Accurate precipitation forecasting can better reflect climate change trends, provide timely and effective environmental information for management decisions, and prevent flood and drought disasters. In this paper, we propose a short-term regional precipitation prediction model based on wind-improved spatiotemporal convolutional network. Among them, the improved Graph Convolution Network (GCN) integrates the effects of wind direction and geographic location at past moments to capture the spatial dependence, whilst the Gated Recurrent Unit (GRU) captures the temporal dependence by learning the dynamic changes of data. The spatio-temporal memory flow module and attention module are added to capture spatial deformation and temporal variation more accurately, thereby better matching the physical properties of precipitation. Experimental results on real data sets show that the proposed model can handle complex spatial dependence and temporal dynamic changes, better learn the temporal and spatial characteristics of precipitation data, and achieve better prediction results.