Search

Understanding why sharks bite humans is essential for developing strategies to prevent these incidents. Here, we use bite wound characteristics and eye witness descriptions of shark behavior to determine the likely motivation for several bites perpetrated by an oceanic whitetip (OWT) shark Carcharhinus longimanus on an adult female snorkeler off Moorea island (French Polynesia) in October 2019. The victim was snorkeling with others in pelagic waters as part of an organized whale-watching tour when the shark-without any warning behavior-bit her at least three times resulting in severe injuries with substantial loss of soft tissue from the chest and both forearms. The victim survived these injuries thanks to rapid and effective first aid provided by her companions. The sudden, unprovoked and repeated bites with substantial tissue removal are consistent with predatory behavior although the dominance hypothesis cannot be fully ruled out. This would be the first case of a predatory shark bite ever documented in French Polynesia in over 70 years of data collection. Given the routine association of OWT sharks with cetaceans, in-water whale watching activities should adopt appropriate risk management strategies in regions hosting this species of shark.

BackgroundGreat hammerhead sharks (Sphyrna mokarran) routinely swim on their sides and periodically roll from side to side. A previous study used wind tunnel tests with a rigid model hammerhead shark to demonstrate that the rolling behavior could improve swimming efficiency using the tall first dorsal fin as a lift-generating surface. Scalloped hammerhead sharks (Sphyrna lewini) also have proportionally taller dorsal fins compared to pectoral fins than most shark species and similar to that of great hammerhead sharks, and thus might exhibit similar rolling behavior. This was assessed by deploying multi-sensor accelerometer instrument packages on free-swimming adult scalloped hammerhead sharks to directly measure swimming depth, body orientation and swimming performance. Specific objectives were to (1) determine whether scalloped hammerhead sharks exhibit side swimming and rolling behavior, (2) characterize the patterns of these behaviors, and (3) evaluate the purpose of these behaviors.ResultsWe obtained 196.7 total days (4720 h) of data from 9 free-swimming adult scalloped hammerhead sharks equipped with multi-instrument biologgers with deployment durations ranging from 7 to 29 days. All sharks exhibited rolling behavior throughout the entire period of observation. The roll angle magnitude and periodicity of rolling showed a clear diel pattern. During daytime, the sharks spent an average of 48% of the time swimming at a roll angle > 30°, with an average roll angle of 41° and rolling periodicity of around 4 min. At night, the sharks spent an average 82% of their time at an angle > 30°, with an average roll angle of 60° and rolling periodicity of around 13 min. In addition to an increase in degree of roll and roll duration, overall dynamic body acceleration (ODBA) also increased at night, and tailbeat frequency was more regular and consistent than during daytime.ConclusionWe observed rolling behavior in scalloped hammerhead sharks similar to that observed in great hammerhead sharks. The diel changes in roll angle and periodicity were accompanied by other changes in swimming behavior. These changes are possibly due to interplay between reducing cost of transport and social interactions with conspecifics.

The application of genome-wide cytonuclear molecular data to identify management and adaptive units at various spatio-temporal levels is particularly important for overharvested large predatory organisms, often characterized by smaller, localized populations. Despite being “near threatened”, current understanding of habitat use and population structure of Carcharhinus galapagensis is limited to specific areas within its distribution. We evaluated population structure and connectivity across the Pacific Ocean using genome-wide single-nucleotide polymorphisms (~7200 SNPs) and mitochondrial control region sequences (945 bp) for 229 individuals. Neutral SNPs defined at least two genetically discrete geographic groups: an East Tropical Pacific (Mexico, east and west Galapagos Islands), and another central-west Pacific (Lord Howe Island, Middleton Reef, Norfolk Island, Elizabeth Reef, Kermadec, Hawaii and Southern Africa). More fine-grade population structure was suggested using outlier SNPs: west Pacific, Hawaii, Mexico, and Galapagos. Consistently, mtDNA pairwise Φ(ST) defined three regional stocks: east, central and west Pacific. Compared to neutral SNPs (F(ST) = 0.023–0.035), mtDNA exhibited more divergence (Φ(ST) = 0.258–0.539) and high overall genetic diversity (h = 0.794 ± 0.014; π = 0.004 ± 0.000), consistent with the longstanding eastern Pacific barrier between the east and central–west Pacific. Hawaiian and Southern African populations group within the west Pacific cluster. Effective population sizes were moderate/high for east/west populations (738 and 3421, respectively). Insights into the biology, connectivity, genetic diversity, and population demographics informs for improved conservation of this species, by delineating three to four conservation units across their Pacific distribution. Implementing such conservation management may be challenging, but is necessary to achieve long-term population resilience at basin and regional scales.

