| Summary: | Movement and habitat use are essential population processes fundamental for the management and conservation of animal species. Hence, the ability to track and infer specific animal behaviours are problems that have concerned scientists for many years. Technological advances have contributed to the improvement of tracking methods over the last century. Among them, ARGOS system has become an important tool for tracking animal movement globally, however it is limited to only providing geolocation. Therefore, it is still necessary to develop a methodology capable of inferring the behaviour of an animal using the available information. The inherent complexity of animal movements and limitations in tracking systems have made necessary to develop statistical methods able to cope with such constraints. Recently, State-Space models (SSMs) have been presented as an approach capable of integrating the treatment of both limitations. In this project, it is presented a two-stages method able to determine, from movement metrics, the distribution of the animals and, more specifically, the foraging behaviour of Danish Wadden Sea population of harbour seals. Firstly, this methodology handles ARGOS inaccuracy positioning and, secondly, assigns behavioural states to each location. Despite of the improvement of technologies, it is always challenging to estimate the spatial distribution of wildlife, but it becomes even more complicated when referring to marine life. This project aim to satisfactory applies a State Space Models to deal with biological and statistical complexities and determine the geographical distribution of the animals and site fidelity in terms of haul-out and foraging positions
|
|---|