An alternative split-window algorithm for retrieving land surface temperature from Visible Infrared Imaging Radiometer Suite data. Issue 5 (19th March 2019)

• Record Type: Journal Article
• Authors: Wang, Chenguang; Duan, Si-Bo; Zhang, Xiaoyu; Wu, Hua; Gao, Mao-Fang; Leng, Pei
• Is Part Of: International journal of remote sensing. Volume 40:Issue 5/6(2019)
• Page Start: 1640
• Page End: 1654
• Publication Date: 2019-03-19
• Subjects: Remote sensing -- Periodicals; Télédétection -- Périodiques; 621.3678
• DOI: 10.1080/01431161.2018.1492180
• Languages: English
• ISSNs: 0143-1161

Abstract

ABSTRACT: Land surface temperature (LST) is a key parameter for surface water and energy budget calculations. To improve the accuracy of LST retrieval over bare soil surfaces, we proposed an alternative split-window algorithm to retrieve LST from Visible Infrared Imaging Radiometer Suite (VIIRS) thermal infrared data. Atmospheric water vapour content (WVC) and surface emissivity are key input parameters in the split-window algorithm. The National Centres for Environmental Prediction (NCEP) reanalysis data in combination with the Moderate Resolution Transmittance Code (MODTRAN) were used to estimate atmospheric WVC. The Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Emissivity Dataset (ASTER GED) product was used to correct emissivity effects. The accuracy of the retrieved LST was evaluated using in situ measurements collected from the Hailar and Urad Front Banner sites in China. The root mean square error (RMSE) values of approximately 1.10 and 1.70 K are obtained for grass/snow surfaces in the Hailar site and sand surfaces in the Urad Front Banner site, respectively. The VIIRS LST product is also validated using the in situ LST in the two sites. The RMSE values of approximately 2.50 and 5.80 K are achieved for the Hailar and Urad Front Banner sites, respectively. The results indicate that the proposed LST algorithm with surface emissivity estimated from the ASTER GED product has better accuracy of LST retrieval.