1950—2010年中国森林火灾时空特征及风险分析

doi:10.11707/j.1001-7488.20150110

摘要/Abstract

摘要：

[目的]开展森林火灾时空特征和风险分析,有利于把握森林火灾发生规律,为预测和防治森林火灾、制定差异化森林火灾财政补贴和保险政策提供科学依据.[方法]根据年鉴数据,借助主成分分析、聚类分析和信息扩散理论,对我国1950—2010年森林火灾的时空分布特征和风险状况进行研究,全面分析森林火灾发生的时空特征,结合灾情指标对全国各省进行评价并分类,并定量分析风险水平.[结果]1) 1950—2010年间,年均发生火灾12 683次,年均火场面积67.48万hm²,年均火灾次数和火场面积均呈下降趋势;1988—2010年间年均成灾面积7.95万hm²,总体表现为上升趋势;1988—2010年间年均成林蓄积损失167.5万m³,没有明显年际变化;1988—2010年间年均伤亡214人,总体呈波浪式下降趋势.2) 人为火源占已查明火源的80%以上,烧荒烧炭是最主要的生产性用火火源(38.47%),上坟烧纸是最主要的非生产性用火火源(19.92%).3) 华东、西南地区相比于其他地区火灾次数明显较大,东北地区火场面积明显较大.4) 按照一般和较大火灾次数、重大和特大火灾次数、火场面积、受害森林面积、成林蓄积、幼林株数、受伤人数、死亡人数8个指标,通过主成分分析对31个省(区、市)的火灾状况进行评价并排序,然后根据主成分得出的综合得分进行聚类分析,根据受灾程度将31个省区按照严重、较重、中度、一般和轻度分为5类,其中,湖南受灾最严重,其次是黑龙江、贵州和云南等.5) 基于信息扩散理论进行森林火灾风险分析,全国每年发生3 000~9 000次一般和较大火灾的概率较大(>0.482 4),每年发生0~30次重大和特大火灾概率较大(>0.430 2),每年受灾森林面积在15万hm²以下概率较大(>0.407 2).[结论]我国森林火灾频发、面积大,火灾高发与气候变化息息相关.引发森林火灾的主要原因是人为火源,必须严格控制和管理野外火源,规范生产、生活用火行为.森林火灾具有显著的区域性特征,我国森林资源面临的火灾风险压力较大.建议在准确把握全国森林火灾时空分布特征和风险基础上,加强宣传教育,采取科学的森林防火、预报、监测措施,制定合理的森林火灾财政补贴和森林保险政策,维护森林健康和国家生态安全.

关键词: 森林火灾, 时空分布, 风险分析, 科学防控

Abstract:

[Objective]Forest fires cause heavy loss to forest resources and society and economy and directly imperil the sustainable development of forestry and national ecological security. Study on the temporal and spatial distribution of forest fires and their risk analyses were helpful to understand the law of forest fires, to provide a scientific basis for forecasting, preventing forest fires and establishing different fiscal subsidy and insurance policies. The study is important to promote forest sustainable development and to maintain the national ecological safety. [Method]In order to find out the basic situation of forest fires in China, the forest fire temporal and spatial distribution characteristics and risk analysis were studied based on the yearbook data from 1950 to 2010 and by means of principal component analysis, cluster analysis and information diffusion theory. The temporal and spatial characteristics of forest fires in each province were comprehensively analyzed conducted, and further evaluated and classified by innovatively combining the disaster index. The fire risk level was quantitatively analyzed to provide scientific basis for forest fire prediction and prevention, to establish the fiscal subsidy policy, and to provide the basic support to forest safety risk prevention and risk decision. [Result]1) nnual average fire number was 12 683 times and burned an area of 674 800 hm² per year from 1950 to 2010, and the annual average fire number and the burned area showed a downward trend. Average annual fire-caused disastrous area was 79 500 hm² and showed an increase trend overall from 1988 to 2010; Annual average stand volume loss was 1.675 million m³ from 1988 to 2010 without obvious year-to-year variations; Annual average casualties was 214 people from 1988 to 2010 and showed overall a downward trend. 2) Human-caused fires accounted for more than 80% of the identified fires. Burning grass on waster land and burning charcoal were one of the main productive causes (38.47%), and sacrificial fires at graves were the main non-productive (19.92%). 3) Fire number in Eastern China and Southwestern China was significantly more than the other regions in China, while the burned area was significantly larger in Northeastern China than that in the other regions in China. 4) According to parameters of the general and large fire number, the major and catastrophic fire number, the burned area, the disastrous area, the stand volume loss, the young stand loss, and the casualties, this paper evaluated the fire situation in 31 provinces (The data of Hongkong, Macau and Taiwan are not included)using PCA and ranked them. Then the cluster analysis was conducted with comprehensive scores obtained by PCA. The 31 provinces were divided into 5 types: serious, heavy, moderate, general and light, and Hunan was most serious and followed by Heilongjiang, Guizhou and Yunnan. 5) With the information diffusion theory, the occurrence probability of general and large fire number in 3 000-9 000 was (> 0.482 4), the major and catastrophic fire number in 0-30 was (> 0.430 2) and the disastrous area under 150 000 hm² was (>0.407 2). [Conclusion]Frequent forest fires, large area in China were closely linked with climate change. The main reason causing the forest fire was man-made fire, and we must strictly control and manage field fire sources, and regulated productive and life fire-used behavior. Forest fires were with obvious regional feature and fire risk pressure was high in China. According to the results, it is suggested that it is important to strengthen the propaganda and education, take scientific measures for preventing, predicting and monitoring forest fires, establish reasonable fiscal subsidies and forest fire insurance policy, and maintain the forest health and national ecological security based on the accurate grasp of the national forest fires distribution and risk.

