Spatiotemporal forest cover change monitoring of phytogeographic regions of Türkiye with a machine learning hybrid classification method

Abstract

This article presents a novel and innovative approach to mapping the forest cover of T & uuml;rkiye over the last 30 years. The authors have developed a machine learning-based hybrid classification method named 'Combination of the Best Hybrid Classification Method (CBHCM)', which combines three different machine learning image processing techniques to provide a more comprehensive and accurate representation of the country's forest cover changes. The study's results, as evidenced by the high kappa coefficient of 0.806, demonstrate the success and reliability of this new approach. Furthermore, it provides insight into the changes experienced by different phytogeographic regions over the same period. This study highlights the increasing trend of forest cover in all three phytogeographic regions in T & uuml;rkiye but also raises the question of the structural status of these forests. This research offers a valuable contribution to remote sensing and geographic information science community and provides essential information for decision-makers in managing European and Turkish forests.