Telemetry reveals potential mating aggregation behavior of tiger sharks (Galeocerdo cuvier) in Hawaiʻi

Tiger sharks (Galeocerdo cuvier) are marine top predators that play a crucial role in ecosystem function across their temperate and tropical circumglobal distribution^1,2,3. By regulating prey abundance and behavior, suppressing mesopredators, scavenging large carrion, and transporting nutrients across habitats, they influence community structure and cross-ecosystem energy flow^4,5. They are typically solitary in nature and rarely form aggregations except in the context of feeding events or possibly in gestation-related groupings^{6,7,8,9,10,11}. Significant knowledge gaps remain in our understanding of their spatial ecology, especially regarding seasonal movements, reproductive behavior, and the occurrence of aggregations³. Although tiger sharks are known to mate seasonally^12,13 it is unclear whether mating relies on opportunistic encounters between males and females, whether they aggregate at specific sites, or whether both factors contemporaneously drive their reproductive strategies. This lack of understanding complicates the identification of key habitats and hinders the prediction of behaviors that are important for informing their conservation and management.

Meyer et al.¹⁴ observed increases in tiger shark presence around Maui that align with the mid-winter mating season, suggesting a possible link to reproductive behavior. However, the observed detection peaks may instead reflect foraging, as tiger sharks in other regions aggregate in response to seasonal prey pulses such as fledgling seabirds and nesting sea turtles^5,15. Waters surrounding Maui also experience an influx of humpback whales (Megaptera novaeangliae) during the winter months¹⁶, offering potential seasonal scavenging opportunities from whale carcasses, placental remains, or neonate calves^17,18,19. The coincidence of Hawaiʻi’s tiger shark mating season with the humpback whale arrival period makes it difficult to distinguish movements driven by mating from those linked to whale-related foraging.

Meyer et al.¹⁴tagged tiger sharks with long-lasting acoustic tags, collecting a continuous six-year time series of detections around Maui and O ‘ahu from 2013 to 2019. This extensive dataset provides an unprecedented opportunity to assess whether seasonal movements to Maui are driven by mating, by foraging, or a combination of the two, as these may not be mutually exclusive activities. Tiger sharks could utilize the shelf of Maui Nui (the region encompassing the islands of Maui, Moloka ‘i, Lānaʻi, and Kahoʻolawe), for both purposes, and both could produce similar patterns of acoustic detections (i.e. clusters of individuals in space and time). The maturity status of the sharks present, as well as the temporal predictability of the clusters could provide further insight into the function of these aggregations. For example, a mating aggregation requires sexually mature adults of both sexes to be present simultaneously at a predictable time each year (tiger shark size at first reproduction: males ~ 292 cm TL, females ~ 330 cm TL¹³;). In contrast, a cluster of detections consisting of only one sex and/or immature individuals that closely tracks prey abundance may be more indicative of foraging activity.

In this study we addressed five questions: (1) Do tiger sharks tagged around O ‘ahu and Maui exhibit repeated inter-island movements? (2) Are these movements seasonal? (3) What are the demographics of seasonal inter-island migrants? (4) Do specific Maui sites show seasonal shark aggregations and overlapping use by both sexes? (5) Does tiger shark seasonal presence correlate with humpback whale indicators (song intensity and calf counts)? By addressing these questions, we aim to clarify the ecological drivers behind tiger shark movements and possible aggregations, thereby advancing our understanding of tiger shark behavior and identifying potentially important habitats in the Hawaiian Islands.

Study site

The Main Hawaiian Islands (MHI) consist of eight high volcanic islands. Shark capture and tagging efforts in this study were concentrated around O ‘ahu and Maui (Fig. 1). The MHI are each bordered by an insular shelf that gradually descends from the shoreline to a shelf break, occurring at depths between 100 and 200 m (Fig. 1). The width of this insular shelf differs across the islands, with the Maui Nui complex (Maui, Moloka ‘i, Lānaʻi, and Kahoʻolawe) having a more extensive shelf than the islands of Ni ‘ihau, Kaua ‘i, O ‘ahu, and Hawai ‘i combined (Fig. 1). The insular shelf hosts a variety of photic and mesophotic coral reefs, macroalgal beds, and sandy habitats^20,21, and is the preferred habitat for tiger sharks in Hawaii waters¹⁴.

Olowalu, on Maui’s west coast (Fig. 2), features a diverse benthic habitat (coral reefs, sandy bottom, macroalgal beds) and a gentle slope to ~ 60 m depth^22,23,24. Importantly, Olowalu’s sheltered waters host breeding, calving, and nursing humpback whales during their overwintering period^25,26,27, making it an important seasonal habitat for this whale population.

