Quantifying SMART drumline catches in an Australian bather protection program

Abstract

The global incidence of shark bites on humans has increased over time, driving the development of shark mitigation strategies to reduce the frequency of human-shark interactions. Traditional lethal shark control programs have used fishing gear aimed at reducing local shark populations. There has, however, been a shift towards more ecologically sustainable, non-lethal approaches. One such method is the SMART (Shark Management Alert in Real Time) drumline, which is designed to capture, tag and relocate target sharks. SMART drumlines have been implemented in 20 locations in New South Wales (NSW, Australia) since 2022. Here, we analyse the temporal variability of SMART drumline catches along the NSW coast from May 2022 to May 2025, focusing on target species Carcharodon carcharias (White shark), Galeocerdo cuvier (Tiger shark), and Carcharhinus leucas (Bull shark), and the most abundant bycatch species. Over the three-year study, target sharks showed a combined CPUE of 0.66 sharks per 1000 hook-hours of fishing, while bycatch species recorded a CPUE of 1.23. Captures were generally higher in the northern and central NSW coast, reflecting distinct spatial hotspots for both target and bycatch species. Catches were examined in relation to temporal, spatial and environmental factors. C. carcharias catches were higher in the winter/spring and with relatively colder water temperatures (< 21 °C) over all seasons, while G. cuvier catches were higher during the summer/autumn and relatively warmer water temperatures (> 20 °C) over all seasons and calm seas. Bycatch species also followed seasonal trends, with Carcharhinus obscurus and Sphyrna zygaena preferring warmer water (> 19 °C) and Carcharhinus brachyurus cold (< 21 °C) water. Overall, seasonality and environmental factors were found to influence catch distribution. This information can assist management of SMART drumline effort to target specific shark species, although the data demonstrates that target sharks were captured in NSW nearshore waters across all seasons, consistent with year-round coastal presence.

Highlights

• C. carcharias CPUE was greater in colder water

• G. cuvier captures were higher in warm water and calm sea conditions

• Bycatch captures followed seasonal trends and were influenced by water temperature

1 Introduction

Globally, the number of reported shark bites on people has increased over time (ISAF, 2025). Despite an annual incidence of shark bite deaths of around five people per year between 2010-2019 (ISAF, 2025), these events evoke strong responses from communities with the media often raising awareness disproportionately to the actual low risk of shark bites (Pepin-Neff and Wynter, 2018; Simmons and Mehmet, 2018). The shark species that are typically deemed to be target sharks of mitigation programs are Carcharhinus leucas (Valenciennes, 1839) [Bull shark], Carcharodon carcharias (Linnaeus, 1758) [White shark] and Galeocerdo cuvier (Péron and Lesueur, 1822) [Tiger shark], as these species are responsible for most unprovoked bites (McPhee, 2014; ISAF, 2025).

Shark bites often prompt communities to request government intervention, leading to the adoption of mitigation strategies to detect, deter or remove target sharks (Gibbs and Warren, 2015; Neff, 2015). Fishing-based methods are commonly used to catch and euthanise, or to catch, relocate and release target sharks. These approaches include mesh nets, traditional drumlines and longlines (Dudley, 1997; Hazin and Afonso, 2014). However, in recent times, there has been a move towards non-lethal shark management. Since 2014, the MLI-s “Catch-A-Live”^® buoy (Marine Instruments ^©) has been used in conjunction with traditional drumlines, providing a solar-powered GPS beacon that is triggered with the dislocation of a magnetic pin linked to a baited fishing trace (Guyomard et al., 2019). This system, named ‘Shark Management Alert in Real Time’ (SMART) was first used in La Réunion (Guyomard et al., 2019). SMART drumlines were further developed as part of the shark-bite mitigation program off New South Wales (NSW, Australia) (Tate et al., 2019; 2021; 2021; Butcher et al., 2023, 2025). As well as removing target sharks from the local vicinity (Butcher et al., 2023), SMART drumlines provide an opportunity to tag, relocate and track target sharks (Spaet et al., 2020). After seven years of trials, in May 2022 SMART drumlines were implemented as an operational program at 20 locations along the NSW coastline (Tate et al., 2021a; SharkSmart, 2025; Cappa et al., unpublish¹).

Despite the adoption of SMART drumlines as a non-lethal shark mitigation tool, there are only a few studies quantifying the main drivers of the captures along the NSW coastline (Tate et al., 2021a; 2021). Understanding the seasonal occurrence of catches in the NSW SMART drumline program is essential to assess its effectiveness and any potential future refinement. Furthermore, environmental factors such as sea temperature, current speeds, sea swell and rainfall are known to influence catch composition (Niella et al., 2017), yet these relationships remain poorly understood in the context of large temporal and spatial scale deployment along the NSW coast. By assessing spatial, temporal and key environmental variables this study identifies how these factors influence SMART drumline catches, providing insight into both the operational performance of the program and the ecological factors that drive catches. Specifically, the present research aimed to quantify the spatial and temporal trend in SMART drumline catches in NSW from 2022 to 2025 and to assess the environmental characteristics that influenced SMART drumline catches of target sharks (C. leucas, C. carcharias and G. cuvier) and the most abundant bycatch species.

