The Featured Research page is designed to highlight current research in the Large Pelagics Research Laboratory. The page will change often, highlighting the findings of graduate students, post-doctoral fellows, research scientists, research technicians and undergraduates involved in the LPRL. Our first installment is by Dr. Andrew Myers, post-doctoral fellow in leatherback turtle research. The turtle group has used unique technology to learn about movement patterns from nesting beaches in St. Croix, USVI.

Daily Diary devices in leatherback turtle research
Andrew Myers, PhD

In the 1930's, the Norwegian Scientist, Per Scholander attached a rudimentary depth recorder to a harpoon which was fired into the blubber of an unfortunate fin whale. Thus began the evolution of a new branch of ecological research. Biotelemetry (the study of animal behaviour using data logging devices) has now evolved to the point whereby multiple sensors can be incorporated into devices. Through links to our collaborators at the University of Wales Swansea , we are conducting research which is at the cutting edge of this field of research.

The Daily Diary device, developed by Professor Rory Wilson has been used extensively on several different terrestrial and marine species. During May 2007 we deployed five of these devices on adult female turtles nesting at Sandy Point Wildlife Refuge, St Croix , USVI. Daily diaries are multi-sensor data logging devices with the capacity to record on up to 14 channels. The devices used in our study recorded depth, temperature, speed, light, compass heading, triaxial acceleration and mouth opening behaviour. The volume of data recorded by these devices is immense, each channel recording at between 1 and 8 times per second, for a total of ~4 million recordings each day.

The post-hoc processing of these data confirm the adage that the whole is greater than the sum of the parts, particularly with respect to the acceleration channels. Three acclerometers on the daily diaries provided us with information on the orientation of the animal. These sensors are positioned so as to record movement on the dorso-ventral (heave), anterior-posterior (surge) and lateral (sway) axes (fig 1.).

Various interesting movement patterns were observed by these turtles during the internesting interval including barrel rolls (fig 2a) and a period of half an hour when the turtle was periodically spending about a minute upside down at the surface before conducting a short dive and then returning upside down to the surface (fig 2b). We believe that this second behaviour is indicative of predator evasion, with the turtle displaying its carapace to a potential predator.

The data from the acceleration channels is particularly noisy, this noise is primarily due to the spikes in acceleration due to movement (dynamic acceleration). By extracting this noise from the raw acceleration data we are able to ascertain the magnitude of movement on the three axes. If we then add together the dynamic acceleration from all three channels we are able to derive the overall dynamic body acceleration (ODBA). In a controlled experiment ODBA was found to be an accurate proxy for oxygen consumption (and hence energy expenditure) in cormorants. Prior to the application of accelerometers, energy expenditure was derived using either prohibitively costly (doubly labeled water) or invasive (heart rate or electromyogram) techniques. Although ODBA only provides us with information regarding the amount of energy used during movement and does not include basal metabolic rate, specific dynamic action, the majority of the energetic demands of reptiles are for the purpose of movement. As a dive begins ODBA is elevated as the turtle is overcoming positive buoyancy (fig 3). At a depth of ~20m ODBA decreases as the turtle becomes neutrally and then negatively buoyant. During the dive ascent ODBA is temporarily elevated as the turtle overcomes negative buoyancy. As the turtle reaches the surface ODBA, once again, decreases as the animal floats to the surface.

This research is allowing us to gain a hitherto unseen glimpse into the activity specific energetic demands of leatherback sea turtles. We currently have a manuscript in preparation detailing the foraging behaviour and related movements of these leatherbacks in terms of sub-surface swimming. This summer we are hoping to attach these devices to leatherbacks in the foraging grounds off the coast of New England which will allow us to conduct a much needed comparative study between the two regions.

Collaborators on this project are:
Rory Wilson & Niko Liebsch http://www.swan.ac.uk/staff/academic/EnvironmentSociety/BiologicalSciences/wilsonroryp/
WIMARCS http://www.wimarcs.org/