Spatial patterns and driving forces of non-grain production of cultivated land in Guizhou Province

The expansion of non-grain cultivated land poses challenges to sustainable land use, particularly in fragile karst mountainous areas. This study focuses on Guizhou Province, which has the most extensive karst landforms, to investigate the spatial patterns and driving mechanisms of NGPCL in karst plateau mountainous areas. Based on the difficulty of restoration and cropping attributes, non-grain cultivated land was classified into four categories: planted non-grain crops (PNGC), unplanted cultivated land (UCL), engineering recoverable (ENR), and immediately recoverable (IMR). Using spatial autocorrelation, hotspot analysis, and the geographic detector method, the spatial distribution and clustering patterns of each type were revealed, and their driving factors were explored. The results show that: (1) non-grain cultivated land accounts for 27.78% of total cultivated land in Guizhou, with ENR having the largest area and readily recoverable land the smallest; the conversion of cultivated land to forest and orchard land is significant; (2) non-grain cultivated land exhibits strong spatial correlation and clustering, with hotspots in the central and northeastern regions, cold spots in the southeast regions, and no significant clustering elsewhere; (3) among multiple driving factors, distance to residential areas is the main factor influencing the spatial distribution of non-grain cultivated land, with different dominant drivers for each type, though some internal correlations exist. This study provides differentiated regional references and policy implications for sustainable agriculture and land use in karst plateau mountainous areas.