Accurate identification of the interaction between vegetation and soil is an important premise of wetland restoration and protection. Hydrologic situation is the key factor affecting the distribution of wetland vegetation, while vegetation distribution pattern can affect the accumulation and occurrence form of soil nutrients. Based on soil environmental factors, optimized Random Forests was used to predict the distribution of Phalaris arundinacea Linn and Triarrhena lutarioriparia L. Liu, which grow in low elevation and high elevation of shoaly wetlands of Poyang Lake respectively. And then we analyzed soil nutrient accumulation under the two vegetation. Results showed that the predication accuracy by Random Forest was 89.6% for Phalaris arundinacea Linn and 89.3% for Triarrhena lutarioriparia L. Liu. According to the model, The importance of soil factors which closely related to Phalaris arundinacea Linn was in order of total potassium > ammonia nitrogen > organic matter > soil water content > total nitrogen > available phosphorus > total phosphorus > pH > nitrate nitrogen, and that for Triarrhena lutarioriparia L. Liu was in order of total potassium > pH > organic matter > total nitrogen > total phosphorus > nitrate nitrogen > ammonia nitrogen > soil water content > available phosphorus. From the partial dependent plot, pH value under Phalaris arundinacea Linn was significantly higher than that under Triarrhena lutarioriparia L. Liu. Distribution of Phalaris arundinacea Linn was negatively correlated to total nitrogen and ammonia nitrogen; distribution of Triarrhena lutarioriparia L. Liu was positively correlated to total nitrogen while week relationship between Triarrhena lutarioriparia L. Liu and ammonia nitrogen was found. Total phosphorus was positively correlated to Phalaris arundinacea Linn while negatively correlated to Triarrhena lutarioriparia L. Liu. Weak relationships were found between available phosphorus and the two kinds of vegetation. Total potassium was negatively correlated with Phalaris arundinacea Linn while positively correlated with Triarrhena lutarioriparia L. Liu. Soil water content was positively correlated to Phalaris arundinacea Linn while negatively correlated to Triarrhena lutarioriparia L. Liu. Random Forests is suitable for simulating complex nonlinear relation, and can show the partial dependence relationship between individual soil factors and vegetation, so can explain the results in the ecological sense. Random Forests is of great value in the study of the interaction between wetland vegetation and environmental factors.