The past six decades have seen the emergence of new electronic tag technologies enabling scientists to remotely study the behavior of fishes in a wide variety of aquatic habitats. This revolution began in the 1950s and 1960s with the first studies utilizing acoustic and radio transmitters to actively track fish movements. Subsequent decades saw the development of passive monitoring systems enabling researchers to scale up the number of individuals followed, and the duration and spatial extent of tracking studies. Recent decades also have witnessed the increasingly-widespread use of sophisticated satellite transmitters to quantify fish movements in remote parts of the ocean beyond the range of fixed listening arrays. Most recently, high-frequency sampling devices incorporating accelerometers and other novel sensors have given us unparalleled new and detailed insights into fish behavior. All of these technologies have yielded important insights into fish spatial behaviors, and shown that many species are most active at night, or in deep waters, but we need better ground-truthing to reveal the full context and purpose of these spatial patterns. Recent advances in animal-mounted cameras have provided important ground-truthing breakthroughs, allowing us to see behavior and habitat use from the fishes' own perspective. However, current generations of animal-mounted cameras are reliant on natural light, or use artificial light overlapping the visual spectrum of most fishes. To see what fishes are doing at night, we need to develop new animal-mounted cameras using far red spectrum light, or high-frequency sound, to illuminate the view field without impacting natural behavior.

Aim Broadly distributed reef fishes tend to have high gene flow mediated by a pelagic larval phase. Here, we survey a reef-associated fish distributed across half the tropical oceans, from the Red Sea to the central Pacific. Our goal is to determine whether genetic structure of the broadly distributed Yellowstripe Goatfish (Mulloidichthys flavolineatus) is defined by biogeographical barriers, or facilitated via larval dispersal. Location Red Sea, Indian Ocean, Pacific Ocean Methods Specimens were obtained at 19 locations from the Red Sea to Hawai'i. Genetic data include mtDNA cytochrome b (n = 217) and 12 microsatellite loci (n = 185). Analysis of molecular variance (AMOVA), structure, a parsimony network and coalescence analyses were used to resolve recent population history and connectivity. Results Population structure was significant (mtDNA ϕST = 0.68, P

Sharks zigzag vertically through the water in a series of alternating ascending and descending segments, changing depth by a few tens of meters over a period of a few hundred seconds. This 'yo-yo' like behavior has several characteristic patterns, identifiable by the way the swimming and vertical velocities vary along the dive. We suggest that these patterns represent different optimal strategies minimizing the cost of locomotion under different constraints; moreover, these constraints can be inferred by matching the pattern of a dive with a (standard) optimal swimming strategy for which the constraints are known. We used three sets of constraints and two definitions of the 'cost of locomotion' to analytically generate four standard optimal strategies; we have used high resolution tracking data from four tiger sharks to identify two different yo-yo diving patterns. These patterns seem to match two of the standard strategies: one that maximizes range, given an alternating power supply (e.g., swimming actively on ascents and lazily on descents); and the other that maximizes range, given an alternating vertical velocity (implying an 'intentional' up-and-down motion).

Summary 1. Animal search patterns reflect sensory perception ranges combined with memory and knowledge of the surrounding environment. 2. Random walks are used when the locations of resources are unknown, whereas directed walks should be optimal when the location of favourable habitats is known. However, directed walks have been quantified for very few species. 3. We re-analysed tracking data from three shark species to determine whether they were using directed walks, and if so, over which spatial scales. Fractal analysis was used to quantify how movement structure varied with spatial scale and determine whether the sharks were using patches. 4. Tiger sharks performed directed walks at large spatial scales (at least 6–8 km). Thresher sharks also showed directed movement (at scales of 400–1900 m), and adult threshers were able to orient at greater scales than juveniles, which may suggest that learning improves the ability to perform directed walks. Blacktip reef sharks had small home ranges, high site fidelity and showed no evidence of oriented movements at large scales. 5. There were inter- and intraspecific differences in path structure and patch size, although most individuals showed scale-dependent movements. Furthermore, some individuals of each species performed movements similar to a correlated random walk. 6. Sharks can perform directed walks over large spatial scales, with scales of movements reflecting site fidelity and home range size. Understanding when and where directed walks occur is crucial for developing more accurate population-level dispersal models.

