Keeping pace with movement analysis
Article first published online: 4 AUG 2006
Journal of Animal Ecology
Volume 75, Issue 5, page 1045, September 2006
How to Cite
Hays, G. (2006), Keeping pace with movement analysis. Journal of Animal Ecology, 75: 1045. doi: 10.1111/j.1365-2656.2006.01147.x
- Issue published online: 4 AUG 2006
- Article first published online: 4 AUG 2006
I.D. Jonsen, R. A. Myers & M. C. James (2006) Robust hierarchical state–space models reveal diel variation in travel rates of migrating leatherback turtles. Journal of Animal Ecology, 75, 1046–1057.
Over the last 10 years there has been an explosion of studies where animals have been tracked over a range of spatial and temporal scales. In some cases animals have been followed for over 1 year while they have moved across 1000s of km while, on the other hand, localised movements of small insects are now being resolved over scales of a few metres. While new technology has provided the tools for this wide range of studies, analytical methods for processing the resulting tracking data-sets are still being developed. A central problem to all tracking techniques is that animal locations have an error associated with them and this error may make it difficult to tease apart details of the animal's behaviour. A case in point is satellite tracking using the Argos system where location errors can range from a few hundred metres to many tens of km.
In this paper, Jonsen et al. use Argos data to investigate the speed of travel for leatherback turtles, including diel patterns, as they travel the length of the North Atlantic. This seems like a trivial task, but a problem emerges because the distance moved in a single night (up to about 30 km) is small compared to the inaccuracy of locations. By employing state–space models, Jonsen et al. are able to extract a clear biological signal from this noisy data-set and show how turtles slow down at night on their long journeys, possibly to feed on migrating zooplankton that come towards the surface at that time of day. The study forms an important step in the ongoing development of analytical approaches that will help ecologists get the most from their hard-earned tracking data-sets. Over the next few years ecologists can hope to have an increasingly sophisticated and useful ‘analytical toolbox’ at their disposal.