2 Materials and methods

This study used SMART drumline data from the NSW Shark Management Program (SharkSmart, 2025) where they were deployed from 15 May 2022 to 15 May 2025 at 20 unchanged locations across ~1200 km of coastline (Figure 1). In each location, 15 (20 at Ballina) SMART drumlines (Figure 2) were deployed daily approximately 1 hour after sunrise (mean ± SD of 6:46 ± 0:30 h) and retrieved 2 hours before sunset (15:55 ± 1:39 h). Gear wasn’t deployed daily where poor weather, swell or restricted river access prevented safe navigation or the ability to handle sharks. SMART drumlines were baited with a whole ~ 0.75–1 kg Mugil cephalus (Linnaeus, 1758) [Sea mullet], except at Kingscliff where, to avoid dolphin bait depredation (Sumpton et al., 2010; Lawson et al., 2025), hooks were baited with the same weight and size shark flesh (Carcharhinus spp.). For each marine animal captured on a SMART drumline, the following information was recorded: date, time hooked, location (with related latitude and longitude) and species. Captured animals were identified at the species level whenever possible. If identification at this level was not possible, the highest taxonomic resolution possible, such as genus or family, was recorded.

Figure 1

Figure 2

Butcher et al., 2025).

2.1 Catch per unit effort (CPUE)

The Catch Per Unit Effort (CPUE) metric was determined by dividing the number of individuals caught by the total number of effort hours the SMART drumlines were used, providing a standardised measure of fishing success. To standardise the unit measure, the sampling unit used for CPUE calculations was a daily aggregate per species and fishing location. The soak time was defined as the period from deployment to the time the boat arrived to release the animal, or, in case of no capture, the time between deployment and the retrieval of the SMART drumline. The overall daily fishing effort was then calculated as the sum of each single SMART drumline soak time per each location. The daily CPUE metric was then calculated as:

2.2 Environmental variables

Environmental characteristics used as potential explanatory variables influencing SMART drumline catches were obtained from multiple sources. Current speed data were extracted from the Australian Ocean Data Network (AODN, 2025) from wave buoys deployed at fixed locations along the study region. To ensure maximum data accuracy, the closest wave buoy to each catch location within a radius of 20 km was selected. Daily rainfall levels were obtained from the Australian Government Bureau of Meteorology (BOM), using data from the closest weather stations to each SMART drumline fishing location. Due to the dominance of zero rainfall days in the data (64% of cases), rainfall was categorised arbitrarily as a three-level factor with classes ‘0 mm’, ‘1-20 mm’ and ‘>20 mm’ intended to represent ‘dry’, ‘showery’ and ‘rainy/stormy’ weather, respectively. Sea surface temperatures (SST) were recorded on-site at the time of the capture or gear retrieval. Wave (daily swell mean) height data were obtained from the NSW Government Manly Hydraulics Laboratory (MHL) using data from the closest wave height recorder to each fishing location.

2.3 Statistical analysis

Daily catch records from the SMART drumline program were collated across 20 locations from 15 May 2022 to 15 May 2025. Daily observations at each location included the number of animals caught (classified into 31 species), total fishing effort (calculated as total hours fished per hook), and a suite of environmental covariates, including sea surface temperature, current speed, mean swell, and rainfall.

The number of animals captured per day was treated as the observable outcome from an underlying rate of capture that varied according to these factors. From this stochastic framework, and with the R package ‘mgcv’ (Wood, 2011) a series of generalised additive models (GAMs) with increasing complexity were fitted to quantify potential changes in catch rate associated with each candidate factor. Specifically, the following model components were considered: location effect, long-term trend (three-year period) at each location, seasonal (cyclic) trend at each location, smooth responses to the environmental covariates in sequence: water temperature, current speed, mean swell and rainfall. The models assumed a Poisson response distribution with log transformed hours of effort included as an offset covariate. The full GAM evaluated for each species was formulated as:

where i is the observation index, Y_i is the number of captures at observation i, effort_i is the fishing effort (hook hours) at observation i, l_i is the fishing location associated with observation i, α₁ is the effect of fishing location l, day_i is the Julian day of observation i (1 = 2022-05-15), yearday_i is the day of year for observation i (1 = January 1, 365/366 = December 31), is the value of covariate p at observation i (p = water temperature, current speed, mean swell, rainfall), is the smooth function for location l, is the cyclic smooth function for location l enforcing periodicity by continuity at the boundaries, is the smooth function for covariate p (Wood, 2017).

An analysis of variance was conducted at each step and terms associated with a significant change in likelihood retained at criterion p< 0.05. Also arbitrarily, seasonal models were only fitted for locations with >10 captures in total of the species of interest. This analysis structure was applied to each of the three target species (C. leucas, C. carcharias and G. cuvier) and for the three most captured bycatch species (Carcharhinus brachyurus, Carcharhinus obscurus and Sphyrna zygaena).

To describe patterns in the capture time of day, a smoothed approximation to the empirical density function of time hooked for the three target species and the three most abundant bycatch species was calculated by kernel density estimation as implemented by the R function ‘density’. Since day length over the study area and period ranged from 9.7 hours to 14.7 hours, the timing of events was better expressed relative to photoperiod (Vazquez et al., 2019). The time of sunrise and sunset on each capture date at each location was calculated with the R ‘suntools’ package (Bivand and Luque, 2023). Daylight standardised hooked time was then calculated as:

So that the time scale used in this analysis has sunrise at time = 0 and sunset at time = 1.

The statistical analyses were performed in R statistical software version 2024.02.29 (R Core Team, 2024). All tests were considered significant if the p-value was< 0.05.

