Data collection and statistical investigation were employed to summarize the types, distribution, and scale of forestry economy in the southwest alpine canyon region. This paper analyzed and predicted the amount of soil erosion in different months and periods based on the influencing factors of soil erosion and the CSLE model. Results have shown that: (1) The total area of forest floor economy in the high mountain and canyon area in the southwest exceeded 240000 hm². The main models included forest floor planting, forest floor breeding, and forest floor collection, among which the area of forest floor planting was 247500 hm², the scale of forest floor breeding was 3 million heads, and the scale of forest floor collection was 85900 tons per year. According to the different types of forest floor planting and breeding, it could be divided into 92 types. The most widely distributed was forest floor collection, and the largest scale was forest floor planting. The development of forest floor economy still had huge potential; (2) The surface disturbance caused by understory economic development mainly included surface preparation, soil loosening and weeding, harvesting disturbance, etc., which increased the risk of soil erosion by reducing surface coverage and destroying soil structure; (3) The predicted annual soil erosion modulus values for the low disturbance undergrowth economic model 1 (13.42 t), low disturbance undergrowth economic model 2 (2.88 t), moderate disturbance undergrowth economic model 1 (20.37 t), moderate disturbance undergrowth economic model 2 (9.95 t), and high disturbance undergrowth economic model (61.15 t) were compared to the incremental values for undeveloped forestland: 12.30 t, 1.75 t, 19.25 t, 8.83 t, and 60.02 t, with the corresponding increases being 1094%, 155.85%, 785.07%, 1670.14%, and 5337.5%. The soil erosion increments were distributed across different periods and months, with the highest increments in June for low disturbance models 1 and 2, moderate disturbance model 2, and high disturbance model, while for moderate disturbance model 1, the highest increment occured in September. In low disturbance model 1, the seedling stage (January-August) accounted for 66.72% of the total soil erosion modulus increment, while in low disturbance model 2, the off-season period (April-September) accounted for 98.86% of the annual increment. In moderate disturbance model 1, the seedling stage (June-August) accounted for 45.84% of the annual increment, and in moderate disturbance model 2, the land preparation and sowing seedling stage (November-December) accounted for 63.25%. These increments were distributed during the overlapping period of soil disturbance and the rainy season. The research findings are of significant guiding value for enhancing the understanding of soil erosion risks in undergrowth economic systems and promoting the rational application and development of soil and water conservation measures in undergrowth development.