Abstract:[Purposes] In order to alleviate the seasonal and regional water shortage problems, improve the development and utilization efficiency of regional water resources, and enhance the ability to ensure water supply security in the northern part of Dongting Lake Area (NDLA), this study proposes an optimal plan for allocating water resources scientifically and reasonably. [Methods] This study aims at achieving the minimum regional comprehensive water shortage rate with the polder as the basic water allocation unit. This study constructed the optimal water resources allocation model for the multi-gate & pump system in the NDLA based on the 0-1 planning strategy, simulated the demand-supply relationship of regional water resources under different scenarios, and analyzed the optimal water resources allocation schemes and spatiotemporal features of water shortages. [Findings] The total water shortage in the NDLA is small, but the spatial and temporal characteristics of the water shortages are obvious. The most serious water shortage occurs in Anhua Polder, Nanding Polder, Jichen Polder, Yule Polder and Hekang Polde with the water shortage rate of 70%, 29%, 29%, 27% and 20% respectively in the current extremely dry years (P=90%). Water shortage mainly occurs in the beginning (April and May) and the ending (October) of the irrigation period. Considering the impacts of short-term and long-term water supplement projects, the spatial extent and duration of water shortage in the NDLA are greatly reduced with significant benefits. The rational analyses indicate that the total water supply of each optimized water supply scheme meet the red lines of water consumption in the study area. [Conclusions] The seasonal and regional water shortage in the NDLA is serious under the current conditions. Considering the benefits of the short-term and long-term water supplement projects, the contradiction between water demand and supply will be alleviated, and a balance between water demand and supply will be basically achieved in the NDLA.