Earth Research from SpaceEarth Research from SpaceИсследование Земли из космоса0205-96143034-5405The Russian Academy of Sciences65916210.31857/S0205961424010012GNEJRHМЕТОДЫ И СРЕДСТВА ОБРАБОТКИ И ИНТЕРПРЕТАЦИИ КОСМИЧЕСКОЙ ИНФОРМАЦИИResearch ArticleForest Fire Risk Assessment and Mapping Using Remote Sensing and GIS Techniques: A Case Study in Nghe An Province, VietnamКартографирование и оценка риска лесных пожаров с использованием методов дистанционного зондирования и ГИС: тематическое исследование в провинции Нгеан, ВьетнамDoanThi Nam PhuongДоанТ. Н. Ф.
Viet Nam

Geomatics in Earth Sciences Research Group

исследовательская группа “Геоматика в науках о Земле”

trinhlehung@lqdtu.edu.vnTrinhLe HungЧиньЛ. Х.
Viet Nam
trinhlehung@lqdtu.edu.vnZablotskiiV. R.ЗаблоцкийВ. Р.
Russian Federation
trinhlehung@lqdtu.edu.vnNguyenVan TrungНгуенВ. Ч.
Viet Nam

Geomatics in Earth Sciences Research Group

исследовательская группа “Геоматика в науках о Земле”

trinhlehung@lqdtu.edu.vnTranXuan TruongЧанС. Ч.
Viet Nam

Geomatics in Earth Sciences Research Group

исследовательская группа “Геоматика в науках о Земле”

trinhlehung@lqdtu.edu.vnPhamThi Thanh HoaФамТ. Т. Х.
Viet Nam

Geomatics in Earth Sciences Research Group

исследовательская группа “Геоматика в науках о Земле”

trinhlehung@lqdtu.edu.vnLeThi Thu HaЛеТ. Т. Х.
Russian Federation

Geomatics in Earth Sciences Research Group

исследовательская группа “Геоматика в науках о Земле”

trinhlehung@lqdtu.edu.vnLeVan PhuЛеВ. Ф.
Viet Nam
trinhlehung@lqdtu.edu.vnHanoi University of Mining and GeologyХанойский горно-геологический университетLe Quy Don Technical UniversityТехнический университет им. Ле Куй ДонMoscow State University of Geodesy and CartographyМосковский государственный университет геодезии и картографии15012024131520022025Copyright ©; 2024, Russian Academy of SciencesCopyright ©; 2024, Российская академия наук2024Russian Academy of SciencesРоссийская академия наукhttps://journals.eco-vector.com/0205-9614/article/view/659162

This paper presents the results of modeling the risk of forest fires in the west of Nghe An Province (north-central Vietnam) using remote sensing and GIS data. The nine factors influencing the risk of forest fires, including vegetation cover (NDVI vegetation index), surface evapotranspiration, elevation (DEM), slope (slope), aspect, wind speed, ground surface temperature, average monthly precipitation and population density are used to build a forest fire risk mapping model based on machine learning methods, including Random Forest (RF), Suppor Vector Machine (SVM), and Classification and Regression Trees (CART). Various parameters are tested in the RF, SVM, CART algorithms to select the algorithm with the highest accuracy in forest fire risk prediction. The obtained results show that the RF algorithm with the value of the numberOfTrees parameter equal to 100 has the highest accuracy in predicting the risk of forest fires in the study area, expressed through the location of the distribution of forest fire points, as well as the AUC value on the ROC curve. The results obtained in the study can be effectively used for monitoring and early warning of forest fire danger in settlements, helping to reduce damage from forest fires.

В работе представлены результаты моделирования риска возникновения лесных пожаров на западе провинции Нгеан (северо-центральная часть Вьетнама), полученные на основе данных дистанционного зондирования и ГИС. С помощью методов машинного обучения: случайного леса (Random Forest), опорных векторов (Support Vector Machine), деревьев классификации и регрессии (Classification and Regression Trees) были построены модели возникновения лесных пожаров. В моделях учитывались девять основных факторов, определяющих вероятность возникновения лесных пожаров, среди них: количество фитомассы растительного покрова, поверхностная эвапотранспирация, высота местности над уровнем моря, наклон и экспозиция склона, скорость ветра, температура земной поверхности, среднемесячное количество осадков, плотность населения на территории. Различные значения параметров в алгоритмах машинного обучения были исследованы для выбора модели, наиболее точно предсказывающей возникновение лесных пожаров. Установлено, что метод случайного леса со значением параметра “количество деревьев решений”, равным 100, имеет наибольшую точность прогнозирования риска лесных пожаров на исследуемой территории.

