SPATIAL ANALYSIS OF ANTHROPOGENIC FACTOR-BASED AGRICULTURAL SUITABILITY USING RANDOM FOREST: THE CASE OF HATAY, TÜRKIYE

This study focuses on anthropogenic factors that can be managed in the short and medium term within land-use planning processes, unlike the natural environmental factors commonly used in agricultural land suitability analyses. The primary objective of the study is to evaluate anthropogenic criteria identified through expert opinions using a machine learning-based spatial modeling approach and to reveal the relative effects of these factors on agricultural suitability. In this respect, the study aims to provide a methodological and practical contribution to anthropogenic-focused approaches that have received limited attention in the literature on Anthropogenic Agricultural Land Suitability (AALS). The evaluation factors used in the analysis were derived from factors identified through the Delphi technique based on expert opinions. Within the scope of the study, spatial data layers were generated using Geographic Information Systems (GIS) and Remote Sensing (RS) techniques, and an agricultural land inventory was created using CORINE land cover data. Using the obtained dataset, a machine learning-based model was developed with the Random Forest (RF) algorithm. The results indicated that the RF model exhibited successful classification performance with an F1-score of 79% and an AUC value of 90%. The findings of this study demonstrate that machine learning-based spatial analyses provide an effective tool for supporting the protection of agricultural lands, reducing land-use conflicts, and facilitating sustainable landscape planning decisions.