Introduction

Sharks and rays are predators within the marine environment and many species are thought to play an important role in coastal and oceanic ecosystems worldwide (Bascompte et al. 2005; Ferretti et al. 2008; Heithaus et al. 2008). Several studies have suggested that the removal of large predators from marine systems can lead to drastic and lasting effects in the form of trophic cascades and/or the collapse of marine communities (Dulvy et al. 2000; Shepherd and Myers 2005; Polovina et al. 2009). To date, most studies examining predator removal from marine systems have focused on the effects of overfishing (Stevens et al. 2000; Baum and Worm 2009; Polovina et al. 2009). However, other drivers exist that may affect the distribution and abundance of sharks and rays within marine ecosystems and, by association, the broader marine community. Abiotic factors are easily measured and may form some of the main drivers for behavioural patterns within marine ecosystems. Biotic factors such as prey density and availability (Heithaus 2001; Heithaus et al. 2002; Torres et al. 2006; Goetze and Fullwood 2013), and predator avoidance (Heupel and Hueter 2002; Collins et al. 2007; Heithaus et al. 2009) have also been shown to play a role in the spatial ecology of elasmobranchs. These studies illustrate that a range of drivers influence shark and ray populations and the ecosystems where they occur.

Recent studies have shown that abiotic factors may be significant drivers of movement among sharks and rays. Both acute (e.g. severe weather events; Heupel et al. 2003; Matich and Heithaus 2012) and chronic (e.g. seasonal temperature change; Hopkins and Cech 2003; Heupel 2007; Carlisle and Starr 2009) changes have been linked to movement. For example, Matich and Heithaus (2012) found that juvenile bull sharks, Carcharhinus leucas, either permanently left an estuarine system or died during an extreme “cold snap.” In Tomales Bay, California, the bat ray, Myliobatis californica, leopard shark, Triakis semifasciata, and brown smooth-hound shark, Mustelis henlei, were observed to emigrate from the bay in response to a seasonal decrease in water temperature (Hopkins and Cech 2003). In addition to temperature, abiotic factors such as salinity (Collins et al. 2008; Heupel and Simpfendorfer 2008; Ubeda et al. 2009; Simpfendorfer et al. 2011), dissolved oxygen (Parsons and Hoffmayer 2005; Heithaus et al. 2009; Craig et al. 2010; Espinoza et al. 2011), tide (Medved and Marshall 1983; Ackerman et al. 2000; Knip et al. 2011a; Campbell et al. 2012), photoperiod (Grubbs et al. 2005; Heupel 2007; Kneebone et al. 2012; Nosal et al. 2014), barometric pressure (Heupel et al. 2003; Udyawer et al. 2013) and pH (Ortega et al. 2009) have all been reported to play a role in the location and movement of sharks and rays.

Understanding how species respond to changes in their environment is increasingly important given reports of growing human-mediated environmental effects on marine systems. Sharks and rays currently face an array of anthropogenic threats, especially in coastal ecosystems (Knip et al. 2010; Dulvy et al. 2014). For example, chemical pollutants and pesticides found in river run-off have been shown to have a negative impact on both ecosystem health and productivity (Fabricius 2005; Hutchings et al. 2005; Schaffelke et al. 2005). Most notably, Gelsleichter et al. (2005) documented a possible link between exposure to organochloride contaminants found in Florida estuaries and reproductive health in the bonnethead shark, Sphyrna tiburo. Intense coastal development such as dredging, construction and deforestation has also been linked to wide-scale habitat alteration and destruction (Edgar et al. 2000; Lotze et al. 2006). Perhaps unsurprisingly, recent studies have reported changes in abundance and distribution of several elasmobranch species associated with human-altered habitats (Jennings et al. 2008; Jirik and Lowe 2012; Werry et al. 2012; Curtis et al. 2013). It is important to note, however, that the effects of utilising modified habitats for sharks and rays are not necessarily negative. Research into use of restored estuarine habitat by round stingrays, Urobatis halleri, suggested that female rays may have benefited from the significantly higher mean water temperatures found in restored estuarine basins than in adjacent natural areas (Jirik and Lowe 2012). Collectively, these studies highlight how human-mediated environmental changes have the potential to alter local environmental conditions, affecting the health and spatial ecology of sharks and rays, both positively and negatively.

Compounding anthropogenic changes in coastal systems is the fact that key inshore habitats for many species are often characterised by fluctuating environmental conditions such as high freshwater run-off/low salinity events (Mann 2000; Devlin and Schaffelke 2009), tropical storms (Conner et al. 1989) and large and/or rapid changes in temperature (Mann 2000). With the global population expected to reach an estimated 9 billion by 2050 (Cohen 2003, 2005), and with future climate change scenarios predicting substantial changes in a range of environmental parameters (Emanuel 2005; Webster et al. 2005; Hoegh-Guldberg et al. 2007; Rahmstorf 2007), most of these environmental changes are expected to be exacerbated.