3 Results

3.1 Overall effort and catch composition

SMART drumlines were deployed for a total of 2,257,345 hours incorporating a mean ± SD of 112,867.27 ± 10,728.60 fishing hours per location (Figure 1). During the study period, 1478 target sharks were caught (35% of total captures), consisting of 985 C. carcharias, 413 G. cuvier and 80 C. leucas. The remaining 65% of total captures consisted of bycatch species, with the most abundant bycatch species being C. obscurus (n = 1212), C. brachyurus (n = 677) and S. zygaena (n = 218). Target species showed a combined averaged CPUE of 0.66 sharks per 1000 hook-hours of fishing, mostly during the austral winter. The average CPUE of bycatch species was 1.23 animals per 1000 hook-hours of fishing distributed evenly across the year. Table 1 shows the total catch and CPUE for all species.

3.2 Overall location effects

There were clear latitudinal trends in capture rates for C. leucas and C. carcharias, accompanied by some off trend effects associated with specific locations (Figure 3a). For example, C. leucas was predominantly caught in the northern regions, with notable absences along the central and southern coast. C. carcharias also showed higher capture rates in the north, and Port Macquarie represented a hotspot compared to neighbouring locations. G. cuvier instead, showed relatively even captures along the NSW coastline, yet had peak captures in Coffs Harbour, with Port Macquarie and Bondi as secondary hotspots.

Figure 3

Among bycatch species, C. obscurus had higher capture rates along the central coast of NSW, with Lake Macquarie and Central Coast North identified as hotspots. C. brachyurus was most frequently caught off Batemans Bay, and relatively high captures at Coffs Harbour, Port Macquarie and Central Coast North. S. zygaena was also most frequently caught off Batemans Bay but generally showed higher catch rates along the northern and mid-north coast (Figure 3b).

3.3 Seasonal and environmental effect

3.3.1 Target sharks

The seasonal effect for C. leucas was minimal (Figure 4a). Only Ballina and Yamba recorded sufficient catches for seasonal analysis, with most individuals caught between January and June. Sea surface temperature was the only significant predictor of C. leucas CPUE (p< 0.001), with higher catch rates observed with temperature >21 °C (Figure 4b).

Figure 4

C. carcharias CPUE peaked in August and September at northern locations, shifting toward more winter-dominated captures in the south (Figure 5a). Sea surface temperature was the only significant environmental predictor of C. carcharias CPUE (p< 0.001; Figure 5b), with catch rates highest between 15 and 21 °C and declining above 21 °C when other factors were equally influential. Mean swell, current speed and rainfall were not significant predictors of C. carcharias captures.

Figure 5

G. cuvier catches were concentrated in summer and autumn at southern locations, shifting towards more winter-dominated catches in the north (Figure 6a). Water temperature and mean swell were significant predictors (p< 0.001), with CPUE increasing at temperatures above 20 °C (Figure 6b) and when mean swell was below 1.3 m (Figure 6c). Current speed also had a significant albeit lower effect (p< 0.01), with most catches occurring at lower speed (< 0.2 m s^-1) (Figure 6d). Rainfall was not a significant predictor of G. cuvier catches.

Figure 6

3.3.2 Bycatch

C. obscurus catches showed clear seasonal variation along the New South Wales coast, with catches concentrated in spring in the few northern locations in which they were caught, and shifting towards summer and autumn dominance but also caught in winter south of Newcastle (Figure 7a). Water temperature (p< 0.001) was a significant environmental predictor of catches, with captures occurring mainly between 19 and 26 °C (Figure 7b). Mean swell also tested significant (p< 0.001), with higher catch rates at swell heights above 1.25 m (Figure 7c). Current speed and rainfall were not significant predictors for C. obscurus catch rates.

Figure 7

Catches of C. brachyurus peaked between September and November in the few northern locations in which they were caught, while from Newcastle southwards they were highest between June and October, except at Merimbula, where CPUE increased during summer (Figure 8a). Water temperature and mean sea swell were significant predictors (p< 0.001), with higher catch rates between 14 and 21 °C (Figure 8b) and at swell heights > 2 m (Figure 8c). Catch rates declined significantly (p< 0.05) when rainfall exceeded 20 mm (Figure 8d). Current speed was not a significant predictor of C. brachyurus catches.

Figure 8

S. zygaena catches were highest in northern locations from September to December, shifting towards summer-dominated catches along the central and southern coast (Figure 9a). Water temperature was a significant predictor (p< 0.001), with higher catch rates observed between 19 and 26 °C (Figure 9b). Rainfall was significant (p< 0.05), with catches increasing with greater precipitation (Figure 9c). Current speed and mean swell height were not significant predictors of S. zygaena catches.

Figure 9

3.4 Time hooked

Time hooked was missing for 356 captures across all species due to the magnet not being triggered. The number of captures for each species of interest with time hooked data available is given in Table 2. Overall, catches for all species were more likely to occur in the morning, decreasing through the latter half of the day. C. carcharias and C. brachyurus also exhibited a weak, bimodal distribution, with local maxima of capture probability density in the morning (t = 0.3) and a weaker one in the early afternoon (t = 0.6) (Figures 10b, e). C. leucas, G. cuvier and C. obscurus showed strong morning peaks between t = 0.2 and 0.4, with few captures later in the day (Figures 10a, c, d). S. zygaena exhibited a dominant and almost exclusively morning peak (t = 0.2) followed by a rapid decline of capture probability density toward sunset (Figure 10f). Kernel density estimate was extrapolated to time points outside the typical line setting and retrieval times due to the data and the underlying Gaussian kernel. These results suggest that C. leucas, C. obscurus, G. cuvier, and S. zygaena are highly active foragers at night and dawn, and there is likely a high probability of these species interacting with gear at night and through the early morning. However, daytime-only sampling in this study precludes direct quantification of nocturnal behaviour, and it may be difficult to distinguish specific patterns of daytime fishing with SMART drumlines from shark fishing in general. Moreover, it should be noted that approximately 3% of fishing days during this study were terminated by midday, meaning the lack of afternoon captures may be partially influenced by reduced fishing effort.

