DETERMINATION OF TEMPORAL CHANGES ON HETEROGENEOUS STRUCTURED FORESTS BY USING THE POST-CLASSIFICATION COMPARISON TECHNIQUE

In this study, it is aimed at determining of temporal changes occurred in forests having different structural characteristics such as tree species, succession age, canopy closure etc. in Western Black Sea region, Turkey. Supervised classification using a maximum likelihood algorithm was applied to the Landsat multidate image set relating to the years of 1987 and 2000. Accuracies of classified images were evaluated by using error matrices and the overall kappa statistics were calculated as 0.8543 and 0.9038, respectively. When the transitions between different main information classes in change matrices were investigated it was determined that 29.77 % of study area had been changed. According to results provided, it was found out that by using Landsat images and post-classification comparison method, the temporal changes in the heterogeneous structured forests of study areas could be determined with high accuracy and limited detail. Keywords: Post-Classification comparison, Change analysis, Remote sensing