In light of current and projected environmental changes on marine systems, and given the important role predators such as sharks and rays play in the trophic dynamics of coastal and oceanic ecosystems globally, it is vital to understand the drivers behind movement. As movement plays a critical role in defining the use and importance of key habitats for sharks and rays, resolving whether abiotic factors drive movement will provide essential information on spatial ecology and provide data valuable to the successful management of these species and habitats. This review paper outlines and discusses: (1) abiotic factors linked to movement of sharks and rays; (2) complexities associated with determining environmental drivers of movement; and (3) the role of environmental change on spatial ecology.

Abiotic factors linked to movement of sharks and rays

Abiotic factors may influence movement either indirectly (e.g. altering patterns of abundance and distribution of principal prey species) or directly. Given that the energetic demands of key metabolic and physiological processes (e.g. digestion, osmoregulation) for sharks and rays are known to fluctuate in response to changes in abiotic factors such as temperature and salinity (Bernal et al. 2012), the primary way in which abiotic factors are thought to influence elasmobranch movement is directly, via their effect on physiology. Specifically, movement may occur in response to a physiological limit being reached or to a physiological preference, where individuals move to maintain abiotic factors at levels that provide them with some form of advantage.

As mobile marine predators, most sharks and rays have the ability to leave an area should local environmental conditions change sufficiently that the energetic costs of maintaining key metabolic and physiological processes become too demanding. Individuals that leave, however, risk the inability to find new suitable habitat and an increased chance of predation, while those that stay must exist in a stressed state or, at worst, suffer death. Whether or not individuals that leave return to their original habitat or relocate permanently often depends on the length and severity of environmental fluctuations. The following section outlines abiotic factors currently linked to movement of sharks and rays.

Temperature

Temperature is well-known to have an effect on the physiology of ectotherms (Fry 1971; Crawshaw 1977; Bernal et al. 2012). The rate of important metabolic and physiological functions such as digestion, somatic growth, and reproduction is determined by the core body temperature of fish which, in turn, is directly controlled by the temperature in their surrounding environment. This is in contrast to endothermic species that are able to regulate their internal body temperature and help maintain metabolic rates using specialized circulatory mechanisms that retain heat (Anderson and Goldman 2001; Dickson and Graham 2004; Donley et al. 2007). Most sharks and rays are ectothermic with the only exceptions found within the family Lamnidae, comprising just five species (Bernal et al. 2012). Given the role that ambient temperature plays in influencing key metabolic and physiological processes for ectotherms, it is perhaps unsurprising that many sharks and rays are sensitive to changes in temperature.

Temperature-related mortality events (i.e. “cold kills”) have been documented for some sharks and rays in response to extreme drops in temperature (Snelson and Bradley 1978; Poulakis et al. 2011; Matich and Heithaus 2012). In contrast, laboratory-held Atlantic stingrays, Dasyatis sabina, acclimated to temperatures at or lower than those experienced in their natural habitat (Fangue and Bennett 2003). The authors suggested that by being able to adapt to temperatures lower than those present in their natural habitat rays were able to exploit a wider range of resources than competitors and/or use thermal gradients as a refuge from predators. While sharks and rays that choose to remain within their home range area when temperatures fluctuate must either adapt to the new conditions or suffer death, many species respond through movement.

Temperature-mediated seasonal (Dunbrack and Zielinski 2003; Hopkins and Cech 2003; Heupel 2007; Vaudo and Heithaus 2009) and diel (Carey and Scharold 1990; Economakis and Lobel 1998; Matern et al. 2000; Sims et al. 2006) movements are well-documented in the literature. For example, juvenile blacktip sharks, Carcharhinus limbatus, in nearshore waters of the Gulf of Mexico used decreases in water temperature as a cue to leave their summer nursery area—a seasonal exodus to avoid lethal winter temperatures (Heupel 2007). In addition to movement in response to seasonal temperature change, many species are thought to use movement to actively seek out preferred temperatures throughout the day. Common thresher sharks, Alopias vulpinus, displayed diel patterns in depth distribution and activity indicating a nocturnal preference for the warmer mixed layer after daytime activity in deeper, cooler water (Cartamil et al. 2010). Collectively, studies documenting temperature-mediated seasonal and diel movements indicate that temperature often works on shark and ray populations at a variety of temporal scales.

Movement to locate a spatially variable preferred temperature range (i.e. behavioural thermoregulation) may be important to foraging (Carey and Scharold 1990; Matern et al. 2000; Sims et al. 2006; Thums et al. 2013) and reproductive (Economakis and Lobel 1998; Hoisington and Lowe 2005; Hight and Lowe 2007; Jirik and Lowe 2012; Speed et al. 2012) strategies among sharks and rays. Thermoregulatory behaviour may confer biological advantages to the individual that offset movement costs. Benefits often include energy savings that could be allocated to growth or other biological functions (e.g. reproduction).

