Integrated spatial generalized additive modeling for forest fire prediction: a case study in Fujian Province, China

Abstract

The increasing frequency of extreme weather events raises the likelihood of forest wildfires. Therefore, establishing an effective fire prediction model is vital for protecting human life and property, and the environment. This study aims to build a prediction model to understand the spatial characteristics and piecewise effects of forest fire drivers. Using monthly grid data from 2006 to 2020, a modeling study analyzed fire occurrences during the September to April fire season in Fujian Province, China. We compared the fitting performance of the logistic regression model (LRM), the generalized additive logistic model (GALM), and the spatial generalized additive logistic model (SGALM). The results indicate that SGALMs had the best fitting results and the highest prediction accuracy. Meteorological factors significantly impacted forest fires in Fujian Province. Areas with high fire incidence were mainly concentrated in the northwest and southeast. SGALMs improved the fitting effect of fire prediction models by considering spatial effects and the flexible fitting ability of nonlinear interpretation. This model provides piecewise interpretations of forest wildfire occurrences, which can be valuable for relevant departments and will assist forest managers in refining prevention measures based on temporal and spatial differences.

References

[1] Abid F. A survey of machine learning algorithms based forest fires prediction and detection systems. Fire Technol, 2021, 57(2): 559-590.

[2] Amraoui M, Pereira MG, DaCamara CC, Calado TJ. Atmospheric conditions associated with extreme fire activity in the Western Mediterranean region. Sci Total Environ, 2015, 524 32-39.

[3] Andela N, van der Werf GR. Recent trends in African fires driven by cropland expansion and El Niño to La Niña transition. Nat Clim Change, 2014, 4 791-795.

[4] Benini E, Ponza R. Nonparametric fitting of aerodynamic data using smoothing thin-plate splines. AIAA J, 2010, 48(7): 1403-1419.

[5] Biswas S, Vadrevu KP, Lwin ZM, Lasko K, Justice CO. Factors controlling vegetation fires in protected and non-protected areas of Myanmar. PLoS ONE, 2015, 10(4): e0124346.

[6] Boubeta M, Lombardía MJ, Marey-Pérez MF, Morales D. Prediction of forest fires occurrences with area-level Poisson mixed models. J Environ Manag, 2015, 154 151-158.

[7] Boyd K, Eng KH, Page CD Blockeel H, Kersting K, Nijssen S, Železný F. Area under the precision-recall curve: point estimates and confidence intervals. Lecture notes in computer science, 2013 Berlin Springer 451-466.

[8] Brillinger DR, Preisler HK, Benoit JW Goldstein DR Statistics and science: a festschrift for terry speed, 2003 Beachwood Institute of Mathematical Statistics 177-196.

[9] Brillinger DR, Preisler HK, Benoit JW. Probabilistic risk assessment for wildfires. Environmetrics, 2006, 17(6): 623-633.

[10] Bugallo M, Esteban MD, Marey-Pérez MF, Morales D. Wildfire prediction using zero-inflated negative binomial mixed models: application to Spain. J Environ Manag, 2023, 328 116788.

[11] Castro FX, Tudela A, Sebastià MT. Modeling moisture content in shrubs to predict fire risk in Catalonia (Spain). Agric for Meteorol, 2003, 116(1–2): 49-59.

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

[13] 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.

[14] Chen MH. Geostatistical analysis on soil and water loss in Fujian province. Chin J Nat Resour, 2011, 26(8): 1394-1400 (in Chinese)

[15] Chuvieco E, Cocero D, Riaño D, Martin P, Martínez-Vega J, de la Riva J, Pérez F. Combining NDVI and surface temperature for the estimation of live fuel moisture content in forest fire danger rating. Remote Sens Environ, 2004, 92(3): 322-331.

[16] Chuvieco E, Aguado I, Jurdao S, Pettinari ML, Yebra M, Salas J, Hantson S, de la Riva J, Ibarra P, Rodrigues M, Echeverría M, Azqueta D, Román MV, Bastarrika A, Martínez S, Recondo C, Zapico E, Martínez-Vega FJ. Integrating geospatial information into fire risk assessment. Int J Wildland Fire, 2014, 23(5): 606.

[17] Collins L, McCarthy G, Mellor A, Newell G, Smith L. Training data requirements for fire severity mapping using Landsat imagery and random forest. Remote Sens Environ, 2020, 245 111839.

[18] Costafreda-Aumedes S, Comas C, Vega-Garcia C. Human-caused fire occurrence modelling in perspective: a review. Int J Wildland Fire, 2017, 26(12): 983-998.

[19] Cutler DR, Edwards TC Jr, Beard KH, Cutler A, Hess KT, Gibson J, Lawler JJ. Random forests for classification in ecology. Ecology, 2007, 88(11): 2783-2792.

