Introduction - Cotton is Benin's leading export crop grown in different climatic zones. As a result of poor farming practices in agroecosystems leading to a decline in soil fertility, coupled with the phenomenon of climatic hazards, seed cotton yields of cultivated varieties are low. Given the context of soil degradation, agroecological practices are proposed as an alternative to conventional cropping systems. The emerging challenge for breeders is therefore to select new cultivars adapted to innovative cropping systems (Gouleau et al., 2021). Crop simulation models can be used to assess growth and yield of different crop genotypes (G) in different environments (E) by using environment-specific weather, soil, and crop management (M) practices (Boote et al., 2001) and help understanding the direct effect of each plant trait or a combination of traits on the G by E interactions for crop yield and other complex traits. The objectives of this study were (i) to evaluate the relative contribution of plant traits driving the G by E by M interactions for cotton yield using CSM-CROPGRO-cotton model from DSSAT, and (ii) to identify morpho-physiological traits for yield enhancement in agroecological cropping systems in Benin. Materials and Methods - The experimental study was conducted from 2020 to 2023, in two sites (Savalou and Soaodou) representative of contrasted cotton-growing regions of Benin. Approach consisted of (1) carrying out field experiments for the calibration and validation of CSM-CROPGRO-cotton model and determining the yield performance of cotton genotypes (2) launching a sensitivity analysis of the genetic coefficients from the crop model to determine the best values for traits adapted to agroecological cropping systems. According to a split plot design, the experiment compared four cropping systems with cowpea/maize//cotton rotation, namely Conventional Tillage (CT) with a plough, Conventional Tillage with Incorporation of biomass (CTI), Conservation Agricu