A Dual-TransUNet Deep Learning Framework for Multi-Source Precipitation Merging and Improving Seasonal and Extreme Estimates

2026-02-04

AI Summary

This study addresses the spatial heterogeneity biases and limited capability in simulating extreme precipitation that hinder the hydroclimatic applications of multi-source precipitation products. To overcome these limitations, we propose the first two-stage TransUNet-based fusion framework (DDL-MSPMF): the first stage predicts the probability of precipitation occurrence, while the second stage integrates multiple precipitation products with near-surface physical variables from ERA5 to generate a 0.25° daily precipitation estimate for China over 2001–2020. The method substantially improves accuracy in capturing both seasonal patterns (R = 0.75, RMSE = 2.70 mm/day) and extreme events (>25 mm/day), enhances equitable threat scores for heavy rainfall in eastern China, and successfully reproduces the spatial structure of the 2021 Zhengzhou rainstorm. It also demonstrates robust performance in data-scarce regions such as the Tibetan Plateau, exhibiting strong physical interpretability (via SHAP analysis) and generalization capability.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Multi-source precipitation products (MSPs) from satellite retrievals and reanalysis are widely used for hydroclimatic monitoring, yet spatially heterogeneous biases and limited skill for extremes still constrain their hydrologic utility. Here we develop a dual-stage TransUNet-based multi-source precipitation merging framework (DDL-MSPMF) that integrates six MSPs with four ERA5 near-surface physical predictors. A first-stage classifier estimates daily precipitation occurrence probability, and a second-stage regressor fuses the classifier outputs together with all predictors to estimate daily precipitation amount at 0.25 degree resolution over China for 2001-2020. Benchmarking against multiple deep learning and hybrid baselines shows that the TransUNet - TransUNet configuration yields the best seasonal performance (R = 0.75; RMSE = 2.70 mm/day) and improves robustness relative to a single-regressor setting. For heavy precipitation (>25 mm/day), DDL-MSPMF increases equitable threat scores across most regions of eastern China and better reproduces the spatial pattern of the July 2021 Zhengzhou rainstorm, indicating enhanced extreme-event detection beyond seasonal-mean corrections. Independent evaluation over the Qinghai-Tibet Plateau using TPHiPr further supports its applicability in data-scarce regions. SHAP analysis highlights the importance of precipitation occurrence probabilities and surface pressure, providing physically interpretable diagnostics. The proposed framework offers a scalable and explainable approach for precipitation fusion and extreme-event assessment.

Problem

• multi-source precipitation
• spatially heterogeneous biases
• extreme precipitation estimation
• hydroclimatic monitoring
• precipitation fusion

Innovation

• TransUNet
• multi-source precipitation merging
• extreme precipitation estimation
• dual-stage deep learning
• SHAP interpretability