[20] Gao J, Shi Y, Zhang H, Chen X, Zhang W, Shen W, Xiao T, Zhang Y (2023) China regional 250 m normalized difference vegetation index data set (2000–2022). National Tibetan Plateau/Third Pole Environment Data Center. https://doi.org/10.11888/Terre.tpdc.300328

[21] Gaughan AE, Stevens FR, Huang ZJ, Nieves JJ, Sorichetta A, Lai SJ, Ye XY, Linard C, Hornby GM, Hay SI, Yu HJ, Tatem AJ. Spatiotemporal patterns of population in mainland China, 1990 to 2010. Sci Data, 2016, 3(1): 160005.

[22] Giglio L, Randerson JT, van der Werf GR. Analysis of daily, monthly, and annual burned area using the fourth-generation global fire emissions database (GFED4). JGR Biogeosci, 2013, 118(1): 317-328.

[23] González-Olabarria JR, Mola-Yudego B, Coll L. Different factors for different causes: analysis of the spatial aggregations of fire ignitions in Catalonia (Spain). Risk Anal, 2015, 35(7): 1197-1209.

[24] Guisan A, Edwards TC, Hastie T. Generalized linear and generalized additive models in studies of species distributions: setting the scene. Ecol Model, 2002, 157(2–3): 89-100.

[25] Guo FT, Innes JL, Wang GY, Ma XQ, Sun L, Hu HQ, Su ZW. Historic distribution and driving factors of human-caused fires in the Chinese boreal forest between 1972 and 2005. J Plant Ecol, 2015, 8(5): 480-490.

[26] Guo FT, Selvalakshmi S, Lin FF, Wang GY, Wang WH, Su ZW, Liu AQ. Geospatial information on geographical and human factors improved anthropogenic fire occurrence modeling in the Chinese boreal forest. Can J for Res, 2016, 46(4): 582-594.

[27] 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.

[28] 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, 25(5): 505.

[29] Guo FT, Su ZW, Tigabu M, Yang XJ, Lin FF, Liang HL, Wang GY. Spatial modelling of fire drivers in urban-forest ecosystems in China. Forests, 2017, 8(6): 180.

[30] 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.

[31] Hastie T, Tibshirani R, Firedman J The elements of statistical learning: data mining, inference and prediction, 2001 New York Springer Publishing.

[32] Hjort J, Luoto M (2013) Statistical methods for geomorphic distribution modeling. In: Shroder Jr J (Editor in Chief), Baas ACW (Ed.) Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 2, Quantitative Modeling of Geomorphology, pp. 59–73.

[33] Hu TY, Zhou GS. Drivers of lightning- and human-caused fire regimes in the Great Xing’an Mountains. For Ecol Manag, 2014, 329 49-58.

[34] Huang ZX, Huang X, Fan JC, Eichhorn M, An F, Chen BQ, Cao L, Zhu ZL, Yun T. Retrieval of aerodynamic parameters in rubber tree forests based on the computer simulation technique and terrestrial laser scanning data. Remote Sens, 2020, 12(8): 1318.

[35] Justice CO, Giglio L, Korontzi S, Owens J, Morisette J, Roy D, Descloitres J, Alleaume S, Petitcolin F, Kaufman Y. The MODIS fire products. Remote Sens Environ, 2002, 83(1): 244-262.

[36] Keilwagen J, Grosse I, Grau J. Area under precision-recall curves for weighted and unweighted data. PLoS ONE, 2014, 9(3): e92209.

[37] Koutsias N, Xanthopoulos G, Founda D, Xystrakis F, Nioti F, Pleniou M, Mallinis G, Arianoutsou M. On the relationships between forest fires and weather conditions in Greece from long-term national observations (1894–2010). Int J Wildland Fire, 2013, 22(4): 493-507.

[38] Krawchuk MA, Moritz MA, Parisien MA, Van Dorn J, Hayhoe K. Global pyrogeography: macro-scaled models for understanding the current and future distribution of fire. PLoS ONE, 2009, 4(4): e5102.

[39] Kuhn M (2008) Building predictive models inR Using the caret Package. J Stat Soft 28(5): 1–26. https://doi.org/10.18637/jss.v028.i05

[40] Li YD, Feng ZK, Chen SL, Zhao ZY, Wang FG. Application of the artificial neural network and support vector machines in forest fire prediction in the Guangxi autonomous region, China. Discrete Dyn Nat Soc, 2020, 2020 1-14.

[41] 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.

[42] Loepfe L, Martinez-Vilalta J, Piñol J. An integrative model of human-influenced fire regimes and landscape dynamics. Environ Model Softw, 2011, 26(8): 1028-1040.

[43] Magnussen S, Taylor SW. Prediction of daily lightning- and human-caused fires in British Columbia. Int J Wildland Fire, 2012, 21(4): 342.