Many pelagic fishes exhibit ‘yo-yo’ diving behavior, which may serve several possible functions, including energy conservation, prey searching and navigation. We deployed accelerometers and digital still cameras on 4 free-ranging tiger sharks Galeocerdo cuvier, to test whether their vertical movements are most consistent with energy conservation or prey searching. All sharks swam continuously, with frequent vertical movements through the water column at mean swimming speeds of 0.5 to 0.9 m s–1. Tail-beating was continuous except for sporadic, powerless gliding during descents (from 0 to 18% of total descent time). At a given tailbeat frequency, swimming speeds were lower during ascent than descent (consistent with negative buoyancy). Burst swimming events, which might represent prey pursuits, were observed during all phases of vertical movements. Camera images showed a variety of potential prey and the possible capture of a unicornfish. Collectively, results suggest that yo-yo diving by tiger sharks is not primarily for energy conservation, but probably represents an effective search strategy for locating prey throughout the water column.

The main Hawaiian Islands (MHI) are the principal breeding ground for humpback whales in the North Pacific. Over the past 3 decades, population recovery from whaling-era losses has resulted in a steady increase in the number of whales wintering in Hawaiian waters and a geo- graphic expansion of their distribution in the MHI. Until recently, no existing evidence showed that this expansion included the islands, atolls, and banks of the Northwestern Hawaiian Islands (NWHI). To better understand the occurrence of humpback whales in the NWHI, 9 ecological acoustic recorders (EARs) were deployed at sites throughout the archipelago to record the occurrence of humpback whale song, an indicator of winter breeding activity. Song was found to be prevalent at Maro Reef, Lisianski Island, and French Frigate Shoals but was also recorded at Kure Atoll, Midway Atoll, and Pearl and Hermes Atoll. Both the timing and abundance of song at several locations closely followed trends observed on Oahu, which is one of the MHI, strongly suggesting that humpback whales use the NWHI as a wintering area. This finding is of particular relevance in light of recent suggestions that, based on photo-identification data, a yet undocumented wintering area exists somewhere in the central North Pacific. We propose that the NWHI could be that area.

We equipped individual tiger (Galeocerdo cuvier Peron and Lesueur, 1822) and Galapagos (Carcharhinus galapagensis Snodgrass and Heller, 1905) sharks with both acoustic and satellite transmitters to quantify their long-term movements in the Papahanaumokuakea Marine National Monument (Northwestern Hawaiian Islands). Tiger sharks exhibited two broad patterns of behavior. Some individuals were detected at French Frigate Shoals (FFS) year round, whereas others visited FFS atoll in summer to forage on fledging albatross, then swam thousands of kilometers along the Hawaiian chain, or out into open ocean to the North Pacific transition zone chlorophyll front, before returning to FFS in subsequent years. These patterns suggest tiger sharks may use cognitive maps to navigate between distant foraging areas. Different patterns of spatial behavior may arise because cognitive maps are built up through individual exploration, and each tiger shark learns a unique combination of foraging sites. Galapagos shark detections were all associated with FFS, suggesting these sharks may be more resident around oceanic islands. Both Galapagos and tiger sharks primarily used the mixed layer (

We tracked the long-term movements of 70 parrotfishes, surgeonfishes and goatfishes captured inside a small (1.3 km2) marine protected area (MPA: Kealakekua Bay Marine Life Conservation District, Hawaii) by implanting them with small transmitters and deploying underwater monitoring devices inside the bay and along 100 km of the adjacent west Hawaii coastline. Individual fish were detected inside Kealakekua Bay for up to 612 days but many were detected for much shorter periods (median = 52 days). There were species-specific differences in the scale of movements and habitats used, but most fish utilized between 0.2 and 1.6 km of coastline, and individuals of each species showed some degree of diel habitat shift. A wide variety of reef fishes captured inside the MPA swam back and forth across an MPA boundary intersecting continuous reef (i.e., this boundary was porous to reef fish movements), but only 1 of 11 species tagged crossed a wide sandy channel inside Kealakekua Bay suggesting that this feature may function as a natural barrier to movements. Results indicate relatively small MPAs (

