Abstract: Considering the negative effects of current integrated soil erosion control, assessing soil and water conservation strategies is essential for choosing suitable and effective ecological management measures. Using Yongsheng County, located in the Southwest Alpine Canyon Area, as a case study, and leveraging GIS as support, a multi-model evaluation mode was developed. This model integrated the weight of evidence model, Logistic regression model and evidence—Logistic regression coupling model to identify, evaluate the existing the economic forest measures and in terms of site suitability and feasibility of social development. The findings of the study are as follows: (1) Through the radar chart of frequency distribution of economic forest measures, 12 evaluation indicators exhibiting a significant fluctuating trend across different classification intervals were identified. (2) Pearson"s correlation analysis, A-C test, variance inflation factor (VIF) and tolerance (TOL), the independence of 12 evaluation indicators has been verified. Through the fitting of indicators coefficient results, the weight values of slope and land use type indicators are larger. (3) The Sridevi Jadi empirical probability method, ten-fold cross-validation and ROC curve approach were applied to assess the model"s accuracy, demonstrating that all three models accurately represent the suitability of economic forestry practices. Among them, the result of the evidence—Logistic regression coupling model shows better evaluation ability than the two single models. The slope indicator plays a leading role (RD=1.39). (4) The economic forest measures implemented in Yongsheng County shows a good trend as a whole, the spatial distribution pattern is high in the southwest and low in the northeast, and the suitability is characterized by staggered distribution of high value and low value, the proportion of the economic forest measures and the suitability grade are increasing step by step. The variation of the proportion of suitability grade in the unsuitable—suitable range, evidence—Logistic regression coupling model with a dramatic increase of 17.83% which shows superior partitioning performance. This study provides a reference for the evaluation of soil and water conservation measures at the county scale in the Southwest Alpine Canyon Area, and has guiding significance for the control of soil erosion and the construction of ecological restoration.