The attributes of landscapes, abiotic and biotic, influence the behaviour of animals. Understanding the reciprocal effects between the landscape, habitats and the animals are an essential tool in sustainable management of natural resources as well as conservation. A source of information about how the environment shapes the behaviour of animals is movement data. The advent of cheap GPS devices has facilitated data collection. The fine scale spatio-temporal resolution allows the identification of complete home-ranges, even habitat that is only transiently used. Fine-scale movement data can also be used as proxy for behaviour-types; different behaviour states cause different movement tracks. I test, extend and apply the multi-change point analysis (MCPA), developed by Gurarie et al. (2009) for the identification of behavioural change points in irregular movement data. The method relies on conventional time-series analysis methods, rather than a Bayesian framework, such as the state space models. After thorough testing of the MCPA, we apply the method and its extension to the GPS data of four red deer and 27 golden eagles. The red deer data were analysed to detect behavioural changes at a seasonal scale and to serve as a test-bed for our extension of the MCPA. The comparison of movement bouts revealed that the expression of behaviour was on a gradient rather than the discreet states. The study of the golden eagle data highlighted the necessity to choose the appropriate sampling regime of movement. If the intervals between the location-fixes are too big, valuable information about important small scale behaviour will be missed.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:577582 |
Date | January 2012 |
Creators | Konrad, Christoph |
Publisher | University of Aberdeen |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Source | http://digitool.abdn.ac.uk:80/webclient/DeliveryManager?pid=196265 |
Page generated in 0.0014 seconds