Alternative methods of analyzing GPS data.

Funding Sources: NSF-NIH Ecology Infectious Disease Program. Awarded to Wayne Getz.

Collaborators: Wayne Getz (UC Berkeley) , Leo Polansky, Shirli Bar-David (UC Berkeley)


High-resolution animal location data are increasingly available, but currently available methods may not yet realize the full potential of these datasets. For home range analyses, my collaborators and I developed a non-parametric approach that improves upon kernel density estimators in defining hard boundaries such as rivers, roads, cliffs, and human development. For movement analyses, some areas may be used repeatedly at various timescales (e.g. annual migrations or daily visits to the same water source). We have also developed techniques to identify these areas and timescales of recursion. Finally, we have borrowed from time series approaches to use the auto-correlated structure of GPS datasets to identify behavioral patterns. Behavioral patterns are likely to vary over time and space, which motivates the importance of wavelet analyses that can detect how movement patterns change. In addition to generally introducing animal ecologists to these methods, we also developed a stochastic movement model with time varying parameters to evaluate these methods under different behavioral complexities and sampling intervals, and to compare observed patterns in data with those from simulated null hypotheses about expected behavior. I hope that these methods provide additional tools for ecologists to more fully realize the potential of GPS datasets.


Citations:
Bar-David, S., I. Bar-David, P. C. Cross, S. J. Ryan, C. U. Knechtel, and W. M. Getz. 2009. Methods for assessing movement path recursion with application ot African buffalo in South Africa. Ecology 90:2467-2479.

Getz, WM, S Fortmann-Roe, PC Cross, AJ Lyons, SJ Ryan, CC Wilmers. 2007. LoCoH: Non-parameteric kernel methods for constructing home ranges and utilization distributions. PLoS One 2(2): e207.

Polansky, L., G. Wittemyer, P.C. Cross, C. Tambling & W.M. Getz. 2010. From moonlight to movement and synchronized randomness: Fourier and wavelet analyses of animal location time series data. Ecology: Vol. 91, No. 5, pp. 1506-1518. doi: 10.1890/08-2159.1



Return to Disease Ecology Research Projects