[44] Martinez AI, Labib SM. Demystifying normalized difference vegetation index (NDVI) for greenness exposure assessments and policy interventions in urban greening. Environ Res, 2023, 220 115155.

[45] 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.

[46] Mohajane M, Costache R, Karimi F, Bao Pham Q, Essahlaoui A, Nguyen H, Laneve G, Oudija F. Application of remote sensing and machine learning algorithms for forest fire mapping in a Mediterranean area. Ecol Indic, 2021, 129 107869.

[47] Morandini F, Silvani X, Dupuy JL, Susset A. Fire spread across a sloping fuel bed: flame dynamics and heat transfers. Combust Flame, 2018, 190 158-170.

[48] National Forestry and Grassland Administration (2016) National forest fire prevention program (2016–2025). Beijing: [EB/OL]. (in Chinese)

[49] 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.

[50] Pang YQ, Li YD, Feng ZK, Feng ZM, Zhao ZY, Chen SL, Zhang HY. Forest fire occurrence prediction in China based on machine learning methods. Remote Sens, 2022, 14(21): 5546.

[51] Parisien MA, Snetsinger S, Greenberg JA, Nelson CR, Schoennagel T, Dobrowski SZ, Moritz MA. Spatial variability in wildfire probability across the western United States. Int J Wildland Fire, 2012, 21(4): 313.

[52] Pedersen EJ, Miller DL, Simpson GL, Ross N. Hierarchical generalized additive models in ecology: an introduction with MGCV. PeerJ, 2019, 7 e6876.

[53] Pham BT, Jaafari A, Avand M, Al-Ansari N, Du Dinh T, Yen HPH, Phong TV, Nguyen DH, Le HV, Mafi-Gholami D, Prakash I, Thi Thuy H, Tuyen TT. Performance evaluation of machine learning methods for forest fire modeling and prediction. Symmetry, 2020, 12(6): 1022.

[54] Phelps N, Woolford DG. Guidelines for effective evaluation and comparison of wildland fire occurrence prediction models. Int J Wildland Fire, 2021, 30(4): 225.

[55] 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.

[56] Preisler HK, Brillinger DR, Burgan RE, Benoit JW. Probability based models for estimation of wildfire risk. Int J Wildland Fire, 2004, 13(2): 133-142.

[57] Preisler HK, Chen SC, Fujioka F, Benoit JW, Westerling AL. Wildland fire probabilities estimated from weather model-deduced monthly mean fire danger indices. Int J Wildland Fire, 2008, 17(3): 305-316.

[58] Preisler HK, Westerling AL, Gebert KM, Munoz-Arriola F, Holmes TP. Spatially explicit forecasts of large wildland fire probability and suppression costs for California. Int J Wildland Fire, 2011, 20(4): 508-517.

[59] Quintanilha JA, Ho LL. Analyzing wildfire threat counts using a negative binomial regression model. Environmetrics, 2006, 17(6): 529-538.

[60] Read N, Duff TJ, Taylor PG. A lightning-caused wildfire ignition forecasting model for operational use. Agric For Meteorol, 2018, 253–254 233-246.

[61] Rodrigues M, Jiménez-Ruano A, Peña-Angulo D, de la Riva J. A comprehensive spatial-temporal analysis of driving factors of human-caused wildfires in Spain using geographically weighted logistic regression. J Environ Manag, 2018, 225 177-192.

[62] Rodríguez-Pérez JR, Ordóñez C, Roca-Pardiñas J, Vecín-Arias D, Castedo-Dorado F. Evaluating lightning-caused fire occurrence using spatial generalized additive models: a case study in central Spain. Risk Anal, 2020, 40(7): 1418-1437.

[63] Sachdeva S, Bhatia T, Verma AK. GIS-based evolutionary optimized gradient boosted decision trees for forest fire susceptibility mapping. Nat Hazards, 2018, 92(3): 1399-1418.

[64] Sakr GE, Elhajj IH, Mitri G. Efficient forest fire occurrence prediction for developing countries using two weather parameters. Eng Appl Artif Intell, 2011, 24(5): 888-894.

[65] Sevinc V, Kucuk O, Goltas M. A Bayesian network model for prediction and analysis of possible forest fire causes. For Ecol Manag, 2020, 457 117723.

[66] Simpson GL. Modelling palaeoecological time series using generalised additive models. Front Ecol Evol, 2018, 6 149.

[67] Stevens FR, Gaughan AE, Linard C, Tatem AJ. Disaggregating census data for population mapping using random forests with remotely-sensed and ancillary data. PLoS ONE, 2015, 10(2): e0107042.

[68] Su ZW, Tigabu M, Cao QQ, Wang GY, Hu HQ, Guo FT. Comparative analysis of spatial variation in forest fire drivers between boreal and subtropical ecosystems in China. For Ecol Manag, 2019, 454 117669.

