Research On Inversion And Spatiotemporal Variation Of Surface Temperature In Northeast Permafrost Region Based On Passive Microwave Data

The northeast permafrost region is the main high-latitude permafrost region in my country.In recent years,the southern boundary of the permafrost distribution area has gradually moved northward,and the permafrost has been seriously degraded.Surface temperature is a key parameter in the surface energy balance.Accurate inversion of the surface temperature in this region can provide basic data for the study of climate change and permafrost degradation in the northeast permafrost region.At present,the surface temperature inversion method based on optical remote sensing is relatively mature,but the optical image is seriously affected by the atmosphere.Compared with optical remote sensing,passive microwave remote sensing can be observed all-weather and is almost unaffected by clouds and rain,but its spatial resolution is lower.Therefore,this paper uses the characteristics of the Aqua Earth observation satellite to carry MODIS and AMSR-E/AMSR2 sensors simultaneously,and establishes the linear regression analysis equation of AMSR-E/AMSR2 brightness temperature and MODIS surface temperature products.The surface temperature in 2020 is inverted,and the Dis Trad downscaling method is used to downscale the inversion temperature to improve the resolution of the data.Then,from 2002 to 2020,the surface temperature in March,June,September and December in large continuous permafrost areas,island-shaped thawing areas,permafrost areas,island-shaped permafrost areas and seasonally frozen soil areas was studied,and the The temporal variation characteristics and spatial distribution laws,the main conclusions are as follows:(1)The correlation between the obtained surface temperature and the real surface temperature is good,the correlation coefficient R~2is above 0.8,and most of the points are located near the 1:1 line,which proves that the adopted surface temperature inversion method has good performance.feasibility.The inversion accuracy at night is higher than that in the daytime,with the highest inversion accuracy in summer and the lowest in winter.(2)The Dis Trad model is used to downscale the resolution of the inverted surface temperature data to 500 m.The downscaled results describe the spatial differences of the surface temperature more precisely and reflect the details of temperature changes.There are differences in the spatial distribution of temperature grades in different seasons.In spring,the temperature in the southwest is high and the temperature in the northeast is low;in summer,the temperature in the west is high and the temperature in the east is low;in autumn,the temperature in the south is high,and the temperature in the central and western regions is low;in winter,the temperature gradually decreases from south to north.(3)During the 19 years from 2002 to 2020,the study area was dominated by the medium-temperature area,and the area proportion of each temperature class changed with the year and season.The extremely low-temperature area in winter accounted for the largest proportion of the four seasons,and the Extremely high temperature areas account for the most.With the change of time,the area of extremely low temperature area and low temperature area gradually decreased,and the area of high temperature area and extremely high temperature area gradually showed an upward trend.(4)Through analysis,it is found that the temperature of the four types of the seasonally frozen regions,isolated patches of thawing permafrost regions,island permafrost regions,continuous permafrost regions area all show a slow upward trend with time.