Figure 10

4 Discussion

The extensive use of SMART drumlines for over two million hours of fishing effort during a three-year period provides critical insight into the spatial, temporal and environmental drivers of shark distribution along the NSW coast. While there were some operational constraints that partially influenced SMART drumline fishing effort, such as adverse weather conditions and seasonal daylight variation, the overall consistent spatial coverage of SMART drumlines provides a solid basis for evaluating species-specific catch performance and patterns. Identifying how environmental variables such as sea surface temperature, current speed, mean swell and rainfall drive catch rates goes beyond baseline catch summaries as it is crucial for predicting shark preferences and refining SMART drumline deployment. Overall, this study highlights the operational reliability of SMART drumlines while demonstrating the complex environmental variables that should be integrated into long-term, non-lethal shark mitigation programs.

4.1 Catch composition

Overall, target shark species accounted for approximately one third of all captures, with C. carcharias being the most frequently caught target species. The catch composition further confirms that SMART drumlines can be an effective tool at catching target sharks while maintaining low bycatch rates especially when compared with other shark-mitigation fishing gears such as conventional bottom-set longlines (Cappa et al., unpublish¹). Among the target sharks, C. carcharias was caught consistently within its period of known seasonal migrations and peak of relative abundance along the New South Wales coast (Bruce et al., 2019). As expected (Butcher et al., 2025), G. cuvier was captured along the entire NSW coastline, although it was more abundant in the central and northern part of the coast. The relatively low CPUE of C. leucas (0.035) indicates that this species may be less abundant in the monitored areas, given that NSW represents the southern distributional limit of the species (Smoothey et al., 2023), and/or less responsive to SMART drumline baits set during daylight hours when they are less likely to forage (Guyomard et al., 2020). Detections of tagged C. leucas on all listening stations (DPIRD, unpublished results) adjacent to the SMART drumlines of this study suggest it is more likely attributable to the latter. The composition of bycatch sharks also reflects the regional composition of sharks in the temperate zone of eastern Australia, dominated by C. obscurus and C. brachyurus (Butcher et al., 2025). The limited number of teleosts, batoids and turtles confirms the high selectivity of SMART lines compared to more traditional shark mitigation gears (Tate et al., 2021a; Cappa et al., unpublish¹).

4.2 Spatial patterns in captures

Catches of the three target shark species were generally higher in the mid north and northern New South Wales coast, with the highest catch rates of C. carcharias from Ballina to Forster. This spatial pattern of capture was found to be driven by multiple factors and reflects ecological and oceanographic gradients along the coast. These results indicate that northern NSW waters provide suitable foraging conditions for all three target species at certain times of the year, with thermal conditions favourable for C. leucas and G. cuvier, which prefer warmer waters (Payne et al., 2018; Smoothey et al., 2019; Niella et al., 2022a), and are known to be sympatric (Niella et al., 2021). Catches of C. leucas were almost entirely north of Forster and with sporadic catches occurring further south, limited to three locations. Despite previous reports of C. leucas regularly being caught on night-set setlines and detected in Sydney Harbour over summer and autumn (Smoothey et al., 2019), very few individuals were caught on SMART drumlines along the entire NSW coast in this study. These relatively low catches are almost certainly due to the species foraging and being generally more active at night (Niella et al., 2022b), i.e. when SMART drumlines are not in the water. C. carcharias catches were also consistently higher from Forster north, with a high CPUE at Port Macquarie, consistent with, and adding spatial resolution to, known juvenile aggregation sites previously documented by acoustic tracking (Bruce et al., 2019; Spaet et al., 2020). While G. cuvier catch rates peaked at Coffs Harbour, Port Macquarie and Bondi, captures were consistent along the entire NSW coastline. Although previous satellite tagging studies have shown that G. cuvier mainly utilises the central and northern length of the New South Wales coast (Holmes et al., 2014; Lipscombe et al., 2020), the similar daytime catch rates between the far north and the far south highlights a broader daytime foraging area, covering the entire NSW coastline, than previously recognised.

Bycatch also showed clear spatial patterns. C. obscurus is widely but discontinuously distributed along the eastern Australian coast (Last and Stevens, 2009; Barnes et al., 2016) and recorded higher catch rates along the central coast. These areas are characterised by large estuaries that provide nutrient-rich waters, which are likely to increase prey availability and aggregation opportunities (Dyer et al., unpublish²). C. brachyurus recorded peak catches in Central Coast North and the Eurobodalla Shire, reflecting a preference for temperate waters (Drew et al., 2016) and a possible alignment with prey movements or habitat selection. The captures of S. zygaena presented a more complex pattern. While traditionally considered a temperate species with expected high abundances in cooler southern waters (Compagno, 1984), it was most frequently caught at Batemans Bay in the south and across northern locations. The spatial distribution of captures of S. zygaena suggests that rather than following latitudinal thermal gradients, other factors, potentially localised upwelling zones or specific bathymetric features, may drive local aggregations through combination of food and predator protection. These patterns warrant further investigation.

