Statistical approaches towards analysing ungulate movement patterns in the Kruger National Park

[摘要] In this thesis I investigate the application of various statistical approaches towardsanalysing time series data collected using GPS collars placed on three ungulatespecies in the same region of the Kruger National Park, South Africa. Animalmovement tracking is a rapidly advancing area of ecological research and largedatasets are being collected with GPS locations of the animal, with shorter periodsbetween successive locations. A statistical challenge is to segment the movementpaths into groups which correspond to different behavioural activities. The aim ofmy study was to investigate and compare alternative statistical approaches foranalysing GPS data and to establish the best statistical framework for interpretingthese large herbivore movements. The focus was on which methods are the mostappropriate for these animals and the comparison of the movement patterns acrossspecies and season. Independent Mixture, Hidden Markov and Bayesian State-Space Models were used to analyse the hourly and daily movements of sable antelope, buffalo and zebra. Mixture Models provide a basic clustering technique to segment themovement paths and identify different underlying groups within the data assumedto correspond to different behavioural states. Posterior probabilities of groupmembership are used to allocate movements between successive locations todifferent states. This method ignores the dependence between successivemovements. Hidden Markov models (HMMs) use a time series technique andinclude a dependency between successive observations via a Markov process.Extensions to the HMMs were applied to allow for the inclusion of seasonalcovariates and irregular time gaps between successive observations caused bymissing locations. A Bayesian state-space model fits a random walk using MCMCmethods. The results were very similar to the HMMs but were more challengingto fit and required much more processing time. In the absence of informative prior information, the Bayesian method does not provide any improvement on the HMMs. The HMMs perform slightly better in terms of state allocation accuracy than the Mixture Models. However Mixture Models perform acceptably if only a straightforward clustering of the observations is required. However, if a more robust method is required, the HMMs are relatively easy to fit and extend, allow for investigation of the state switchingprobabilities and are recommended as the best method for analysing this type of data.

[发布机构] University of the Witwatersrand