Abstract: Cultivated land abandonment posed a growing challenge to land protection and food security in economically developed regions undergoing rapid urbanization and non-agricultural transformation. Previous studies had often paid insufficient attention to the long-term identification of abandoned cultivated land, the assessment of associated production capacity loss, and the determination of governance priorities across spatial scales. Taking Guangdong Province as the study area, this study identified cultivated land abandonment from 2000 to 2024 using a long-term land use/land cover dataset with a spatial resolution of 30 m. A temporal consistency test and a four-year moving-window procedure were applied to reduce classification errors and distinguish abandonment from cultivated land converted to construction land. County-level cultivated land abandonment rates were then calculated. Exploratory spatial data analysis, global and local spatial autocorrelation analysis, and hot and cold spot analysis were jointly employed to examine spatial dependence, clustering patterns, and temporal changes. Net primary productivity (NPP) was converted into crop production capacity by incorporating crop-specific moisture content, harvest index, and aboveground biomass allocation fraction, thereby estimating production capacity loss from cultivated land abandonment. A cultivated land protection urgency index was further introduced, and a dual-scale zoning-based regulation framework linking plot-level diagnosis with county-level governance was constructed.The results show that: 1) From 2000 to 2024, county-level cultivated land abandonment rates in Guangdong exhibited a phased trajectory of low-level fluctuation, rapid increase, marked decline, and renewed growth. The abandonment rate reached a high point during the study period in 2015, declined substantially in 2020, and rebounded in 2024. Spatially, cultivated land abandonment gradually shifted from a scattered distribution to increasing concentration in the mountainous areas of northern Guangdong, parts of western Guangdong, the hilly areas of eastern Guangdong, and the periphery of the Pearl River Delta. 2) Cultivated land abandonment in Guangdong exhibited significant positive spatial autocorrelation. Low-low clusters remained dominant over the long term, whereas high-high clusters intensified markedly during 2010–2015 and showed a localized resurgence in 2024. Hot spots repeatedly occurred in the mountainous areas of northern Guangdong, parts of western Guangdong, and the Leizhou Peninsula, while the core areas of the Pearl River Delta were predominantly identified as cold spots or statistically insignificant areas. 3) Production capacity loss generally changed in the same direction as the abandonment rate, but the two spatial patterns did not fully overlap because abandoned cultivated land differed in its original productive capacity. Northern Guangdong showed the clearest overlap between high abandonment and high production capacity loss, whereas some counties in eastern Guangdong exhibited relatively high losses despite not consistently recording the highest abandonment rates. 4) The dual-scale zoning results indicate that the governance of cultivated land abandonment in Guangdong should shift from a single area-oriented approach toward an approach emphasizing protection value, production capacity loss, and governance priority. At the plot scale, abandoned cultivated land was classified into five governance.At the county scale, the study area was divided into four governance zones, each matched with a corresponding resource allocation strategy. This study provides a scientific reference for the precise identification of cultivated land abandonment, the prevention and control of production capacity loss, and differentiated governance in economically developed regions.