A Temperature and Emissivity Separation Algorithm for Landsat-8 Thermal Infrared Sensor Data

Remote Sensing

Songhan Wang¹  Longhua He³  Wusheng Hu¹  Ruiliang Pu²  Richard Müller² 

[1] Southeast University, School of Transportation, No.2, Sipailou Road, Nanjing 210096, China; E-Mail:;Southeast University, School of Transportation, No.2, Sipailou Road, Nanjing 210096, China; E-Mail;Chinese Academy of Sciences, Nanjing Institute of Geography and Limnology, Key Laboratory of Watershed Geographic Sciences, No.73, East Beijing Road, Nanjing 210008, China

关键词: Landsat-8;    TIRS;    land surface temperature (LST);    land surface emissivity (LSE);    minimum-maximum emissivity difference method (MMD method);    emissivity log difference method (ELD method);    MODTRAN;    SURFRAD;   

DOI  :  10.3390/rs70809904

来源: mdpi

【 摘 要 】

On-board the Landsat-8 satellite, the Thermal Infrared Sensor (TIRS), which has two adjacent thermal channels centered roughly at 10.9 and 12.0 μm, has a great benefit for the land surface temperature (LST) retrieval. The single-channel algorithm (SC) and split-window algorithm (SW) have been applied to retrieve the LST from TIRS data, which need the land surface emissivity (LSE) as prior knowledge. Due to the big challenge of determining the LSE, this study develops a temperature and emissivity separation algorithm which can simultaneously retrieve the LST and LSE. Based on the laboratory emissivity spectrum data, the minimum-maximum emissivity difference module (MMD module) for TIRS data is developed. Then, an emissivity log difference method (ELD method) is developed to maintain the emissivity spectrum shape in the iterative process, which is based on the modified Wien’s approximation. Simulation results show that the root-mean-square-errors (RMSEs) are below 0.7 K for the LST and below 0.015 for the LSE. Based on the SURFRAD ground measurements, further evaluation demonstrates that the average absolute error of the LST is about 1.7 K, which indicated that the algorithm is capable of retrieving the LST and LSE simultaneously from TIRS data with fairly good results.

【 授权许可】

CC BY   
© 2015 by the authors; licensee MDPI, Basel, Switzerland.