The use of behavioural thermoregulation as a foraging strategy may be the reason why deep dives by blue sharks, Prionace glauca, were punctuated by frequent short periods at the surface (Carey and Scharold 1990). Since muscle warms more quickly than it cools, blue sharks were thought to re-warm quickly at the surface between dives, allowing them to extend their foraging time below the thermocline. Similarly, a recent study investigating movement of whale sharks, Rhincodon typus, off the coast of Western Australia found post-dive surface duration to be negatively correlated with the minimum temperature of dives, thought to be a thermoregulatory strategy where sharks re-warm at the surface after foraging in cooler, deep waters (Thums et al. 2013). Movement into cooler water after feeding has been suggested to decrease the rate of gastric evacuation and increase assimilation efficiency for several species (Matern et al. 2000; Sims et al. 2006; Di Santo and Bennett 2011). A 10 °C decrease in water temperature, for example, resulted in a 30 % increase in overall food absorption in Atlantic stingrays (Di Santo and Bennett 2011). By adopting a “hunt warm, rest cool” strategy male dogfish, Scyliorhinus canicula, were thought to lower their daily energetic costs by just over 4 % (Sims et al. 2006). Increases in assimilation efficiency and extended foraging times are examples of how sharks and rays use temperature-mediated movement to conserve energy and improve individual foraging success.

In contrast, movement of female sharks and rays into warm, shallow water is thought to provide reproductive benefits (Economakis and Lobel 1998; Hoisington and Lowe 2005; Hight and Lowe 2007; Jirik and Lowe 2012; Nosal et al. 2013, 2014). Female leopard sharks off the coast of California actively sought out the warmest part of an embayment as temperatures fluctuated throughout the day (Hight and Lowe 2007). The authors estimated that elevated core body temperatures could increase metabolic rates by up to 17 % and augment physiological functions, possibly increasing the rate of embryonic development and decreasing gestation period. By shortening the gestation period, female sharks would have more time to replenish energy reserves following parturition before the onset of winter and the next reproductive cycle (Wallman and Bennett 2006; Jirik and Lowe 2012). Jirik and Lowe (2012) suggested that preferential use of warm, restored estuarine habitat by female round stingrays during gestation may increase the size of offspring at birth, presumably enhancing the survival rates of newborn rays. Alternatively, as females of some shark species are thought to reach sexual maturity at a size greater than their male conspecifics, it has been suggested that aggregations of female sharks in warm waters may increase somatic growth rates, allowing them to reach reproductive maturity more quickly (Economakis and Lobel 1998; Robbins 2007). Whether due to a physiological limitation or as a thermoregulatory strategy in response to physiological preferences, it is clear that temperature is an important driver of movement and space use.

Salinity

Salinity is also well-known to have a strong influence on physiology (Pang et al. 1977; Bernal et al. 2012). Although some species are euryhaline and able to exploit a wide range of salinities (Thorson 1972, 1974; Montoya and Thorson 1982; Hazon et al. 2003), the vast majority of sharks and rays are strictly stenohaline and occupy a narrow salinity range (Froeschke et al. 2010; Martin et al. 2012). Regardless of individual salinity tolerances, however, optimal habitat requirements for sharks and rays must be balanced against the energetic costs of osmoregulation, resulting in species-specific responses to changes in salinity.

Salinity has been reported to influence both distribution (Simpfendorfer et al. 2005; Collins et al. 2008; Ubeda et al. 2009; Knip et al. 2011b; Francis 2013) and local abundance (Hopkins and Cech 2003; Carlisle and Starr 2009; Poulakis et al. 2011) of sharks and rays. For example, juvenile pigeye sharks, Carcharhinus amboinensis, in a tropical nearshore environment shifted their home range areas during extreme wet season flooding events to avoid freshwater inflow and decreased salinity (Knip et al. 2011b). Similarly, cownose rays, Rhinoptera bonasus, monitored in a southwest Florida estuary occurred farther upriver during periods of decreasing flow and increasing salinity (Collins et al. 2008). Given that most sharks and rays are stenohaline, movement in response to salinity change may simply be a means to avoid physiological stress and possible mortality when salinity levels fall outside of individual tolerances. It is important to note that, as a driver of movement, salinity most likely has a greater influence on nearshore species than species that occur further from shore as sharks and rays utilising inshore habitats are more frequently exposed to freshwater runoff and associated salinity fluctuations.

Similar to behavioural thermoregulation, some species have been shown to use movement to actively seek out spatially variable preferred salinity ranges, a type of behavioural osmoregulation thought to confer a biological advantage (Collins et al. 2008; Heupel and Simpfendorfer 2008; Froeschke et al. 2010; Simpfendorfer et al. 2011). Juvenile bull sharks within a Florida estuary selected habitat based on salinity, avoiding areas of low salinity and moving to remain within a preferred salinity range (Simpfendorfer et al. 2005; Heupel and Simpfendorfer 2008). Selection for a preferred salinity range may minimize the energetic costs associated with osmoregulation, energy that presumably could be allocated to growth or other physiological processes (Simpfendorfer et al. 2005; Heupel and Simpfendorfer 2008; Froeschke et al. 2010).

