A New Long-term Radar Reflectivity Nowcasting Method based on DeepRNN

An accurate nowcasting on mesoscale weather events is significant for many social situations. Good prediction of radar reflectivity is essential for the development of mesoscale-weather-event forecast.

Different from traditional radar exploration methods, a new method named DeepRNN which is tailored on nowcasting problem is proposed.This study converts the nowcasting task to an images series prediction task.

Unlike straight input an images series to the model, we split the images prediction to spatial learning and temporal predictions. The spatial information is extracted by a convolutional neural network (CNN) to learn the static information of the cloud in single images such as structure, intensity and location. The motion information is transformed from a recurrent neural network (RNN) model to learn the variation of the extracted high-level information between frames.

Eventually, the Representation Bridge Model (RBM) module in our DeepRNN framework further splits the spatial and temporal information flow, and connects them by bridges.

To ensure the gradient flow smoothly we create temporal dropout(TD) rate to randomly dropout some connections. Traditional statistic scores are calculated to evaluate the performance of the proposed method. The results show that this method improves around 10% on each target (CSI, POD, FAR) than other state-of-the-art and conventional radar exploration methods on 2 hours forecasting of radar echo. In the future , this new method will be improved to incorporate the information from the NWP.