Mobile cleaning stations: trevally scrape on sharks in a tropical reefscape

Abstract

Inter‐species scraping has been reported between sharks and some teleost fish species to reduce ectoparasite loads and thereby increase fish health in an otherwise featureless pelagic environment. Here we report observations of interspecific scraping behaviour in reef environments between three species of trevally and three species of shark in the Seychelles Archipelago. These include the first scraping interactions for bull sharks Carcharhinus leucas and tiger sharks Galeocerdo cuvier observed in the wild. The interactions occurred despite the availability of cleaning stations and rough substrates suitable for self‐scraping. These observations further our understanding of inter‐species cleaning relationships in reef‐associated fish and sharks, and suggest that sharks may play a role in supporting reef fish health.

Marine ectoparasites are widely distributed in the ocean and affect hosts by attaching to their external surfaces causing tissue damage, impairing growth and ultimately reducing overall health (Finley & Forrester, 2003; Triki et al., 2016). Fish have limited options to remove ectoparasites or other nuisance items such as dead skin in the absence of limb‐like appendages. Instead, many species rely on the assistance of specialized cleaner species to clean themselves (Caves, 2021; Floeter et al., 2007; Vaughan et al., 2017). Mutualistic ‘cleaning stations’ are common in tropical reefs, whereby ‘client’ species visit to have ectoparasites and nuisance material removed by cleaner wrasses (e.g. Labroides dimidiatus), gobies (e.g. Elacatinus spp.) or cleaner shrimp (Côté, 2000; Vaughan et al., 2017). These interactions are generally considered low‐risk and high‐reward, as the client species typically exhibit docile behaviour such as posing when engaging with cleaner specialists (Caves, 2021; Côté, 1998).

It has been proposed that the scraping behaviour used by fish along the surface of another fish is used as an alternative to cleaner species as a way to clean themselves (Papastamatiou et al., 2007). Sharks provide ideal cleaning surfaces for both inter‐ and intra‐species scraping due to their unique dermal denticles that create a rough skin texture well‐suited for removing ectoparasites (Green et al., 2023; Williams et al., 2022). Various tuna species have been observed using sharks as scraping surfaces in pelagic environments (Thompson & Meeuwig, 2022). Various species of fish have scraped on great white sharks Carcharodon carcharias and basking sharks Cetorhinus maximus in near‐coastal environments in South Africa and the United States (Williams et al., 2022). Rainbow runner Elegatis bipinnulata and trevally (Carangidae spp.) have used Galapagos sharks Carcharhinus galapagensis and grey reef sharks Carcharhinus amblyrhynchos as scraping surfaces in Palmyra Atoll's coral reefs (Papastamatiou et al., 2007). Similar behaviour between bigeye trevally Caranx sexfasciatus and grey reef sharks has also been reported adjacent to a cleaning station in the Maldives (Williams et al., 2022). Cleaning interactions involving sharks are mostly commensal as there is no evidence of the sharks benefitting from the interaction and it has been reported that the sharks were irritated when scraped on and accelerated or turned away from the scraping fish (Papastamatiou et al., 2007).

Here we report scraping interactions in a tropical reefscape between three species of trevally—bluefin trevally Caranx melampygus, golden trevally Gnathanodon speciosus, and yellowspotted trevally Turrum fulvoguttatum—on three reef‐associated shark species—grey reef shark, bull shark Carcharhinus leucas, and tiger shark Galeocerdo cuvier. While this behaviour has previously been observed in grey reef sharks (Papastamatiou et al., 2007), our observations on bull sharks and tiger sharks represent, to our knowledge, the first documented instances for these species. These observations update our current understanding of cleaning behaviour in reef environments and fish‐shark scraping interactions.

Inter‐species scraping behaviour was observed between 2022 to 2024 within the Seychelles archipelago in the western Indian Ocean. Our observations took place at two locations: (1) the Inner Islands Group, specifically at northwest Mahé Island and Conception Island and (2) the Amirantes Islands of the Outer Islands Group, specifically at St Joseph Atoll and a remote reef south of D'Arros Island. A total of six craping events were recorded across 516 seabed baited remote underwater video system (BRUVS; see Langlois et al., 2020 for methodology) samples and during a recreational SCUBA dive. In total, over 510 h of video footage were collected. During analysis of BRUVS footage for species identification, abundance and size, we identified four instances of scraping behaviour among 286 samples where both sharks and trevally were present. These samples were re‐analysed to record quantitative and descriptive data on the species involved in scraping interactions. For each interaction, we documented the MaxN (defined as the maximum number of individuals of a taxa simultaneously seen in the field of view) and fork length of the trevally and sharks, counted the number of individuals directly involved in each scraping event and the number of scrapes, and recorded the affected body areas (Table 1 and Figure 1). For SCUBA diving footage, lengths were estimated visually. All research was carried out under The University of Western Australia's Animal Ethics Permit 2021/2024/ET000694.

