Drivers of human-caused fire ignitions in China’s forest regions: a comparative analysis

Abstract

Human-caused fires account for over 90% of wildfires in China, with drivers varying across forest regions. Identifying and comparing these drivers is critical for effective fire management and prevention. This study employs a binary logistic regression (BLR) model, utilizing verified human-caused fire data from 2000 to 2022, to analyze fire ignition drivers and map fire probabilities in three forested regions of China: northeastern (NEF), southeastern (SEF) and southwestern (SWF). Results indicate that fire weather and vegetation dominate NEF and SEF, whereas anthropogenic factors are more influential in SWF. In the NEF, regions with very high (VH) and high (H) fire risk are mainly concentrated along the eastern national border and at the junction of the Daxing’an and Xiaoxing’an Mountains, encompassing 33.33% of the forested area. In the SEF, VH and H fire risk zones are primarily situated in the far southern and northern parts, accounting for 35.54% of the forest area. In the SWF, VH and H fire risk areas are predominantly located in Yunnan Province, covering 26.43% of the forested region. Within regions, fire risk varies with vegetation type but aligns with similar levels of human activity. In NEF, high fire risk correlates with increased broadleaf and decreased mixed broadleaf-conifer cover. In SEF and SWF, high-risk areas exhibit reduced broadleaf presence, with SWF also showing less shrubland, emphasizing the critical role of fuel characteristics. Analysis of climate change scenarios revealed no significant trends, underscoring the multifaceted nature of fire ignition drivers.

Keywords

Human-caused fire ignition / Fire drivers / Logistic regression model / Fire risk map

References

[1] Balch JK, Bradley BA, Abatzoglou JT, Nagy RC, Fusco EJ, Mahood AL. Human-started wildfires expand the fire niche across the United States. Proc Natl Acad Sci U S A, 2017, 114(11): 2946-2951

[2] Bergeron Y, Gauthier S, Flannigan M, Kafka V. Fire regimes at the transition between mixedwood and coniferous boreal forest in northwestern Quebec. Ecology, 2004, 85(7): 1916-1932

[3] Bradstock RA. A biogeographic model of fire regimes in Australia: current and future implications. Glob Ecol Biogeogr, 2010, 19(2): 145-158

[4] Buntaine M, Mullen R, Lassoie J. Human use and conservation planning in Alpine areas of Northwestern Yunnan, China. Environ Dev Sustain, 2007, 9(3): 305-324

[5] Cardille JA, Ventura SJ. Occurrence of wildfire in the northern Great Lakes Region: effects of land cover and land ownership assessed at multiple scales. Int J Wildl Fire, 2001, 10(2): 145-154

[6] Cary GJ, Keane RE, Gardner RH, Lavorel S, Flannigan MD, Davies ID, Li C, Lenihan JM, Rupp TS, Mouillot F. Comparison of the sensitivity of landscape-fire-succession models to variation in terrain, fuel pattern, climate and weather. Landsc Ecol, 2006, 21(1): 121-137

[7] Catry FX, Rego FC, Bação FL, Moreira F. Modeling and mapping wildfire ignition risk in Portugal. Int J Wildl Fire, 2009, 18(8): 921-931

[8] Chang Y, Zhu ZL, Bu RC, Chen HW, Feng YT, Li YH, Hu YM, Wang ZC. Predicting fire occurrence patterns with logistic regression in Heilongjiang Province, China. Landsc Ecol, 2013, 28(10): 1989-2004

[9] Chang Y, Zhu ZL, Bu RC, Li YH, Hu YM. Environmental controls on the characteristics of mean number of forest fires and mean forest area burned (1987–2007) in China. For Ecol Manag, 2015, 356: 13-21

[10] Chen J, Di XY. Forest fire prevention management legal regime between China and the United States. J for Res, 2015, 26(2): 447-455

[11] Chen SS, Zang SY, Sun L. Characteristics of permafrost degradation in Northeast China and its ecological effects: a review. Sci Cold Arid Reg, 2020, 12: 1-11

[12] Copernicus Climate Change Service Climate Data Store (2019) Fire danger indices historical data from the Copernicus Emergency Management Service. https://doi.org/10.24381/cds.0e89c522