Acoustic monitoring system

We used Vemco VR2W acoustic receivers (69 kHz frequency) to monitor tiger shark presence around Maui and O ‘ahu. These small, self-contained underwater receivers (∼ 34 cm length, 6 cm diameter) detect coded acoustic transmitters. Each tagged shark carried a V16-6H transmitter (16 × 94 mm, 14 g in water) that emits a unique ‘ping’ sequence lasting 3–5 s, with a random silent interval of 20–230 s between sequences. Each decoded transmission is logged by receivers with a timestamp and the shark’s ID code. Transmitters had nominal battery lives of 2–10 years (10-year tags were used on Maui sharks; O ‘ahu sharks had mixed 2–10 year tags), enabling multi-year tracking.

We determined receiver detection ranges using a boat-mounted Vemco VR100 hydrophone and test transmitters. We dropped a transmitter from the surface directly above each VR2W receiver, recorded ~ 10 transmissions, then incrementally moved ~ 100 m away (up to 1.5 km) and repeated. By cross-referencing the VR100 GPS log with detections recorded on the stationary VR2W, we identified that VR2W receivers could detect tags at distances up to ~ 900 m.

We deployed receivers at twenty-six monitoring sites around Maui (15 sites total, with data recovered from 14) and O ‘ahu (12 sites) (Fig. 2). The array spanned the depth range of the insular shelf, with inshore units deployed at depths of 5 to 20 m and offshore units deployed in deeper waters (100 to 200 m) up to several kilometers offshore. This array design allowed for the comparison of tiger shark presence between deep and shallow areas, between different coasts of the same island, and between Maui and O ‘ahu. Receivers were actively monitoring around Maui from October 2013 until April 2019, and around O ‘ahu throughout and beyond this period. Receivers were deployed on subsurface moorings.

Shark capture and tagging

Shark handling and tagging activities were carried out in accordance with the animal use protocols of the University of Hawai’i at Manoa Institutional Animal Care and Use Committee (IACUC) and were approved under IACUC protocol #05–053. Sharks were captured and tagged around Maui from Oct 2013 to Feb 2015 (26 individuals: 3 M, 23 F) and around O ‘ahu from Apr 2013 to Oct 2018 (16 individuals: 6 M, 10 F). In addition, 17 ‘legacy’ tiger sharks (3 M, 14 F) originally tagged off O ‘ahu prior to this study were still transmitting during all or part of the monitoring period covered by the Maui receiver array. Following the methods of Meyer et al.¹⁴, we used demersal longlines baited with large tuna heads, which soaked for 2–4 h at depths of 10–100 m. Targeting these shallower depths typically resulted in a strong female bias in capture rates, as tiger sharks in Hawaiʻi exhibit sexual segregation—with females more commonly occupying nearshore habitats and males more frequently found offshore¹⁴. Captured sharks were brought alongside a 6 m skiff, tail-roped, and placed into tonic immobility by inversion. Acoustic transmitters were surgically implanted into the body cavity through a small incision in the abdominal wall, and the incision was closed with interrupted sutures²⁸. Sharks were also tagged with unique external identification tags (Hallprint, Hindmarsh Valley, Australia), then released. The use of internal acoustic transmitters is widespread in shark movement studies and has been shown to have minimal impact on the study subjects²⁹. The size of our acoustic transmitters was small (16 × 94 mm, weight in water 14 g) compared to the size range of the sharks tagged (183–464 cm TL), and surgical implantation procedures were quick and efficient (between 5 and 15 min duration). Our use of tonic immobility as an anesthetic has several advantages, including: rapid induction, minimal disruption to respiration, and immediate and full recovery³⁰. Tiger sharks have been shown to be especially resilient to capture stress. Post-release mortality (PRM) in tiger sharks is very low (0–2% observed^31,32,33,34;), and they rank as the least disturbed shark species in terms of capture stress³⁵.

Field observations

Sharks captured in the field were photo-documented for evidence of recent mating activity, characterized as clasper abrasions or chafing on males and tooth gouges on females.

Data analysis

Maui receiver site selection

We used a Gantt chart (Fig. S1) to visualize receiver deployment timelines and identify core sites with the longest continuous monitoring periods and the greatest temporal overlap. Limiting analyses to these core sites enabled us to compare seasonal patterns across the full six years of the study. Of the 14 receiver sites around Maui, six had the most consistent coverage throughout the monitoring period: Kalama Deep, Mākena Pt., Olowalu Shallow, Kalama Shallow, Honokōwai Shallow, and Palauea Shallow (Fig. 2). These sites were selected as the focus for further comparative analyses.