Key words: forest fire, temporal and spatial distribution, risk analysis, scientific prevention and control

中图分类号:

S762.2

苏立娟, 何友均, 陈绍志. 1950—2010年中国森林火灾时空特征及风险分析[J]. 林业科学, 2015, 51(1): 88-96.

Su Lijuan, He Youjun, Chen Shaozhi. Temporal and Spatial Characteristics and Risk Analysis of Forest Fires in China from 1950 to 2010[J]. Scientia Silvae Sinicae, 2015, 51(1): 88-96.

参考文献

陈建忠, 刘剑斌, 肖应忠. 2010. 建阳市森林火灾时空分布特征. 福建林学院学报, 30(2): 119-122.
(Chen J Z, Liu J B, Xiao Y Z. 2010. Study on the spatio-temporal distribution of forest fire in Jianyang County. Journal of Fujian College of Forestry, 30(2): 119-122.)
何晓群. 2012. 多元统计分析.3版. 北京:中国人民大学出版社.
(He X Q. 2012.Multivariate statistical analysis.3^rd ed. Beijing: China Renmin University Press.)
胡海清,金森. 2002. 黑龙江林火规律研究Ⅱ林火动态与格局影响因素的分析. 林业科学, 38(2): 98-102.
(Hu H Q, Jin S. 2002. Study on forest fire regime of Heilongjiang ProvinceⅡ Analysis on factors affecting fire dynamics and distributions. Scientia Silvae Sincae, 38(2): 98-102.)
胡海清,魏书精,孙龙. 2012. 大兴安岭2001—2010年森林火灾碳排放的计量估算.生态学报,(17):5373-5386.
(Hu H Q, Wei S J, Sun L. 2012. Estimating carbon emissions from forest fires during 2001 to 2010 in Daxing'anling Mountain. Acta Ecologica Sinica, (17):5373-5386.)
胡远方,屈宜春. 1998. 黑龙江省林火面积分布格局. 森林防火, (2): 19-23.
(Hu Y F, Qu Y C. 1998. Distribution patterns of forest area in Heilongjiang Province. Forest Fire Prevention, (2): 19-23.)
黄崇福. 1995. 模糊信息优化处理技术及其应用. 北京: 北京航空航天大学出版社.
(Huang C F. 1995.The fuzzy information optimization processing technology and its application. Beijing: Beihang University Press.)
黄玉霞,王宝鉴,许东蓓. 2006. 甘肃森林火灾的分布特征及其与归一化差分植被指数的关系. 西北农林科技大学学报:自然科学版, 34(5): 126-130.
(Huang Y X, Wang B J, Xu D P. 2006.Distribution features of forest fires and relations between forest fires and vegetation index in Gansu Province. Journal of Northwest A&F University(Natural Science Edition), 34(5): 126-130.)
寇晓军, 冯玉芬, 董俊华. 1997. 黑龙江省林火时空格局分析——时间分布. 森林防火, (4): 13-15.
(Kou X J,Feng Y F, Dong J H. 1997. Forest fire spatio-temporal distribution of Heilongjiang Province——temporal distribution. Forest Fire Prevention, (4): 13-15.)
李楠. 2011.伊春地区森林火灾时空分布格局. 哈尔滨: 东北林业大学硕士学位论文.
(Li N. 2011.Spatial and temporal distribution patterns of forest fires in Yichun Area. Harbin: MS thesis of Northeast Forestry University.)
刘亚彬,刘黎明,许迪,等. 2010. 基于信息扩散理论的中国粮食主产区水旱灾害风险评估. 农业工程学报, (8): 1-7.
(Liu Y B, Liu L M, Xu D, et al. Risk assessment of flood and drought in major grain-producing areas based on information diffusion theory. Transactions of the CSAE, (8): 1-7.)
田国华,杨松. 2013. 我国31个地区森林火灾时空分布特征. 森林防火, (2): 10-14.
(Tian G H, Yang S. 2013. Forest fire spatio-temporal distribution of 31 regions in our country. Forest Fire Prevention, (2): 10-14.)
田晓瑞,舒立福,王明玉,等. 2007. 西藏森林火灾时空分布规律研究. 火灾科学, 16(1): 10-13.
(Tian X R, Shu L F, Wang M Y, et al. 2007. Study on the spatial and temporal distribution of forest fire in Tibet. Fire Safety Science, 16(1): 10-13.)
王荷秀,慕建利,侯建忠,等. 1998. 陕西省林火的时空分布及火源特点.西北林学院学报, 13(l): 75-78.
(Wang H X, Mu J L,Hou J Z. et al. 1998. The spatial and temporal distribution of forest fire and the cause characteristics in Shaanxi Province. Journal of Northwest Forestry University, 13(1): 75-78.)
