Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting

TL;DR

This paper introduces the ConvLSTM, a novel neural network architecture that effectively models spatiotemporal data for short-term precipitation forecasting, outperforming previous methods.

Contribution

The paper proposes the convolutional LSTM (ConvLSTM), extending traditional LSTM with convolutional structures to better capture spatiotemporal correlations in weather data.

Findings

ConvLSTM outperforms FC-LSTM in precipitation nowcasting.

ConvLSTM surpasses the operational ROVER algorithm.

ConvLSTM effectively models spatiotemporal dependencies.

Abstract

The goal of precipitation nowcasting is to predict the future rainfall intensity in a local region over a relatively short period of time. Very few previous studies have examined this crucial and challenging weather forecasting problem from the machine learning perspective. In this paper, we formulate precipitation nowcasting as a spatiotemporal sequence forecasting problem in which both the input and the prediction target are spatiotemporal sequences. By extending the fully connected LSTM (FC-LSTM) to have convolutional structures in both the input-to-state and state-to-state transitions, we propose the convolutional LSTM (ConvLSTM) and use it to build an end-to-end trainable model for the precipitation nowcasting problem. Experiments show that our ConvLSTM network captures spatiotemporal correlations better and consistently outperforms FC-LSTM and the state-of-the-art operational ROVER algorithm for precipitation nowcasting.