Seasonal migration patterns and demography (Rayleigh and F-tests)

We conducted a Rayleigh test using the R (v4.4.3; R Core Team 2025) package ‘circular’³⁶ to evaluate whether seasonal migratory patterns differed between sharks captured and tagged around Maui and O ‘ahu. First, we calculated the total observed number of unique individuals from each tagging location (Maui or O ‘ahu) detected around Maui each month. The Rayleigh test identifies departures from uniformity in circular data, where one calendar year is represented as a complete circle. Under the null hypothesis of no seasonal migration, shark detections would be uniformly distributed across months (equal in each month). This test also assumes that any departure from the null hypothesis will be unimodal reflecting a single peak in the distribution. In addition to the Rayleigh test, we calculated the coefficient of variation, variance, and standard deviation to quantify the variability in the number of sharks detected each month for Maui- and O ‘ahu-tagged individuals. These statistics provided insight into the degree of seasonal variation, where high variability in the monthly number of sharks detected may indicate shifts in habitat use and possibly migratory behavior. We used an F-test to compare the variances between the two tagging locations, assessing whether the seasonal variation in the number of sharks detected monthly around Maui differed between Maui and O ‘ahu-tagged individuals. The maturity status of sharks making inter-island movements was determined using the estimated size at first detection, derived from capture size and adjusted based on average annual growth rates for Hawaiian tiger sharks (from³⁷) (Table 1).

Seasonal comparison using T-test for paired samples, independent T-test, and Rayleigh test

Months were categorized as “Summer” (April through September) or “Winter” (October through March), and monthly numbers of sharks detected at each receiver site were summed separately for each season. Before conducting the t-test, we verified key assumptions. Normality was assessed using the Shapiro–Wilk and Kolmogorov–Smirnov tests, supported by visual inspections of histograms and Q-Q plots³⁸. Independence of observations was confirmed: for the independent t-test, we ensured no correlation between shark detections across sites³⁹, and for the paired t-test, we verified logical pairing of Summer vs. Winter detections within each site⁴⁰. Equality of variances was tested with Levene’s test⁴¹ and outliers were identified using box plots and z-scores⁴². A paired t-test assessed seasonal differences across all six receiver sites combined, while independent t-tests and Rayleigh tests evaluated seasonal patterns at individual sites. P-values were calculated for each test, and sites were ranked by probability (Tables 2, 3).

Discrete Fourier transformation using fast Fourier transformation

To examine seasonal patterns at the top four ranked receiver sites (Olowalu, Kalama Deep, Kalama Shallow, and Honokōwai), we applied a Discrete Fourier Transformation (DFT) using the Fast Fourier Transformation (FFT) algorithm to time series data representing monthly shark detection counts over the full monitoring period. The DFT converts a time-domain signal into its constituent frequencies, enabling identification of underlying cyclical trends⁴³. The FFT efficiently computes the DFT, allowing rapid analysis of complex temporal signals⁴⁴. Peaks in the resulting magnitude spectrum indicate dominant frequencies and their corresponding amplitudes, highlighting the strongest periodic patterns in shark detections⁴⁵.

Seasonal trend decomposition using Loess

We used Seasonal Trend Decomposition using Loess (STL) to analyze the time series data of shark detections at Olowalu, the site that showed the most prominent seasonal philopatry. STL is a flexible and robust decomposition method that breaks down time series data into three distinct components: trend, seasonality, and residual⁴⁶. The trend component reflects the overall direction of the data over time, the seasonality component captures recurring, predictable patterns that occur at fixed intervals (e.g., monthly or yearly), and the residual component represents the random fluctuations or noise that remain after removing the trend and seasonal elements. For our analysis, monthly counts of sharks detected were binned to create the time series data. STL was applied to understand how these counts varied over time, allowing us to distinguish long-term trends and recurring seasonal patterns from random variations.

Analysis of diel patterns of shark presence at Olowalu

We used kernel density estimation (KDE)⁴⁷ to quantify diel overlap in habitat use at Olowalu between male and female tiger sharks. Raw detection data, including detection timestamps, unique transmitter IDs, and sex, were analyzed for the entire monitoring period. To reduce pseudo-replication, we excluded consecutive detections of the same individual at the same receiver within 3 min. From each timestamp, we extracted the hour (0–23) as a numeric column. The kernel density of activity for each sex was estimated in R using the density function, which calculates probability density for activity times and produces a smooth distribution. Both densities were interpolated onto a shared 0–23 h grid and normalized to ensure total probabilities summed to 1 for each sex. Overlap was assessed using the Bhattacharyya Coefficient⁴⁸, calculated as the sum of square roots of the product of the two densities at each point. This coefficient, ranging from 0 (no overlap) to 1 (perfect overlap), measures distribution similarity. Kernel density estimates were visualized with ggplot2⁴⁹.