Overall, our findings demonstrate that SMART drumline shark catches, both target and bycatch, along the coast of New South Wales exhibit distinct spatial hotspots, determined by the species-specific thermal preferences, diurnal foraging behaviours and key coastal geomorphological features, such as estuary mouths and deep-water upwelling zones.

4.3 Seasonal and environmental factors

Environmental conditions and physical characteristics of coastal areas play a key role in the distribution and capture of many marine species (Niella et al., 2017). We confirmed that such factors, combined with seasonality, increased the CPUE of animals caught on SMART drumlines. Among the variables considered, sea surface temperature, current speed, swell height and rainfall were key predictors for catch rates of target and bycatch species, highlighting the complex interactions among physiology, behaviour and environmental constraints.

G. cuvier and C. leucas were more frequently caught in warmer waters, consistent with their known affinity for higher temperatures and coastal habitats during summer and early autumn (Smoothey et al., 2019; Niella et al., 2020). The southward expansion of G. cuvier during the warmer months likely corresponds to migratory behaviour in search of prey (Holmes et al., 2014; Payne et al., 2018), while C. leucas tends to remain near estuaries or rivers, where temperatures are moderate and prey availability is high (Smoothey et al., 2016; Werry et al., 2018; Smoothey et al., 2019). Seasonal availability of prey likely reinforces these trends. Many prey species such as teleosts and elasmobranchs exhibit temperature-driven migrations or coastal aggregations during warmer months, providing concentrated foraging opportunities that attract predatory sharks (Rogers et al., 2012). The peaks observed in the CPUE of C. leucas and G. cuvier during summer and early autumn may not only reflect thermal preferences, but also the spatial distribution of prey resources. C. carcharias showed species-specific seasonal and thermal preferences, being caught predominantly in colder waters, with peak catches occurring in late winter and spring. However, the spatial concentrations of C. carcharias catches in northern locations indicates that their distribution is not only dictated by thermal preferences alone (e.g., Bruce and Bradford, 2012), but they reflect seasonal prey resources (Grainger et al., 2020). The opportunistic foraging strategy of C. carcharias (Colefax et al., 2020; Tate et al., 2021b) allows it to exploit active prey and scavenging opportunities throughout the year, supporting its presence in south-eastern Australian waters during the colder months when other sharks may move northwards (Spaet et al., 2020).

The remaining physical oceanographic factors also influenced catches of target sharks. Increases in wave height and current velocity were associated with reduced catch rates of G. cuvier, suggesting that high-energy conditions may reduce foraging efficiency by limiting G. cuvier’s ability to locate and capture prey and increasing the energy expenditure required for foraging in turbulent waters (Niella et al., 2021). Such limitations highlight that catch rates not only reflect abundance, but also the catchability of sharks under variable environmental conditions, a critical consideration for the design of effective shark management programs.

Bycatch also showed clear environmental preferences. C. obscurus and S. zygaena were more abundant in moderately warm waters, while C. brachyurus preferred colder conditions, reflecting both specific thermal niches (Compagno, 1984; Lucifora et al., 2005) and known seasonal migrations (Rogers et al., 2013; Drew et al., 2016). C. obscurus and C. brachyurus also showed higher catch rates in poor sea conditions, indicating a tolerance for more dynamic waters and the capacity to exploit baits when other sharks were less likely be caught (e.g., G. cuvier). The effects of precipitation were species-specific, with catches of C. brachyurus decreasing during periods of high precipitation, while those of S. zygaena increased, potentially reflecting behavioural adaptations such as vertical or coastal movements that offer camouflage from predators or access to prey in cloudy or rainy conditions. Prey dynamics may also contribute to these patterns, as bycatch species follow prey that respond differently to temperature and hydrodynamic changes (Dudley and Cliff, 2010).

Sea surface temperature was a key factor of shark distribution (as determined by catches on SMART drumlines) in New South Wales coastal waters, while other environmental variables, such as swell, currents and rainfall, influence species-specific behaviour, foraging and hence, catchability from SMART drumlines. With ongoing climate change and rising ocean temperatures, a shift in distribution of marine species, including sharks, is expected (Cheung et al., 2010; Tanaka et al., 2021). G. cuvier and C. leucas may expand further south and for longer duration of the year (Payne et al., 2018; Niella et al., 2021) while the abundance of C. carcharias may decline in the north (Gooden et al., 2025). Nonetheless, it is likely that at least one of the target species will be present throughout the year in New South Wales waters, highlighting the need for year-round shark mitigation measures.

4.4 Time of the day as a predictor of catch probability

The observed suggests that target sharks and bycatch species exhibit similar diurnal activity, with the highest catch probabilities recorded in the morning and reduced activity in the afternoon. Such temporal variation may correspond to natural foraging behaviour, as many shark species adapt their hunting activity to coincide with periods when prey is most available or most detectable under particular light conditions (Hammerschlag et al., 2016; Niella et al., 2022b). However, our results showed a weak bimodal afternoon peak in C. carcharias and C. brachyurus which may reflect two distinct phases of foraging activity related to prey movements, while the other species have shown capture probabilities concentrated mainly before mid-photoperiod. Integrating these temporal patterns with environmental factors such as sea surface temperature, swell, and current speed can improve predictions of shark presence and behaviour. However, it is important to note that SMART drumlines in New South Wales are used exclusively during daylight hours to align with peak periods of water user activity. As a result, captures are limited to daytime periods, and potential nocturnal activity patterns remain unrecorded. Comparing between these daytime fishing periods and traditional bather-protection gillnets, such as those used during the north coast net trials (Broadhurst and Cullis, 2020), highlights this operational difference; as nets operated continuously during night time, they captured a broader range of species (especially batoids), in low-light conditions and at night. Similarly, SMART drumlines were not deployed during adverse weather conditions creating a further unknown and baits could be depredated during daylight hours so reducing fishing capacity (Lawson et al., 2025).

