| With global warming,the frequency of fires in the Arctic region has shown a significant increase.The occurrence of fires will release a large amount of carbon dioxide,which intensifies further global warming.The warming Arctic increases the thickness of the active permafrost layer,which in turn accelerates the release of methane from the frozen soil.Therefore,based on active fire location data of FIRMS(Fire Information for Resource Management System)MODIS C6(MODIS NRT 1km active fire products,MCD14DL)from 2000 to 2019 and VIIRS V1(VIIRS NRT 375m active fire products,VNP14IMGTDL_NRT)from 2012 to 2019,according to different time periods(annual,Monthly,hourly)and spatial(national)scales analyze the dynamic changes of active fires and carbon emissions in the Arctic region.Based on GIS,it quantitatively analyzes the related characteristics of the 9 fire risk factors related to the frequency of active fires and the temperature of air at 2m above the surface of land,precipitation,wind speed at 10m above the surface of the Earth,NDVI,vegetation types,altitude,slope,distance to populated places and distance to roads,based on Logistic regression establish a fire risk assessment model.Analyze the impact of fire on the ecosystem based on the postfire net primary productivity and vegetation restoration.It aims to provide a basis for the prediction and management of active fires in the Arctic region,reduce the extinction of rare animals and plants,and reduce fire carbon emissions.The results show that:(1)From 2012 to 2019,the trend of active fires frequency of MODIS C6 and VIIRS V1 is consistent,with the maximum and minimum values appearing in 2019 and 2015 respectively.From 1997 to 2016,the maximum annual carbon emissions caused by fires appeared in 2003,which is closely related to the frequency of active fires and the area of fire.Active fires phenomena and carbon emissions in 7 countries in the Arctic region(Russia,United States,Canada,Denmark(Greenland),Norway,Sweden,and Finland)are mainly concentrated in June to August,and the active fires observation period is mainly concentrated around 12 o’clock local time in each country.The countries with the highest frequency of active fires and cumulative carbon emissions in the Arctic region are Russia,and the countries with the lowest are Norway and Denmark(Greenland).Fires in the Arctic region do not happen randomly,and are affected by certain factors.(2)From 2012 to 2019,the frequency of active fires in MODIS C6 and VIIRS V1 has the same trend with fire risk factors.The frequency of active fires has basically the same relationship with NDVI,altitude,slope,temperature of air at 2m above the surface of land,and precipitation,while the frequency of active fires There is a "double peak" structure at wind speed at 10m above the surface of the Earth,and the frequency of active fires generally decreases with the distance from the road and the distance from the populated places.Active fires are concentrated in open shrublands and savannas with vegetation types,NDVI around 0.4-0.8,altitude below 600m,slope below 10°,temperature of air at 2m above the surface of land at about 10-20℃,precipitation at about 0-3mm,wind speed at 10m above the surface of the Earth at about 2-6m/s,a distance of about 0-20km from roads and a distance of about 0-5km from populated places.Both the accuracy evaluation results and the risk evaluation results indicate that the risk model based on Logistic has good accuracy.(3)The NPP value decreased to the lowest in the first year after the fire,and then gradually recovered.The average recovery time of NPP for all fires with a fire area greater than 1 X 104 ha is about 11 years.The NDVI of vegetation with different burn levels showed an overall increasing trend,and their NDVI recovery ability was light burned,moderate burned,and severe burned in order from strong to weak.The recovery time of vegetation NDVI of mild and moderate fire is about the second year after the fire,while the recovery time of the severe fire is about the 16th year after the fire.Figure[36]table[10]reference[92]... |