LONG-TERM GIS MODELING OF THE FOREST FIRE RISK IN THE SAKHA REPUBLIC (YAKUTIA) (RUSSIAN FEDERATION)

The article discusses the features of forest fires in Russia and Yakutia, their role in global climate warming, factors affecting forest fires, suggests a GIS structure for monitoring forest fires. The source material was satellite images of different spatial and spectral resolution (Landsat 5, Modis TERRA, GMTED2010, VIIRS), vector data (NextGIS), various meteorological data (WORLDCLIM), the results of expeditions and surveys of the local population. For the first time, a structure and a GIS database of fires for 2001-2018 were developed for the territory of Yakutia, which contains information about the relief, climate, combustible materials, and anthropogenic activity. For the first time, based on an analysis of literary sources and world experience of assessing the risk of forest fires, a method was developed for determining macroeconomic and microscale factors influencing fire risk as well as a method for determining this risk in the Republic of Sakha (Yakutia) using artificial intelligence and machine learning.

MaxEnt, wildfires, remote sensing, risk modeling, GIS, multi-scale analysis, Yakutia, Siberia, artificial intelligence, machine learning, random forest, MaxEnt

1. Kasischke, E.S. Fire, climate change, and carbon cycling in the boreal forest. Vol. 138. - Springer Science & Business Media, 2012. - 464 p.

2. Troeva, E.I. The Far North: Plant Biodiversity and Ecology of Yakutia. Vol. 3. / E.I. Troeva, A.P. Isaev, M.M. Cherosov, N.S. Karpov // Springer Science & Business Media, 2010. - 385 p.

3. San-Miguel-Ayanz, J. Current methods to assess fire danger potential / J. San-Miguel-Ayanz, J.D. Carlson, M. Alexander, K. Tolhurst, G. Morgan, R. Sneeuwjagt, M. Dudley // Wildland fire danger estimation and mapping: The role of remote sensing data. - 2003. - P. 21-61.

4. Жданко В.A., Гриценко В.А. Метод анализа лесопожарных сезонов: практические рекомендации. - Л.: ЛНИИЛХ, 1980. - 19 с.

5. Gralewicz, N. J. Factors influencing national scale wildfire susceptibility in Canada / N.J. Gralewicz, T.A. Nelson, M.A. Wulder // Forest Ecology and Management. - 2012. - Vol. 265. - P. 20-29.

6. Davies, D.K. Fire information for resource management system: archiving and distributing MODIS active fire data / D.K. Davies, S.Ilavajhala, M.M. Wong, C.O. Justice // IEEE Transactions on Geoscience and Remote Sensing. - 2009. - Vol. 47, No. 1. - P. 72-79.

7. Fick, S.E. WorldClim 2: new 1 km2 spatial resolution climate surfaces for global land areas / S.E. Fick, R.J. Hijmans, J. Robert // International Journal of Climatology. - 2017. - Vol. 37, No. 12. - P. 4302-4315.

8. Goward, C.R. Expression and purification of a truncated recombinant streptococcal protein G / C.R. Goward, J.P. Murphy, T. Atkinson, D.A. Barstow // Biochemical Journal. - 1990. - Vol. 267, No. 1. - P. 171-177.

9. Illera, P. Temporal evolution of the NDVI as an indicator of forest fire danger / P. Illera, A. Fernandez, J.A. Delgado // International Journal of remote sensing. - 1996. - Vol. 17, No. 6. - P. 1093-1105.

10. Parisien, M.A. Environmental controls on the distribution of wildfire at multiple spatial scales / M.A. Parisien, M.A. Mortiz // Ecological Monographs. - 2009. - Vol. 79, No. 1. - P. 127-154.

11. Renard Q. Environmental susceptibility model for predicting forest fire occurrence in the western Ghats of India / Q. Renard, R. Pelissier, B.R. Ramesh, N. Kodandapani // International Journal of Wildland Fire. - 2012. - Vol. 21, No. 4. - P. 368-379.

12. Cutler, D.R. Random forests for classification in ecology / D. R. Cutler, et al. // Ecology. - 2007. - Vol. 88, No. 11. - P. 2783-2792.

13. Cortez, P.A., Morais, A.D. J.R. Data mining approach to predict forest fires using meteorological data // New trends in artificial intelligence: proceedings of the 13th Portuguese Conference on Artificial Intelligence (EPIA 2007).- 2007. - P. 512-523.