混合效应模型在林火发生预测中的适用性

doi:10.13287/j.1001-9332.202206.026

应用生态学报 ›› 2022, Vol. 33 ›› Issue (6): 1547-1554.doi: 10.13287/j.1001-9332.202206.026

混合效应模型在林火发生预测中的适用性

张珍^1,2, 杨凇^1,2, 朱贺^1,2, 王光玉³, 郭福涛^1,2, 孙帅超^1,2*

¹福建农林大学林学院, 福州 350002;
²3S技术与资源优化利用福建省高校重点实验室, 福州 350002;
³不列颠哥伦比亚大学林学院亚洲森林研究中心, 加拿大温哥华 BC V6T 1Z4

收稿日期:2021-08-23 接受日期:2022-03-24 发布日期:2022-12-15
通讯作者: *E-mail: sun_sc@yeah.net
作者简介:张珍, 女, 1996年生, 硕士研究生。主要从事林火生态与管理研究。E-mail: zhen17803477566@163.com
基金资助:
国家重点研发计划战略性国际科技创新合作重点专项(2018YFE0207800)资助。

Applicability of mixed effect model in the prediction of forest fire

ZHANG Zhen^1,2, YANG Song^1,2, ZHU He^1,2, WANG Guang-yu³, GUO Fu-tao^1,2, SUN Shuai-chao^1,2*

¹College of Forestry, Fujian Agricultural and Forestry University, Fuzhou 350002, China;
²3S Technology and Resource Optimization Utilization Key Laboratory of Fujian Universities, Fuzhou 350002, China;
³Asia Forest Research Centre, Faculty of Forestry, University of British Columbia, Vancouver, BC V6T 1Z4, Canada

Received:2021-08-23 Accepted:2022-03-24 Published:2022-12-15

摘要/Abstract

摘要： 林火是森林生态系统的重要影响因子,建立科学准确的林火预测预报模型对林火管理工作至关重要。本研究以不同气象因子为主要预测变量,基于Logistic回归和广义线性混合效应模型建立福建省林火发生预测模型,通过对比Logistic基础模型和广义线性混合效应模型的拟合度和预测精度,研究混合效应模型在林火预报中的适用性。结果表明: Logistic基础模型的受试者工作特征曲线下面积(AUC)值为0.664,验证准确率为60.4%。添加随机效应后,模型的拟合和检验精度均获得了提升。其中,考虑行政区划和海拔差异效应的两水平混合效应模型的表现最优,其AUC值和验证准确率分别比基础模型提升0.057和6.0%。用此混合效应模型对福建省各地区的林火发生概率进行预测的结果表明,福建省西北部和南部为林火中高发区域,西南部和东部为林火低发区域,与实际观测的火点分布一致。混合效应模型在数据拟合和林火发生预测方面均优于Logistic基础模型,可作为林火预测和管理的重要工具。

关键词: 林火预测, 随机效应, Logistic回归, 混合效应

Abstract: Fire is an important influencing factor in forest ecosystems. Establishing an accurate forest fire forecasting model is important for forest fire management. We used different meteorological factors as predictors to construct a forest fire prediction model in Fujian Province, based on Logistic regression and generalized linear mixed effect model. We compared the fitness and prediction accuracy of the two models, judged the applicability of the mixed effect model in forest fire forecasting. The results showed that the AUC and accuracy values of the Logistic base model were 0.664 and 60.4%, respectively. Models considering random effects gave better fitting and validating statistics. Among them, the two-level mixed model containing both area and altitude difference effects performed best, with increases of 0.057 and 6.0% for the AUC and accuracy values, respectively. By applying the model to predict the probability of forest fires in Fujian Province, we found that the middle-incidence and high-incidence areas of forest fires distributed in northwest and south Fujian, whereas the low-incidence areas of forest fires distributed in southwest and east Fujian, which was consistent with the observed data. The data fitting and forest fire prediction of the mixed effects model was better than those of the Logistic basic model. Therefore, it could be used as an important tool for forest fire prediction and management.

Key words: forest fire forecasting, random effect, Logistic regression, mixed effect

张珍, 杨凇, 朱贺, 王光玉, 郭福涛, 孙帅超. 混合效应模型在林火发生预测中的适用性[J]. 应用生态学报, 2022, 33(6): 1547-1554.

ZHANG Zhen, YANG Song, ZHU He, WANG Guang-yu, GUO Fu-tao, SUN Shuai-chao. Applicability of mixed effect model in the prediction of forest fire[J]. Chinese Journal of Applied Ecology, 2022, 33(6): 1547-1554.