SUMMARYShark cage diving is both popular and controversial, with proponents citing educational value and non-extractive use of natural resources and opponents raising concerns about public safety and ecological impacts. Logbook data collected 2004–2008 from two Oahu (Hawaii) shark cage diving operations were analysed to determine whether such voluntary records provide useful insights into shark ecology or ecotourism impacts. Operators correctly identified common shark species and documented gross seasonal cycles and long-term trends in abundance of Galapagos (Carcharhinus galapagensis), sandbar (Carcharhinus plumbeus) and tiger sharks (Galeorcerdo cuvier). Annual cycles in shark abundance may indicate seasonal migrations, whereas long-term trends suggest gradual exclusion of smaller sandbar sharks from cage diving sites. Numerically dominant (> 98%) Galapagos and sandbar sharks are rarely implicated in attacks on humans. Negligible impact on public safety is supported by other factors such as: (1) remoteness of the sites, (2) conditioning stimuli that are specific to the tour operations and different from inshore recreational stimuli and (3) no increase in shark attacks on the north coast of Oahu since cage diving started. Tracking studies are required to validate logbook data and to determine whether sharks associated with offshore cage diving travel into inshore areas used for in-water recreation.

Little is known about the long-term movement patterns of most marine apex predators. A network of acoustic receivers was used to quantify the long-term movements of transmitter- equipped tiger sharks Galeocerdo cuvier Peron & Lesueur, 1822 in the Main Hawaiian Islands. Tiger sharks were wide-ranging, swam between islands and patrolled up to 109 km of contiguous coastline. Visits to specific acoustic receiver sites were typically brief (mean duration 3.3 min), unpredictable and interspersed by absences of weeks, months or years. This pattern may be an optimal foraging strategy for capturing risk-averse prey. Tiger sharks may have to move on soon after arriving in an area because the element of surprise is quickly lost and potential prey become wary and difficult to catch. Juvenile tiger sharks were significantly wider-ranging and less frequently detected than mature females. Juveniles may be avoiding predation by larger individuals, or explor- ing to find suitable home ranges. Tiger sharks may also switch movement patterns and foraging strategies to take advantage of different prey types, restricting their movements to exploit seasonally abundant and naive prey. Further empirical studies are required to directly link movement patterns with foraging.

We quantified spatial dynamics and substrate impacts of snorkelers and SCUBA divers within four Hawaiian MPAs to determine: (1) whether coral reefs in these areas are being damaged by recreational activities, and (2) how damage might be mitigated. Observers secretly followed snorkelers and SCUBA divers, and used handheld Global Positioning System (GPS) units to record their geographic tracks and substrate contacts. Most activities occurred within relatively small, well-defined areas associated with access points. Despite heavy use, recreation impact was low at Hawaiian MPAs because most fragile corals occurred below the maximum depth of the dominant recreational activity (snorkeling). SCUBA diving was only common at one MPA with physically durable benthic habitats. GPS tracking provided useful insights into how recreational impacts to MPAs could be reduced. General topography could be used to predict where visitors will go, and designated access points could be used to focus substrate contact away from fragile habitats.

Gaps in our knowledge of basic fish ecology have provided impetus for development of novel "ecology tags" to detect and quantify hard to observe behaviors such as spawning, schooling and feeding. The acoustic envi- ronment is one source of potentially useful information about these behaviors. We implanted an acoustic recording tag (Bioacoustic Probe) into the gut cavity of a blacktip reef shark to determine whether an implanted tag could success- fully record external and internal sounds. The tag successfully recorded reef fish vocalizations, boat engine noise, the sound of the shark feeding and unidentified rhythmic sounds that may derive from shark tail beats. Technical challenges remain, but sound recording tags have the potential to provide novel insights into shark and fish ecology.

Little is known about the feeding habits of large free ranging fish, due in large part to lack of an appropriate technique for quantifying feeding variables. A previous study demonstrated that changes in gastric pH can be used as a proxy for feeding events in free-ranging sharks. Here we describe the development of a new acoustic pH transmitter to remotely measure gastric pH in sharks in the field. The transmitter consists of a dual sensor (pH and temperature) continuous pinger, and was tested in captive adult blacktip reef sharks (Carcharhinus melanopterus). The transmitter was retained in the shark's stomach for between 5-12 days. The empty stomach had a low pH (1.6 ± 0.2) and feeding induced a rapid increase in gastric pH, which was clearly distinguishable from baseline levels. Meal size showed a significant linear relationship with the magnitude of the pH changes. Measurement accuracy of the pH transmitter ranged from 0.05-0.9, although resolution of the VR100 receiver that decoded the transmitter signal was 0.1 units. The pH transmitter can be used to determine when free-ranging sharks in the field are feeding and hence quantify feeding chronology, frequency and daily ration.