forest fire riskremote sensingGISsupport vector machine algorithmNghe An provinceлесные пожарыоценки риска возникновениядистанционное зондированиеГИСметоды машинного обученияпровинция НгеанХанойский университет горного дела и геологииHUMG – Hanoi University of Mining and GeologyT23–38ИСТОЧНИК ФИНАНСИРОВАНИЯ Работа выполнена в рамках научного проекта: “Исследование модели прогнозирования риска лесных пожаров с использованием геопространственных технологий на примере западного региона провинции Нгеан”, код: T23–38. Авторы благодарят Ханойский университет горного дела и геологии (HUMG – Hanoi University of Mining and Geology) за финансирование проекта.Bondur V. G., Gordo K. A., Kladov V. L. Spacetime distributions of wildfire areas and emissions of carbon-containing gases and aerosols in northern Eurasia according to satellite-monitoring data // Izvestiya, Atmospheric and Oceanic Physics. 2017. Vol. 53. No. 9. P. 859–874. DOI: 10.1134/S0001433817090055.Бондур В.Г., Гордо К. А., Кладов В. Л. Пространственно-временные распределения площадей природных пожаров и эмиссий углеродсодержащих газов и аэрозолей на территории северной Евразии по данным космического мониторинга // Исследование Земли из космоса. 2016. № 6. С. 3–20. DOI: 10.7868/S0205961416060105.Bondur V. G., Gordo K. A. Satellite monitoring of burnt-out areas and emissions of harmful contaminants due to forest and other wildfires in Russia // Izvestiya, Atmospheric and Oceanic Physics. 2018. Vol. 54. No. 9. P. 955–965. DOI: 10.1134/S0001433818090104.Бондур В. Г., Гордо К. А. Космический мониторинг площадей, пройденных огнем, и объемов эмиссий вредных примесей при лесных и других природных пожарах на территории Российской Федерации // Исследование Земли из космоса. 2018. № 3. С. 41–55. DOI: 10.7868/S020596141803003X.Bondur V. G. Satellite monitoring of trace gas and aerosol emissions during wildfires in Russia // Izvestiya, Atmospheric and Oceanic Physics. 2016. Vol. 52. No. 9. P. 1078–1091. DOI: 10.1134/S0001433816090103.Бондур В. Г. Космический мониторинг эмиссий малых газовых компонент и аэрозолей при природных пожарах в России // Исследование Земли из космоса. 2015. № 6. С. 21–35. DOI: 10.7868/S0205961415060032.Bondur V. G., Ginzburg A. S. Emission of Carbon-Bearing Gases and Aerosols from Natural Fires on the Territory of Russia Based on Space Monitoring // Doklady Earth Sciences. 2016. Vol. 466. No. 2. P. 148–152. DOI: 10.1134/S1028334X16020045.Бондур В. Г., Гинзбург А. С. Эмиссия углеродсодержащих газов и аэрозолей от природных пожаров на территории России по данным космического мониторинга // Доклады академии наук. 2016. Т. 466. № 4. С. 473–477. DOI: 10.7868/S0869565216040186.Arpaci A., Malowerschnig B., Sass O., Vacik H. Using multi variate data mining techniques for estimating fire susceptibility of Tyrolean forests // Applied Geography. 2014. V. 53. P. 258–270.Beals E. A. Weather forecasts in the problem of protecting forests from fire // Monthly Weather Review. 1914. V. 42. P. 111–119.Bui T. D., Bui Q. T., Nguyen Q. P., Pradhan B., Nampak H., Phan T. T. A hybrid artificial intelligence approach using GIS-based neural-fuzzy inference system and particle swarm optimization for forest fire susceptibility modeling at a tropical area, Agricultural and Forest Meteorology. 2017. V. 233. P. 32–44.Bui T. D., Le T. K.T., Nguyen V. C., Le H. D., Revhaug I. Tropical Forest Fire Susceptibility Mapping at the Cat Ba National Park Area, Hai Phong City, Vietnam, Using GIS-Based Kernel Logistic Regression // Remote Sensing. 2016. V. 8. P. 347. DOI:10.3390/rs8040347.Bui T. D., Le T. K.T., Nguyen V. C., Le H. D., Revhaug I. Tropical Forest Fire Susceptibility Mapping at the Cat Ba National Park Area, Hai Phong City, Vietnam, Using GIS-Based Kernel Logistic Regression // Remote Sensing. 