In addition to reducing the energetic costs of osmoregulation, movement to remain within a specific salinity range may also be a means of predator avoidance for some species (Poulakis et al. 2011; Simpfendorfer et al. 2011). Simpfendorfer et al. (2011) found that young smalltooth sawfish, Pristis pectinata, demonstrated an affinity for different salinities than those of most marine predators. Within this estuarine system large bull sharks used salinities ranging from 7 to 20 psu (dimensionless “units” of the Practical Salinity Scale) while juvenile sawfish occupied salinities from 18 to 24 psu. By moving to remain within a different salinity range than that used by bull sharks, young sawfish presumably lowered their risk of predation and increased survival. Whether as a means to reduce the energetic costs associated with osmoregulation, or to avoid predation, it is apparent that salinity plays a role in moderating distribution and movement patterns of some sharks and rays.

Dissolved oxygen

Although comparatively fewer studies have focused on the role of dissolved oxygen in spatial ecology, this abiotic factor has been shown to influence both distribution (Grubbs and Musick 2007; Carlisle and Starr 2009; Espinoza et al. 2011; Knip et al. 2011b; Drymon et al. 2013) and abundance (Parsons and Hoffmayer 2005; Heithaus et al. 2009) of several shark and ray species. Heithaus et al. (2009) found dissolved oxygen concentrations to be the best predictor of bull shark abundance within a Florida estuary. The authors suggested that, over small spatial and temporal scales movement was driven largely by the effort of individuals to remain within optimal dissolved oxygen conditions. Similarly, habitat use and movement of the gray smooth-hound shark, Mustelus californicus, in a newly restored estuary in southern California, indicated individuals avoided the warmest inner basin during the day due to spatial differences in dissolved oxygen levels (Espinoza et al. 2011). These studies indicate that fluctuations in dissolved oxygen levels may be a significant driver of movement for some species. It is important to note, however, that as a driver of movement, dissolved oxygen most likely has a greater influence on those species utilising comparatively lower oxygen habitats.

In contrast to studies documenting movement in response to changes in dissolved oxygen concentrations, several studies have shown that some species display a degree of tolerance to fluctuations in dissolved oxygen levels. Epaulette sharks, Hemiscyllium ocellatum, provide perhaps the best example of hypoxia tolerance in elasmobranchs (Wise et al. 1998; Routley et al. 2002; Nilsson and Renshaw 2004; Speers-Roesch et al. 2012). A study by Wise et al. (1998) found epaulette sharks to be tolerant to both mild hypoxia (20 % of normoxia) and to cyclic exposure to extreme hypoxia (5 % of normoxia). As a resident on shallow reef platforms characterised by large tidal fluctuations, epaulette sharks are exposed to progressively longer and more severe hypoxic conditions as tides become lower, a natural preconditioning regimen that elicits an enhanced physiological response to hypoxia, allowing it to better exploit its niche environment (Routley et al. 2002; Nilsson and Renshaw 2004).

Laboratory studies on the bonnethead shark (Parsons and Carlson 1998; Carlson and Parsons 2001, 2003) and blacknose shark, Carcharhinus acronotus (Carlson and Parsons 2001), determined that these species were able to tolerate at least moderate hypoxic conditions. The authors speculated that behavioural (i.e. increase in swimming activity, mouth gape) and associated physiological (i.e. increase in oxygen uptake) mechanisms employed by these two species may allow them to exploit warm shallow environments (where dissolved oxygen levels can be variable) more effectively than other species (Parsons and Carlson 1998; Carlson and Parsons 2001). Similarly, the observed ability of the scalloped hammerhead shark, Sphyrna lewini (Jorgensen et al. 2009), and cownose ray (Craig et al. 2010) to penetrate hypoxic environments may mean these species have access to prey not accessible to other predators. Therefore, dissolved oxygen levels may play an important role in the spatial ecology of some sharks and rays, with individuals driven by a need to remain within optimal dissolved oxygen conditions or as a means to exploit habitats and/or resources inaccessible to others and hence reduce competition.

