|Author: CaixiaGao,XiaoguangJiang,HuaWu,BohuiTang,ZiyangLi,and Zhao Liang Li
Land surface temperature (LST) is an important parameter in climatological and environmental research and can be derived from satellite thermal infrared data using various LST retrieval methods. Due to the differences in retrieval methods and input data sources, the LST generated from the Spinning Enhanced Visibleand Infrared Imager (SEVIRI)onboardthe MSG-2 satellite is different from that retrieved from the MODIS onboard Terra satellite. This study aims toevaluate the SEVIRI LST retrieved using the generalised split-window method with the land surface emissivity (LSE) estimated using the day/night Temperature-Independent Spectral Indices based method against the MODIS-derived LST extracted from theMOD11B1 V5 product during 7 clear-sky days. The results show that (1) discrepancies exist between the two LST products, with a maximum of 4.9 K on average; (2) these differences are considered to be time-dependent, i.e., higher discrepancies are observed during the daytime;(3) these differences are land-cover dependent, i.e., bare areas generally present larger differences than vegetated areas; (4) these differences depend on the view zenith angle (VZA) differences, inversely proportional to VZA differences. Finally, the main sources of LST differences are investigated and identified in terms of LSE, instrumental noise equivalent temperature difference (NEΔT) and mis-registration ofthe two LST products. These discrepancies may mainly result from errors in LSE, which are caused primarily by the atmospheric correction error for the SEVIRI-derived LST;the LST differences arising from NEΔT andmis-registration are within 0.4 K.