Inflow combination forecast of reservoir based on SWAT, XAJ model and Bayes model averaging method

Abstract Inflow forecast plays an indispensable role in reservoir operation. Accuracy and effectiveness of model prediction play a decisive it. In this paper, the certainty coefficient, root mean square error (RMSE), absolute deviation (MAE) Nash-Suthcliffe coefficient (NSE) are used to consider effect Soil Water Assessment Tool (SWAT) Xin'anjiang (XAJ) on inflow Jinxi Reservoir. Results indicate that XAJ is 90.95% SWAT 84.98%. The NSE 84.42% 82.33% which superior model. view this, based two hydrological models results, Bayes Model Averaging (BMA) method study combination forecast. SWAT-XAJ coupling BMA proposed. During calibration period verification period, compared with single model, maximum increment deterministic 2.87%, decrement RMSE 82.00, 6.80%. This shows new can effectively overcome shortcomings give full advantages different forecasting models.