Tide

Tidally-influenced movement has been observed for both sharks (Medved and Marshall 1983; Ackerman et al. 2000; Wetherbee and Rechisky 2000; Carlisle and Starr 2010) and rays (Smith and Merriner 1985; Whitty et al. 2009; Campbell et al. 2012). In separate studies examining movement patterns of juvenile sandbar sharks, Carcharhinus plumbeus, off the east coast of the United States, young sharks were observed to move predominantly in the direction of tidal current flow (Medved and Marshall 1983; Wetherbee and Rechisky 2000). Tidally-driven movement patterns are thought to be related to foraging tactics (Smith and Merriner 1985; Ackerman et al. 2000; Campos et al. 2009; Carlisle and Starr 2010), energy conservation strategies (Ackerman et al. 2000; Ortega et al. 2009; Whitty et al. 2009; Campbell et al. 2012), and predator avoidance (Wetherbee et al. 2007; Knip et al. 2011a; Guttridge et al. 2012). For example, species typically found in nearshore waters such as the leopard shark (Ackerman et al. 2000; Carlisle and Starr 2009, 2010), brown smooth-hound shark (Campos et al. 2009) and cownose ray (Smith and Merriner 1985) move into shallow water on incoming tides, presumably to forage. Movement with the tide allows individuals to maximise their foraging area by utilising regions only available at high tide (Smith and Merriner 1985; Gilliam and Sullivan 1993; Ackerman et al. 2000; Carlisle and Starr 2010) as well as providing access to comparatively rich prey resources found only in intertidal areas (Carlisle and Starr 2009).

Movement with the tide may also serve to minimize energy expenditure for some sharks (Ackerman et al. 2000; Ortega et al. 2009; Conrath and Musick 2010) and rays (Whitty et al. 2009; Campbell et al. 2012). Conrath and Musick (2010) found that 53.6 % of the net movements of juvenile sandbar sharks were with the tide. Due to strong tidal currents within the area the authors speculated that it was energetically costly for young sharks to swim against tidal currents. Energy conservation was also suggested to explain why 98 % of young-of-the-year (0+) freshwater sawfish, Pristis microdon, moved in the direction of the tide (Whitty et al. 2009). In addition, Ackerman et al. (2000) estimated that tidally-assisted swimming could potentially conserve up to 6 % of total energy expenditure for leopard sharks, suggesting that energy savings acquired by young sharks and rays through tidally-directed movement may be considerable.

Movement with the tide may also be a means for juvenile sharks and rays to avoid larger predators (Wetherbee et al. 2007; Guttridge et al. 2012). Movement patterns of juvenile lemon sharks, Negaprion brevirostris, on a Brazilian atoll revealed the smallest individuals were most strongly influenced by tides, restricting their movements to the shallowest tide pools at low tide (Wetherbee et al. 2007). Similarly, Knip et al. (2011a) concluded that changes in water depth associated with the tide had the strongest influence on the youngest individuals within a population of pigeye sharks. It is important to note that depth is associated with tide and both authors speculated that, as small individuals are likely to be the most vulnerable to predation, tidally-based movement may be a mechanism to avoid predators by remaining in shallow water. These studies indicate that tide may be an important environmental driver of movement among sharks and rays, as a refuging strategy, a foraging tactic, or as a means to conserve energy.

Photoperiod and light

Few studies have identified photoperiod as a potential driver of movement for sharks and rays (Grubbs et al. 2005; Heupel 2007; Kneebone et al. 2012; Nosal et al. 2014). However, long-term movements of juvenile sandbar sharks indicated that winter migration from a summer nursery area was highly correlated with decreasing day length (Grubbs et al. 2005). As an abiotic driver of movement, photoperiod is perhaps more important for species found in temperate regions that are subject to greater differences in seasonal day length. It is also important to note that photoperiod on a seasonal scale is probably an indicator for correlated temperature changes. For example, Grubbs et al. (2005) speculated that the return of juvenile sandbar sharks to nursery areas in the spring was triggered by both photoperiod and temperature with day length a signal to begin northward migrations and temperature the primary driver stimulating movement into the nursery. Similarly, emigration of juvenile sand tiger sharks, Carcharias taurus, from a Massachusetts estuary was correlated to both day length and water temperature (Kneebone et al. 2012). Although both factors may be influential, the authors speculated that photoperiod was a stronger, more consistent cue driving movement.

In addition to day length, light intensity has been suggested to drive movement for some species (Nelson et al. 1997; Cartamil et al. 2003; Andrews et al. 2009; Whitty et al. 2009). Andrews et al. (2009) suggested that changes in light intensity during crepuscular periods may initiate nocturnal foraging behaviour in sixgill sharks, Hexanchus griseus. Similarly, diel patterns of movement observed in the Hawaiian stingray, Dasyatis lata, were thought to be more influenced by light intensity than temperature or tidal stage (Cartamil et al. 2003). Finally, Nelson et al. (1997) revealed crepuscular vertical migrations of the megamouth shark, Megachasma pelagios, closely tracked specific isolumes. The authors concluded that depth selection was related to light level. As potential drivers of movement, both photoperiod and light intensity have been shown to play a key role in the spatial ecology of several elasmobranch species, but remain relatively understudied to date.