[69] Su ZW, Xu ZH, Lin L, Chen YM, Hu HH, Wei SJ, Luo SS. Exploration of the contribution of fire carbon emissions to PM_2.5 and their influencing factors in Laotian tropical rainforests. Remote Sens, 2022, 14(16): 4052.

[70] Su ZW, Lin L, Xu ZH, Chen YM, Yang LM, Hu HH, Lin ZP, Wei SJ, Luo SS (2023) Modeling the effects of drivers on PM₂.5 in the Yangtze River Delta with geographically weighted random forest. Remote Sens 15(15): 3826. https://doi.org/10.3390/rs15153826

[71] Sun YY, Zhang FQ, Lin HF, Xu SW. A forest fire susceptibility modeling approach based on light gradient boosting machine algorithm. Remote Sens, 2022, 14(17): 4362.

[72] Syphard AD, Radeloff VC, Keeley JE, Hawbaker TJ, Clayton MK, Stewart SI, Hammer RB. Human influence on California fire regimes. Ecol Appl, 2007, 17(5): 1388-1402.

[73] Syphard AD, Radeloff VC, Keuler NS, Taylor RS, Hawbaker TJ, Stewart SI, Clayton MK. Predicting spatial patterns of fire on a southern California landscape. Int J Wildland Fire, 2008, 17(5): 602.

[74] Turco M, Llasat MC, von Hardenberg J, Provenzale A. Impact of climate variability on summer fires in a Mediterranean environment (northeastern Iberian Peninsula). Clim Change, 2013, 116(3): 665-678.

[75] Viedma O, Urbieta IR, Moreno JM. Wildfires and the role of their drivers are changing over time in a large rural area of west-central Spain. Sci Rep, 2018, 8(1): 17797.

[76] Vilar L, Woolford DG, Martell DL, Martín MP. A model for predicting human-caused wildfire occurrence in the region of Madrid. Spain Int J Wildland Fire, 2010, 19(3): 325.

[77] Wood SN. Stable and efficient multiple smoothing parameter estimation for generalized additive models. J Am Stat Assoc, 2004, 99(467): 673-686.

[78] Wood SN, Augustin NH. GAMs with integrated model selection using penalized regression splines and applications to environmental modelling. Ecol Model, 2002, 157(2–3): 157-177.

[79] Wood SN, Pya N, Säfken B. Smoothing parameter and model selection for general smooth models. J Am Stat Assoc, 2016, 111(516): 1548-1563.

[80] Woolford DG, Bellhouse DR, Braun WJ, Dean CB, Martell DL, Sun J. A spatio-temporal model for people-caused forest fire occurrence in the Romeo Malette forest. J Environ Stat, 2011, 2(1): 1-26

[81] Woolford D, Dean CB, Martell DL, Cao JG, Wotton BM. Lightning-caused forest fire risk in Northwestern Ontario, Canada, is increasing and associated with anomalies in fire weather. Environmetrics, 2014, 25 406-416.

[82] Woolford DG, Martell DL, McFayden CB, Evens J, Stacey A, Wotton BM, Boychuk D. The development and implementation of a human-caused wildland fire occurrence prediction system for the province of Ontario, Canada. Can J for Res, 2021, 51(2): 303-325.

[83] Wu ZW, He HS, Yang J, Liang Y. Defining fire environment zones in the boreal forests of Northeastern China. Sci Total Environ, 2015, 518 106-116.

[84] Yang J, He HS, Shifley SR, Gustafson EJ. Spatial patterns of modern period human-caused fire occurrence in the Missouri Ozark Highlands. For Sci, 2007, 53(1): 1-15.

[85] Yathish H, Athira KV, Preethi K, Pruthviraj U, Shetty A. A comparative analysis of forest fire risk zone mapping methods with expert knowledge. J Indian Soc Remote Sens, 2019, 47(12): 2047-2060.

[86] 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.

[87] Zeng AC, Yang S, Zhu H, Tigabu M, Su ZW, Wang GY, Guo FT. Spatiotemporal dynamics and climate influence of forest fires in Fujian province. China For, 2022, 13(3): 423.

[88] Zhang HJ, Qi PC, Guo GM. Improvement of fire danger modelling with geographically weighted logistic model. Int J Wildland Fire, 2014, 23(8): 1130.

[89] Zhang Y, Lim S, Sharples JJ. Modelling spatial patterns of wildfire occurrence in South-Eastern Australia. Geomat Nat Hazards Risk, 2016, 7(6): 1800-1815.

[90] Zhong YQ, Ning P, Yan S, Zhang CN, Xing J, Shi JW, Hao JM. A machine-learning approach for identifying dense-fires and assessing atmospheric emissions on the Indochina Peninsula, 2010–2020. Atmos Res, 2022, 278 106325.

RIGHTS & PERMISSIONS

Northeast Forestry University