I used shoreline creel surveys to quantify fishing activities in and around a small (0.34 km 2 ) Marine Protected Area (MPA) in Hawaii (Waikiki Marine Life Conservation District). Spear fishing and shoreline pole & line fishing (angling) were the dominant fishing activities at Waikiki. Spear fishing had a greater overall impact on reef fishes than shoreline pole & line fishing, accounting for 70% of the total reef fish harvest at Waikiki, despite accounting for only 25% of fishing activities observed. Fishing activities at Waikiki were unevenly distributed in space and time. The MPA experienced minor illegal fishing and was located between an area of high diurnal spear fishing effort and an area of generally low fishing effort. This pattern of fishing activities allows jacks and goatfishes to evade capture despite nightly excursions from the MPA into fished areas, and may partly explain why these fishes remain more abundant and larger inside the Waikiki MPA than in surrounding fished areas. Quantifying fishing activities at MPA sites can provide valuable insight into how these areas function, and this information can be used to improve MPA design and effectiveness.

Empirical data quantifying the long-term movement patterns of coral reef top predators are needed in order to design marine protected areas (MPAs) that will provide these fishes with effective, long-term protection. Acoustic telemetry was used to quantify the movements of a large coral reef top predator (Aprion virescens, Lutjanidae; Hawaiian name ‘uku’) at five atolls in the Northwestern Hawaiian Islands Marine National Monument (NWHIMNM) from May 2005 to September 2006. The study atolls were located between 23.8°N, 166.2°W and 28.5°N, 178.3°W, and were separated from their nearest receiver-equipped neighbor by distances ranging from 100 to 478 km. No inter-atoll movements by uku were detected but individuals were seasonally site-attached to core activity areas of up to 12 km in length, and ranged up to 19 km across atolls. Within their core areas, tagged uku exhibited diel and tidal habitat shifts, with the latter resulting in round trips of up to 24 km in 24 h. Seasonal uku migrations resulted in extended winter (October–April) absences from summer (May–September) core activity areas and may be linked to summer spawning. Large MPAs (i.e., entire islands, atolls or banks) would probably be required for full protection of resident populations of adult uku, but such ‘island-scale’ MPAs will not benefit fisheries unless there is significant larval supply from MPAs to neighboring fished areas, or adult emigration over time scales exceeding the 16-month monitoring period of this study. A mixed management strategy of combining smaller MPAs with conventional measures (e.g., minimum size limits, catch, and effort restrictions) may be the best approach for sustaining uku fisheries. However, this would still require relatively large MPAs (12 km in length) to contain uku short-term (diel and tidal) movements, and seasonal migrations would still take uku beyond the boundaries of MPAs of this size. These fluxes across MPA boundaries could supply fish to fisheries but, if high exploitation rates exist, fishing could eliminate key MPA benefits such as increased numbers of large, highly fecund individuals.

We quantified bluespine unicornfish, Naso unicornis, movement patterns, home range size and habitat preferences in a small Hawaiian marine reserve. Bluespine unicornfish were site-attached to home ranges situated within the reserve boundaries and their movements were aligned with topographic features. Two different diel movement patterns (‘commuting’ and ‘foraying’) were observed. Commuters made crepuscular migrations of several hundred meters between daytime foraging areas and nighttime refuge holes. Foraying fish did not partake in crepuscular migrations and utilized refuge holes both day and night. Two bluespine unicornfish were also nocturnally active. There was little direct evidence of dispersal from the reserve but differences in bluespine unicornfish abundance and size among reef habitat zones were consistent with ontogenetic habitat shifts. The influence of habitat topography on bluespine unicornfish movements suggests that gross habitat characteristics could be used to predict reef fish movements. This could provide a simple method for setting marine reserve boundaries at sites for which empirical fish movement data are unavailable.

The feasibility of long-term acoustic monitoring of coral reef fish movements was evaluated by quantifying transmitter retention times for captive bluefin trevally Caranx melampygus and detection patterns of transmitter-implanted free-ranging bluefin trevally. Wax-coated dummy transmitters (9 × 22 mm; 3.6 g) were surgically implanted into the gut cavity of four captive bluefin trevally. The fish retained the dummy transmitters for 17–22 months, but at the end of the study transintestinal expulsion was in progress in one individual. Five bluefin trevally captured at Honokohau, Hawaii, were surgically implanted with long-life (1-year) coded transmitters and released. Their movements were monitored with the use of 27 underwater receivers deployed along a 115-km stretch of coastline. The five bluefin trevally were detected by the receivers over periods ranging from 1 to 173 d (mean, 60 d), and their movements spanned up to 10.2 km of coastline. Daily movements were consistent with those observed by p...

