[1]苏漳文,曾爱聪,蔡奇均,等.基于Gompit回归模型的大兴安岭林火预测模型及驱动因子研究[J].林业工程学报,2019,4(04):135-142.[doi:10.13360/j.issn.2096-1359.2019.04.020]
 SU Zhangwen,ZENG Aicong,CAI Qijun,et al.Study on prediction model and driving factors of forest fire in Da Hinggan Mountains using Gompit regression method[J].Journal of Forestry Engineering,2019,4(04):135-142.[doi:10.13360/j.issn.2096-1359.2019.04.020]
点击复制

基于Gompit回归模型的大兴安岭林火预测模型及驱动因子研究()
分享到:

《林业工程学报》[ISSN:1001-8081/CN:32-1160/S]

卷:
4
期数:
2019年04期
页码:
135-142
栏目:
森林工程与土建
出版日期:
2019-07-09

文章信息/Info

Title:
Study on prediction model and driving factors of forest fire in Da Hinggan Mountains using Gompit regression method
文章编号:
2096-1359(2019)04-0135-08
作者:
苏漳文1曾爱聪2蔡奇均2胡海清1*
1.东北林业大学林学院,哈尔滨 150040; 2.福建农林大学林学院,福州 350002
Author(s):
SU Zhangwen1 ZENG Aicong2 CAI Qijun2 HU Haiqing1*
1.College of Forestry, Northeast Foresty University, Harbin 150040, China; 2.College of Forestry, Fujian Agriculture and Foresty University, Fuzhou 350002, China
关键词:
大兴安岭 Gompit回归 林火预测模型 驱动因子
Keywords:
Da Hinggan Mountains Gompit regression the prediction model of forest fire importance of driving factor
分类号:
S762.2
DOI:
10.13360/j.issn.2096-1359.2019.04.020
文献标志码:
A
摘要:
本研究基于2000—2016年林火数据,选取气象、地形、植被、人为活动等因素作为林火预测变量,采用Gompit回归模型对林火发生的主要驱动因子进行分析,并建立大兴安岭地区林火发生预测模型。结果表明:大兴安岭地区林火受气象因素(日累计降水、日平均相对湿度)的影响最大且与林火发生均呈显著负相关; 此外,大兴安岭林火多发生于缓坡、远离居民区、铁路、公路等人为活动较为频繁地区。模型结果表明:Gompit回归模型的预测效果较好(准确率77%),ROC检验结果表明模型的拟合度较高(效果值为0.868); 而独立样本的检验显示,预测准确率为75.3%,模型具有较高的适用性。大兴安岭近17年的火险等级总体呈南高北低、东高西低的地理分布,其中高火险和中火险区主要集中在南部、东南部等地,占整个研究区域的24.2%; 同时南部和东南部存在大面积低估区,表明模型对这些地区的预测能力不高。
Abstract:
In this study, the Gompit regression method was applied to analyze the driving factors of forest fires in Da Hinggan Mountains and establish the forest fire prediction model based on forest fire data from 2001 to 2016 and meteorological, topographic, vegetation, human activity factors that are associated with the occurrence of forest fires in the region.The results of investigation showed that the meteorological factors(daily cumulative precipitation and daily average relative humidity)had the most powerful and important influence on forest fire occurrence.The daily cumulative precipitation and daily average relative humidity were negatively related to the fire occurrence in Da Hinggan Mountains.Multitude of forest fires occurred in the areas with slow slope and far away from residential regions, railways, roads and other human activities.The results of model fitting showed that the Gompit regression method had a good prediction accuracy of 77%, and the area under the ROC curve(AUC)derived from the ROC test was 0.868, indicating a high-level goodness of fit for the Gompit regression method.Same as the modeling result, the result of validation test also indicated a high prediction accuracy(75.3%)and good applicability of the Gompit regression method for fire prediction in Da Hinggan Mountains.In general, the geographical distribution of the fire risk level was high in the South and East parts, low in the North and West parts of Da Hinggan Mountains from 2001 to 2016.The high and medium fire risk level areas were mainly concentrated in the south and southeast of study region, accounting for 24.2% of the entire Da Hinggan Mountains.At the same time, the result of model residual examination revealed that multitude of underestimated areas existed in the southern and southeastern Da Hinggan Mountains, which indicated a weak prediction ability of Gompit regression method employed in these areas.

