Classification of kangaroo habitat distribution using three GIS models |
| |
| Authors: | Andrew K Skidmore Alison Gauld Paul Walker |
| |
| Institution: | 1. Centre for Remote Sensing and GIS, School of Geography, University of New South Wales , Sydney, NSW, 2052, Australia E-mail: email: A.Skidmore@unsw.edu.au;2. CSIRO Division of Wildlife Ecology , Gungahlin, ACT, Australia |
| |
| Abstract: | Abstract In this paper, we compare three techniques for mapping wildlife habitat, termed BIOCLIM, CART and a new classification method based on nonparametric techniques. These techniques model dependent map layers of species distributions, where the areas to be mapped are large and the plot data is sparse. The techniques recognise pattern in the (independent) plot data, available to natural resource managers. In this case, the independent data set comprised 12 climate surfaces, that attempt to represent the range of temperature and precipitation important in determining the habitat of kangaroos across Australia. With this particular data set, the CART (decision tree) model was most accurate, but more time consuming to initialise. The relative performance of these models depends on the quality of the data set, and skill of the GIS analyst. Where possible, GIS analysts should implement all available methods, and compare output. |
| |
| Keywords: | Salinity Risk Condamine Murray‐Darling Basin Geodatabase |
|
|
