Spatio-temporal Variations Of Land Surface Temperature And Cold Island Effect In Ejina Oasis

As an important parameter for the surface energy balance,Land Surface Temperature(LST)is a key factor of surface physical process on regional and global scales.LST plays a signifancant role in global climate change,urban thermal environment and drought monitoring.Nearly 30% of the land's surface is in arid and semi-arid areas,and oasis,as the unique but significant component of arid area,is the foundation of regional economy and people's lives.With the rapid development of remote sensing technology in recent years,this technology has been used to deduce and monitor the surface temperature of the oasis area.Besides,it has been used in the study of spatial distribution pattern of oasis.Remote sensing technology has a theoretical and practical significance for the sustainable development of ecological economy in western areas of China.Therefore,this paper selects Ejina Oasis,a place in Inner Mongolia Autonomous Region,as the study area.Considering the actual situation of the study area,this paper establishes a quantitative relationship between atmospheric water vapor content and transmittance.Three Landsat images from 2009 to 2017 were used to retrieve LST of the Ejina oasis by the single-window algorithm.The surface temperatures of Ejina Oasis in the year of 2009,2013 and 2017 are deduced respectively with the help of a single window algorithm in this paper.This paper also analyzes spatial distribution characteristics and space-time change characteristics of LST.The cold island ratio index of Ejina Oasis is calculated and the changes of it in three different periods are discussed in the paper.Normalized Difference Vegetation Index(NDVI)and Modified Normalized Difference Water Index(MNDWI)are used to analyze the effect of vegetation and water on the surface temperature.This paper adopts the normalized surface temperature to calculate landscape index and analyzes landscape pattern of cold island effect in three different periods.Besides,this paper predict the LST of Ejina Oasis in the year of 2021.Main conclusions are as follows:(1)On the spatial scale,LST presents the pattern of water body <forest land < grassland < saline-alkali land < bare land,indicating an obvious oasis-centered "cold island effect" distribution pattern.On the temperal scale,LST of each land use type increased from 2009 to 2017 and the area of bare land and saline-alkali land increased with the decrease of grassland area.It is demonstrated that the increase of LST can be attributed to the aggravating of land desertification and salinizationin this region(2)Cold island indexes decreased from 2009 to 2013 and then increased from 2013 to 2017 with values of 0.0031,0.0020 and 0.0051,respectively.LST has negative correlations with NDVI and MNDWI.For every 0.1 unit increase of NDVI and MNDWI,LST will reduced about 4.46 K and 4.34 K,respectively.It is obvious thatboth vegetation and water have a certain cooling effect on LST,while the effect of vegetation is greater.(3)The Ejina normalized LST is divided into 7 levels.From 2009 to 2017,the area of the sub-high LST zone,high LST zone and ultra-high LST zone increases year by year,and the dominant patches develops to the high LST.The diversity and vergence index show that the high LST zone in 2009 The polymerization degree is the largest and the distribution is the most concentrated.In 2013,the degree of fragmentation was the highest and the complexity was the lowest.In 2017,the degree of fragmentation was the lowest and the complexity was the highest.The increase of the ultra-high LST zone has led to a high landscape index in 2017.From 2009 to 2017,the landscape indices showed a trend of decreasing first and then rising,which was the same trend as the cold island effect.(4)Prediction of the Ejina Oasis in 2021 shows that the area of the sub-middle LST zone,the middle LST zone in the oasis and the peripheral high LST zone and the ultra-high LST zone adjacent the oasis are reduced,but the area of the sub-middle LST zone adjacent the oasis is increased.Therefore,the cold island effect in 2021 is lower than that of 2017.Special high temperature areas distributed more extentively,leading a more complicated landscape pattern of the cold island effect in 2021 The aggregation degree and combination degree of each temperature plaque are lower,,indicating high fragmentation and abundance of various temperature plaques in 2021.