5 Conclusion

The SMART drumline program implemented along the east coast of NSW, Australia demonstrated that catch per unit effort (CPUE) for both target and bycatch species is strongly influenced by spatial, temporal, and environmental variability. Catch patterns differed among species, with sea surface temperature identified as the primary predictor of CPUE across both target sharks and bycatch assemblages. Additional environmental variables, including swell, current speed, and rainfall, further influenced species-specific catchability. Clear seasonal and thermal preferences were evident among target species, with Carcharodon carcharias more frequently captured in cooler waters during late winter and spring on the mid north and north coast, while Galeocerdo cuvier and Carcharhinus leucas were more commonly encountered in warmer waters and calmer conditions during summer and early autumn.

The current year-round deployment strategy of SMART drumlines in NSW appears to be effective in capturing target sharks and is broadly aligned with the seasonal and thermal preferences of target species. This approach supports the operational objectives of capturing, relocating, and releasing individuals while minimising risk to both sharks and water users. Continued refinement of deployment strategies should incorporate environmental drivers and species-specific movement patterns to further optimise program performance. Such improvements have the potential to enhance selectivity, reduce bycatch, and strengthen the long-term sustainability and precision of shark mitigation initiatives in NSW waters.

References

• 1

  AODN (2025). Australian Ocean Data Network. Available online at: https://portal.aodn.org.au/ (Accessed May 16, 2025).

  • Google Scholar
  • View reference in article

• 2

  BarnesC.ButcherP.MacbethW.MandelmanJ.SmithS.PeddemorsV. (2016). Movements and mortality of two commercially exploited carcharhinid sharks following longline capture and release off eastern Australia. Endang. Species. Res.30, 193–208. doi: 10.3354/esr00730

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  BivandR.LuqueS. (2023). Suntools: Calculate sun position, sunrise, sunset, solar noon and twilight. doi: 10.32614/CRAN.package.suntools

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  BruceB. D.BradfordR. W. (2012). “ Habitat use and spatial dynamics of juvenile white sharks, Carcharodon carcharias, in eastern Australia,” in Global Perspectives on the Biology and Life History of the White Shark, 225–254.

  • Google Scholar
  • View reference in article

• 5

  BruceB.HarastiD.LeeK.GallenC.BradfordR. (2019). Broad-scale movements of juvenile white sharks Carcharodon carcharias in eastern Australia from acoustic and satellite telemetry. Mar. Ecol. Prog. Ser.619, 1–15. doi: 10.3354/meps12969

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  BroadhurstM. K.CullisB. R. (2020). Mitigating the discard mortality of non-target, threatened elasmobranchs in bather-protection gillnets. Fisheries Research. 222, 105435. doi: 10.1016/j.fishres.2019.105435

  • CrossRef
  • Google Scholar

• 7

  ButcherP. A.BroadhurstM. K.PeddemorsV. M.McGartyA. J.CullisB. R. (2025). Hooking efficiencies of SMART drumlines and their possible deployment rates vs gillnets for bather protection. Front. Mar. Sci.11, 1513232. doi: 10.3389/fmars.2024.1513232

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  ButcherP. A.LeeK. A.BrandC. P.GallenC. R.GreenM.SmootheyA. F.et al. (2023). Capture response and long-term fate of white sharks (Carcharodon carcharias) after release from SMART drumlines. Biology12, 1329. doi: 10.3390/biology12101329

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  CheungW. W. L.LamV. W. Y.SarmientoJ. L.KearneyK.WatsonR.ZellerD.et al. (2010). Large‐scale redistribution of maximum fisheries catch potential in the global ocean under climate change. Global Change Biol.16, 24–35. doi: 10.1111/j.1365-2486.2009.01995.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  ColefaxA. P.KelaherB. P.PagendamD. E.ButcherP. A. (2020). Assessing white shark (Carcharodon carcharias) behavior along coastal beaches for conservation-focused shark mitigation. Front. Mar. Sci.7, 268. doi: 10.3389/fmars.2020.00268

  • CrossRef
  • Google Scholar

• 11

  CompagnoL. J. (1984). FAO Species Catalogue. Vol. 4. Sharks of the World. An Annotated and Illustrated Catalogue of Shark Species Known to Date. Part 1. Hexanchiformes to Lamniformes ( FAO). Available online at: https://openknowledge.fao.org/items/5c6e0631-ebe6-4c02-a186-443ffc29073e.