[13] Fusco EJ, Abatzoglou JT, Balch JK, Finn JT, Bradley BA. Quantifying the human influence on fire ignition across the western USA. Ecol Appl, 2016, 26(8): 2390-2401

[14] Goss M, Swain DL, Abatzoglou JT, Sarhadi A, Kolden CA, Williams AP, Diffenbaugh NS. Climate change is increasing the likelihood of extreme autumn wildfire conditions across California. Environ Res Lett, 2020, 15(9 094016

[15] Guo FT, Su ZW, Wang GY, Sun L, Lin FF, Liu AQ. Wildfire ignition in the forests of southeast China: identifying drivers and spatial distribution to predict wildfire likelihood. Appl Geogr, 2016, 66: 12-21

[16] Guo FT, Wang GY, Su ZW, Liang HL, Wang WH, Lin FF, Liu AQ. What drives forest fire in Fujian, China? Evidence from logistic regression and random forests. Int J Wildland Fire, 2016, 255): 505-519

[17] Guo FT, Su ZW, Wang GY, Sun L, Tigabu M, Yang XJ, Hu HQ. Understanding fire drivers and relative impacts in different Chinese forest ecosystems. Sci Total Environ, 2017, 605–606: 411-425

[18] Guo M, Yao QC, Suo HQ, Xu XX, Li J, He HS, Yin S, Li JN. The importance degree of weather elements in driving wildfire occurrence in mainland China. Ecol Indic, 2023, 148 110152

[19] Hantson S, Andela N, Goulden ML, Randerson JT. Human-ignited fires result in more extreme fire behavior and ecosystem impacts. Nat Commun, 2022, 13(1 2717

[20] Hocking RR (1976) A biometrics invited paper. the analysis and selection of variables in linear regression. Biometrics 32(1): 1. https://doi.org/10.2307/2529336

[21] Hong HY, Tsangaratos P, Ilia I, Liu JZ, Zhu AX, Xu C. Applying genetic algorithms to set the optimal combination of forest fire related variables and model forest fire susceptibility based on data mining models. The case of Dayu County, China. Sci Total Environ, 2018, 630: 1044-1056

[22] Jin RY (2022) Exploring human-caused fire occurrence prediction. Undergraduate Student Research Internships Conference

[23] Laschi A, Foderi C, Fabiano F, Neri F, Cambi M, Mariotti B, Marchi E. Forest road planning, construction and maintenance to improve forest fire fighting: a review. Croat J for Eng J Theory Appl for Eng, 2019, 401): 207-219

[24] Li XW, Zhao G, Yu XB, Yu Q. A comparison of forest fire indices for predicting fire risk in contrasting climates in China. Nat Hazards, 2014, 70(2): 1339-1356

[25] Liu ZH, Yang J, Chang Y, Weisberg PJ, He HS. Spatial patterns and drivers of fire occurrence and its future trend under climate change in a boreal forest of Northeast China. Glob Change Biol, 2012, 18(6): 2041-2056

[26] Liu H, Gong P, Wang J, Clinton N, Bai YQ, Liang SL. Annual dynamics of global land cover and its long-term changes from 1982 to 2015. Earth Syst Sci Data, 2020, 12(2): 1217-1243

[27] Liu ZH, Wang WJ, Ballantyne A, He HS, Wang XG, Liu SG, Ciais P, Wimberly MC, Piao SL, Yu KL, Yao QC, Liang Y, Wu ZW, Fang YT, Chen AP, Xu WR, Zhu JJ. Forest disturbance decreased in China from 1986 to 2020 despite regional variations. Commun Earth Environ, 2023, 4 15

[28] Luo KW, Wang XL, de Jong M, Flannigan M. Drought triggers and sustains overnight fires in North America. Nature, 2024, 627(8003): 321-327

[29] Martell DL, Otukol S, Stocks BJ. A logistic model for predicting daily people-caused forest fire occurrence in Ontario. Can J for Res, 1987, 17(5): 394-401