TABLE 1.

Deployment details of baited remote underwater video system (BRUVS) and SCUBA dive observations of scraping behaviour recorded within the Seychelles archipelago.

Method	Year	Site	Habitat complexity	Habitat cover	Species	MaxN	Size (cm)	No. of scrapers	No. of scrapes
SCUBA	2022	D'Arros	MPR	Fan coral	Bull shark	1	200	‐	‐
					Golden trevally	7	50	2	7
SCUBA	2022	D'Arros	MPR	Fan coral	Bull shark	1	220	‐	‐
					Yellowspotted trevally	31	40–50	2	2
BRUVS	2022	St. Joseph	LPR	Macro algae	Grey reef shark	1	122	‐	‐
					Bluefin trevally	1	35	1	1
BRUVS	2024	St. Joseph	LPR	Rubble	Grey reef shark	1	118	‐	‐
					Golden trevally	7	35–42	2	3
BRUVS	2022	Conception	LPR	Bare	Tiger shark	1	131	‐	‐
					Yellowspotted trevally	31	20–25	2	3
BRUVS	2022	NW Mahé	LPR	Turf algae	Grey reef shark	1	115	‐	‐
					Yellowspotted trevally	2	35–36	1	2

Note: Habitat complexity is classified as low‐profile reef (LPR; 0–10° elevation) or medium‐profile reef (MPR; 10–20° elevation). Habitat cover included bare substrate (Bare), coral rubble (Rubble), fan coral (Fan coral), macro algae (Macro) and turf algae (Turf). Scraping interactions are shown with the scrapees listed above the scraper species, and the number of individuals recorded in the field of view is recorded (MaxN). SCUBA measurements are based on visual estimates. The number of individual scrapers and the number of scraping events are reported in the final two columns.

FIGURE 1.

Scraped body regions are organized by fish (scraper) and shark (scrapee) species. Colour shading between the scraper and scrapee indicates the specific body parts of the trevally that made contact during scraping. In interactions with the tiger shark, trevally were observed scraping multiple regions of their bodies along the shark's flank, represented as a colour gradient.

Trevally‐shark scraping interactions were observed between multiple species and occurred consistently within low‐complexity habitats. Bull sharks were scraped on by two different trevally species: golden trevally in the first event and yellowspotted trevally in the second event. Grey reef sharks were scraped on by a different trevally species across three separate events: bluefin, golden and yellowspotted trevally. Lastly, we observed a single event in which a juvenile tiger shark was scraped on by a yellowspotted trevally. Up to 31 trevally per species were present in the field of view during these events but only a maximum of two individuals scraped at any given time. Each interaction involved one to seven scraping instances. All observations occurred in low or medium profile reef environments, characterized by low‐complexity habitat cover, including bare substrate, coral rubble, fan coral gardens turf algae and macroalgae (Table 1).

Our observations revealed scraping behaviour that was consistent across different trevally‐shark species combinations. The trevally followed the shark, either individually or in groups, then a single fish would accelerate alongside the shark. The fish scraped against the shark's flanks, moving forward along the shark's body toward the nose, against the grain of the shark's dermal denticles. The fish tilted its body to facilitate scraping of specific areas (Figure 1). Scraping occurred exclusively on the lateral flanks of the shark, behind the fifth gill slit and below the dorsal fin. No scraping was observed on the shark forward of the gills, on its ventral side or on any fins. The sharks showed no reaction to the scraping process and displayed no antagonistic behaviour toward the scraper, such as speeding up, turning away or chasing (see Video S1).