  • Google Scholar

• 12

  DrewM.RogersP.HuveneersC. (2016). Slow life-history traits of a neritic predator, the bronze whaler (Carcharhinus brachyurus). Mar. Freshw. Res.68, 461–472. doi: 10.1071/MF15399

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13

  DudleyS. F. J. (1997). A comparison of the shark control programs of New South Wales and Queensland (Australia) and KwaZulu-Natal (South Africa). Ocean Coast. Manage.34, 1–27. doi: 10.1016/S0964-5691(96)00061-0

  • CrossRef
  • Google Scholar

• 14

  DudleyS. F.CliffG. (2010). Influence of the annual sardine run on catches of large sharks in the protective gillnets off KwaZulu-Natal, South Africa, and the occurrence of sardine in shark diet. Afr. J. Mar. Sci.32, 383–397. doi: 10.2989/1814232X.2010.502641

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  GibbsL.WarrenA. (2015). Transforming shark hazard policy: Learning from ocean-users and shark encounter in Western Australia. Mar. Policy58, 116–124. doi: 10.1016/j.marpol.2015.04.014

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  GoodenA.KelaherB.NiellaY.ButcherP. A. (2025). Environmental factors drive differences in activity between sexes of a large marine predator. Sci. Total Environ.980, 179436. doi: 10.1016/j.scitotenv.2025.179436

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  GraingerR.PeddemorsV. M.RaubenheimerD.Machovsky-CapuskaG. E. (2020). Diet composition and nutritional niche breadth variability in juvenile white sharks (Carcharodon carcharias). Front. Mar. Sci.7, 422. doi: 10.3389/fmars.2020.00422

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  GuyomardD.LeeK. A.PerryC.JaquemetS.CliffG. (2020). SMART drumlines at Réunion Island do not attract bull sharks Carcharhinus leucas into nearshore waters: Evidence from acoustic monitoring. Fish. Res.225, 105480. doi: 10.1016/j.fishres.2019.105480

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  GuyomardD.PerryC.TournouxP. U.CliffG.PeddemorsV.JaquemetS. (2019). An innovative fishing gear to enhance the release of non-target species in coastal shark-control programs: The SMART (shark management alert in real-time) drumline. Fish. Res.216, 6–17. doi: 10.1016/j.fishres.2019.03.011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  HammerschlagN.SkubelR. A.CalichH.NelsonE. R.ShiffmanD. S.WesterJ.et al. (2016). Nocturnal and crepuscular behavior in elasmobranchs: a review of movement, habitat use, foraging, and reproduction in the dark. BMS93, 355–374. doi: 10.5343/bms.2016.1046

  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  HazinF. H. V.AfonsoA. S. (2014). A green strategy for shark attack mitigation off Recife, Brazil: Shark attack mitigation off Recife. Anim. Conserv.17, 287–296. doi: 10.1111/acv.12096

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  HolmesB. J.PepperellJ. G.GriffithsS. P.JaineF. R. A.TibbettsI. R.BennettM. B. (2014). Tiger shark (Galeocerdo cuvier) movement patterns and habitat use determined by satellite tagging in eastern Australian waters. Mar. Biol.161, 2645–2658. doi: 10.1007/s00227-014-2536-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  ISAF (2025). International Shark Attack File. Available online at: https://www.floridamuseum.ufl.edu/shark-attacks/ (Accessed March 6, 2025).

  • Google Scholar
  • View reference in article

• 24

  LastP. R.StevensJ. D. (2009). Sharks and Rays of Australia. 2nd Edn (Collingwood: CSIRO Publishing).

  • Google Scholar
  • View reference in article

• 25

  LawsonC. L.MorrisS. G.GreenM.ButcherP. A. (2025). Minimising dolphin-induced bait depredation in an Australian bather protection program. Wildl. Res.52, WR25099. doi: 10.1071/WR25099

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  LipscombeR. S.SpaetJ. L. Y.ScottA.LamC. H.BrandC. P.ButcherP. A. (2020). Habitat use and movement patterns of tiger sharks (Galeocerdo cuvier) in eastern Australian waters. ICES J. Mar. Sci.77, 3127–3137. doi: 10.1093/icesjms/fsaa212

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  LuciforaL. O.MenniR. C.EscalanteA. H. (2005). Reproduction and seasonal occurrence of the copper shark, Carcharhinus brachyurus, from north Patagonia, Argentina. ICES J. Mar. Sci.62, 107–115. doi: 10.1016/j.icesjms.2004.09.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  McPheeD. (2014). Unprovoked shark bites: Are they becoming more prevalent? Coast. Manage.42, 478–492. doi: 10.1080/08920753.2014.942046

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  NeffC. (2015). The Jaws Effect: How movie narratives are used to influence policy responses to shark bites in Western Australia. Aust. J. Political Sci.50, 114–127. doi: 10.1080/10361146.2014.989385

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  NiellaY.ButcherP.HolmesB.BarnettA.HarcourtR. (2022a). Forecasting intraspecific changes in distribution of a wide-ranging marine predator under climate change. Oecologia198, 111–124. doi: 10.1007/s00442-021-05075-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  NiellaY. V.HazinF. H. V.AfonsoA. S. (2017). Detecting multispecific patterns in the catch composition of a fisheries-independent longline survey. Mar. Coast. Fish.9, 388–395. doi: 10.1080/19425120.2017.1347115

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  NiellaY.SmootheyA.PeddemorsV.HarcourtR. (2020). Predicting changes in distribution of a large coastal shark in the face of the strengthening East Australian Current. Mar. Ecol. Prog. Ser.642, 163–177. doi: 10.3354/meps13322

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  NiellaY.SmootheyA. F.TaylorM. D.PeddemorsV. M.HarcourtR. (2022b). Environmental drivers of fine-scale predator and prey spatial dynamics in Sydney Harbour, Australia, and adjacent coastal waters. Estuaries Coasts45, 1465–1479. doi: 10.1007/s12237-021-01020-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  NiellaY.WiefelsA.AlmeidaU.JaquemetS.LagabrielleE.HarcourtR.et al. (2021). Dynamics of marine predators off an oceanic island and implications for management of a preventative shark fishing program. Mar. Biol.168, 42. doi: 10.1007/s00227-021-03852-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  PayneN. L.MeyerC. G.SmithJ. A.HoughtonJ. D. R.BarnettA.HolmesB. J.et al. (2018). Combining abundance and performance data reveals how temperature regulates coastal occurrences and activity of a roaming apex predator. Global Change Biol.24, 1884–1893. doi: 10.1111/gcb.14088

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 36

  Pepin-NeffC.WynterT. (2018). Shark bites and shark conservation: An analysis of human attitudes following shark bite incidents in two locations in Australia. Conserv. Lett.11, e12407. doi: 10.1111/conl.12407

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  R Core Team (2024). R: A Language and Environment for Statistical Computing (Vienna, Austria: R Foundation for Statistical Computing). Available online at: https://www.R-project.org/.

