The Requisition–Compensation Balance of Cropland (RCBC) policy is important for ensuring food security. Previous studies have mainly focused on the quantity and quality of cropland when assessing the impacts of this policy. In terms of morphology, previous studies have primarily relied on landscape indicators. Therefore, this study aims to thoroughly analyze the impacts of the RCBC policy on the quality and morphology of cropland (especially morphological spatial pattern analysis, MSPA) in the Pearl River Delta (PRD) during 1996–2021. To this end, we constructed a comprehensive evaluation index system by combining MSPA, landscape indicators, and field research. The results show that the cropland quality in the PRD has exhibited a consistent improvement trend. High-quality cropland is spreading from central cities to the periphery, and the spatial distribution is becoming more even. Nonetheless, MSPA reveals an increasing trend of cropland fragmentation. The results indicate a decline in the area of the “core”, an increase in the area of the “island”, and a decrease in the connectivity of the cropland. Our field research confirms that the RCBC policy has indirectly exacerbated cropland fragmentation. In many regions of the PRD, the fragmentation of cropland hinders the application of agricultural mechanization and increases the cost of cultivation, resulting in severe cropland abandonment. Therefore, local governments should implement rigorous planning and prioritize cropland morphology when compensating cropland. Our findings are expected to provide empirical evidence for improving the RCBC policy and protecting cropland.

In the context of economic globalization, the issue of imbalanced regional development has become increasingly prominent. Misreporting in traditional economic censuses has made it difficult to accurately reflect economic conditions, increasing the demand for precise GDP estimation. While nighttime light data, point of interest (POI) data, and street-view imagery (SVI) have been utilized in economic research, each data source has limitations when used independently. Furthermore, previous studies have rarely used high-resolution (over 30 m) nighttime light data. To address these limitations, we constructed both random forest and decision tree models and compared different indicator combinations for estimating GDP at the town scale in Dongguan: (1) Qimingxing-1 nighttime light data only; (2) Qimingxing-1 nighttime light and SVI data; and (3) Qimingxing-1 nighttime light, SVI, and POI data. The random forest model performed better than the decision tree, with its correlation coefficient improving from 0.9604 (nighttime light only) to 0.9710 (nighttime light and SVI) and reaching 0.9796 with full integration. Moreover, the Friedman test and SHAP values further demonstrated the reliability of our model. These findings indicate that the integrated model provides a more accurate reflection of economic development levels and offers a more effective tool for regional economic estimation.