Region- and duration-specific disaster-rainstorm criteria are central to flood defense in humid hilly monsoon catchments. In the Poyang Lake Basin (largely coterminous with Jiangxi Province), 55 disaster-verified events (2010–2022) are combined with 190 rainstorm events (2011–2020) from 22 hourly gauges in Yichun. Four dimensions are examined—temporal clustering, duration structure, spatial differentiation and extreme indices—and a robust lower envelope combined with a coping-capacity ranking yields prefecture-level multi-duration warning thresholds. The events cluster in May–July (81%) and peak in June (36%), with daily maxima mainly from night to late morning; 76% last no more than 36 h and split into short- (24%), intermediate- (69%) and long-duration (7%) patterns. Ganzhou and Shangrao hold 51% of events, 37 counties are hot spots, point extremes cluster on the windward ranges of northeastern and northwestern Jiangxi, and the Yichun gauges show heterogeneous, right-skewed indices. Province-wide thresholds are 35, 70, 100, 120 and 140 mm for 1, 3, 6, 12 and 24 h, event totals exceed 200 mm, and differentiated values for all 11 prefectures attain a 100% back-evaluation hit rate on the 55 calibration events; this is an inherent fit property on the calibration dataset, not independent validation. Bootstrap 90% intervals span only 9–11% of the thresholds, indicating stable baselines. Because independent non-disaster control samples are not yet available and FAR/CSI remain to be computed, the thresholds should be treated as preliminary references rather than ready-to-use operational criteria. The approach is transferable to data-sparse humid mountainous basins.
No takes yet. Share an insight, caveat, or question.
liao et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: