Key points are not available for this paper at this time.
The optimization of new cropland allocation is crucial for promoting the efficient use of cropland resources and safeguarding food security. However, existing studies primarily take suitability as the optimization objective and lack the consideration of stable utilization potential, which may lead to subsequent unstable use. To address this gap, this study quantified the stable utilization potential of new cropland using a machine learning model and integrated it with the ant colony optimization (ACO) model to develop a spatial allocation framework. This framework was validated in Guangdong Province, China, a region characterized by diverse resource endowments and pronounced regional heterogeneity. The results indicated that, under the specified objective-weighting scheme and compared with the business-as-usual (BAU) scenario, the optimized scenario achieved regional cropland quantity balance. It also increased the average stable utilization potential of new cropland and overall utility by 34.84% and 12.74%, respectively, while reducing the mean cost and mean ecological benefit loss per unit area by 10.22% and 41.36%, respectively. Overall, under the specified constraints, the proposed framework offers a promising approach for new cropland planning and provides a basis for governments and land management authorities to improve future allocation practice.
Zhao et al. (Sat,) studied this question.