西南山地区域森林火险综合预报模型研究——以重庆市为例  

Research on the construction of composite risk prediction model for forest fire in the mountainous area of southwestern China——Taking Chongqing city as an example

作　　者：郑忠 高阳华 杨庆媛[3] 唐云辉 徐永进[3] 陈艳英 ZHENG Zhong;GAO Yanghua;YANG Qingyuan;TANG Yunhui;XU Yongjin;CHEN Yanying(Chongqing Institute of Meteorological Sciences,Chongqing Meteorological Bureau,Chongqing 401147,China;College of Resources and Environment,Chengdu University of Information Technology,Chengdu 610225,China;School of Geographical Sciences,Southwest University,Chongqing 400715,China)

机构地区：[1]重庆市气象科学研究所,重庆401147 [2]成都信息工程大学资源环境学院,四川成都610225 [3]西南大学地理科学学院,重庆400715

出　　处：《自然灾害学报》2020年第1期152-161,共10页Journal of Natural Disasters

基　　金：国家自然科学基金(41801315);重庆市气象局开放式研究基金(KFJJ-201705);四川省教育厅科研项目(18ZB0128);成都信息工程大学校引进人才项目(KYTZ201742)联合资助~~

摘　　要：森林火灾严重破坏森林生态系统的结构和功能,针对其风险预报开展研究具有重要的理论意义和应用价值。通过融合气象、地表覆盖和人类活动等要素的方式构建森林火险综合预报模型,是西南山地区域森林火险综合预报业务化应用的重要发展方向。基于《森林火险气象等级》国家标准(GB/T 36743-2018),本文研究首先构建了重庆市森林火险气象风险预报模型。然后通过融合可燃物因素和人为因素的方式,进一步构建重庆市森林火险综合预报模型。最后利用历史火点数据针对上述两类模型的精度进行对比和分析。研究结果表明:重庆市森林火灾发生次数较多,且呈现出季节性规律;重庆市森林火灾的发生不仅受到气象条件的影响,而且受到地表覆盖和人类活动等多种要素条件的共同影响;与森林火险气象风险预报结果相比,森林火险综合预报结果在空间分布上具有更高的精细程度;森林火险综合预报模型能有效提高森林火险等级预报精度。Forest fire could seriously damage the structure and function of forest ecosystems.From theoretical and practical perspectives,it has great significant to conduct research on the risk prediction of forest fires.In the mountainous area of southwestern China,risk prediction models based on weather,land cover,and human activity factors,are expected to have good performance on predicting the forest fire risk.Based on the new national standard for Forest Fire Danger Weather Ratings published on September 2018 in China,a weather data-based risk prediction model of forest fire was developed at first.On the basis,a composite risk prediction model was further proposed in this study by integrating the land cover and human activity factors.Following that,performances of these models were verified by using historical data of forest fire.Results show that forests in Chongqing city are frequently affected by fires and the occurrence of forest fires presents seasonal patterns.Both weather,land cover,and human activity can lead to the occurrence of forest fires.Moreover,the spatial distribution of results predicted from the proposed composite risk model are more detailed compared to that of weather data-based model.Further,performance verifications from historical forest fire data indicate that the composite risk model can improve the prediction accuracy of forest fire risk.

关 键 词：森林火灾 气象风险 综合风险 预报 西南山地区域 重庆市 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] S127[自动化与计算机技术—控制科学与工程]