Median daily root-mean-squared sound pressure levels at Olowalu vs. sharks at Olowalu

We used whale song chorusing data provided by Oceanwide Science Institute’s Ecological Acoustic Recorders (EARs) in the 0.1–1.5 kHz frequency band as a proxy for humpback whale abundance at Olowalu^26,27. Median daily root-mean-squared sound pressure levels (RMS SPL) were overlaid with monthly shark detection data at Olowalu to compare the timing of presence and abundance for both species (Fig. 8).

To assess time-lagged relationships, we applied a Cross-Correlation Function (CCF) to the monthly maximum RMS SPL and the number of tiger sharks detected per month. This analysis identifies statistically significant correlations and the lag at which they are strongest, revealing whether peaks in shark presence typically precede or follow peaks in acoustic activity.

We also conducted a linear regression analysis to directly evaluate the relationship between maximum monthly RMS SPL and monthly tiger shark detections at Olowalu. Model assumptions of normality and homoscedasticity were assessed using residual diagnostic plots. Statistical analyses were performed in R (v4.4.3; R Core Team 2025) using the base lm() function, with significance evaluated at α = 0.05 and model fit assessed via R² and residual standard error.

Land-based humpback whale calf survey data vs. sharks at Olowalu

Data from land-based visual scan surveys obtained by Kügler et al.²⁷ were used as a proxy for humpback whale calf abundance near Olowalu. Observations were conducted from a cliff site approximately 1 km northwest of Olowalu. Calf counts were normalized by survey effort and overlaid with the monthly time series of tiger shark detections at Olowalu to assess the timing and potential overlap in species presence and abundance.

We conducted a linear regression analysis comparing the number of individual tiger sharks detected per month with the average number of whale calves observed per hour, based on multiple daily 30 min scans. Data from 13 months spanning peak whale season (December–April) across three years (2017–2019) were included. Calf observations were normalized to account for variable survey effort. A linear model was fit to test whether monthly variation in tiger shark detections corresponded with changes in calf abundance. Model assumptions of normality and homoscedasticity were assessed using residual diagnostic plots. Statistical analyses were performed in R (v4.4.3; R Core Team 2025) using the base lm() function. Significance was set at α = 0.05, with model fit evaluated using R² and the residual standard error.

We also applied a Cross-Correlation Function (CCF) to the normalized monthly calf averages and tiger sharks detected to assess time-lagged relationships and determine whether peaks in shark presence consistently preceded or followed calf abundance.

Overview

From 2013 to 2019, the Maui receiver array detected a total of 44 individual tiger sharks: 8 males (223–408 cm TL) and 36 females (183–464 cm TL). Of these, 21 sharks (48%) had been originally captured and tagged around O ‘ahu (5 of the 8 males and 16 of the 36 females). Detection spans (time between first and last detection on the array) ranged from 1 to 1,990 days (mean = 900 days) for individual sharks, and total detections per shark ranged from 2 to 5,899 (mean = 1,114). Only 5 Maui-tagged tiger sharks were ever detected on the O ‘ahu receiver array, and those detections were sparse (only 3–14 detections each), underscoring the rarity of inter-island movements for Maui-tagged sharks.

