[M-IS01] Environmental, Socio-Economic and Climatic Changes in Northern Eurasia

[MIS01-P04] Assessment of the long-term dynamics of forest fires and disturbance degree in southern Siberia using MODIS data

*Evgeny Shvetsov^1,2、Alexey Golyukov ^1,3,4、Vyacheslav Kharuk^1,3,4 (1.V.N. Sukachev Institute of Forest of the Siberian Branch of the Russian Academy of Sciences、2.Khakassian State University、3.Siberian Federal University、4.Tomsk State University)

キーワード：wildfires, fire radiative power, vegetation indices, climate, Siberia

Wildfire is one of the main disturbances affecting forest dynamics, succession, and the carbon budget in many Siberian regions including forests of southern Siberia. We analyzed spatial and temporal dynamics of burned area in the southern regions of Central Siberia. Using MODIS wildfire data (2002 – 2021) we considered burning rate within different forest types: dark needle conifers (DNC) composed by Siberian pine, spruce and fir, light-needle composed by Scotch pine and larch, and mix wood (composed by birch, aspen and conifers). To estimate the relationship between burning rate and climatic variables we used long-term (1982 – 2021) climatic variables from the ERA5-Land database and we also gravimetric (GRACE) data (2002 – 2021) for estimation soil moisture content. We found a significant decreasing trend in burned area. On average, about 978.7±447.5 thousand ha including 257.1±154.3 thousand ha of forests are disturbed by fire annually. Light coniferous forests are characterized by the highest degree of the fire disturbance (relative burned area, RBA, was ca. 0.83% per year); whereas DNC stands have the lowest RBA (ca. 0.15% per year). We found a decreasing logarithmic trend in burned area between plains and low mountains and high mountains due to lower forest fuel loads and increased amount of precipitation on higher elevations. The long-term dynamics of burned area in the region was related to the variation of climatic variables, such as soil moisture (r² = 0.70, p < 0.01), moisture content (r²= 0.63, p < 0.01) and drought index (r²= 0.3, p < 0.05). We also found that the highest tree mortality rate was observed in the DNC forests. Almost complete tree cover loss was observed on more than 50% of burned area in such stands. Statistically significant (p<0.05) differences between fire severity estimated using the dNBR index and corresponding tree mortality rate was observed for all dominant forest types within the study region.
The study was supported by the Russian Science Foundation and the Government of Republic of Khakassia (grant #22-17-20012, https://rscf.ru/en/project/22-17-20012/).