摘要
【目的】为科学地预防森林火灾和进行应急资源配置,运用空间logistic森林火灾风险概率模型对湖南省森林火灾风险等级进行划分,为县级以下区域进行森林火灾风险区划提供技术方法。【方法】采用logistic回归分析方法,利用湖南省2008-2018年的地表温度数据、植被指数数据、气象数据、人文数据、植被类型数据、森林火灾监测数据等,构建森林火灾风险概率模型,通过森林火灾风险概率分布区间对森林火灾风险等级进行划分。【结果】建立的湖南省空间logistic森林火灾风险概率模型拟合效果较好,在0.05的显著性水平下,经混合检验和Wald检验,logistic森林火灾风险概率模型能显著反映森林火险发生概率;模型相对运行特征(ROC)值为0.779;经栅格图层计算得到湖南省森林火灾风险概率值,并根据概率大小将湖南省森林火险分为极低、低、中、高和极高5个等级。湖南省森林火灾风险概率≥0.6的极高森林火险等级和高森林火险等级主要分布在邵阳市、衡阳市、永州市、株洲南部以及郴州市部分地区;中森林火险等级主要分布在怀化市、娄底市、湘潭市以及长沙市;低森林火险等级和极低森林火险等级主要分布在吉首市、常德市、张家界市、益阳市以及岳阳市。影响湖南省森林火险等级的主要因子为温度植被干旱指数(TVDI)、高程(GC)、年平均温度(TEM),其中TVDI因子的影响最为显著。【结论】构建的空间logistic森林火灾风险概率模型能够科学有效地对湖南省森林火灾风险等级进行划分,为森林火灾的预防和应急资源配置提供科学依据。
【Objective】In order to scientifically prevent forest fires and carry out emergency resource allocation,the spatial logistic forest fire risk probability model is used to divide the forest fire risk levels in Hunan province,and to provide technical methods for forest fire risk zoning below the county level.【Method】Logistic regression analysis method was used to build a forest fire risk probability model by surface temperature data,vegetation index data,meteorological data,humanistic data,vegetation type data,forest fire monitoring data of Hunan province from 2008 to 2018.The forest fire risk level was grade by the forest fire risk probability distribution interval.【Result】The researches show that the established spatial logistic forest fire risk probability model in Hunan province has a good fitting effect.At the significance level of 0.05,the mixed test and Wald test show that the logistic forest fire risk probability model can significantly reflect the occurrence probability of forest fire risk.The value of relative operating characteristic(ROC)is 0.779.The probability value of forest fire risk in Hunan province was calculated by grid layer,and according to the probability of the forest fire risk,Hunan province is divided into 5 levels:very low,low,medium,high and very high.The high forest risk level and extremely high fire risk level with a forest fire risk probability≥0.6 in Hunan province are mainly distributed in Shaoyang city,Hengyang city and Yongzhou city,south of Zhuzhou city and parts of Chenzhou city.Middle forest fire risk is mainly distributed in Huaihua city,Loudi city,Xiangtan city and Changsha city.Low forest fire risk and very low forest fire risk are mainly distributed in Jishou city,Changde city,Zhangjiajie city,Yiyang city and Yueyang city.The main factors affecting the fire risk level in Hunan province are temperature vegetation drought index(TVDI),elevation(GC),and annual average temperature(TEM),among of them,TVDI factor has the most significant influence.【Conclusion】The e
作者
王双
张贵
谭三清
王平
吴鑫
WANG Shuang;ZHANG Gui;TAN Sanqing;WANG Ping;WU Xin(College of Forestry,Central South University of Forestry&Technology,Changsha 410004,Hunan,China)
出处
《中南林业科技大学学报》
CAS
CSCD
北大核心
2020年第9期88-95,共8页
Journal of Central South University of Forestry & Technology
基金
湖南省科技创新平台与人才计划项目(2017TP1022)
湖南省重点研发计划项目(2016SK2025)
湖南省教育厅项目(18C0283)。
关键词
森林火灾风险概率模型
森林火灾风险评价
湖南省
forest fire risk probability model
forest fire risk assessment
Hunan province