参考文献/References:

[1] MCKENZIE D, SHANKAR U, KEANE R E, et al.Smoke consequences of new wildfire regimes driven by climate change[J].Earth's Future, 2014, 2(2): 35-59.DOI:10.1002/2013ef000180.
[2] GUO F T, INNES J L, WANG G Y, et al.Historic distribution and driving factors of human-caused fires in the Chinese boreal forest between 1972 and 2005[J].Journal of Plant Ecology, 2015, 8(5): 480-490.DOI:10.1093/jpe/rtu041.
[3] GUO F T, SELVALAKSHMI S, LIN F F, et al.Geospatial information on geographical and human factors improved anthropogenic fire occurrence modeling in the Chinese boreal forest[J].Canadian Journal of Forest Research, 2016, 46(4): 582-594.DOI:10.1139/cjfr-2015-0373.
[4] GUO F T, WANG G Y, SU Z W, et al.What drives forest fire in Fujian, China? Evidence from logistic regression and Random Forests[J].International Journal of Wildland Fire, 2016, 25(5): 505.DOI:10.1071/wf15121.
[5] GUO F T, SU Z W, WANG G Y, et al.Understanding fire drivers and relative impacts in different Chinese forest ecosystems[J].Science of the Total Environment, 2017, 605/606: 411-425.DOI:10.1016/j.scitotenv.2017.06.219.
[6] 石晶晶.浙江省林火发生格局及预测模型研究[D].杭州: 浙江农林大学, 2014.
SHI J J.Study on occurring space and forecasting model of forest fires in Zhejiang province[D].Hangzhou: Zhejiang A&F University, 2014.
[7] OLIVEIRA S, OEHLER F, SAN-MIGUEL-AYANZ J, et al.Modeling spatial patterns of fire occurrence in Mediterranean Europe using Multiple Regression and Random Forest[J].Forest Ecology and Management, 2012, 275: 117-129.DOI:10.1016/j.foreco.2012.03.003.
[8] PRADHAN B, DINI HAIRI BIN SULIMAN M, ARSHAD BIN AWANG M.Forest fire susceptibility and risk mapping using remote sensing and geographical information systems(GIS)[J].Disaster Prevention and Management: an International Journal, 2007, 16(3): 344-352.DOI:10.1108/09653560710758297.
[9] VILAR L, WOOLFORD D G, MARTELL D L, et al.A model for predicting human-caused wildfire occurrence in the region of Madrid, Spain[J].International Journal of Wildland Fire, 2010, 19(3): 325.DOI:10.1071/wf09030.
[10] GURI 瘙 塁 S, ÇAGˇLAYAN E, UN T.Estimating of probability of home-ownership in rural and urban areas: Logit, probit and Gompit model[J].European Journal of Social Sciences, 2011,21(3):405-411.
[11] MINAB M, OIA M, ASL M G.Analysis of the effect of corporate governance on the financial health of companies Using Logit, Probit and Gompit models[C]//The 12th Iranian National Conference on Accounting, Iran, 2014.
[12] JUSTICE C O, GIGLIO L, KORONTZI S, et al.The MODIS fire products[J].Remote Sensing of Environment, 2002, 83(1/2): 244-262.DOI:10.1016/s0034-4257(02)00076-7.
[13] CHANG Y, ZHU Z L, BU R C, et al.Predicting fire occurrence patterns with logistic regression in Heilongjiang Province, China[J].Landscape Ecology, 2013, 28(10): 1989-2004.DOI:10.1007/s10980-013-9935-4.
[14] 胡海清.林火生态与管理[M].北京:中国林业出版社, 2005.
HU H Q.Forest fire ecology and management [M].Beijing:China Forestry Publishing House, 2005.
