Global Precipitation Nowcasting of Integrated Multi-satellitE Retrievals for GPM: A U-Net Convolutional LSTM Architecture

TL;DR

This paper introduces a deep learning model combining U-Net and ConvLSTM for near-global 30-minute precipitation nowcasting with a 4-hour lead, leveraging satellite and GFS data to improve extreme event prediction.

Contribution

It proposes a novel U-Net ConvLSTM architecture trained on IMERG and GFS data, analyzing the effects of different loss functions and physical variables on precipitation nowcasting performance.

Findings

Regression network captures light precipitation well.

Classification network outperforms in extreme precipitation nowcasting.

Including physical variables improves long-term nowcasting accuracy.

Abstract

This paper presents a deep learning architecture for nowcasting of precipitation almost globally every 30 min with a 4-hour lead time. The architecture fuses a U-Net and a convolutional long short-term memory (LSTM) neural network and is trained using data from the Integrated MultisatellitE Retrievals for GPM (IMERG) and a few key precipitation drivers from the Global Forecast System (GFS). The impacts of different training loss functions, including the mean-squared error (regression) and the focal-loss (classification), on the quality of precipitation nowcasts are studied. The results indicate that the regression network performs well in capturing light precipitation (below 1.6 mm/hr), but the classification network can outperform the regression network for nowcasting of precipitation extremes (>8 mm/hr), in terms of the critical success index (CSI).. Using the Wasserstein distance, it is shown that the predicted precipitation by the classification network has a closer class probability distribution to the IMERG than the regression network. It is uncovered that the inclusion of the physical variables can improve precipitation nowcasting, especially at longer lead times in both networks. Taking IMERG as a relative reference, a multi-scale analysis in terms of fractions skill score (FSS), shows that the nowcasting machine remains skillful (FSS > 0.5) at the resolution of 10 km compared to 50 km for GFS. For precipitation rates greater than 4~mm/hr, only the classification network remains FSS-skillful on scales greater than 50 km within a 2-hour lead time.