[30] Martínez J, Vega-Garcia C, Chuvieco E. Human-caused wildfire risk rating for prevention planning in Spain. J Environ Manage, 2009, 90(2): 1241-1252

[31] Martínez-Fernández J, Chuvieco E, Koutsias N. Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression. Nat Hazards Earth Syst Sci, 2013, 13(2): 311-327

[32] Matthew WJ, John TA, Sander V, Niels A, Gitta L, Matthias F, Adam JPS, Chantelle B, Richard AB, Guido RVDW, Stephen S, Josep GC, Cristina S, Crystal K, Stefan HD, Corinne LQ (2022) Global and regional trends and drivers of fire under climate change. Rev Geophys 60: e2020RG000726. https://doi.org/10.1029/2020rg000726

[33] Milanović S, Marković N, Pamučar D, Gigović L, Kostić P, Milanović SD. Forest fire probability mapping in eastern Serbia: logistic regression versus random forest method. Forests, 2021, 12(1): 5

[34] O’brien R. A caution regarding rules of thumb for variance inflation factors. Qual Quant, 2007, 41(5): 673-690

[35] Oliveira S, Oehler F, San-Miguel-Ayanz J, Camia A, Pereira JMC. Modeling spatial patterns of fire occurrence in Mediterranean Europe using multiple regression and random forest. For Ecol Manag, 2012, 275: 117-129

[36] Oliveira AS, Silva JS, Guiomar N, Fernandes P, Nereu M, Gaspar J, Lopes RFR, Rodrigues JPC. The effect of broadleaf forests in wildfire mitigation in the WUI–a simulation study. Int J Disaster Risk Reduct, 2023, 93 103788

[37] Parisien MA, Moritz MA. Environmental controls on the distribution of wildfire at multiple spatial scales. Ecol Monogr, 2009, 79(1): 127-154

[38] Pausas JG, Paula S. Fuel shapes the fire–climate relationship: evidence from Mediterranean ecosystems. Glob Ecol Biogeogr, 2012, 21(11): 1074-1082

[39] Phelps N, Woolford DG. Comparing calibrated statistical and machine learning methods for wildland fire occurrence prediction: a case study of human-caused fires in Lac La Biche, Alberta, Canada. Int J Wildland Fire, 2021, 30(11): 850-870

[40] Prichard SJ, Hessburg PF, Hagmann RK, Povak NA, Dobrowski SZ, Hurteau MD, Kane VR, Keane RE, Kobziar LN, Kolden CA, North M, Parks SA, Safford HD, Stevens JT, Yocom LL, Churchill DJ, Gray RW, Huffman DW, Lake FK, Khatri-Chhetri P. Adapting western North American forests to climate change and wildfires: 10 common questions. Ecol Appl, 2021, 31(8 e02433

[41] Sachdeva S, Bhatia T, Verma AK. GIS-based evolutionary optimized gradient boosted decision trees for forest fire susceptibility mapping. Nat Hazards, 2018, 923): 1399-1418

[42] Scott AC, Bowman DMJS, Bond WJ, Pyne SJ, Alexander ME. Fire on earth: an introduction, 2014, Hoboken, New Jersey, USA, Wiley Blackwell

[43] Seger C (2018) An investigation of categorical variable encoding techniques in machine learning: binary versus one-hot and feature hashing

[44] Senande-Rivera M, Insua-Costa D, Miguez-Macho G. Spatial and temporal expansion of global wildland fire activity in response to climate change. Nat Commun, 2022, 13(1): 1208

[45] Syphard AD, Keeley JE. Location, timing and extent of wildfire vary by cause of ignition. Int J Wildland Fire, 2015, 24(1): 37-47

[46] Syphard AD, Keeley JE, Pfaff AH, Ferschweiler K. Human presence diminishes the importance of climate in driving fire activity across the United States. Proc Natl Acad Sci U S A, 2017, 114(52): 13750-13755

[47] Vasconcelos M, Silva S, Tomé M, Alvim M, Pereira J. Spatial prediction of fire ignition probabilities: comparing logistic regression and neural networks. Photogramm Eng Remote Sens, 2001, 67(1): 73-81

