Estimation of 30 m land surface temperatures over the entire Tibetan Plateau based on Landsat-7 ETM+data and machine learning methods  

作　　者：Xian Wang Lei Zhong Yaoming Ma 

机构地区：[1]Laboratory for Atmospheric Observation and Climate Environment Research,School of Earth and Space Sciences,University of Science and Technology of China,Hefei,People's Republic of China [2]CAS Center for Excellence in Comparative Planetology,Hefei,People's Republic of China [3]Jiangsu Collaborative Innovation Center for Climate Change,Nanjing,People's Republic of China [4]Key Laboratory of Tibetan Environment Changes and Land Surface Processes,Institute of Tibetan Plateau Research,Chinese Academy of Science,Beijing,People's Republic of China [5]CAS Center for Excellence in Tibetan Plateau Earth Sciences,Beijing,People's Republic of China [6]College of Earth and Planetary Sciences,University of Chinese Academy of Science,Beijing,People's Republic of China

出　　处：《International Journal of Digital Earth》2022年第1期1038-1055,共18页国际数字地球学报（英文）

基　　金：supported by the Second Tibetan Plateau Scientifc Expedition and Research(STEP)Program[grant number:2019QZKK0103];Strategic Priority Research Program of Chinese Academy of Sciences[grant number:XDA20060101];National Natural Science Foundation of China[grant number 41875031,41522501,41275028,91837208];The Chinese Academy of Sciences[grant number QYZDJSSW-DQC019];CLIMATE-TPE[grant number:32070]in the framework of the ESA-MOST Dragon 4 Programme.

摘　　要：Land surface temperature(LST)is an important parameter in land surface processes.Improving the accuracy of LST retrieval over the entire Tibetan Plateau(TP)using satellite images with high spatial resolution is an important and essential issue for studies of climate change on the TP.In this study,a random forest regression(RFR)model based on different land cover types and an improved generalized single-channel(SC)algorithm based on linear regression(LR)were proposed.Plateau-scale LST products with a 30 m spatial resolution from 2006 to 2017 were derived by 109,978 Landsat 7 Enhanced Thematic Mapper Plus images and the application of the Google Earth Engine.Validation between LST results obtained from different algorithms and in situ measurements from Tibetan observation and research platform showed that the root mean square errors of the LST results retrieved by the RFR and LR models were 1.890 and 2.767 K,respectively,which were smaller than that of the MODIS product(3.625 K)and the original SC method(5.836 K).

关 键 词：Google Earth Engine remote sensing machine learning land surface temperature random forest 

分 类 号：P41[天文地球—大气科学及气象学] TP39[自动化与计算机技术—计算机应用技术]