中国生态农业学报（中英文）2025，Vol.33Issue(5)：939-948,10.DOI:10.12357/cjea.20240585
基于Google Earth Engine的玉米洪涝灾害精细化评估
Refined evaluation of maize flood disaster based on Google Earth Engine:A case study of the"23·7"heavy precipitation process in Baoding City,Hebei Province
摘要
Abstract
Considering the frequent occurrence of extreme weather events,the assessment of agricultural disasters caused by large-scale flooding is crucial for food production,agricultural insurance,and disaster prevention and mitigation.From July 29 to August 2,2023,Baoding City experienced significant rainfall and flooding("23.7"flood disaster),resulting in substantial losses.In this study,we focused on this event,aiming to develop a rapid assessment method for maize disaster damage using remote sensing technology,specifically leveraging the Google Earth Engine(GEE)platform.The goal of this study is to provide a scientific basis for agricultural disaster management and post-disaster recovery.Relying on the GEE platform,this study utilized Landsat satellite data and agricul-tural statistics from 2016 to 2020 to examine the correlation between maize yield and two vegetation indices:NDVI(normalized dif-ference vegetation index)and EVI2(enhanced vegetation index).The extent of maize damage was delineated by analyzing the differ-ences in vegetation indices between normal and disaster years.Furthermore,Sentinel-2 data were incorporated to classify the maize complete yield loss and reduction grades in the affected areas using supervised classification of remote sensing images and the natur-al breakpoint technique.The results indicated that:1)NDVI was strongly positively correlated with maize yield,with a correlation coefficient of 0.841(P＜0.01),demonstrating its suitability for maize yield inversion.2)Spectral analysis revealed varying degrees of maize yield reduction across Baoding.The eastern part of the city(Zhuozhou and Gaobeidian)experienced the most severe yield re-ductions,whereas the central and southern parts of the city were less affected.3)The"23.7"flood disaster caused a complete yield loss of maize over an area of nearly 45 000 hm2 in Baoding,representing about 5％of the total farmland.Additionally,approximately 66％of farmland experienced yield reductions,highlighting the widespread impact of the disaster.This study underscores the import-ance of leveraging remote sensing technology and cloud-based platforms such as GEE for improving agricultural resilience in the face of increasing climate variability and extreme weather events,and provides a rapid and reliable methodological framework for assess-ing maize damage from heavy rainfall and mapping damage distribution.This offers valuable insights for large-scale crop damage as-sessments and serves as a reference for future disaster response.关键词
灾害评估/洪涝/Google Earth Engine/卫星遥感/植被指数/玉米Key words
disaster assessment/flood/Google Earth Engine/satellite remote sensing/vegetation index/maize分类
信息技术与安全科学引用本文复制引用
周琛,司丽丽,赵亮,郎紫晴,付真真..基于Google Earth Engine的玉米洪涝灾害精细化评估[J].中国生态农业学报（中英文）,2025,33(5):939-948,10.基金项目
中国气象局创新发展专项(CXFZ2023J071)资助 This study was supported by the Innovation Development Project of China Meteorological Administration(CXFZ2023J071). （CXFZ2023J071）
