An attention mechanism based convolutional network for satellite precipitation downscaling over China

TL;DR

This paper introduces an attention-based convolutional neural network designed to improve the spatial resolution of satellite-derived precipitation data over China, integrating surface characteristics and in-situ measurements for enhanced accuracy.

Contribution

The study develops a novel end-to-end attention mechanism convolutional network with a specialized loss function for effective precipitation downscaling from coarse satellite data.

Findings

Significantly outperforms baseline methods in accuracy

Effectively incorporates surface features and in-situ data

Enhances spatial resolution of precipitation estimates

Abstract

Precipitation is a key part of hydrological circulation and is a sensitive indicator of climate change. The Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (GPM) mission (IMERG) datasets are widely used for global and regional precipitation investigations. However, their local application is limited by the relatively coarse spatial resolution. Therefore, in this paper, an attention mechanism based convolutional network (AMCN) is proposed to downscale GPM IMERG monthly precipitation data. The proposed method is an end-to-end network, which consists of a global cross-attention module, a multi-factor cross-attention module, and a residual convolutional module, comprehensively considering the potential relationships between precipitation and complicated surface characteristics. In addition, a degradation loss function based on low-resolution precipitation is designed to physically constrain the network training, to improve the robustness of the proposed network under different time and scale variations. The experiments demonstrate that the proposed network significantly outperforms three baseline methods. Finally, a geographic difference analysis method is introduced to further improve the downscaled results by incorporating in-situ measurements for high-quality and fine-scale precipitation estimation.