王丽娜, 王惠东, 李伟光. 2006. 近25年黑龙江省森林火灾发生趋势的研究. 森林防火, (3): 16-17.
(Wang L N, Wang H D, Li W G. 2006. Analysis of the trend of forest fires in Heilongjiang Province in nearly 25 years. Forest Fire Prevention, (3): 16-17.)
王志成,刘绍卓,张金山. 2006. 黑龙江省夏季林火时空分布特点. 林业科技, 31(2): 37-40.
(Wang Z C, Liu S C, Zhang J S. 2006.The spatial and temporal distribution of forest fires in summer in Heilongjiang Province. Forestry Science & Technology, 31(2): 37-40.)
魏鹏,杨松,田国华. 2012. 基于主成分分析法的全国森林火灾综合评价分析. 绿色科技, (12): 40-43.
(Wei P, Yang S,Tian G H. 2012. Comprehensive evaluation analysis on national forest fire based on principal component analysis. Journal of Green Science and Technology, (12): 40-43.)
魏书精,罗碧珍,魏书威,等. 2014. 黑河市森林火灾碳排放的计量估算研究.南京林业大学学报:自然科学版, 38(1): 70-76.
(Wei S J,Luo B Z, Wei S W, et al. 2014. Estimates of carbon emissions in Heihe City to forest fires. Journal of Nanjing Forestry University (Natural Sciences Edition), 38(1): 70-76.)
文东新,张明军,龙定华,等. 2007. 广西森林火灾的时空分布及其影响因素. 中南林业科技大学学报,27(5):83-86.
(Wen D X, Zhang M J, Long D H, et al. 2007. The influence of environment condition in Guangxi on the temporal and spatial distribution of forest fire. Journal of Central South University of Forestry & Technology, 27(5):83-86.)
吴宽让,崔铁成. 1991. 陕西森林火灾的特点与规律. 陕西林业科技, (3): 57-59.
(Wu K R, Cui T C. 1991. Features and regularities of forest fires in Shaanxi Province. Shaanxi Forest Science and Technology, (3): 57-59.)
张冬有,邓欧,李亦秋,等. 2012. 黑龙江省1980—2005年森林火灾时空特征.林业科学, 48(2):175-179.
(Zhang D Y, Deng O, Li Y Q,et al.2012. Temporal and spatial characteristics of forest fires in Heilongjiang Province between 1980 and 2005. Scientia Silvae Sincae, 48(2):175-179.)
张继权, 刘兴朋. 2007. 基于信息扩散理论的吉林省草原火灾风险评价. 干旱区地理, 30(4): 590-594.
(Zhang J Q, Liu X P. 2007.Risk assessment of grassland fire in Jilin Province based on information diffusion theory. Arid Land Geography, 30(4): 590-594.)
赵辉. 2012. 福建三明2000—2009年森林火灾时空分布及影响因素研究. 北京:北京林业大学硕士学位论文.
(Zhao H. 2012. Spatial and temporal distribution of forest fires and it's impact in Sanming of Fujian Province from 2000 to 2009. Beijing: MS thesis of Beijing Forestry University.)
周雪,张颖. 2014. 中国森林火灾风险统计分析. 统计与信息论坛, (1): 34-39.
(Zhou X, Zhang Y. 2014.Statistical analysis of forest fire risk in China. Statistics & Information Forum, (1): 34-39.)
Clemens W, Christian S, Michael L, et al. 2012. Recent climate change: Long-term trends in meteorological forest fire danger in the Alps. Forest Ecology and Management, 162(15): 1-13.
Igor D, Mats N, Hans W L. 2012. Forest fire activity in Sweden: Climatic controls and geographical patterns in 20th century. Forest Ecology and Management, 154(15): 174-186.
Liu X, Zhang J, Cai W, et al. 2010. Information diffusion-based spatio-temporal risk analysis of grassland fire disaster in northern China. Knowledge-Based Systems, 23(1): 53-60.
Margarita H, Javier L, Alicia P, et al.2009. Assessment of forest fire seasonality using MODIS fire potential: A time series approach. Forest Ecology and Management, 149(11): 1946-1955.

