Macquarie University

File(s) stored somewhere else

Please note: Linked content is NOT stored on Macquarie University and we can't guarantee its availability, quality, security or accept any liability.

Swim with the tide: tactics to maximise prey detection by a specialist predator, the greater sea snake (Hydrophis major)

posted on 2022-06-10, 02:36 authored by Vinay Udyawer, Claire Goiran, Olivier Chateau, Richard Shine
The fitness of a predator depends upon its ability to locate and capture prey; and thus, increasing dietary specialization should favor the evolution of species-specific foraging tactics tuned to taxon-specific habitats and cues. Within marine environments, prey detectability (e.g., via visual or chemical cues) is affected by environmental conditions (e.g., water clarity and tidal flow), such that specialist predators would be expected to synchronize their foraging activity with cyclic variation in such conditions. In the present study, we combined behavioral-ecology experiments on captive sea snakes and their prey (catfish) with acoustic tracking of free-ranging sea snakes, to explore the use of waterborne chemical cues in this predator-prey interaction. In coral-reef ecosystems of New Caledonia, the greater sea snake (Hydrophis major) feeds only upon striped eel catfish (Plotosus lineatus). Captive snakes became more active after exposure to waterborne chemical cues from catfish, whereas catfish did not avoid chemical cues from snakes. Movement patterns of tracked snakes showed that individuals were most active on a rapidly falling tide, which is the time when chemical cues from hidden catfish are likely to be most readily available to a foraging predator. By synchronizing foraging effort with the tidal cycle, greater sea snakes may be able to exploit the availability of chemical cues during a rapidly falling tide to maximize efficiency in locating and capturing prey. 


This repository has three sets of data:  1) Experimental trials to assess response of snakes to catfish chemical cues: We captured four H. major (snout-vent length [SVL] 53–130 cm, mass 44–1150 g) from the Baie des Citrons and Anse Vata in January-March 2020, and maintained them for periods of 6 to 19 days in a large (2 x 2 m, 18 cm deep) tank with circulating seawater at the Aquarium des Lagons. During those periods, we conducted trials on the responses of snakes to chemical cues from catfish. To quantify distances moved by snakes, we defined 16 equally sized 0.5m2 quadrats on the floor of the tank by stretching two sets of cords, perpendicular to each other, across the tank. As a measure of activity, we scored the number of times a snake’s head crossed a cord within a 2-minute period after we introduced either chemical cues from fish, or a control stimulus. Where possible, we also scored tongue-flicking rates. The donors of those fish chemicals were a swarm of nine Plotosus lineatus (body lengths 7.5 to 12.5 cm), captured from nearby bays by staff at the Aquarium des Lagons and maintained at the Aquarium in a tank containing 30 L of seawater. At the beginning of a trial we moved a single snake from its holding tank into the large experimental tank, waited 15 minutes to allow the animal to acclimate and settle down, then took 10 L of water from the catfish tank and gently poured it into one corner of the large tank. For control trials, we took the same volume of water from an empty tank, and poured it into the test arena in the same way. Because the experimental tank had circulating seawater, chemical cues from each experiment were removed between successive trials by opening both the inlet and outlet valves of the tank to allow water replacement. Trials with different snakes were separated by 15 to 30 min, with each trial commencing after the focal snake had been inactive for at least 1 min. The order in which the stimuli were presented (control versus fish-scented water) was randomized. All trials were conducted during daylight hours (1000–1700 h). 2) Experimental trials to assess response of catfish to sea snake chemical cues: To quantify responses of catfish to chemical cues from snakes, we used the same fish as described above, and conducted behavioral trials in the home tank within which fish were held. We added four plastic tubes (5 cm diameter, 15 cm long) as shelters, one to each corner of the catfish’s home tank (160 x 50 x 50 cm) with the open ends of the tubes all pointing towards the center of the tank. Inside each shelter, we placed a square piece of cheesecloth. Two of those squares had been moistened and then vigorously rubbed along the body of a live H. major for 30 s, whereas two served as controls (moistened but not rubbed on a snake). For each trial, the two scented squares of cheesecloth were rubbed on different snakes each time. After 5 min, we scored which shelter had been chosen as a retreat, lifted the tubes out of the tank, and replaced them in the same positions. The catfish inside the tube were gently poured back into the tank before the tube was replaced in position. After another 5 min, we scored which retreat had been chosen. After five such trials, we changed the locations of treatment (snake chemical) versus control shelters, then repeated the above procedure another five times at 5-minute intervals, then again changed the locations of shelters and repeated the trials five more times. In total, this experiment provided 15 replicates for catfish selection of shelter-sites as a function of the presence of chemical cues from sea snakes. 3) Acoustic telemetry data from telemetered sea snakes monitored within two bays in Noumea, New Caledonia We hand-captured 19 Hydrophis major (15 males, 4 females; SVL range 63 to 125 cm) in the Baie des Citrons and Anse Vata in January and October 2017, and a qualified veterinarian surgically implanted acoustic transmitters (V9P-2H; Vemco Ltd., Bedford, Nova Scotia) into the snakes under gaseous anesthesia. Transmitters weighed <1% (mean ± SE; 0.65 ± 0.17%) of snake body mass and were neutrally buoyant; snake behavior and locomotor ability were assessed during a 24-hour post-surgery recovery period and appeared to be unaffected by the implantation. The animals were released at their sites of capture the day after surgery, and their movements over the following 349 days were monitored with an array of 18 acoustic receivers that recorded detections of individuals when they were within detection range of each receiver. The detection range of receivers within the array was tested by comparing the expected and observed number of detections from range-testing tags placed at multiple locations within the array over a period of 1 week at the start of the study, and was measured to be on average 150 m. Receiver stations were arranged ~300 m apart through relatively shallow (<3 m depth) areas of the two bays, with all receivers remaining fully submerged and active across the full tidal range within the study site. We also obtained hourly data on tidal height (data from Service Hydrographique de la Marine Nationale) from which change in tidal height (henceforth hourly tidal range) and direction of flow (i.e., rising vs. falling tide) was calculated for each 60-minute observation period coinciding with movement records of telemetered snakes.  

Usage Notes

Experimental trial datasets are in an excel spreadsheet with two worksheets a) Hydrophis response to cues; and b) Catfish response to cues. Acoustic telemetry data is in a R data object in the .RDS format that can be opened and accessed in the R statistical environemnt. The R object contained within is an ATT object to be used with the Animal Tracking Toolbox in the VTrack packge ( The ATT object is a list object containing three components a) Raw tag detection data from the acoustic array in Noumea, b) tag metadata from animals tracked in this study, and c) receiver array metadata for the array used to monitor the telemetered snakes. See the following vignette on how movement metrics and activity space metrics can be obtained using this ATT object:


Laboratoire d'Excellence Corail*

Laboratoire d'Excellence Corail

Australian Research Council : FL120100074

PADI Foundation : 28454


FAIR Self Assessment Rating

  • Unassessed

Data Sensitivity

  • General



Usage metrics

    Macquarie University Research Data Repository



    Ref. manager