Other factors

Two other abiotic factors have been identified as drivers of movement among sharks and rays, but neither has been examined in detail. The first of these is barometric pressure. Changes in barometric pressure associated with tropical storms can trigger movement in sharks (Heupel et al. 2003; Udyawer et al. 2013). Heupel et al. (2003) found that juvenile blacktip sharks left a nursery area in response to a decrease in barometric pressure associated with a tropical storm. Findings by Udyawer et al. (2013), however, indicate movement in response to extreme storm events may be species-specific. Five species of coastal shark (Carcharhinus limbatus, C. tilstoni, C. melanopterus, C. sorrah, and C. amboinensis) responded differently to changes in barometric pressure before, during and after a severe tropical storm. A short-term flight response was observed in all monitored species except the blacktip reef shark (C. melanopterus) which did not depart or change location. The authors speculated that blacktip reef sharks are perhaps more tolerant to the adverse conditions associated with tropical storms or that the benefits gained by individuals who remain in their preferred habitat outweigh the benefits of departing.

The second rarely examined environmental driver of movement for elasmobranch species is pH (Ortega et al. 2009). Juvenile bull sharks within a Florida river were tracked before and after an influx of freshwater revealing two distinct movement patterns. Movement and location of individuals were either primarily nocturnal and correlated with salinity, temperature and dissolved oxygen, or diurnal in nature and correlated with temperature, dissolved oxygen, turbidity and pH (Ortega et al. 2009). Although under-represented in the literature, changes in barometric pressure and pH may play a role in influencing movement and their role should be explored further.

Complexities associated with determining environmental drivers of movement

Although there are multiple studies linking shark and ray movement with abiotic factors, these studies often present unique problems. For example, links between movement and environment are inherently correlative and do not provide conclusive evidence that abiotic factors and not some other driver (e.g. predator avoidance, prey distribution/availability) are behind such movement. In addition, if abiotic factors are shown to be responsible for movement, it often remains unclear whether species are actively seeking specific conditions or merely reacting to them. Finally, abiotic factors rarely occur in isolation, making it difficult to determine which factor is of highest importance or whether synergistic interactions between different factors are occurring. The influence of other factors confounds interpretation of abiotic cues for movement including biological and physiological processes and geographic variability. Here we discuss some of the complexities associated with determining environmental drivers of movement for elasmobranchs.

Biotic factors

Several biotic factors have been shown to trigger movement and changes in behaviour among sharks and rays. In particular, prey density and availability (Heithaus et al. 2002; Shepard et al. 2006; Torres et al. 2006; Jaine et al. 2012) and predator avoidance (Heupel and Hueter 2002; Collins et al. 2007; Heithaus et al. 2009) are thought to influence movement patterns and habitat choice for many species. Movements of several planktivorous sharks and rays, for example, have been linked to tidally-concentrated plankton aggregations (Sims and Quayle 1998; Shepard et al. 2006; Priede and Miller 2009; Jaine et al. 2012). Similarly, spatial patterns in abundance for the gray smooth-hound shark (Espinoza et al. 2011), tiger shark (Heithaus 2001) and sixgill shark (Andrews et al. 2010) were shown to be influenced by both water temperature and prey availability, most likely due to these species following a seasonal shift in prey resources. As biotic factors may work alongside and/or mask the effects of abiotic factors, it is often a challenge to decipher the exact effects of abiotic factors on movement.

Variation between seasons

Seasonal variability in abiotic factors may complicate resolution of important drivers of movement for sharks and rays. For example, Campbell et al. (2012) observed that while movement of the freshwater whipray, Himantura dalyensis, was dominated by the diel cycle during the wet season, whipray movement patterns during the dry season were related to the tidal/lunar cycle. The authors concluded that the observed shift from tidal/lunar to diel periodicity in movement was related to the seasonal suppression of tidal flow during the wet season. Similarly, seasonal differences in diving patterns were observed for the blue shark where individuals made frequent diurnal excursions to depth resting at the surface between dives during winter, but did not show this pattern in summer (Carey and Scharold 1990). The authors speculated that seasonal changes in the type or availability of prey, or sexual activity associated with the onset of mating season, may be responsible for the observed change in swimming behaviour. Since conditions often vary between seasons, it is important that studies include long-term monitoring of individuals when possible to encompass seasonal differences.

Variation between sexes

Sexual segregation is a common characteristic of shark and ray populations (Klimley 1987; Sims 2005; Robbins 2007; Wearmouth and Sims 2008; Mucientes et al. 2009) so it is perhaps unsurprising that research has also shown variability in response to abiotic factors between sexes. Male Atlantic stingrays, for example, exhibited little change in preferred temperature while females showed thermal preferences based on feeding and reproductive state (Wallman and Bennett 2006). The authors concluded that female selection for a preferred temperature range may be due to higher energetic demands to maintain a larger body size, the reproductive cost of yolking eggs, or to meet nutritional demands of pups during gestation. Sexual segregation was also observed in round stingrays monitored in a restored Californian estuary (Jirik and Lowe 2012). Preferential use of warmer, shallow waters of the restored basin by female stingrays was thought to provide some reproductive benefit. Given that energetic demands vary due to sex-specific costs in reproduction, it may not be uncommon for environmental drivers of movement to differ between sexes, making consideration of sex-based differences important to data interpretation.

