海南省万泉河流域山洪灾害预警复盘分析：以2024年台风“潭美”为例

Post-event analysis of flash flood early warning in Wanquan River Basin, Hainan Province: A case study of typhoon “Trami” (2024)

摘要

【目的】为评估极端天气背景下山洪灾害监测预警系统的响应效果,以海南省万泉河流域为研究区,以2024年第20号台风“潭美”残涡与冷空气共同影响引发的暴雨洪水过程为案例,基于中国山洪水文模型(CNFF)与海南省山洪灾害监测预警平台,构建了涵盖“预报—监测—预警发布—灾情验证”四阶段的预警评估框架。【方法】研究集成降雨监测数据、水文模型模拟结果及96个重点村庄的实地调研信息,以行政村为评价单元,对警戒(准备转移)及以上等级预警的命中情况进行分析。【结果】结果表明:平台共发布115次“准备转移(警戒)”及以上等级的预警,覆盖70个防治行政村,其中50个村庄实际发生灾情,对应预警命中率为71.43%。【结论】监测预警平台在重点风险区具备较强的趋势识别能力,但在村级响应适配性和阈值设定合理性方面存在不足。研究提出预警阈值动态调整、洪水过程模拟优化以及多源数据融合等改进方向,以提升系统在极端气象条件下的适应性与预警效果。可为热带季风区山洪灾害监测预警系统优化及风险治理能力提升提供参考。

Abstract

[Objective] To evaluate the response effectiveness of the flash flood monitoring and early warning system under extreme weather conditions, the Wanquan River Basin in Hainan Province is taken as the study area, and the rainstorm flood process triggered by the combined influence of the residual circulation of Typhoon “Trami”(2024, No. 20) and a cold air mass is used as a case study. Based on the China National Flash Flood Hydrological Model(CNFF) and the Hainan Flash Flood Monitoring and Early Warning Platform, a four-stage early warning evaluation framework covering “forecasting-monitoring-warning issuance-disaster verification” is established. [Methods]Rainfall monitoring data, hydrological model simulation result, and field survey information from 96 key villages were integrated. Using the administrative village as the evaluation unit, the hit rate of warnings at the alert level(“ prepare for evacuation”) and above was analyzed. [Results] The result showed that: the platform issued a total of 115 warnings at the level of “prepare for evacuation(alert)” and above, covering 70 administrative villages under prevention and control. Among them, 50 villages actually experienced disasters, corresponding to a warning hit rate of 71. 43%. [Conclusion] The monitoring and early warning platform has relatively strong capability in trend identification in key risk areas, but deficiencies exist in village-level response adaptability and the rationality of threshold settings. Improvement directions are proposed, including dynamic adjustment of early warning thresholds, optimization of flood process simulation, and multi-source data fusion, so as to enhance the adaptability and early warning effectiveness of the system under extreme meteorological conditions. The findings can provide a reference for the optimization of flash flood monitoring and early warning systems and the improvement of risk governance capacity in tropical monsoon regions.

关键词

山洪灾害预警 / CNFF模型 / 台风暴雨 / 预警体系 / 预警效果评估 / 预警阈值 / 万泉河流域 / 影响因素

Key words

flash flood early warning / CNFF model / typhoon rainstorm / early warning system / early warning effectiveness evaluation / early warning threshold / Wanquan River Basin / influencing factors

基金资助

国家自然科学基金项目(42271095)

国家自然科学基金项目(52239006)