基于随机森林模型的海南岛热带森林火灾发生概率预测
收稿日期: 2024-07-16
修回日期: 2024-10-28
网络出版日期: 2025-04-18
基金资助
海南省科技计划项目-省属科研院所技术创新专项“海南岛森林火灾时空分布、驱动机制及预测模型研究”(SQKY2022-0021)
Predicting the Probability of Tropical Forest Fires in Hainan Island Based on Random Forest Model
Received date: 2024-07-16
Revised date: 2024-10-28
Online published: 2025-04-18
分析海南岛热带森林火灾的驱动因素,构建适用性强的预测模型,为森林火灾的精准预防提供技术支撑。基于地面调查的历史森林火灾数据和MOD14A火灾数据集,结合气候、植被、地形及人类活动等数据,利用随机森林模型构建森林火灾发生概率预测模型。结果表明:1)月均气温是影响海南岛森林火灾发生风险的最强因子,其次为月均降水量。2)随机森林模型预测森林火灾发生概率的曲线下面积(AUC)值为1.00,高于地理加权逻辑斯蒂回归模型的AUC值(0.88),表明随机森林模型在海南岛的热带森林火灾风险预测中表现更高的适用性。3)海南岛森林火灾风险的高发区域主要集中在西部。总体来看,随机森林模型在构建热带森林火灾风险预测模型方面,比地理加权逻辑斯蒂回归模型具有更高的适用性。
陈小花 , 陈宗铸 , 杨青青 , 雷金睿 , 吴庭天 , 李苑菱 , 潘小艳 . 基于随机森林模型的海南岛热带森林火灾发生概率预测[J]. 林草资源研究, 2024 , 0(6) : 140 -145 . DOI: 10.13466/j.cnki.lczyyj.2024.06.016
In the context of global climate change,the forest fire prevention situation on Hainan Island is becoming increasingly severe,and it is urgently needed to analyze the driving factors of tropical forest fires on Hainan Island and build a strong predictive model that is applicable.Utilizing historical forest fire data compiled by the forestry department from ground surveys and MOD14A fire detection,a comprehensive dataset was established for Hainan Island.This dataset was combined with climate,vegetation,topography,and human activity data to construct a predictive model using the random forest methodology.1)The average monthly temperature is the most influential factor on forest fire risk in Hainan Province,followed by the average monthly precipitation.2)Comparative model analysis shows the random forest model,with an AUC value of 1,outperforms the geographically weighted logistic regression model,which has an AUC value of 0.88,indicating that the random forest model is more suitable for predicting the probability of tropical forest fires on Hainan Island than the geographically weighted logistic regression model.3)The spatial distribution of forest fire risk on Hainan Island mainly occurs in the west.This study believes that the random forest model is more applicable than the geographically weighted logistic regression model in building a predictive model for tropical forest fire risk.
| [1] | HUANG Fang, LIU Xiangnan, YUAN Jinguo. Study on forest fire danger model with remote sensingbased on GIS[J]. Chinese Geographical Science, 2000, 10(1):61-67. |
| [2] | SACHDEVA S, BHATIA T, VERMA A K. GIS-based evolutionary optimized gradient boosted decision trees for forest fire susceptibility mapping[J]. Natural Hazards, 2018,92:1399-1418. |
| [3] | VAN L H, HOANG D A, TRAN C T, et al. A new approach of deep neural computing for spatial prediction of wildfire danger at tropical climate areas[J]. Ecological Informatics, 2021,63:101300. |
| [4] | 刘国华, 傅伯杰, 方精云. 中国森林碳动态及其对全球碳平衡的贡献[J]. 生态学报, 2000, 20(5):733-740. |
| [5] | 何锐, 陆恒, 晋子振, 等. 基于随机森林算法的中国西南地区林火发生预测模型构建及驱动因子[J]. 生态学报, 2023, 43(22):9356-9370. |
| [6] | 胡汉舟, 孙守亮. 中国环境统计年鉴[M]. 北京: 中国统计出版社, 2021,97-99. |
| [7] | ZHOU Tianjun, CHEN Ziming, CHEN Xiaolong, et al. Interpreting IPCC AR6:Future global climate based on projection under scenarios and on near-term information[J]. Advances in Climate Change Research, 2021, 17(6):652-663. |
| [8] | HU Yilun, JI Guoxu, LI Jihong, et al. Interpretation of IPCC AR6:Terrestrial and freshwater ecosystems and their services[J]. Advances in Climate Change Research, 2022, 18(4):395-404. |
| [9] | LI Wenhui, XU Quanli, YI Junhua, et al. Predictive model of spatial scale of forest fire driving factors:a case study of Yunnan Province,China[J]. Scientific reports, 2022, 12(1):19029. |
| [10] | ACHU A L, THOMAS J, AJU C D, et al. Machine-learning modelling of fire susceptibility in a forest-agriculture mosaic landscape of southern India[J]. Ecological Informatics, 2021,64:101348. |
| [11] | BJANES A, DELAFUENTE R, MENA P. A deep learning ensemble model for wildfire susceptibility mapping[J]. Ecological Informatics, 2021,65:101397. |
