Research | Hawksbill Turtles

Hawksbill Turtle foraging grounds

David Rowat and Jeanne Mortimer met with us to explore possibilities to help them in their work with the Marine Conservation Society of the Seychelles. Jeanne is a turtle expert and had some fascinating data from 1998 when she attached Argos satellite tags to five nesting turtles and then tracked them after they left the beaches. Amazingly, none left the Seychelles banks, and at least one stayed within a very tight area for over 560 days! Jeanne very much wanted to see just what was out there at this site that made it so special for this turtle to hang out there for well over a year!

We began by doing some statistical analysis on the turtle positions. The position reports were indeed very closely-clustered, over 50% of them within 580 m of the centroid of all the points. So we took Jocara out and ran a survey grid over the area, about 1.2 km long and 600 m wide. We logged the depth and recorded the GPS NMEA data stream on the Bohemian Rhapsody, then cubic-spline interpolated the data (if your eyes are glazing over at this point, don't worry, this is as technical as it gets) to get a bathymetric map. It turned out that there was a deep gulley cutting through the seabed, deepening it from an average 30-40 m to over 50 m.

The water depth is indicated in colour, with the key given on the right, varying from 30-55 m. The axes are in decimal degrees, the latitude shown as negative since we are south of the equator.

So now we were all fired up to go see what the seabed might look like, and what was growing there to feed this turtle! Caroline and John geared up for a mild decompression dive to go take a look. The idea was to inspect the seabed just outside of the gulley, then to work our way down into the gulley to see what changes might be found. At the starting point we found a smoothly-undulating sand bottom with scattered outcrops of sponge and coral without significant features or abundant food sources.

Here's a photograph of the most colourful and intensely-colonised piece of seabed we could find, at about 38 m depth, taken with the F828 and colour-corrected.

Moving downslope into the gulley, we found that vegetation coverage decreased, until almost no life could be seen. The length of the dive track was 100 m. Then, at just over 50 m depth, we began to encounter numerous pits, apparently excavated by some creature. Here's an image of a pit, taken from a non colour-corrected video still.

These pits often contained small stones that may have been captured by the pit after it was excavated. Most, if not all, pits looked weathered and old, often with algae on the slopes. Some had small holes in the sides or bottom consistent with some type of burrowing worm occupation, although the pits were too large to have been dug by worms, being in the range 1.0 – 1.5 m in diameter.

That was it for the first dive, we had already incurred some decompression obligation. Although we missed the anchor chain for our ascent, we did our decompression stop free-floating and Alex (primed to look for us 35 minutes after we went down) picked us up in the dinghy when we surfaced.

What could these pits be for? First of all we thought that maybe they were feeding pits, excavated by the turtles to get molluscs or other food. Then Jeanne suggested that they might be sleeping pits, turtles apparently need to get deep enough to become negatively buoyant to sleep. Perhaps the gulley was important for that reason? Checking, Jeanne found that 96% of the best-quality satellite positions were received just after dawn! Ah ha! So on Caro's birthday off we went to be back at 06:00 to check for surfacing turtles. It was a lovely dawn, but no turtles. Undaunted, we planned out second dive, to see if we could find anything different from the first dive.

The second dive we began again at the gulley margin, some 40m to the SE of the first dive, and again worked our way into the gulley. The ground track was 120m. Same old story, with an even more marked decrease in vegetation and onset of excavated pits below 50 m depth. At 54m, where the pits were most densely-packed together, we collected samples of the material in the pit bottom, The samples contained coarse sand and silt with many small fragments of coral, sponge and shells. then it was high time to get back to the anchor chain and decompress. That's when we noticed the current was against us. We should have realised before. It was also 20 m further back than we had thought. Sure enough, we ran short on air and had a bit of drama, but you can read all about that in the chatty log! No harm done, lessons learnt...

Here's Caroline, with the Sony PDX10 video camera in its Sea&Sea video housing, collecting a seabed material sample at 54 m depth in a ‘ziplock’ bag.

Sightings of adult Hawksbill turtles by recreational divers are rare. This suggests that adult turtles are either occupying non-recreational depths (perhaps 30m and over) and/or non recreational sites (i.e. without attractive fish and coral life). That this one turtle apparently spent 560 days in a very confined area, showing up within 580 m of a single point once every week on average, is certainly intriguing.

There are three main contenders to explain the pits:

1 The pits are dug by turtles to sleep in
2 The pits are dug by turtles foraging for food
3 The pits are dug by other creatures

If the pits are sleeping pits, this makes sense if the turtle requires a minimum depth in order to guarantee negative buoyancy during sleep. The gulley provides such depth, whereas much of the surrounding terrain does not, and may provide the right sediment consistency for digging. If so, one can imagine that this gulley could be in high demand as a turtle dormitory, with turtles returning again and again once they discovered this desirable site. This explanation also explains why 96% of the high-quality position reports for this turtle were obtained in such a tight area around the gulley and at times just after dawn (presuming that this timing is not already accounted-for by limited satellite coverage).

With this explanation, this turtle may have spent considerable time at other locations that went unreported due to limited tag surfacing times during feeding and free-swimming. The true foraging area may therefore be somewhere else. We may know even less than we thought about foraging grounds!

If these pits are for feeding, they are presumably dug to extricate buried worms, crustaceans and molluscs. This seems a less plausible explanation, unless a rich supply of such food is only to be found at depths over 50 m and/or the sand in the gulley is of a different consistency that favours the prey population to inhabit it. Given the lack of other obvious food sources, the prey unearthed in such excavations must have been sufficient to sustain the turtle for the extended period of well over a year. Loggerhead turtles have been observed in Australia to dig trenches to unearth molluscs but this has not been observed for Hawksbill turtles. Further, why was this turtle only digging shortly after dawn, unless perhaps the prey behaviour is synchronized in some way to the dial cycle?

Alternatively, perhaps the pits are dug by some other creatures, such as the giant marbled stingray. In which case, we are left with the question of what this turtle was doing in this tightly-confined region and what is so attractive about it, since there is no obvious abundance of food. Furthermore, why were high-quality position reports almost exclusively obtained just after dawn?

As so often with science, we generate more questions than answers...