfilled across distant habitats (Frederiksen et al., 2004; Richardson et al., 2016). As mentioned above, movement patterns in C. leucas appear highly context-dependent, with large-scale migrations and residency, likely driven by reproductive philopatry, seasonal environmental change and resource availability across disparate habitats (Heupel et al., 2015; Niella et al., 2022; Lubitz et al., 2023). In contrast, limited dispersal, and affinity for shallow, inshore habitats of S. mokarran suggests that critical life-history requirements are performed over smaller spatial scales. Indeed, growing evidence from the Atlantic suggests S. mokarran utilise inshore pupping grounds (Barker et al., 2017; Macdonald et al., 2021). Although pups are rarely reported for Australia (Harry et al., 2011), young-of-the-year (YOY) have been caught in Cleveland Bay (Simpfendorfer et al., 2014) and observed in the Whitsunday Islands (Lubitz, personal observations), suggesting habitat overlap with mature individuals.

The smaller and less connected home ranges in S. mokarran imply reduced connectivity among regions. Yet, genetic connectivity occurs across the Australian distribution (Chin et al., 2017). However, only a few migrants per generation are needed to maintain genetic connectivity, while demographic connectivity may remain limited (Mills and Allendorf, 1996). This could make S. mokarran more vulnerable to local extinctions (Hueter et al., 2005). If genetic connectivity equates demographic connectivity needs further investigation (Lowe and Allendorf, 2010). Across taxa, narrow niche breadth is positively correlated to extinction risk (Kotiaho et al., 2005; Slatyer et al., 2013) and coastal, inshore habitats are most affected by pollution, habitat degradation and climate change, while small-scale movements may increase risk of targeted harvest and bycatch (Chin et al., 2010; Hughes et al., 2017; Scanes et al., 2020). Thus, limited demographic connectivity, a reliance on shallow inshore habitats and small home ranges may render S. mokarran particularly vulnerable to localized depletions particularly through targeted fishing, and climate change, raising concerns regarding recovery potential of this Critically Endangered species (Chin et al., 2010; Rigby et al., 2019). Reef and sand flats likely exhibit faster warming, while hammerhead physiology and specialisation could make them less adaptable to climate change (Wosnick et al., 2019). In contrast, the behavioural flexibility of C. leucas, including use of multiple migration strategies, a generalized diet, and high connectivity may facilitate adaptability to anthropogenic stressors and increased resilience (Chin et al., 2010).

Studies investigating connectivity and essential habitats for S. mokarran life history requirements, and the anthropogenic stressors they are subjected to in coastal habitats are required to evaluate the impacts of climate change and assess extinction risks (Chin et al., 2010). Further trophic studies using high resolution dietary markers through fatty acid analysis of both predator and prey may help to further solidify the predator-prey relationship between S. mokarran and batoids. This may facilitate a better understanding of how predation pressure and risk effects impact behaviour and distribution of batoids (Heithaus et al., 2012).

In conclusion, while it would have been difficult using tracking or trophic information alone, the multi-method approach allowed us to establish a link between limited dispersal and smaller home ranges in S. mokarran with a potential feeding specialisation on batoids in benthic food webs. In contrast, C. leucas showed restricted and large-scale movements which could be partially influenced by higher trophic flexibility and reliance on dynamic, pelagic food webs (Tillett et al., 2014; Niella et al., 2022; Lubitz et al., 2023). Therefore, as with their terrestrial counterparts (Tumenta et al., 2013), trophic ecology can play a role in shaping marine predator space-use.

The tracking data used in this study are publicly available and can be accessed from the Australian Animal Acoustic Telemetry Database facilitated by the Integrated Marine Observing System (IMOS) at https://animaltracking.aodn.org.au/. Drone survey data and stingray stable isotope values are available in Crook, 2020 and Martins et al., 2022. Uncorrected shark stable isotope data are currently being used for additional studies in preparation but can be requested from the corresponding author. Video files are available in the Supplementary Material .

Animal capture and tagging was conducted under Great Barrier Reef Marine Park Permit G22/46908.1, general Queensland Fisheries Permit 266351 and ethics permit A2846, approved by the James Cook University ethics committee.

NL: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Writing – original draft. KA: Data curation, Formal Analysis, Writing – review & editing. KC: Conceptualization, Data curation, Writing – review & editing. LC: Funding acquisition, Project administration, Writing – review & editing. AC: Data curation, Funding acquisition, Writing – review & editing. MS: Conceptualization, Supervision, Writing – review & editing. RF: Data curation, Writing – review & editing. AM: Data curation, Writing – review & editing. SB: Data curation, Writing – review & editing. IM: Data curation, Writing –review & editing. AB: Conceptualization, Data curation, Funding acquisition, Supervision, Writing – review & editing.