参考文献

[1] Flannigan M, Stock B, Turetsky M, et al. Impacts of climate change on fire activity and fire management in the circumboreal forest. Global Change Biology, 2010, 15: 549-560
[2] Turetsky MR,Kane ES, Harden JW, et al. Recent acceleration of biomass burning and carbon losses in Alaskan forests and peatlands. Nature Geoscience, 2010, 4: 27-31
[3] 郭福涛, 胡海清, 马志海, 等. 不同模型对拟合大兴安岭林火发生与气象因素关系的适用性. 应用生态学报, 2010, 21(1):159-164
[4] Lozano FJ, Suarez SS, Kelly M,et al. A multi-scale approach for modeling fire occurrence probability using satellite data and classification trees: A case study in a mountainous Mediterranean region. Remote Sensing of Environment, 2008, 112: 708-719
[5] 孙龙, 尚喆超, 胡海清. Poisson回归模型和负二项回归模型在林火预测领域的应用. 林业科学, 2012, 48(5): 126-129
[6] Mallinis G, Petrila M, Mitsopoulos L,et al. Geospatial patterns and drivers of forest fire occurrence in Romania. Applied Spatial Analysis and Policy, 2019, 12: 773-795
[7] 郭福涛, 苏漳文, 马祥庆, 等. 大兴安岭塔河地区雷击火发生驱动因子综合分析. 生态学报, 2015, 35(19): 6439-6448
[8] Martínez FJ, Chuvieco E, Koutsias N. Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression. Natural Hazards and Earth System Science, 2013, 13:311-327
[9] Rodrigues M, de la Riva J, Fotheringham S. Modeling the spatial variation of the explanatory factors of human-caused wildfires in Spain using geographically weighted logistic regression. Applied Geography, 2014, 48: 52-63
[10] Blozis SA, Harring JR. Fitting nonlinear mixed-effects models with alternative residual covariance structures. Sociological Methods & Research, 2021, 50: 531-566
[11] Groom JD, Hann DW, Temesgen H. Evaluation of mixed-effects models for predicting Douglas-fir mortality. Forest Ecology and Management, 2012, 276: 139-145
[12] Liu X, Hao YS, Widagdo FRA, et al. Predicting height to crown base of Larix olgensis in northeast China using UAV-LiDAR data and nonlinear mixed effects models. Remote Sensing, 2021, 13: 1834
[13] Zhou X, Chen Q, Sharma RP, et al. A climate sensitive mixed-effects diameter class mortality model for Prince Rupprecht larch (Larix gmelinii var. principis-rupprechtii) in northern China. Forest Ecology and Management, 2021, 491: 119091
[14] Hegeman EE, Dickson BG, Eochmann LJ. Probabilistic models of fire occurrence across National Park Service units within the Mojave Desert Network, USA. Landscape Ecology, 2014, 29: 1587-1600
[15] Drever CR, Drever MC, Messier C, et al. Fire and the relative roles of weather, climate and landscape characteristics in the Great Lakes-St. Lawrence forest of Canada. Journal of Vegetation Science, 2008, 19: 57-66
[16] Xiao YD, Zhang XQ, Ji P. Modeling forest fire occurrences using Count-Data mixed models in Qiannan autonomous prefecture of Guizhou Provincein China. PLoS One, 2015, 10(7): e0120621
[17] 国家林业和草原局. 中国林业和草原统计年鉴2018. 北京: 中国林业出版社, 2018: 4-5
[18] 梁慧玲, 林玉蕊, 杨光, 等. 基于气象因子的随机森林算法在塔河地区林火预测中的应用. 林业科学, 2016, 52(1): 89-98
[19] Guo FT, Su ZW, Tigabu M, et al. Spatial modelling of fire drivers in urban-forest ecosystems in China. Forests, 2017, 8: 180
[20] 蔡奇均, 曾爱聪, 苏漳文, 等. 基于Logistic回归模型的浙江省林火发生驱动因子分析. 西北农林科技大学学报:自然科学版, 2020, 48(2):102-109
[21] Carvalho A, Flannigan MD, Logan K, et al. Fire activity in Portugal and its relationship to weather and the Canadian Fire Weather Index system. International Journal of Wildland Fire, 2008, 17: 328-338
[22] 李春明. 基于广义线性混合效应模型的蒙古栎林单木枯损建模及影响因子分析. 林业科学研究, 2020, 33(6): 105-113
[23] Rawlings JO, Pantula SG, Dickey DA. Applied Regression Analysis: A Research Tool.New York: Springer, 1998
[24] Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning: Data Mining, Inferenceand Prediction. New York: Springer, 2001
[25] Burnham KP, Anderson DR. Model Selection and Multi-model Inference: A Practical Information-theoretic Approach. New York: Springer, 2002
[26] He ZM, Li LZ, Huang ZM, et al. Quantum-enhanced feature selection with forward selection and backward elimination. Quantum Information Processing, 2018, 17: 154
[27] 曾爱聪, 蔡奇均, 苏漳文, 等. 基于MODIS卫星火点的浙江省林火季节变化及驱动因子. 应用生态学报, 2020, 31(2): 399-406
[28] 苏漳文, 刘爱琴, 梁慧玲, 等. 基于气象因子的福建省森林火险预测模型. 森林与环境学报, 2015, 35(4): 370-376
[29] Wilgen BW, Biggs HC, Regan SP, et al. A fire history of the savanna ecosystems in the Kruger National Park, South. South African Journal of Science, 2000, 96: 167-178
[30] Spessa A, McBeth B, Prentice C. Relationships among fire frequency, rainfall and vegetation patterns in the wet-dry tropics of northern Australia: An analysis based on NOAA-AVHRR data. Global Ecology and Biogeography, 2005, 14: 439-454
[31] 杜军凯. 考虑垂直地带性的山区分布式水文模拟与应用. 博士论文. 北京: 中国水利水电科学研究院, 2019
[32] 国家林业和草原局. 全国森林防火规划(2016—2025年) [EB/OL]. (2016-12-29) [2021-07-06]. https://www.ndrc.gov.cn/fggz/fzzlgh/gjjzxgh/201705/W020191104624246431007.pdf
[33] 王涛, 董利虎, 李凤日. 基于混合效应的杂种落叶松人工幼龄林单木枯损模型. 北京林业大学学报, 2018, 40(10): 1-12
[34] 李春明, 赵丽芳, 李利学. 基于混合效应模型和零膨胀模型方法的蒙古栎林分水平枯损模型. 林业科学, 2019, 55(11):27-36
[35] Guo FT, Zhang LJ, Jin S, et al. Modeling anthropogenic fire occurrence in the boreal forest of China using logistic regression and random forests. Forests, 2016, 7: 250

