中国农业气象 ›› 2025, Vol. 46 ›› Issue (3): 420-431.doi: 10.3969/j.issn.1000-6362.2025.03.013

• 农业气象灾害栏目 • 上一篇    下一篇

基于水动力模型的农作物暴雨洪涝灾害损失评估:以江汉平原为例

秦鹏程,周月华,刘火胜,夏智宏   

  1. 1.武汉区域气候中心,武汉 430074;2.武汉市公众气象服务中心,武汉 430040
  • 收稿日期:2024-04-08 出版日期:2025-03-20 发布日期:2025-03-19
  • 作者简介:秦鹏程,E-mail:qinpengcheng027@163.com
  • 基金资助:
    湖北省自然科学基金项目(2023AFD098);中国气象局创新发展专项项目(CXFZ2023J051);中国气象局青年创新团队项目(CMA2023QN15)

Flood Loss Assessment for Crops Based on Hydrodynamic Modeling: A Case Study in the Jianghan Plain

QIN Peng-cheng, ZHOU Yue-hua, LIU huo-sheng, XIA Zhi-hong   

  1. 1. Wuhan Regional Climate Center, Wuhan 430074, China; 2. Wuhan Public Meteorological Service Center, Wuhan 430040
  • Received:2024-04-08 Online:2025-03-20 Published:2025-03-19

摘要:

洪涝灾害的形成涉及因素多、灾害链复杂,具有较大的时空变异性,构建反映致灾过程空间异质性及其动态变化的机理性模型是准确估算灾害损失的关键。本研究通过将灾害损失表达为淹没水深持续时间的函数构建洪涝脆弱性曲线,利用国际水灾与风险管理中心基于二维扩散波方程构建的降雨−径流−淹没模拟模型(RRI)模拟洪水淹没动态,结合作物种植分布及其所处生育阶段建立格点尺度洪涝损失定量评估模型,以江汉平原地区典型洪涝灾害过程为例对模型进行检验结果表明:RRI模型较好地模拟强降水后洪峰的形成和消退过程及地表淹没动态,流域控制断面径流量模拟误差为−14.8%11.5%,地表淹没范围模拟精度在80%以上,模拟水深匹配率达84.2%87.1%农作物损失空间分布基本合理,地市级农作物受灾率、成灾率和绝收率估算偏差分别为−33.8%6.4%−10.8%9.5%−6.0%1.8%。该方法具有精细化、定量化和动态评估等优点,可用于面向致灾过程的洪涝灾害快速评估、预估及复盘研究。

关键词: 暴雨洪涝, 损失评估, 水动力模型, 江汉平原, 农作物

Abstract:

Crop loss assessment is critical for decision making in flooding management. From the perspective of disaster chain, flooding damage is a complex interaction of hazard factors (e.g., extreme precipitation), local topographic attributes and vulnerability of affected bodies, and thus characterized by large temporal and spatial variations. Developing a physically based modelling chain that can capture the dynamic evolution and spatial heterogeneity of the disaster process is critical for timely and efficient emergency response to flooding prevention. This study presented a modelling framework for estimating crop loss due to flooding, by coupling the flooding vulnerability curves with the Rainfall−Runoff−Inundation (RRI) model developed by the International Center for Water Hazard and Risk Management. The flooding vulnerability curve was shown as a function of inundation depth, duration and crop stage. A quantitative assessment of crop loss at gridded scale was established by integrating the inundation maps, crop distribution, and flooding vulnerability curves. The framework was applied to two representative flooding events on the Jianghan plain to demonstrate its capability to estimate crop losses due to rainstorm−induced flooding. The results showed that the RRI model could reasonably simulate the formation and retreat of the flooding peak as well as the surface inundation dynamics in accordance with the rainstorm, with the simulation error ranging from −14.8% to 11.5% for the runoff, the simulation accuracy exceeding 80% for the inundation area, the matching rate ranging from 84.2% to 87.1% for the inundation depth, and the estimated deviation of crop loss rate were −33.8% to 6.4%, −10.8% to −9.5%, and −6.0% to 1.8% for areas covered, areas affected and areas of total crop failure, respectively. The method proposed in this study provides a fundamental support for the rapid assessment and risk early warning for flooding mitigation and post−disaster reconstruction.

Key words: Rainstorm and flooding, Loss assessment, Hydrodynamic model, Jianghan plain, Crops