Variation between size classes

Ontogenetic shifts in movement, behaviour and habitat use have been widely documented (Gruber et al. 1988; Papastamatiou et al. 2009; Andrews et al. 2010; Heupel et al. 2010) and several studies have shown environmental drivers of movement to vary between size classes (Wetherbee et al. 2007; Heupel and Simpfendorfer 2008; Whitty et al. 2009; Knip et al. 2011a). For example, bull sharks in the Caloosahatchee River, Florida, partitioned habitat among size classes, with the youngest, smallest individuals typically found upriver while older, larger juveniles occurred farther downstream in adjacent embayments (Simpfendorfer et al. 2005; Heupel and Simpfendorfer 2008). As the youngest individuals displayed an affinity for salinities between 7 and 20 psu, and given that the Caloosahatchee River was the only place within the system where salinities <20 psu regularly occurred, the authors speculated that salinity was an important factor in the observed size-based partitioning, possibly as a means to reduce intraspecific predation.

Similarly, studies on movement patterns of juvenile pigeye sharks (Knip et al. 2011a), freshwater sawfish (Whitty et al. 2009) and lemon sharks (Wetherbee et al. 2007) all reported movement by the youngest, smallest individuals into the shallowest habitat with the tides. This behaviour is thought to be a refuging strategy from larger predators. Ontogenetic differences in habitat requirements among sharks and rays may translate into different abiotic drivers of movement between size classes and highlights the need to include a range of age groups within a given sampling pool before population level conclusions are drawn.

Variation between and within geographic locations

To further complicate identification of triggers for movement, environmental drivers may differ between and within regions for a given species. For example, the primary abiotic factor influencing the distribution of bull sharks in the Caloosahatchee River was salinity (Simpfendorfer et al. 2005; Heupel and Simpfendorfer 2008), whereas in the Florida Everglades dissolved oxygen was reported to be the greatest predictor of bull shark abundance (Heithaus et al. 2009). Heithaus et al. (2009) speculated that differences in primary environmental drivers of movement between regions may be due to site-specific differences in the physical structure and hydrological dynamics of the two estuaries. Specifically, the Caloosahatchee River estuary is affected by acute low salinity/high freshwater run-off events while the Everglades estuary is not, resulting in salinity having a lower influence in this region. Similarly, residency and movement patterns of cownose rays in a southwest Florida estuary indicated no relationship between ray activity and tidal stage (Collins et al. 2007), contrasting with previous results from Chesapeake Bay (Smith and Merriner 1985, 1987). Collins et al. (2007) speculated that site-specific differences in foraging profitability may explain the differing drivers of movement observed between these two regions.

Abiotic factors triggering movement and changes in behaviour may also differ over smaller spatial scales. Carlisle and Starr (2010) reported that distribution and movement of leopard sharks within an estuarine system on the coast of California were strongly influenced by the tides, but that the pattern of movement depended on what section of the site sharks were using. In the main channel sharks were observed to move with the tide, maximising their foraging area over a more dispersed prey field, while in the adjacent Elkhorn Slough National Estuarine Research Reserve sharks swam against tidal currents, most likely in an effort to remain close to intertidal mudflats where prey resources were abundant. Similarly, Rechisky and Wetherbee (2003) examined the short-term movements of young sandbar sharks within a nursery ground in Delaware Bay and found movement patterns of sharks to differ between sections of the same bay. On the shallow shelf of western Delaware Bay shark movements were tidally-based while movements on the deeper eastern side of the bay were independent from tidal currents. Studies on spatial ecology of elasmobranchs must consider differences in species response to environmental drivers over both broad and small spatial scales and ensure that all habitat types are represented within a sample to provide sufficient resolution of species behaviour.

The role of environmental change on spatial ecology

A wide range of abiotic factors are expected to fluctuate within the next century as a result of climate change. Current estimates by the Intergovernmental Panel on Climate Change (IPCC) predict a 1–6 °C rise in average sea surface temperature (SST) by 2100 (IPCC 2007). In addition, projected increases in both the frequency and intensity of tropical storms and associated high rainfall events (Emanuel 2005; Webster et al. 2005) will most likely result in acute fluctuations in barometric pressure and salinity levels. Atmospheric carbon dioxide levels are expected to exceed 500 parts per million by the end of this century, dramatically altering the pH of oceans due to ocean acidification (Hoegh-Guldberg et al. 2007). Given that many of the abiotic factors predicted to fluctuate under future climate change scenarios have been shown to trigger movement and changes in behaviour and habitat use of sharks and rays, it is important to consider their responses to future changes in their environment.

