Analyzing The Spatiotemporal Variations Of The Land Surface Temperature On The Tibetan Plateau From 2003-2020 Based On MODIS Data

Land Surface Temperature(LST)is an essential parameter in the research of agricultural drought monitoring,urban heat island and climate change.The Tibetan Plateau plays a significant role in regional and global climate system.Quantitative analysis of the LST helps to understand the energy balance of the Tibetan Plateau and its interaction with climate.Remote sensing satellite can obtain continuous data in the large area with better temporal and spatial resolutions,which solves the problem of uneven distribution of meteorological stations on the Tibetan Plateau.The Moderate Resolution Imaging Spectroradiometer(MODIS)8-day LST products with 1 km resolution from 2003 to 2020 are used in this paper.In order to remove the influence of cloud and other factors,the Harmonic Analysis of Time Series(HANTS)is used to reconstruct the MODIS8-day LST products from 2003 to 2020.Based on this,the spatiotemporal variation of the LST on the Tibetan Plateau is analyzed,including the variation of LST under different land cover types.After that,the relationship among LST,latitude and altitude is analyzed.Finally,the distribution of permafrost on the Tibetan Plateau is simulated by using surface frost number model,and the degradation trend of permafrost is analyzed.The main results of the thesis are as follows:(1)The quality control(QC)files in the MODIS LST products are analyzed,then MOD11A2day,MOD11A2 night,MYD11A2 day and MYD11A2 night are reconstructed using HANTS algorithm.The availability of MYD11A2 night data is the highest,and the proportion of low-quality pixels in 79.7%of the area is below 20%.The average value of the correlation coefficients between the reconstruction annual average LST and the annual average air temperature is 0.96.(2)Performing a linear regression analysis on the average annual LST during the day and night from 2003 to 2020,and the increase rate in LST at night is greater than during the day.The daytime warming area accounts for 2.44%of the entire Tibetan Plateau,and the nighttime warming area accounts for 30.99%.The land cover types of the International Geosphere Biosphere Programme(IGBP)are reclassified from 17 to 9.During the period 2003-2020,the 9 types of land cover show obvious warming trend.Among them,urban and built-up lands are the most affected by the warming.Therefore,in the ecological environment protection of the Tibetan Plateau,attention should be paid to urban and built-up lands,and the speed of their expansion should be appropriately limited.(3)The surface frost number model is used to simulate the distribution of permafrost on the Tibetan Plateau.From 2003 to 2020,the degradation rate of permafrost is about 0.13×10~4 km~2/a(4)During the day,the annual average LST shows a slow downward trend when the altitude is less than 2500 meters or the altitude is between 3000-5000 meters.When the altitude is greater than 5000 meters,the annual average LST shows a rapid downward trend.During the night,the annual average LST decreases slowly when the altitude is less than 2500 meters.When the altitude is higher than 3000 meters,as the altitude increases,the LST decreases linearly,at a rate of about0.46°C/100m.Perform multiple regression on LST,altitude and latitude,and standardize the correlation coefficients.The correlation coefficient between LST and altitude is-0.821,and the correlation coefficient with latitude is-0.337.