Broad-scale seasonal movement patterns

A comparison of monthly tiger shark detections around Maui revealed significant differences in seasonal patterns between sharks tagged on Maui versus those tagged around O ‘ahu. Sharks tagged around Maui were consistently detected across all months, with no significant seasonal variation in the number of unique individuals detected. In contrast, sharks tagged around O ‘ahu exhibited a clear seasonal cycle, with detections of individuals peaking around Maui in February and reaching their lowest point during the summer months. A Rayleigh test confirmed that detections of Maui-tagged sharks did not deviate significantly from a uniform distribution (R = 0.03, p = 0.71), while O ‘ahu-tagged sharks displayed a significant peak, indicating a non-uniform distribution and demonstrating their seasonal presence around Maui (R = 0.28, p = 0.006) (Fig. 3). Additionally, the coefficient of variation for O ‘ahu-tagged sharks was 49.5% compared to 5.7% for Maui-tagged sharks, indicating much higher variability in O ‘ahu sharks’ detection patterns. The variance of O ‘ahu-tagged sharks detected around Maui was 6.9, reflecting a greater spread from the mean, whereas Maui-tagged sharks had a variance of 1.5. The F-test confirmed a significant difference between the variances of the two groups (F(11,11) = 4.5, p = 0.01). Out of the 21 O ‘ahu-tagged sharks that visited Maui, 14 (67%) were later detected back at O ‘ahu at least once. The maximum number of roundtrips observed for any individual shark between O ‘ahu and Maui was four (Fig. S2). Seven O ‘ahu-tagged sharks were not detected anywhere following their detections around Maui; however, most had previously been detected around O ‘ahu before appearing around Maui. Eight O ‘ahu-tagged individuals (36%) were detected in Maui waters during the tiger shark mating season months (January–February), but only one of those appeared in multiple mating seasons.

Demographic patterns of inter-island movements

Of the 21 tiger sharks (16 females and 5 males) captured, tagged in O ‘ahu waters, and later detected around Maui at any time of year, 17 (80%) were sexually mature at the time of their first detection in Maui waters (Table 1). Six individuals were sexually immature when tagged and only detected around Maui after they reached sexual maturity (lengths at detection estimated using tiger shark growth curve from³⁷ and maturity status at detection estimated from maturity data in¹³), 4 were sexually immature when tagged and still sexually immature when detected, and 11 were sexually mature when initially tagged. Among the 13 O ‘ahu individuals (12 females and 1 male) detected at any Maui receiver site during the peak mating season months of January and February 11 were sexually mature.

Evidence for seasonal aggregations at specific Maui locations

The total number of individuals detected seasonally at each of the 6 core Maui monitoring sites ranged from 8 (Honokowai, summer) to 32 (Kalama Deep, winter) (Table 2). The overall mean number of sharks detected at these sites during winter (25.5 ± 5.68) was significantly higher than during summer (20 ± 6.81) (paired t-test, t = − 4.68, df = 5, p = 0.015). A t-test for independent samples comparing summer and winter means at each site found that all six sites exhibited significantly higher mean shark counts during winter than summer months. However, the level of significance for these seasonal differences varied across sites (Table 2).

Aggregate seasonal histograms showing the number of individuals detected in each calendar month across the entire 6 year monitoring period indicate that the seasonal pattern is more clearly defined at Olowalu than at any other site (Fig. 4). Rayleigh tests performed on each individual site confirmed Olowalu as having the most significant peak (Table 3).

Discrete Fourier Transforms (DFTs) performed on the four receiver sites with significant p-values revealed a clear peak in amplitude density at a 1-year (annual) frequency for Olowalu (Fig. 5). This dominant peak indicates that annual cycles are the strongest temporal pattern in the Olowalu time series of tiger shark detections. Kalama Deep showed a similar, though slightly weaker, annual peak, suggesting a comparable cyclical trend. In contrast, Kalama Shallow and Honokōwai exhibited only low-amplitude peaks, indicating a lack of strong or consistent seasonal patterns in shark detections at those sites.

Seasonal Trend Decomposition using Loess also identified a strong cyclical seasonal pattern of shark detections at Olowalu, combined with a long-term trend showing an initial rise in shark detections during the early tagging phase, followed by a gradual decline over the later years (Fig. 6). The remainder was confirmed to be residual noise using Shapiro–Wilk normality test and examination of QQ plot.

Diel activity patterns

A kernel density plot (Fig. 7) revealed substantial temporal overlap between sexes at Olowalu. The Bhattacharyya coefficient of 0.983 for the diel activity patterns of males and females confirmed this overlap. Both sexes were more consistently present during the day than at night, with overlapping peaks in presence at approximately 1500 h. In addition, multiple sharks captured at this site had physical signs of recent mating activity (females with fresh mating scars and males with visibly chafed claspers (Figs. 8, 9)).

Sex, size and maturity status of sharks detected at Olowalu, Maui

Among the O ‘ahu-tagged individuals (8 females and 0 males) detected at Olowalu during the peak tiger shark mating season months of January and February, 7 (87.5%) were sexually mature. Outside of mating season, there were 7 O ‘ahu-tagged tiger sharks (6 females and 1 male) detected at Olowalu, 5 of which were sexually mature.