[1]	孔维尧, 孙权, 刘鑫鑫, 曲丽, 王福友, 姚明远, 邹红菲. 基于红外相机监测的汪清自然保护区东北豹种群动态[J]. 林业科学, 2019, 55(5): 188-196.
[2]	方祥, 王海晖, 陶骏骏, 盛昌栋. 林火蔓延过程中辐射换热和点燃特性分析[J]. 林业科学, 2019, 55(3): 97-105.
[3]	尹赛男, 舒立福, 张大明, 单延龙, 杜帅, 唐抒圆, 张鑫瑶, 张志文. 吉林省森林火灾火源数据特征分析[J]. 林业科学, 2018, 54(7): 165-172.
[4]	杨伟, 姜晓丽. 森林火灾火烧迹地遥感信息提取及应用[J]. 林业科学, 2018, 54(5): 135-142.
[5]	祖笑锋, 覃先林, 李增元, 孙桂芬, 刘树超. 基于幂律分布的森林燃烧生物量卫星遥感估测方法[J]. 林业科学, 2017, 53(10): 90-99.
[6]	张晨, 牛树奎, 陈锋, 邵潇, 王欢. 基于GIS的景观格局对云南省森林火灾的影响[J]. 林业科学, 2016, 52(7): 96-103.
[7]	张德成, 陈绍志, 白冬艳. 森林火灾保险纯费率厘定模型及实证分析[J]. 林业科学, 2016, 52(7): 129-137.
[8]	郑琼;邸雪颖;金森. 伊春地区1980-2010年森林火灾时空格局及影响因子[J]. 林业科学, 2013, 49(4): 157-163.
[9]	田晓瑞;舒立福;赵凤君;王明玉. 大兴安岭雷击火发生条件分析[J]. 林业科学, 2012, 48(7): 98-103.
[10]	张冬有;邓欧;李亦秋;李桂君;冯仲科. 黑龙江省1980—2005年森林火灾时空特征[J]. 林业科学, 2012, 48(2): 175-179.
[11]	胡海清;魏书精;孙龙. 大兴安岭呼中区2010年森林火灾碳排放的计量估算[J]. 林业科学, 2012, 48(10): 109-119.
[12]	郑伟;李亚君;刘诚;王萌. 基于多源卫星遥感数据的森林过火区面积估算方法[J]. 林业科学, 2011, 47(8): 192-195.
[13]	赵尘;王大明;张正雄. 原木采集生产的生物质流特征分析[J]. 林业科学, 2011, 47(5): 90-95.
[14]	肖云丹;鞠洪波;张雄清;纪平. 黔南地区气象因子与森林火灾发生次数之间的关系[J]. 林业科学, 2011, 47(10): 128-133.
[15]	梅志雄;徐颂军王佳. 基于DRNN和ARIMA模型的森林火灾面积时空综合预测方法[J]. 林业科学, 2009, 12(8): 101-107.