Habitat suitability model for red deer (Cervus elaphus Linnaeus, 1758): spatial multi-criteria analysis with GIS application

Monitoring population trends is essential in wildlife management. In fact, to identify those environmental conditions linked to habitat use and selection is a key task in any conservation plan. Our aim in this work was predict red deer (Cervus elaphus Linnaeus, 1758) colonization in the Lombada National Hunting Area (LNHA), one of the largest wild populations in Portugal. The sharply increase in density and range of red deer over the last decades justify the careful monitoring of this population. A multi-criteria model using the Analytical Hierarchic Process (AHP) and GIS weighted spatial analysis with matrix algebra techniques were applied in this approach. This method was developed in three distinct phases: i) setting the objectives, ii) analysing pairs of factors through the application of a comparison matrix and iii) its synthesis with the weight assignment that followed a predetermined numeric range according to the ecological requirements of the study species. The variables used in the red deer habitat suitability model included the hydrographic network, asphalted roads, relief aspects and land use. From October 2010 to March 2011, a total of 129 observations along line transects were recorded and used to validate the final model. By using the AUC method it was obtained a success rate of 72%. Our results suggest that this method would be applied on a larger scale being suitable to predict red deer expansion. To our knowledge, this is the first study performed in Portugal using this methodological approach to assess the red deer-habitat relationships.