  • Google Scholar

• 38

  RogersP. J.HuveneersC.GoldsworthyS. D.MitchellJ. G.SeurontL. (2013). Broad‐scale movements and pelagic habitat of the dusky shark Carcharhinus obscurus off southern Australia determined using pop‐up satellite archival tags. Fish. Oceanogr.22, 102–112. doi: 10.1111/fog.12009

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  RogersP. J.HuveneersC.PageB.HamerD. J.GoldsworthyS. D.MitchellJ. G.et al. (2012). A quantitative comparison of the diets of sympatric pelagic sharks in gulf and shelf ecosystems off southern Australia. ICES J. Mar. Sci.69, 1382–1393. doi: 10.1093/icesjms/fss100

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40

  SharkSmart (2025). Available online at: https://www.sharksmart.nsw.gov.au (Accessed March 6, 2025).

  • Google Scholar

• 41

  SimmonsP.MehmetM. I. (2018). Shark management strategy policy considerations: Community preferences, reasoning and speculations. Mar. Policy96, 111–119. doi: 10.1016/j.marpol.2018.08.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  SmootheyA. F.GrayC. A.KennellyS. J.MasensO. J.PeddemorsV. M.RobinsonW. A. (2016). Patterns of occurrence of sharks in Sydney Harbour, a large urbanised estuary. PloS One11, e0146911. doi: 10.1371/journal.pone.0146911

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  SmootheyA. F.LeeK. A.PeddemorsV. M. (2019). Long-term patterns of abundance, residency and movements of bull sharks (Carcharhinus leucas) in Sydney Harbour, Australia. Sci. Rep.9, 18864. doi: 10.1038/s41598-019-54365-x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  SmootheyA. F.NiellaY.BrandC.PeddemorsV. M.ButcherP. A. (2023). Bull shark (Carcharhinus leucas) occurrence along beaches of south-eastern Australia: Understanding where, when and why. Biology12, 1189. doi: 10.3390/biology12091189

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  SpaetJ. L. Y.PattersonT. A.BradfordR. W.ButcherP. A. (2020). Spatiotemporal distribution patterns of immature Australasian white sharks (Carcharodon carcharias). Sci. Rep.10, 10169. doi: 10.1038/s41598-020-66876-z

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  SumptonW.LaneB.HamA. (2010). Gear modifications and alternative baits that reduce bait scavenging and minimize bycatch on baited drumlines used in the Queensland shark-control program. Proc. R. Soc Queensl.116, 23–34.

  • Google Scholar

• 47

  TanakaK. R.Van HoutanK. S.MailanderE.DiasB. S.GalginaitisC.O’SullivanJ.et al. (2021). North Pacific warming shifts the juvenile range of a marine apex predator. Sci. Rep.11, 3373. doi: 10.1038/s41598-021-82424-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  TateR. D.CullisB. R.SmithS. D. A.KelaherB. P.BrandC. P.GallenC. R.et al. (2019). The acute physiological status of white sharks (Carcharodon carcharias) exhibits minimal variation after capture on SMART drumlines. Conserv. Physiol.7, coz042. doi: 10.1093/conphys/coz042

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  TateR. D.KelaherB. P.BrandC. P.CullisB. R.GallenC. R.SmithS. D. A.et al. (2021a). The effectiveness of Shark‐Management‐Alert‐in‐Real‐Time (SMART) drumlines as a tool for catching white sharks, Carcharodon carcharias, off coastal New South Wales, Australia. Fisheries Manage. Eco28, 496–506. doi: 10.1111/fme.12489

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  TateR. D.KelaherB. P.BrandC. P.GallenC. R.SmithS. D. A.ButcherP. A. (2021b). Shark behaviour and marine faunal assemblage beneath SMART drumlines. Fish. Res.243, 106102. doi: 10.1016/j.fishres.2021.106102

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  VazquezC.RowcliffeJ. M.SpoelstraK.JansenP. A. (2019). Comparing diel activity patterns of wildlife across latitudes and seasons: Time transformations using day length. Methods Ecol. Evol.10, 2057–2066. doi: 10.1111/2041-210X.13290

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  WerryJ. M.SumptonW.OtwayN. M.LeeS. Y.HaigJ. A.MayerD. G. (2018). Rainfall and sea surface temperature: Key drivers for occurrence of bull shark, Carcharhinus leucas, in beach areas. Global Ecol. Conserv.15, e00430. doi: 10.1016/j.gecco.2018.e00430

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53

  WoodS. N. (2011). Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J. R. Stat. Soc. (B)73, 3–36. doi: 10.1111/j.1467-9868.2010.00749.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  WoodS. N. (2017). Generalized Additive Models: An Introduction With R, 2nd Ed ( Chapman and Hall/CRC). doi: 10.1201/9781315370279

  • CrossRef
  • Google Scholar