Comparison of tiger shark detection patterns with humpback whale song intensity

Daily RMS SPL values (used as a proxy for humpback whale presence) plotted against monthly shark detections at Olowalu revealed consistent but slightly offset peaks in abundance for each species (Fig. 10). Cross-correlation analysis confirmed this offset, indicating a one-month time lag in which peak tiger shark detections preceded peak whale call intensity (best lag = 1 month, r = 0.821; Fig. 11). Linear regression analysis showed a significant positive relationship between the number of tiger sharks detected per month and the monthly maximum RMS SPL (β = 0.387, SE = 0.059, t = 6.511, p < 0.001). Residual diagnostics indicated no clear violations of model assumptions, suggesting a good overall fit.

Comparison of tiger shark detection patterns with whale calf survey data

Normalized monthly averages of humpback whale calf counts plotted against monthly individual tiger shark detections at Olowalu show closely aligned peaks across the 13 months of available data (Fig. 12). Cross-correlation analysis confirmed this synchrony, indicating no time lag between the two patterns (best lag = 0 months, r = 0.749; Fig. 13). Linear regression analysis revealed a significant positive relationship between monthly tiger shark detections and average whale calf counts from shore-based surveys (p = 0.00316). The model explained 56.23% of the variance in shark presence (R² = 0.5623, Adjusted R² = 0.5225), suggesting a moderate association between the two variables. However, a residual standard error of 2.272 indicates that additional unmeasured factors likely influence tiger shark presence during peak winter months.

Understanding the ecological drivers shaping apex predator distributions is essential for predicting species interactions, identifying important habitats, and informing effective marine conservation strategies^4,50,51. Here, we combined long-term acoustic telemetry with demographic analyses to investigate the seasonal site-specific presence of tiger sharks in waters around Maui (Hawaiian Islands). Our findings suggest that the seasonal aggregation of tiger sharks around Maui, particularly at Olowalu, is consistent with a combination of reproductive activities and opportunistic foraging on whale associated biomass. Four main lines of evidence support this interpretation: (1) Seasonal inter-island migrations to Maui that coincide with the known tiger shark mating period (mid-winter) and overlap with humpback whale calving season; (2) Demographic patterns indicating migrating sharks are predominantly sexually mature (80% mature at first Maui detection) which is consistent with mating driven movement; (3) Spatially and temporally overlapping occurrences of mature males and females at Olowalu during mid-winter, together with independent evidence of recent mating (fresh mating scars on females and clasper chafing on males; Figs. 8 and 9), provide circumstantial support for potential mating interactions at this site; and (4) A temporal alignment between shark presence and peak whale calf abundance, implying that foraging opportunities (e.g., scavenging whale placentas or carcasses) may also influence shark aggregations.

Seasonal inter-island movements indicating reproductive migrations

Tiger sharks tagged around O ‘ahu showed distinct seasonal peaks near Maui, especially in January and February (mating season), whereas sharks tagged around Maui displayed year-round residency. Most O ‘ahu-tagged sharks returned to O ‘ahu after visiting Maui. Given the three-year gestation period described by Whitney and Crow¹³, successfully mated females would not be expected to revisit mating grounds for at least three years, allowing time to gestate, give birth, and recover. Thus, detection of adult females at suspected mating grounds in consecutive years would contradict the mating migration hypothesis. We observed only one instance of an adult female returning during mating season in back-to-back years (Fig. S3). All other sharks were detected during only one mating season (Fig. S3). Taken together, these observations indicate directed, seasonal partial migrations by O’ahu-tagged tiger sharks that coincide with Hawai ‘i’s mid-winter mating period¹³. This suggests reproduction as a driver for these seasonal migrations; the overlap of which peaks during whale calving events²⁷.

Partial migration describes the behavior of populations composed of individuals with varying degrees of site attachment due to factors such as reproductive status, local competition, predation risk, and body condition⁵². The contrasting patterns observed between O’ahu-tagged and Maui-resident sharks may reflect differences in the physical and ecological settings of each island. O’ahu and Maui differ in oceanography, bathymetry, coastal habitat structure, and levels of human activity, all of which could influence shark residency and movement. Maui’s extensive insular shelf and proximity to whale calving grounds may favor year-round occupancy, whereas O’ahu’s more developed coastline and narrower shelf may limit suitable habitats for tiger sharks, encouraging seasonal migration. There is evidence that some tiger shark migrations are driven by reproduction. In Hawai ‘i, mature females from the Northwestern Hawaiian Islands migrate to the Main Hawaiian Islands during the fall pupping season, highlighting parturition as a key migratory driver⁵³. Similarly, in the northwest Atlantic, mature male tiger sharks migrate from open-ocean habitats to specific reef areas for the mating season, implicating mating as the primary driver⁵⁴. Comparable reproduction-driven partial migrations occur in other shark species, including lemon sharks⁵⁵, in which only a subset of the population migrates to distinct breeding locations.