混合效应模型在林火发生预测中的适用性

Applicability of mixed effect model in the prediction of forest fire

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	曹晓梅, 苗铮, 郝元朔, 董利虎. 基于树种分类的帽儿山阔叶混交林树高-胸径模型 [J]. 应用生态学报, 2024, 35(2): 307-320.
[2]	易达, 李凤日, 马爱云, 林富成, 郝元朔, 董利虎. 基于混合效应模型和分位数回归的长白落叶松枝下高模型构建 [J]. 应用生态学报, 2023, 34(4): 1035-1042.
[3]	高铭伟, 董秀强, 张崇良, 徐宾铎, 纪毓鹏, 任一平, 薛莹. 基于Delta-GAMMs模型研究海州湾5种关键饵料生物被摄食的影响因素 [J]. 应用生态学报, 2023, 34(4): 1137-1145.
[4]	高羽, 谢龙飞, 郝元朔, 董利虎. 考虑随机效应的长白落叶松立木生物量模型构建及精度分析 [J]. 应用生态学报, 2023, 34(2): 333-341.
[5]	佟艺玟, 陈东升, 冯健, 高慧淋. 基于线性分位数混合效应的辽东山区红松冠幅模型 [J]. 应用生态学报, 2022, 33(9): 2321-2330.
[6]	王雅丽, 朱海晨, 栗小东, 徐开达, 周永东, 蒋日进, 李振华. 浙江沿岸春秋季小黄鱼生长异质性 [J]. 应用生态学报, 2022, 33(8): 2244-2250.
[7]	闫明, 刘青青, 刘志萍, 奚为民. 干旱和林分因子对树木死亡的影响——以美国德克萨斯州东部国家森林为例 [J]. 应用生态学报, 2022, 33(11): 2897-2906.
[8]	朱文斌, 朱海晨, 王雅丽, 张亚洲, 卢占晖, 崔国辰. 基于线性混合效应模型的日本鳀幼鱼叉长-体重关系异质性 [J]. 应用生态学报, 2021, 32(12): 4532-4538.
[9]	裴如德, 高春霞, 田思泉, 麻秋云, 范青松. 浙江南部近海小黄鱼生长和性成熟特征的异质性 [J]. 应用生态学报, 2021, 32(1): 333-341.
[10]	罗也, 王君, 杨雨春, 及利, 朱瑞, 杨斌, 张丽杰, 祁永会. 利用随机效应模型模拟东北三省胡桃楸地位指数 [J]. 应用生态学报, 2020, 31(8): 2549-2557.
[11]	王佳慧, 董利虎, 李凤日. 帽儿山阔叶混交林天然更新幼苗幼树地径-树高模型 [J]. 应用生态学报, 2019, 30(11): 3811-3823.
[12]	谢龙飞, 董利虎, 李凤日. 人工长白落叶松立木叶面积预估模型 [J]. 应用生态学报, 2018, 29(9): 2843-2851.
[13]	魏士凯, 范顺祥, 张玉珍, 黄选瑞, 张志东. 塞罕坝自然保护区主要植被类型动态及其驱动力 [J]. 应用生态学报, 2018, 29(4): 1170-1178.
[14]	褚琳,张欣然,王天巍,李朝霞,蔡崇法. 基于CA-Markov和InVEST模型的城市景观格局与生境质量时空演变及预测 [J]. 应用生态学报, 2018, 29(12): 4106-4118.
[15]	王蒙, 李凤日, 金星姬. 基于抚育间伐效应的长白落叶松人工林两阶段枯死模型 [J]. 应用生态学报, 2017, 28(3): 779-788.