| [12] | MONJARAS-VEGA N A, BRIONES-HERRERA C I, VEGA-NIEVA D J, et al. Predicting forest fire kernel density at multiple scales with geographically weighted regression in Mexico[J]. Science of the Total Environment, 2020,718:137313. |
| [13] | YANG Xiajie, SU Zhangwen, TIAN Chao, et al. Optimization of resource distribution of forest fire management in southern Fujian based on ArcGIS[J]. Chinese Journal of Ecology, 2017, 36(4):1142. |
| [14] | 梁慧玲, 王文辉, 郭福涛, 等. 比较逻辑斯蒂与地理加权逻辑斯蒂回归模型在福建林火发生的适用性[J]. 生态学报, 2017, 37(12):4128-4141. |
| [15] | 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(4):135-142. |
| [16] | PIAO Y, LEE D, PARK S, et al. Forest fire susceptibility assessment using google earth engine in Gangwon-do,Republic of Korea[J]. Geomatics,Natural Hazards and Risk, 2022, 13(1):432-450. |
| [17] | BISQUERT M, CASELLES E, SANCHEZ J M, et al. Application of artificial neural networks and logistic regression to the prediction of forest fire danger in Galicia using MODIS data[J]. International Journal of Wildland Fire, 2012, 21(8):1025-1029. |
| [18] | POURHASMAR H R, GAYEN A, LASAPONARA R, et al. Application of learning vector quantization and different machine learning techniques to assessing forest fire influence factors and spatial modelling[J]. Environmental research, 2020,184:109321. |
| [19] | THACH N N, NGO D B T, XUAN P, et al. Spatial pattern assessment of tropical forest fire danger at Thuan Chau area(Vietnam)using GIS-based advanced machine learning algorithms:A comparative study[J]. Ecological informatics, 2018,46:74-85. |
| [20] | SINGH K R, NEETHU K P, MADHUREKAA K, et al. Parallel SVM model for forest fire prediction[J]. Soft Computing Letters, 2021,3:100014. |
| [21] | ABID F. A survey of machine learning algorithms based forest fires prediction and detection systems[J]. Fire technology, 2021, 57(2):559-590. |
| [22] | PEEL M C, FINLAYSON B L, MCMAHON T A. Updated world map of the K?ppen-Geiger climate classification[J]. Hydrology and earth system sciences, 2007, 11(5):1633-1644. |
| [23] | 杨青青, 陈小花, 陈宗铸, 等. 基于MODIS数据的海南岛森林火灾时空分布特征分析[J]. 林业科技通讯, 2024(1):22-26. |
| [24] | 周鹏飞, 王艳霞. 应用机器学习算法分析广西林火发生驱动因素及林火预测[J]. 东北林业大学学报, 2024, 52(11):72-82. |
| [25] | 张国丽, 慈雪伦, 杨雪清, 等. 森林火灾时空分布特征及易发性分析研究[J]. 林草资源研究, 2023(5):48-55. |
| [26] | 李史欣, 张福全, 林海峰. 基于机器学习算法的森林火灾风险评估研究[J]. 南京林业大学学报(自然科学版), 2023, 47(5):49-56. |
| [27] | 朱诗豪, 吴志伟, 李政杰, 等. 赣南马尾松林地表细小死可燃物含水率动态及模型[J]. 林业科学, 2024, 60(5):158-168. |
| [28] | 赵宇虹. 随机森林遥感信息提取研究进展及应用研究[J]. 测绘与空间地理信息, 2021, 44(3):133-136. |
| [29] | SACHDEVA S, BHATIA T, VERMA K A. GIS-based evolutionary optimized Gradient Boosted Decision Trees for forest fire susceptibility mapping[J]. Natural Hazards, 2018, 92(3):1399-1418. |
| [30] | AMIR S N, REZA H P, KARIM A. A comparison between ten advanced and soft computing models for groundwater qanat potential assessment in Iran using R and GIS[J]. Theoretical and Applied Climatology, 2018, 131(3-4):967-984. |
| [31] | RODRIGUES H, GUTIERREZ F M, PONTIUS G R, et al. A Suite of Tools for ROC Analysis of Spatial Models[J]. ISPRS International Journal of Geo-Information, 2013, 2(3):869-887. |
| [32] | 黄永迎, 程志浩, 刘扬, 等. 如何理解受试者工作特征曲线及曲线下面积?[J]. 中国生育健康杂志, 2023, 34(6):586-591. |
| [33] | NAGHIBI A S, POURGHASEMI R H. A Comparative Assessment Between Three Machine Learning Models and Their Performance Comparison by Bivariate and Multivariate Statistical Methods in Groundwater Potential Mapping[J]. Water Resources Management, 2015, 29(14):5217-5236. |
| [34] | MINNICHAL R A, BAHREZ C J. Wildland fire and chaparral succession along the California Baja-California boundary[J]. International Journal of Wildland Fire, 1995, 5(1):13-24. |
| [35] | 何诚, 舒立福, 刘柯珍. 大兴安岭地区夏季森林火灾环境因子特征分析[J]. 西南林业大学学报(自然科学), 2021, 41(3):87-93. |
| [36] | 符国瑷, 冯绍信. 海南五指山森林的垂直分布及其特征[J]. 广西植物, 1995(1):57-69. |
| [37] | 杨小波, 陈宗铸, 李东海. 海南植被分类体系与植被分布图[J]. 中国科学:生命科学, 2021, 51(3):321-333. |
/
| 〈 |
|
〉 |
