The relationships between occurrence probability of european hare (lepus europaeus) based on presence-absence data and satellite data in Lake district

This thesis was carried out to determine relationships between habitats of European hare in Burdur, Ağlasun and Sütçüler Region located in the Lakes Region and satellite data. For this purpose, the relationships between the data of the European hare and variables derived from Landsat 8-OLI satellite image (GLCM Homogeneity, GLCM Entropy, GLCM Correlation, circular mean, edge contrast and maximum difference) were investigated. The presence probability values of the European hare were calculated on 8 different scales: 71 m, 100 m, 158 m, 200 m, 300 m, 500 m, 750 m, and 1000 m radius. Statistical relationships were analyzed by multiple regression analysis and classification tree methods. As a result of the multiple regression analysis, the highest correlation was obtained with the presence probability values calculated for 71 m radius. The R2 value of this model created by circular mean and edge contrast was found to be 0,738. As a result of the classification tree performed to reveal relationships between the presence probability of European hare (in 71 m radius) and the parameters derived from the satellite data, the value of the training data R2 and the test data were found to be 0.82 and 0.77, respectively. The variables forming the model were detected as circular mean, maximum difference and edge contrast. In conclusion, it was clearly seen that the habitats preferred by the European hare can be determined successfully with satellite data.