Answers are grounded in this paper and cite their sources.
Using ERA reanalysis data at 0.25°×0.25° resolution with an hourly interval, this study calculates daily physical variables for Bazhong (Sichuan) during May-September of 1990–2019, identifies physical factors with clear correlations, and constructs a new comprehensive anomaly index to examine whether there is a stable relationship between extreme heavy rainfall and its location and intensity. A Physics-Informed Neural Network (PINN) is further explored for quantitative forecasting of extreme heavy rainfall. Results show that: (1) the comprehensive anomaly index is convenient for qualitative diagnosis of extreme heavy rainfall, but its correlation with rainfall intensity and the rainfall center is weak; (2) for the PINN-based quantitative forecast, out-of-sample case tests indicate that larger observed rainfall tends to yield larger errors. The minimum error is 0.11 mm, the maximum error is -81.91 mm, and the mean absolute error is 24.73 mm. The proportions of cases with absolute errors below 10 mm, 20 mm, and 30 mm are 33.3%, 60.0%, and 73.3%, respectively, suggesting practical reference value.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: