EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics-Constrained Deep Learning with Adaptive Z-R Relationship for Accurate and Interpretable Quantitative Precipitation Estimation

Ting Shu, Huan Zhao, Kanglong Cai, Zexuan Zhu

Remote Sensing · 03.01.2026 · 18 (1) · 156

DOI: 10.3390/rs18010156 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 03.01.2026. Verfasst von Ting Shu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Basic and Applied Basic Research Foundation, National Natural Science Foundation of China, Scientific Foundation for Youth Scholars of Shenzhen University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Quantitative precipitation estimation (QPE) from radar reflectivity is fundamental for weather nowcasting and water resource management. Conventional Z-R relationship formulas, derived from Rayleigh scattering theory, rely heavily on empirical parameter fitting, which limits the estimation accuracy and generalization across different precipitation regimes. Recent deep learning (DL)-based QPE methods can capture the complex nonlinear relationships between radar reflectivity and rainfall. However, most of them overlook fundamental physical constraints, resulting in reduced robustness and interpretability. To address these issues, this paper proposes FusionQPE, a novel Physics-Constrained DL framework that integrates an adaptive Z-R formula. Specifically, FusionQPE employs a Dense convolutional neural network (DenseNet) backbone to extract multi-scale spatial features from radar echoes, while a modified squeeze-and-excitation (SE) network adaptively learns the parameters of the Z-R relationship. The final rainfall estimate is obtained through a linear combination of outputs from both the DenseNet backbone and the adaptive Z-R branch, where the trained linear weight and Z-R parameters provide interpretable insights into the model’s physical reasoning. Moreover, a physical-based constraint derived from the Z-R branch output is incorporated into the loss function to further strengthen physical consistency. Comprehensive experiments on real radar and rain gauge observations from Guangzhou, China, demonstrate that FusionQPE consistently outperforms both traditional and state-of-the-art DL-based QPE models across multiple evaluation metrics. The ablation and interpretability analysis further confirms that the adaptive Z-R branch improves both the physical consistency and credibility of the model’s precipitation estimation.

Förderung

  • Guangdong Basic and Applied Basic Research Foundation
  • National Natural Science Foundation of China
  • Scientific Foundation for Youth Scholars of Shenzhen University
  • Guangdong Meteorological Bureau General Project
  • China Meteorological Administration Youth Innovation Team

Zitieren

APA

Shu, T., Zhao, H., Cai, K., & Zhu, Z. (2026). Physics-Constrained Deep Learning with Adaptive Z-R Relationship for Accurate and Interpretable Quantitative Precipitation Estimation. Remote Sensing, 18(1), 156. https://doi.org/10.3390/rs18010156

Vancouver

Shu T, Zhao H, Cai K, Zhu Z. Physics-Constrained Deep Learning with Adaptive Z-R Relationship for Accurate and Interpretable Quantitative Precipitation Estimation. Remote Sensing. 2026;18(1):156. doi:10.3390/rs18010156

BibTeX
@article{Shu2026,
  author = {Shu, Ting and Zhao, Huan and Cai, Kanglong and Zhu, Zexuan},
  title = {Physics-Constrained Deep Learning with Adaptive Z-R Relationship for Accurate and Interpretable Quantitative Precipitation Estimation},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {1},
  pages = {156},
  doi = {10.3390/rs18010156}
}
Kostenlos starten →