2016. V. 8. P. 347. DOI: 10.3390/rs8040347.Chowdhury H., Hassan K. Use of remote sensingderived variables in developing a forest fire danger forecasting system // Natural Hazards. 2013. V. 67. P. 321–334.Chowdhury H., Hassan K. Use of remote sensing derived variables in developing a forest fire danger forecasting system // Natural Hazards. 2013. V.67. P. 321–334.Dang N. B.T. Study on the risk and warning of forest fire in Son La province based on the application of geoinformation technology // Geography Doctoral Thesis, Hanoi National University. 2021.Doan H. P. Developing algorithms for determining land surface temperature in forest fire warning monitoring based on MODIS satellite images (TERRA and AQUA) in the territory of Vietnam // Thesis of Doctor of Engineering. Hanoi University of Mining and Geology. 2007.Doan H. P. Developing algorithms for determining land surface temperature in forest fire warning monitoring based on MODIS satellite images (TERRA and AQUA) in the territory of Vietnam. Hanoi: Vietnam. Thesis of Doctor of Engineering. Hanoi University of Mining and Geology. 2007.Dong X. Forest fire risk zone mapping from satellite images and GIS for Baihe Forestry Bureau, Jilin, China // Journal of Forestry Research. 2005. V. 16(3). P. 169–174.Fernandez J., Chuvieco E., Koutsias N. Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression // Natural Hazards Earth System Scieces. 2012. V. 12. P. 1–17.Fernandez J., Chuvieco E., Koutsias N. Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression // Natural Hazards Earth System Sciences. 2012. V. 12. P. 1–17.Enoh M., Okeke U., Narinua N. Identification and modelling of forest fire severity and risk zones in the Cross – Niger transition forest with remotely sensed satellite data // The Egyptian Journal of Remote Sensing and Space Science. 2021. V. 24(3). P. 879–887.Hererra V., Soon W., Moreno C., Hererra G., Dubois R., Cruz L., Fedorov V., Estrada S., Bongelli E., Zuniga E. Past and future of wildfires in Northern Hemisphere’s boreal forests // Forest Ecology and Management. 2022. V. 504. 119859.Hoang V. T., Chou T., Fang Y., Nguyen N. T., Nguyen Q. H., Pham X. C., Dang N. B.T., Nguyen X. L., Meadows M. Mapping forest fire risk and development of early warning system for NW Vietnam using AHP and MCA/GIS methods // Applied Sciences. 2020. V. 10(12). 4348.Jaiswal R., Mukherjee S., Raju K., Saxena R. Forest fire risk zone mapping from satellite imagery and GIS // International Journal of Applied Earth Observation and Geoinformation. 2002. V. 4(1). P. 1–10.Iban M., Sekertekin A. Machine learning based wildfire susceptibility mapping using remotely sensed fire data and GIS: A case study of Adana and Mersin provinces, Turkey // Ecological Informatics. 2022. V. 69. 101647.Nguyen N. T., Dang N. B.T., Pham X. C., Nguyen T. H., Bui T. H., Hoang D. N., Bui T. D. Spatial pattern asessment of tropical forest fire danger at Thuan Chau area (Vietnam) using GIS-based advanced machine learning algorithms: A comparative study // Ecological Informatics. 2018. V. 46. P. 74–85.Nguyen N. T., Dang N. B.T., Pham X. C., Nguyen T. H., Bui T. H., Hoang D. N., Bui T. D. Spatial pattern assessment of tropical forest fire danger at Thuan Chau area (Vietnam) using GIS-based advanced machine learning algorithms: A comparative study // Ecological Informatics. 2018. V. 46. P. 74–85.Oliveira S., Oehler F., Ayanz J., Camia A., Pereira J. Modeling spatial patterns of fire occurrence in Mediterranean Europe using Multiple Regression and Random Forest // Forest Ecology and Management. 2012. V. 275. P. 117–129.Pham N. H. Developing a method to predict forest fires Pinus merkusii J. in Quang Ninh province // Thesis of Doctor of Agricultural Science. Hanoi (in Vietnamese). 1988.Pourghasemi H. GIS-based forest fire susceptibility mapping in Iran: A comparison between evidential belief function and binary logistic regression models // Scandinavian Journal of Forest Research. 