Movement patterns, site fidelity and growth were studied for giant trevally ( Caranx ignobilis) inhabiting a marine reserve surrounding Coconut Island in Kaneohe Bay, off the Island of Oahu, Hawaii. Short-term movements of fish were determined by acoustic telemetry and long-term movements and growth were investigated with conventional tag and release methods. Giant trevally appear to move to the Coconut Island marine reserve from the adjacent bay floor upon attaining a size of approximately 20–25 cm fork length. After several years spent frequenting the reserve, giant trevally move out of the bay into deeper water. Tracked fish spent considerable time within the marine reserve moving along the reef slope, but frequently ventured outside of the reserve boundaries. The recapture rate for tagged fish was 11%, with an average time at liberty of 346 days. Nearly one-third of recaptured fish were caught at distances greater than 3 km from the tagging site with maximum values demonstrating long-term (more than 7 years) and long-distance (30 km) movements. The protective function of the Coconut Island marine reserve for giant trevally is limited because the reserve is utilized by only a portion of the population, and even these fish regularly move outside of the refuge. © 2003 Elsevier B.V. All rights reserved.

Suitability of small (< 1 km2) marine reserves for protecting a commercially important endemic Hawaiian goatfish, Parupeneus porphyreus, was examined by quantifying goatfish habitat use, home range size and site fidelity in an existing marine reserve (Coconut Island in Kaneohe Bay, Hawaii). Five goatfish equipped with acoustic transmitters were tracked for up to 93 h each over 3–14 days. Daytime habitat use patterns of two of these fish were continuously monitored for one month using a fixed hydrophone hardwired to an onshore computer. Acoustically tagged fish showed consistent diel patterns of behavior, refuging in holes in the reef by day and moving over extensive areas of sand and coral rubble habitat at night. Remote monitoring of daytime habitat use by two goatfish revealed that the same daytime refuge was used by both fish for at least one month (the battery life of the transmitters). Home ranges of all fish were within the boundaries of the Coconut Island reserve suggesting that even small areas containing suitable habitat can make effective reserves for this species. A relatively low abundance of reproductive size P. porphyreus at Coconut Island in comparison with deeper areas may indicate an ontogenetic shift to deeper habitat in this species.

Externally and internally implanted sonic transmitters were used to track the movements of eight tiger sharks (Galeocerdo cuvier) ranging between 200 and 417 cm total length (TL), captured by hook-and-line on the south coast of Oahu, Hawaii, between 1993 and 1997. Attachment of the transmitters was facilitated by the fact that captured sharks exhibited tonic immobility when restrained and inverted at the side of the tagging vessel. Three common themes emerged from the horizontal movements of the tracked sharks: (1) offshore movements away from the island, (2) extended periods of directed, “straight-line” swimming, (3) orientation to the Penguin Banks – a shallow bank located ≃35 km from the release point. In shallow water ( 300 m) they swam within the mixed layer at depths of ∼80 m. One shark dove briefly to 335 m. The average estimated swimming speed of sharks traversing open water was 0.29 body length (BL) s−1. Two sharks were recaptured after termination of the tracks; one of these sharks was recaptured twice, with a total time at liberty of 377 d. The data suggest that Hawaiian tiger sharks move within large home ranges and that they can efficiently navigate between distant parts of their range, even when this requires crossing open ocean waters.

Sharks are found in association with main Hawaiian Island ocean fish farms more frequently and at higher densities than is typical for coastal Hawaiian waters. Sharks attracted to fish farms could potentially threaten human water users, interact negatively with other fisheries, and seasonal migrations could be disrupted if individuals become entrained around farms throughout the year. We hypothesised that smaller coastal species would reside near farms, whereas more wide-ranging species would associate with farms only for short periods. We utilised passive acoustic telemetry to monitor the movements and behaviour of sandbar (Carcharhinus plumbeus) and tiger (Galeocerdo cuvier) sharks adjacent to two open ocean fish farms in Hawaii. Approximately half the tagged sandbar sharks showed site fidelity to the farms, with some individuals being detected repeatedly for 2.5 years. Sandbar sharks moved seasonally to the west coast of Oahu, suggesting that fish farms are not disrupting natural seasonal cycles in this species. Tiger sharks tagged near the cages were more transient, and showed much shorter residence times although some individuals returned sporadically to the cages over the 3-year period. Ocean fish cages appear to aggregate sandbar sharks, but are only ‘visited’ by tiger sharks. Although threats to public safety are probably minimal, the ecological effects of aggregating top-predators are still unknown.