Abstract: In order to scientifically express the uncertainty and risk information in flood forecasting, this paper implements a probabilistic forecast on the inflowing flood to Danjiangkou Reservoir based on the Copula Bayesian forecast processor.The probabilistic forecast results are evaluated in terms of the accuracy of the expected value, the goodness of the forecast interval and the overall performance, and a risk strategy analysis on the reservoir scheduling is carried out.The results show that the uncertainty of the probabilistic flow forecasts of Danjiangkou Reservoir increases and the accuracy decreases as the forecast period extends.The probabilistic forecast intervals of each forecast period are generally reasonable and reliable, with the coverage rate of the intervals exceeding 0.87.Compared with the deterministic forecast, the accuracy of the probabilistic forecast expectation is slightly improved, and the Continuous Ranked Probability Score(CRPS)values for different forecast periods are always smaller than the Mean Absolute Error(MAE)of the deterministic forecast, with a reduction of more than 25%.The research results can provide risk information on the reservoir's water level exceeding the target level under different scenarios, which can provide a reference for scientific risk scheduling decisions.