[48] Vilar del Hoyo L, Martín Isabel MP, Martínez Vega FJ. Logistic regression models for human-caused wildfire risk estimation: analysing the effect of the spatial accuracy in fire occurrence data. Eur J for Res, 2011, 130(6): 983-996

[49] Vitolo C, Di Giuseppe F, Barnard C, Coughlan R, San-Miguel-Ayanz J, Libertá G, Krzeminski B. ERA5-based global meteorological wildfire danger maps. Sci Data, 2020, 7 216

[50] Van Wagner CE (1987) Development and structure of the canadian forest fire weather index system, Forestry Technical Report. Canadian Forestry Service, Ottawa

[51] Wang Z, Liu C, Alfredo H. From AVHRR-NDVI to MODIS-EVI: advances in vegetation index research. Acta Ecol Sin, 2003, 23(5): 979-987

[52] Wang WW, Wang XL, Flannigan MD, Guindon L, Swystun T, Castellanos-Acuna D, Wu WL, Wang GY. Canadian forests are more conducive to high-severity fires in recent decades. Science, 2025, 387(6729): 91-97

[53] Williams AP, Abatzoglou JT, Gershunov A, Guzman-Morales J, Bishop DA, Balch JK, Lettenmaier DP. Observed impacts of anthropogenic climate change on wildfire in California. Earths Future, 2019, 7(8): 892-910

[54] Wilson N, Yebra M. The role of climate in ignition frequency. Fire, 2023, 6(5 195

[55] Wotton BM. Interpreting and using outputs from the Canadian Forest Fire Danger Rating System in research applications. Environ Ecol Stat, 2009, 162): 107-131

[56] Wotton BM, Martell DL, Logan KA. Climate change and people-caused forest fire occurrence in Ontario. Clim Change, 2003, 60(3): 275-295

[57] Xiong QL, Luo XJ, Liang PH, Xiao Y, Xiao Q, Sun H, Pan KW, Wang LX, Li LJ, Pang XY. Fire from policy, human interventions, or biophysical factors? Temporal–spatial patterns of forest fire in southwestern China. For Ecol Manage, 2020, 474 118381

[58] Ying LX, Han J, Du YS, Shen ZH. Forest fire characteristics in China: spatial patterns and determinants with thresholds. For Ecol Manag, 2018, 424: 345-354

[59] Ying LX, Cheng HJ, Shen ZH, Guan PG, Luo CF, Peng XZ. Relative humidity and agricultural activities dominate wildfire ignitions in Yunnan, Southwest China: patterns, thresholds, and implications. Agric for Meteorol, 2021, 307 108540

[60] Yu Y, Mao JF, Wullschleger SD, Chen AP, Shi XY, Wang YP, Hoffman FM, Zhang YL, Pierce E. Machine learning-based observation-constrained projections reveal elevated global socioeconomic risks from wildfire. Nat Commun, 2022, 13(1 1250

[61] Zacharakis I, Tsihrintzis VA. Integrated wildfire danger models and factors: a review. Sci Total Environ, 2023, 899 165704

[62] Zeng JM (2018) The Classification System of Natural Forests and Its Geographic Distribution in Yunnan. J Southwest For Univ 38(6): 1–18. https://doi.org/10.11929/j.issn.2095-1914.2018.06.001

[63] Zhang ZX, Zhang HY, Li DX, Xu JW, Zhou DW. Spatial distribution pattern of human-caused fires in Hulunbeir grassland. Acta Ecol Sin, 2013, 337): 2023-2031

[64] Zong XZ, Tian XR. The process of vegetation recovery and burn probability changes in post-burn boreal forests in Northeast China. Int J Wildland Fire, 2022, 31(9): 886-900

[65] Zong XZ, Tian XR, Wang XL. An optimal firebreak design for the boreal forest of China. Sci Total Environ, 2021, 781 146822

[66] Zong XZ, Tian XR, Yao QC, Brown PM. An analysis of fatalities from forest fires in China, 1951–2018. Int J Wildland Fire, 2022, 31(5): 507-517

RIGHTS & PERMISSIONS

Northeast Forestry University