The use of sharks as scraping surfaces may reflect calculated behaviour within a familiar ecological relationship rather than an exceptional risk (Papastamatiou et al., 2006, 2007; Sheaves et al., 2023). Trevally have been found in the diets of bull and tiger sharks (Lowe et al., 1996; Snelson et al., 1984) and juvenile trevally frequently associate with different shark species and use them as mobile nurseries (Sheaves et al., 2023). Alternative surfaces such as rock, sand, reef structures and cleaning stations are readily available in reef environments (Berthe et al., 2017; Coward, 2017; Ritter, 2011). However, shark skin is unique, with the tooth‐like structure of their dermal denticles potentially providing a superior scraping surface compared to naturally occurring surfaces (Raschi, 1992; Williams et al., 2022). Dermal denticles are particularly dense along the dorso‐lateral areas of the shark, compared to the smoother skin on their ventral sides (Ankhelyi et al., 2018; Gabler‐Smith et al., 2021). The scrapers consistently targeted the flanks and avoided the head region, where predation risk would be highest (Williams et al., 2022). Fish lead with their head where ectoparasite attachment is most common in pelagic fish (Feist & Longshaw, 2008; Hadfield et al., 2025). This apparent selection of scraping regions against the shark may reduce the number of scraping events required, conserve energy expenditure by the scraper and maximize parasite removal. Scraping on sharks by fish could be driven by factors other than ectoparasite removal, such as ‘mobbing’, whereby scraping is suggested to be employed by fish to drive sharks from an area (Michael, 1993). However, such reports are rare and given the lack of observable responses in sharks, observations here are more consistent with ectoparasite removal as the driver of scraping.

Anthropogenic pressures threaten cleaning services as reef‐associated sharks face immense pressure from targeted and incidental fishing, and some species have become functionally extinct in coral reefs over recent decades (MacNeil et al., 2020). The three observed shark species are found in tropical to subtropical waters circumglobally and are all classified in categories of conservation concern on the IUCN Redlist of Threatened Species (www.iucnredlist.org). Their populations are declining globally, and bull sharks are listed as Vulnerable (Rigby et al., 2021), grey reef sharks are Endangered (Simpfendorfer et al., 2020) and tiger sharks are Near Threatened (Simpfendorfer & Ferreira, 2019). Notably, all three species are caught in the Seychelles artisanal fishery in which their catch is unregulated (Nevill et al., 2007). The combined decline of complex reef habitats and shark populations may have significant implications for parasite removal, and consequently fish health, in tropical reefscapes. While the exact impacts on fish health remain unclear, the loss of these presumed parasite removal pathways in scraping interactions may trigger broader ecosystem and socio‐economic consequences not yet understood.

The growing ecological and conservation concerns for both reefs and sharks highlight the importance of documenting behavioural interactions. Such observations are increasing as access to the underwater environment improves. Both scientists and recreational ocean users increasingly use video technologies, such as cameras and drones, and sampling methods that favour video‐based data collection. Yet, our limited observations of scraping behaviour across more than 500 h of BRUVS deployments, combined with many hours of SCUBA diving, highlight the challenges of observing animal behaviour in the wild and understanding species dynamics. Progress in this field is likely hindered by a lack of recognition and documentation of these interactions. We emphasize the need for behavioural interactions to be more consistently recognized, reported and investigated during underwater surveys to better understand their implications for behavioural ecology and ecosystem function. Our observations contribute to the growing record of this behaviour in the wild and suggest that more shark species may be involved in scraping interactions than currently reported.

AUTHOR CONTRIBUTIONS

N.F.: Investigation, data collection, writing – original draft preparation, writing – review and editing. I.B.S.: Investigation, visualization, writing – review and editing. R.W.B.: Investigation, data collection, writing – review and editing. J.J.M.: Investigation, writing – review and editing. All authors have read and agreed to the published version of the manuscript.

FUNDING INFORMATION

The author(s) declare financial support was received for the research, authorship and/or publication of this article. All equipment and facilities used to undertake this work were supported by the Save Our Seas Foundation under Keystone Special Grant 494. We thank the founder of the Save Our Seas Foundation for providing access and support to all facilities and equipment at the Save Our Seas – D'Arros Research Centre, allowing this work to be undertaken. Fieldwork was supported by the Robson and Robertson award awarded to Nico Fassbender by the Jock Clough Marine Foundation. Sampling in partnership with Nature Seychelles was made possible by additional support from the Coral Reef Restoration Project Grant PIMS5736, financed by the Adaptation Fund through UNDP Mauritius and the Government of Seychelles. Nico Fassbender was supported by the Australian Government Research Training Program Stipend during write‐up of this work.

CONFLICT OF INTEREST STATEMENT

None of the authors have a conflict of interest to disclose.

Supporting information

Video S1. The video of the scraping interactions.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.