The observed seasonal influx of sharks from O ‘ahu combined with the high residency of Maui sharks suggests a spatially structured mating system. Sharks with core habitats located outside Maui Nui appear to specifically migrate to this region for mating opportunities, whereas local sharks remain resident year-round. However, given that Olowalu is also a known hotspot for whale calves and placental resources^56,57, this dual timing and location may not be coincidental. Both mating and foraging may jointly influence tiger shark seasonal movements to Maui. Access to calorie-dense whale blubber during the breeding season could substantially improve female energetic condition, as large sharks readily scavenge whale carcasses and preferentially consume lipid-rich blubber⁵⁸. Fecundity in many sharks scales with female body size (a proxy for energetic condition⁵⁹;), and therefore access to high-calorie prey pulses (e.g. fledgling seabirds at French Frigate Shoals; migratory birds along the Gulf coast; or scavenged whale blubber) can subsidize females’ energy budgets and help meet the substantial costs of viviparous reproduction^5,58,60.

Demographic evidence supports mating hypothesis

Extensive tagging and long-term monitoring, coupled with established tiger shark growth rates in Hawai ‘i³⁷, allowed us to estimate size at first detection on Maui for several O ‘ahu-tagged individuals. We found that 80% of O ‘ahu-tagged sharks detected around Maui were already sexually mature at their initial detection. Notably, several sharks initially tagged as juveniles only appeared in Maui waters after reaching estimated maturity. Similar maturity-dependent migratory behaviors have been documented in tiger sharks from the northwest Atlantic, suggesting that these migrations are reproductive in nature⁵⁴. Such ontogenetic shifts in habitat use and behavior are widespread among elasmobranchs⁶¹ and typically represent transitions from juvenile priorities, such as foraging or predation avoidance, to adult reproductive strategies. However, these patterns do not preclude additional drivers. Not all sharks arriving at Maui were mature, and the presence of some juvenile sharks suggest that factors like foraging could also play a role.

Seasonal aggregation and diel synchrony at Olowalu

Among the six Maui shelf monitoring sites, Olowalu exhibited the strongest and most consistent seasonal pattern of tiger shark detections. Diel activity at Olowalu revealed near-complete overlap (98.3% overlap by kernel density) between male and female presence, possibly indicating synchronized habitat use during daytime that may facilitate mating interactions. This aligns with the current operational definition for aggregation in elasmobranch species, characterized by the co-occurrence of two or more individuals in space and time due to the deliberate use of a common driver⁶². Although males appeared underrepresented in our detections, this may reflect a sampling bias stemming from difficulties capturing mature males in nearshore habitats¹⁴, rather than their actual absence. The fresh mating scars observed on females and chafed claspers on males provide physical evidence of active mating at this site, similar to observations that have supported Fernando de Noronha as a reproductive area for tiger sharks in the South Atlantic⁶³.

Understanding the mating systems of tiger sharks is inherently challenging due to their large size, mobility, and generally solitary nature⁶⁴. Our findings align with theoretical models proposing that even low-density aggregations can substantially enhance reproductive success in wide-ranging solitary species by increasing encounter rates among receptive individuals^65,66. Rather than forming dense clusters, tiger sharks may form diffuse aggregations spanning several kilometers over extended periods. This coordinated yet dispersed aggregation at Olowalu appears to be a functionally significant reproductive strategy that balances mating opportunities against ecological constraints such as the energetic costs of migration, intra-specific competition, and the potential unpredictability of whale-derived food resources^66,67. Such strategies are especially advantageous for sparsely distributed marine predators, reflecting an adaptive compromise between solitary lifestyles and reproductive needs.