A recent integrated risk assessment looking at the vulnerability of sharks and rays on Australia’s Great Barrier Reef (GBR) to climate change found that vulnerability was driven by case-specific interactions of multiple factors (e.g. water temperature, ocean acidification) and species attributes (e.g. habitat specificity, mobility) (Chin et al. 2010). Based on this, the authors concluded that freshwater/estuarine and reef associated sharks and rays are particularly vulnerable to climate change impacts since key habitats for these species are affected by an array of anthropogenic and environmental impacts, thus reducing the overall resilience of these species. Although there is an argument to be made that changing conditions may benefit species by making some habitats available that previously were not, movement to alternative locations may not be possible or desirable for many species. For freshwater/estuarine and reef associated sharks and rays, the risks of moving to find new, suitable habitat (e.g. increased chance of predation, inability to find new suitable habitat) must be weighed against the energetic costs of remaining in an altered environment. Individuals that choose to remain in these habitats may exist in a stressed state, forced to allocate energy to physiological processes such as osmoregulation that could be put towards growth and/or reproduction, resulting in a loss to overall fitness or, at worst, mortality. Furthermore, although laboratory studies have demonstrated that some elasmobranch species are able to acclimate to higher than average temperatures, adaptation to extreme environmental fluctuations remains poorly studied.

Temperature and salinity are the two abiotic factors predicted to have the greatest effects on sharks and rays under current climate change projections (Chin et al. 2010). Given that many species have been shown to behaviourally regulate these two factors, fluctuations in either may cause alteration of movement, changing the way species interact with and utilise habitat, or departure from affected areas should conditions become intolerable. If adverse conditions persist they could induce permanent range contractions or shifts as individuals attempt to locate and/or remain in suitable habitat (Last et al. 2011; Hazen et al. 2013). Range contractions could negatively affect populations by making them more susceptible to localised depletion (i.e. fishing impacts) or mortality due to site-specific degradation and loss of habitat. Range shifts have already been observed for many temperate marine fishes (Holbrook et al. 1997; Perry et al. 2005; Munday et al. 2007) and some sharks (Last et al. 2011; Hazen et al. 2013) in response to higher than average ocean temperatures. Hazen et al. (2012) predicted substantial northward displacement of a range of predators under current climate change scenarios, with an up to 35 % change in core habitat for some species. The authors concluded that the shark guild showed the greatest risk of pelagic habitat loss. Ultimately, selection of preferred environmental conditions via movement may dictate the distribution of not just individuals, but populations and could significantly alter entire ecosystems and their associated marine communities.

Conclusion and future work

This review has shown that many shark and ray species actively select for or exploit specific environmental conditions, often forgoing their home range areas to access a spatially variable resource. Movement to seek out and/or remain within preferred environmental conditions highlights the important role abiotic factors play in the spatial ecology of many marine predators. However, abiotic factors rarely act in isolation and behavioural responses may differ between species, sex, ontogenetic stage, season and geographic location. Movement of sharks and rays is most likely driven by a suite of factors—both abiotic and biotic—with habitat selection by individuals representing a compromise between the need to optimize metabolic/physiological function and maintain access to valuable resources (e.g. food, shelter from predators). Recent studies looking at aggregation behaviour in female leopard sharks, for example, have shown drivers of movement to be complex and influenced by preferred environmental conditions (i.e. low wave intensity, warm water temperatures) as well as proximity to key foraging grounds (Nosal et al. 2013, 2014). Future research should consider both abiotic and biotic factors and their potential roles in shaping movement behaviour.

Resolution of the specific role abiotic factors play in shark and ray movement, behaviour and habitat use may be improved through long-term monitoring studies examining movement at multiple scales and through laboratory research. A long-term acoustic monitoring study on the ray community in Shark Bay, Western Australia, found that previously observed differences in seasonal abundance of rays was most likely due to seasonal changes in habitat use rather than large-scale migrations as was previously thought (Vaudo and Heithaus 2009, 2012). The authors were able to compare movement patterns of tracked individuals to abundance data and speculated that observed decreases in ray density during winter may be due to more time spent in deep water habitats rather than emigration from the bay. Laboratory studies can help clarify response to abiotic factors by separating out and testing individual factors while controlling for others. For example, results from controlled experiments that display species actively seeking out a preferred temperature range (Casterlin and Reynolds 1979; Wallman and Bennett 2006) or improving digestive efficiency through post-feeding thermotaxis (Di Santo and Bennett 2011) serve to provide support for behavioural thermoregulation.

Past studies on elasmobranch response to environmental change have focused primarily on coastal species and are heavily biased to sharks. Reasons for this are multifold, but typically include economic constraints, the highly variable nature of coastal systems and ease-of-access to nearshore shark species. As such, the current body of literature concerning the effects of abiotic factors on movement, behaviour and habitat use of elasmobranchs is not necessarily reflective of all species traits or habitat types. Information on response to abiotic factors for sharks resident on coral reefs, for example, remains relatively unknown despite their identification as some of the species most at risk from climate change effects (Heupel and Simpfendorfer 2014). Future studies examining the role of environmental drivers on elasmobranch spatial ecology should include species from a broad range of habitats to gain a better understanding of the importance of these variables across this diverse family.