Geoinformation Modelling of Forest Fire Risk in the Sakha Republic (Yakutia)

• Published in: Успехи современного естествознания (Advances in Current Natural Sciences)
• Main Authors: Janiec, Pior, Gadal, Sébastien, Ivanova, Svetlana
• Other Authors: Aix Marseille Université (AMU), Études des Structures, des Processus d’Adaptation et des Changements de l’Espace (ESPACE), Université Côte d'Azur (UCA)-Avignon Université (AU)-Université Nice Sophia Antipolis (. - 2019) (UNS), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-COMUE Université Côte d'Azur (2015-2019) (COMUE UCA)-Centre National de la Recherche Scientifique (CNRS)-Aix Marseille Université (AMU), North-Eastern Federal University
• Format: Article in Journal/Newspaper
• Language: Russian
• Published: HAL CCSD 2019
• Subjects: Spatial Modelling; Machine learning; NDVI; Artificial intelligence; Arctic; Russia; Yakutia; GIS; Fire forest forecast; Forest fires; Forest fire risk; Remote Sensing; MODIS; [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV]; [SHS.STAT]Humanities and Social Sciences/Methods and statistics; [SDE.ES]Environmental Sciences/Environmental and Society; [SHS.GEO]Humanities and Social Sciences/Geography; [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation; Sakha; Republic of Sakha; Sakha Republic; Саха; Якути*; Якутия
• Online Access: https://hal.archives-ouvertes.fr/hal-02408358; https://hal.archives-ouvertes.fr/hal-02408358/document; https://hal.archives-ouvertes.fr/hal-02408358/file/37237.pdf; https://doi.org/10.17513/use.37237

International audience This article describes forest fires in Yakutia as one of the main types of natural hazards, as well as the factors of their occurrence and proposes methods of geoinformation modeling of forest fire risk in the Republic of Sakha (Yakutia). Forests of Yakutia occupy a significant area. Forest fires annually cover an area of hundreds of thousands of hectares. The article analyzes the impact of various factors on the risk of forest fires using the linear correlation coefficient and the coefficient of determination. GIS modeling was used to predict the risk of forest fires. It was carried out in stages: the collection of different types of data and their integration into the GIS database, checking the impact of individual factors on the fire occurrence, assessing the risk of forest fires using artificial intelligence and machine learning. The study used global sources of fire information-the fire information management system (FMS). The data available in the service are collected from satellite systems and MODIS Collection 6 Active Fire Product, as well as VIIRS 375 m, 750 m. The article presents the results of assessing the impact of individual factors on the occurrence of forest fires. The analysis showed that the greatest influence was exerted by such factors as solar radiation, maximum temperature (July), NDVI index, altitude, slope, human factor (distance to roads, distance from settlements and rivers). The following artificial intelligence and machine learning methods were used to assess fire risk: random forest and maximum entropy methods. With the help of GIS modeling, geographical maps of the risk of forest fires in Yakutia were created. The maps highlight areas of very low, low, medium, high, very high and extremely high probability of forest fires. В данной статье рассматриваются лесные пожары в Якутии как один из основных видов стихийных угроз природного характера, а также факторы их возникновения и предлагаются методы геоинформационно-го моделирования риска лесных пожаров на ...