Sustainable development is contingent upon the efficient management of land resources for resolving spatial challenges such as land-use conflicts and fragmentation. A land-suitability model offers a potential instrument for assessing land-use/land-cover (LULC) consistency with spatial plans. This study employed a data-driven probabilistic approach using a support vector machine (SVM) algorithm and error-correcting output codes (ECOCs) for incorporating 11 physical parameters to generate spatial grids that reflected land-suitability levels. The probabilistic outputs were derived by calibrating SVM decision values using Platt scaling within the ECOC framework, enabling a reliable estimation of class-wise landsuitability probabilities. The model achieved the highest probability value of 0.9952, with an average of 0.8251; this demonstrated its potential for assessing the consistency of land use/land cover with spatial plans. The model exhibited robust performance and substantial agreement between the predictions and actual data, with an overall accuracy of 88.56% and a kappa index of 0.873. Additionally, the study utilized a land-suitability model and non-weighted overlay relevance matrix to identify discrepancies in Bogor Regency’s spatial plan, quantifying the compliant and noncompliant land areas for each LULC class within specified spatial-plan zones. The evaluation revealed a significant misalignment, with 25–45% of agricultural land uses that included wetland and dryland agriculture, plantations, and inland fish farms being allocated within settlement zones; this indicated a mismatch between spatial plans and land suitability. These findings underscored the importance of evaluating and revising the spatial plan to enhance its alignment with land suitability.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.