Correlations with whale biomass indicate trophic opportunities

Comparing the spatial and temporal patterns of predators and prey enables researchers to assess the likelihood of interactions between species⁶⁸. To evaluate the whale biomass foraging hypothesis, we examined associations between tiger shark presence and proxies of humpback whale biomass—specifically adult humpback whale song chorusing intensity and calf counts. Our analysis revealed a positive relationship between monthly tiger shark abundance and humpback whale acoustic activity, predominantly reflecting adult male singing. However, peak shark abundance occurred one month before peak adult whale acoustic activity. There are multiple potential explanations for this phenomenon, namely (1) adult whales are unlikely to be the primary factor attracting sharks, (2) sharks may position themselves early to secure access to predictable resources, or (3) both sharks and whales may simply be responding to shared environmental cues (e.g. ocean temperature). Further fine-scale monitoring will be required to disentangle anticipatory behavior from coincidental timing. Conversely, tiger shark abundance closely aligned with peak calf numbers, suggesting calves, placentas, or stillborn and vulnerable neonates may provide important, low-risk foraging opportunities. The moderate explanatory power (R² = 0.56) of whale biomass on shark abundance, observed across the 13 months when whale calf data were available within the broader six-year monitoring dataset, is noteworthy for ecological studies⁶⁹. This relationship suggests that whale-derived nutrients may indeed influence shark aggregation, although coincidental timing driven by shared migratory cues again cannot be ruled out. Tiger sharks are known opportunistic scavengers of whale carcasses⁷ and potentially consume placental or neonatal tissues. The predictable seasonal presence of whale-derived resources off Olowalu aligns well with observed shark aggregations. Similar seasonal foraging-driven shark aggregations have been documented elsewhere. One example is at French Frigate Shoals Atoll (Kānemiloha ‘i) in the Northwest Hawaiian Islands where tiger sharks aggregate in early summer specifically to prey upon fledgling albatross chicks, with shark presence closely matching the narrow seasonal and diel timing of fledgling events, primarily between sunrise and noon⁵. Such precise foraging-driven aggregations may be relatively rare within the broader context of tiger shark long-term movement patterns. For example, tiger sharks showed low overall site fidelity and no increase in fidelity over time at an ecotourism provisioning site⁷⁰; individual prey species occurrence patterns did not explain tiger shark residency patterns in the northern Great Barrier Reef⁷¹; tiger shark proximity to a turtle nesting island remained consistent between nesting and non-nesting seasons⁶⁸; and, although tiger sharks preferred shallow habitats where prey was abundant in Australia, prey availability did not explain their broader-scale movements⁷².

Given Olowalu’s role as a hotspot for whale-calf pairs^56,57,73 tiger shark reproductive activities likely coincide with significant foraging opportunities provided by nutrient-rich whale-derived resources. Our findings collectively suggest that tiger shark aggregations around Olowalu are influenced by both reproductive strategies and seasonal availability of whale biomass.

Ecological significance and future studies

Explicit identification of mating aggregations in elasmobranchs remains uncommon, with only a few documented cases, such as nurse sharks (Ginglymostoma cirratum⁷⁴;), Port Jackson sharks (Heterodontus portusjacksoni⁷⁵;), and potentially white sharks (Carcharodon carcharias) at Guadalupe Island⁷⁶. Most studies, however, infer mating aggregations indirectly from observed spatial and temporal patterns^77,78. Beyond identifying mating aggregations, it is also important to consider the ecological attributes that define breeding and nursery areas. Shark breeding and/or nursery sites may reflect a combination of predictable prey availability, benthos type, depth, and predation risk⁶⁶. The year-round residency for Maui-tagged sharks may therefore reflect a convergence of both functions, offering reproductive opportunities and reliable access to food resources that make the region suitable for multiple life-history stages.

The convergence of mature tiger sharks during a defined seasonal window, their pronounced site fidelity to Olowalu—including repeated returns by specific individuals over seven years—and clear diel synchrony between sexes collectively suggest that Olowalu serves as a reproductive site for tiger sharks in Hawai ‘i. Fidelity to mating areas has been documented in only a few shark species⁶⁶; notably, nurse sharks exhibit strong long-term fidelity, returning annually to the Dry Tortugas mating site over periods spanning up to 16 mating seasons and nearly three decades⁷⁹. Similarly, our findings suggest mating-site fidelity among tiger sharks at Olowalu. Future research should adopt multi-modal approaches, such as biologging tags with video capabilities to directly quantify conspecific interactions, verify mating events, and document shark-whale interactions¹⁴. Other promising approaches include cloacal swabbing to detect whale DNA to assess the importance of scavenging events⁸⁰ and energetic modeling to evaluate the relative importance of whale-derived nutrients.

The temporal overlap between tiger shark aggregation and peak whale-calf biomass suggests that Olowalu aggregations may serve dual roles—supporting both reproductive and foraging activities. As generalist and opportunistic predators, tiger sharks can readily exploit these whale-derived resources while concurrently engaging in mating behaviors. Our integrative analysis of movement patterns, demographic structure, diel behavior, and seasonal occurrence provides the first evidence of seasonal mating aggregations for tiger sharks and advances our understanding of their mating behavior in Hawai’i.