2015. P. 40. DOI: 10.1080/02827581.2015.1052750.Rouse, J.W., R. H. Haas, J. A. Schell, and D. W. Deering. Monitoring Vegetation Systems in the Great Plains with ERTS. // Third ERTS Symposium. 1973. NASA SP-351 P. 309–317.Rouse, J.W., R. H. Haas, J. A. Schell, and D. W. Deering. Monitoring Vegetation Systems in the Great Plains with ERTS. // Third ERTS Symposium. 1973. NASA SP-351. P. 309–317.Ruano A., Jolly W., Freeborn P., Nieva D., Vega N., Herrera C., Rodrigues M. Spatial Predictions of Human and Natural-Caused Wildfire Likelihood across Montana (USA) // Forests 2022. V.13(8). 1200.Ruano A., Jolly W., Freeborn P., Nieva D., Vega N., Herrera C., Rodrigues M. Spatial Predictions of Human and Natural-Caused Wildfire Likelihood across Montana (USA) // Forests 2022. V. 13(8). 1200.Sivrikaya F., Kucuk O. Modeling forest fire risk based on GIS-based analytical hierarchy process and statistical analysis in Mediterranean region // Ecological Informatics. 2022. V. 68. 101537.Tran Q. B., Phung N. T., Le N. H. Research on designing new software for early detection of forest firesfrom MODIS satellite image. // Journal of Agriculture and Rural Development. 2016. V. 21. P. 114–120.Tran Q. B., Phung N. T., Le N. H. Research on designing new software for early detection of forest fires from MODIS satellite image. // Journal of Agriculture and Rural Development. 2016. V. 21. P. 114–120.Tran V. H., Vo Q. M., Vo T. G. Geographic information system (GIS) aproach in forest fire warning methodology development for U Minh Ha national park // Science Journal of Can Tho University. 2010. V. 14. P. 97–106.Tran V. H., Vo Q. M., Vo T. G. Geographic information system (GIS) approach in forest fire warning methodology development for U Minh Ha national park // Science Journal of Can Tho University. 2010. V. 14. P. 97–106.Trinh L. H., Zablotskii V. R. The application of Landsat multi-temporal thermal infrared data to identify coal fire in the Khanh Hoa coal mine, Thai Nguyen province, Vietnam // Izvestiya. Atmospheric and Oceanic Physics. 2017. V. 53(9). P. 1181–1188.Trinh L. H. Studies of land surface temperature distribution using multispectral image Landsat // Vietnam Journal of Earth Sciences. 2014. V. 36(1). P. 82–89.Valor E., Caselles V. Mapping land surface emissivity from NDVI. Application to European African and South American areas // Remote Sensing of Environment. 1996. V. 57. P. 167–184.Vasilakos C., Kalabokidis K., Hatzopoulos J., Matsinos T. Identifying wildland fire ignition factors through sensitivity analysis of a neural network // Natural Hazards. 2009. V. 50. P. 125–143.Vo D. T. Methods of forecasting, mapping and zoning the key areas of forest fires in Binh Thuan // Forestry Journal. 1995. V. 10. P. 11–14 (in Vietnamese).Vuong V. Q. Research and develop solutions to prevent and overcome the consequences of forest fires for the U Minh and Central Highlands regions // Summary report of topic KC08.24. 2005.Yassemi S., Dragicevic S., Schmidt M. Design and implementation of an integrated GIS-based cellular automata model to characterize forest fire behavior // Ecological Modelling. 2008. V. 210. P. 71–84.Williams M. Remote sensing, GIS and wildland fire management: A global perspective // Proceedings of the International Workshop on Satellite Technology and GIS for Mediterranean Forest Mapping and Fire Management. 1983.Wimberly M., Reilly M. Assessment of fire severity and species diversity in the southern Appalachians using Landsat TM and ETM+ imagery // Remote Sensing of Environment. 2007. V. 108(2). P. 189–197.https://www.worldclim.org/. Accessed April 12. 2023.https://data.worldpop.org/. Accessed April 12. 2023.https://www.usgs.gov/landsat-missions/landsat-8-data-users-handbook. Accessed March 8. 2023.https://effis.jrc.ec.europa.eu/apps. Accessed April 15. 2023.