[15] CHANDLER C, CHENEY P, TRABAUD L, et al.Fire in forestry.Volume I.Forest fire behavior and effects[M].Florida: Krieger Publishing Company, 1991.
[16] GUTMAN G, IGNATOV A.The derivation of the green vegetation fraction from NOAA/AVHRR data for use in numerical weather prediction models[J].International Journal of Remote Sensing, 1998, 19(8): 1533-1543.DOI:10.1080/014311698215333.
[17] STURTEVANT B R, CLELAND D T.Human and biophysical factors influencing modern fire disturbance in northern Wisconsin[J].International Journal of Wildland Fire, 2007, 16(4): 398.DOI:10.1071/wf06023.
[18] SU Z W, HU H Q, WANG G Y, et al.Using GIS and Random Forests to identify fire drivers in a forest city,Yichun, China[J].Geomatics, Natural Hazards and Risk, 2018, 9(1): 1207-1229.DOI:10.1080/19475705.2018.1505667.
[19] 邓欧, 李亦秋, 冯仲科, 等.基于空间Logistic的黑龙江省林火风险模型与火险区划[J].农业工程学报, 2012, 28(8): 200-205.DOI:10.3969/j.issn.1002-6819.2012.08.031.
DENG O, LI Y Q, FENG Z K, et al.Model and zoning of forest fire risk in Heilongjiang province based on spatial Logistic[J].Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(8): 200-205.
[20] LITTELL J S, MCKENZIE D, PETERSON D L, et al.Climate and wildfire area burned in western USecoprovinces, 1916-2003[J].Ecological Applications, 2009, 19(4): 1003-1021.DOI:10.1890/07-1183.1.
[21] 张波, 代鲁燕, 黄启风, 等.Logistic回归模型中自变量相对重要性的优势分析[J].浙江预防医学, 2012, 24(8): 13-15,29.DOI:10.3969/j.issn.1007-0931.2012.08.005.
ZHANG B, DAI L Y, HUANG Q F, et al.Dominance analysis for the relative importance of independent variable in logistic regression model[J].Zhejiang Journal of Preventive Medicine, 2012, 24(8): 13-15,29.
[22] MENARD S.Six approaches to calculating standardized logistic regression coefficients[J].The American Statistician, 2004, 58(3): 218-223.DOI:10.1198/000313004x946.
[23] WU Z W, HE H S, YANG J, et al.Relative effects of climatic and local factors on fire occurrence in boreal forest landscapes of northeastern China[J].Science of the Total Environment, 2014, 493: 472-480.DOI:10.1016/j.scitotenv.2014.06.011.
[24] PENMAN T D, BRADSTOCK R A, PRICE O.Modelling the determinants of ignition in the Sydney Basin, Australia: Implications for future management[J].International Journal of Wildland Fire, 2013, 22(4): 469-478.
[25] PRESTEMON J P, BUTRY D T.Time to burn: modeling wildland arson as an autoregressive crime function[J].American Journal of Agricultural Economics, 2005, 87(3): 756-770.DOI:10.1111/j.1467-8276.2005.00760.x.
[26] ALDERSLEY A, MURRAY S J, CORNELL S E.Global and regional analysis of climate and human drivers of wildfire[J].Science of the Total Environment, 2011, 409(18): 3472-3481.DOI:10.1016/j.scitotenv.2011.05.032.
[27] POURGHASEMI H R.GIS-based forest fire susceptibility mapping in Iran: a comparison between evidential belief function and binary logistic regression models[J].Scandinavian Journal of Forest Research, 2016, 31(1): 80-98.DOI:10.1080/02827581.2015.1052750.

备注/Memo

备注/Memo:
收稿日期:2018-12-24 修回日期:2019-01-16
基金项目:国家林业公益性行业科研专项(201404402)。
作者简介:苏漳文,男,研究方向为林火预测预报。通信作者:胡海清,男,教授。E-mail: huhq-cf@nefu.edu.cn
更新日期/Last Update: 2019-07-10