SmaAt-UNet: Precipitation Nowcasting using a Small Attention-UNet Architecture
This addresses the problem of accurate precipitation nowcasting for weather forecasting, though it is incremental as it builds on existing UNet architecture.
The paper tackles short-term precipitation forecasting by proposing SmaAt-UNet, a neural network that achieves comparable prediction performance to other models while using only a quarter of the trainable parameters.
Weather forecasting is dominated by numerical weather prediction that tries to model accurately the physical properties of the atmosphere. A downside of numerical weather prediction is that it is lacking the ability for short-term forecasts using the latest available information. By using a data-driven neural network approach we show that it is possible to produce an accurate precipitation nowcast. To this end, we propose SmaAt-UNet, an efficient convolutional neural networks-based on the well known UNet architecture equipped with attention modules and depthwise-separable convolutions. We evaluate our approaches on a real-life datasets using precipitation maps from the region of the Netherlands and binary images of cloud coverage of France. The experimental results show that in terms of prediction performance, the proposed model is comparable to other examined models while only using a quarter of the trainable parameters.