Habitat modelling using remote sensing and geographical information systems: Kentish plover case study

Increasing human activities result pressure on the natural resources. Inventorying the wild life as a one of the most effected natural resource will reveal the magnitude of these detrimental impacts. Wild life inventory together with habitat quality will be informative to the conservation oriented studies. In this respect, habitat of Kentish Plover (Charadrius alexandrinus) bird species which uses Tuzla Lagoon and its surround was modelled considering the human impacts. Three different models including Artificial Neural Networks, Regression Tree and Regression (Multi Linear Regression and Logistic Regression) were used with the support of remote sensing and Geographical Information Systems (GIS). These models were applied to the data set which was collected by a group of biologist collected region between 1996 and 1999. Artificial Neural Networks resulted relatively more accurate than other methods. Within this study the potential of mathematical models for habitat quality estimation was assessed with the support of remote sensing and GIS.