This paper addresses how much flood water can be conserved for use after the flood season through the operation of reservoir by taking into account the residual flood control capacity (the difference between flood conveyance capacity and the expected inflow in a lead time). A two‐stage model for dynamic control of the flood‐limited water level (the maximum allowed water level during the flood season, DC‐FLWL) is established considering forecast uncertainty and acceptable flood risk. It is found that DC‐FLWL is applicable when the reservoir inflow ranges from small to medium levels of the historical records, while both forecast uncertainty and acceptable risk in the downstream affect the feasible space of DC‐FLWL. As forecast uncertainty increases (under a given risk level) or as acceptable risk level decreases (under a given forecast uncertainty level), the minimum required safety margin for flood control increases, and the chance for DC‐FLWL decreases. The derived hedging rules from the modeling framework illustrate either the dominant role of water conservation or flood control or the trade‐off between the two objectives under different levels of forecast uncertainty and acceptable risk. These rules may provide useful guidelines for conserving water from flood, especially in the area with heavy water stress. The analysis is illustrated via a case study with a real‐world reservoir in northeastern China.

Flash floods occur frequently and distribute widely in mountainous areas because of complex geographic and geomorphic conditions and various climate types. Effective flash flood forecasting with useful lead times remains a challenge due to its high burstiness and short response time. Recently, machine learning has led to substantial changes across many areas of study. In hydrology, the advent of novel machine learning methods has started to encourage novel applications or substantially improve old ones. This study aims to establish a discharge forecasting model based on Long Short-Term Memory (LSTM) networks for flash flood forecasting in mountainous catchments. The proposed LSTM flood forecasting (LSTM-FF) model is composed of T multivariate single-step LSTM networks and takes spatial and temporal dynamics information of observed and forecast rainfall and early discharge as inputs. The case study in Anhe revealed that the proposed models can effectively predict flash floods, especially the qualified rates (the ratio of the number of qualified events to the total number of flood events) of large flood events are above 94.7% at 1–5 h lead time and range from 84.2% to 89.5% at 6–10 h lead-time. For the large flood simulation, the small flood events can help the LSTM-FF model to explore a better rainfall-runoff relationship. The impact analysis of weights in the LSTM network structures shows that the discharge input plays a more obvious role in the 1-h LSTM network and the effect decreases with the lead-time. Meanwhile, in the adjacent lead-time, the LSTM networks explored a similar relationship between input and output. The study provides a new approach for flash flood forecasting and the highly accurate forecast contributes to prepare for and mitigate disasters.