Assessment of walnut (Juglans regia L.) cultivation land suitability using a multiple-criteria decision-making method in Southeastern Turkey.

Site selection for agricultural products is critical for agricultural planning, productivity, and farmers. Site selection is also critical for agricultural sustainability, as it helps ensure the efficient use of natural resources and avoids environmental degradation. This research proposes an evaluation model for walnut cultivation in the Savur (Mardin, Turkey) district in the Southeastern Anatolia region. Based on literature data, expert opinion, and local walnut producers' perspectives, a spatial decision support system was developed using Geographic Information System (GIS)-based Multi-Criteria Decision Analysis (MCDA). This model provides farmers and stakeholders with a comprehensive evaluation of the different walnut-producing sites in the Savur district. It allows them to make informed decisions about the most suitable areas for walnut production. The land suitability assessment for walnut orchards included four primary evaluation aspects, ten main criteria, and forty-seven sub-criteria. Evaluation criteria were weighted using the Analytic Hierarchy Process (AHP). A land suitability map was created using the weighted overlay method in overlap with the evaluation criteria. The results showed 442.56 ha of very highly suitable land, 9789.43 ha of highly suitable land, and 1140.76 ha of moderately suitable land in the Savur district. In order to determine the accuracy of the study model, ROC (Receiver Operating Characteristic) analysis was conducted. As a result of the validation analysis, the constructed model had an Area Under the Curve (AUC) of 0.911, which indicates an excellent fit between the model and reality; therefore, the model is acceptable. This study demonstrates that integrating AHP and GIS can aid agricultural land use planning and decision-making.