Abstract: Slope cropland is crucial for agricultural production and soil erosion control in mountainous regions. Revealing its spatial distribution and driving mechanisms is essential for sustainable land resource use and eco-economic synergistic development. However, existing slope extraction methods often fail to accurately reflect the overall slope characteristics of cultivated plots, limiting precise identification of slope cropland and analysis of its distribution across different scales. To address this, this paper proposes an automatic batch-processing algorithm for calculating the overall slope of cultivated plots based on DEM, aiming to improve slope accuracy. Using this algorithm, slope cropland was extracted, and methods including kernel density estimation, spatial autocorrelation, and geodetector were integrated to systematically reveal the inventory characteristics, multi-scale distribution patterns, driving factors, and interaction effects of slope cropland in the Southwest Alpine Canyon Area of China. The results show that: 1) The slope values derived from our algorithm agree well with manual measurements (R²=0.9, MAE=2.5°), effectively overcoming the shortcomings of traditional methods that overestimate slopes for large gentle plots and underestimate slopes for small steep plots. 2) A total of 918,000 slope cropland plots, covering an area of 12,500 km², are distributed in the study area, accounting for 72.6% of all cultivated land. They generally exhibit a “denser in the south, sparser in the north” spatial pattern, with high concentrations in southern counties (occupying 7.4% of the regional area) and watersheds (occupying 13.2% of the regional area) where population is larger and elevation is lower. 3) Slope cropland is mainly distributed in subtropical climate zones, with soil types dominated by red soil (25.9%), brown soil (16.9%), yellow brown soil (15.7%), and yellow soil (10.2%). Notably, 43.3% of slope cropland exceeds the prohibited reclamation slope, and about one-third reaches slopes above 30°. It is concentrated in the mid-mountain belt at 2000–3500 m (56.7%) and the low-mountain river-valley belt below 2000 m (33.2%). 4) At the county scale, the differentiation of slope cropland is primarily controlled by population density, followed by topographic factors; at the watershed scale, it is influenced by the synergistic effects of multiple factors. All factors exhibit mainly interactive enhancement effects on slope cropland distribution, with the explanatory power notably increased when population density interacts with soil, topography, and hydrothermal conditions. This study can provide support for the rational utilization and scientific remediation of slope cropland in mountainous areas such as the Southwest Alpine Canyon Area of China.