基于Landsat 8 TIRS的地表温度反演算法对比分析

Comparison and Analysis of Land Surface Temperature Retrieval Algorithms Based on Landsat 8 TIRS

为研究地表温度反演算法之间的异同,以神东矿区为例,以Landsat 8 TIRS为数据源,利用辐射传导方程法、基于影像算法(image-based method,IB算法)、单窗算法(mono-window algorithm,MW算法)和单通道算法(single-channel method,SC算法)分别反演研究区地表温度,并用MODIS地表温度(land surface temperature,LST)数据进行验证,结果表明辐射传导方程法、MW算法和SC算法分别高于MODIS LST数据0.19 K、0.03 K和0.58 K,IB算法低于MODIS LST数据0.65 K,四种算法与MODIS LST数据的相关系数均达到0.9以上,其中MW算法与MODIS LST数据相关系数为0.971 1,精度最高。

In order to the study of similarities and differences between the surface temperature retrieval algorithm,taking Shendong mining area as an example,using Landsat 8 TIRS as the data source,the radiative transfer equation method,the inversion algorithm based on image（image-based method,IB algorithm）,single window algorithm（mono-window algorithm,MW algorithm） and single channel algorithm（SC algorithm） method were used to land surface temperature inversion.And use MODIS surface temperature data to verify the accuracy,the results showed that the radiation transfer equation method,MW algorithm and SC algorithm were higher than that of MODIS LST data 0.19 K,0.03 K,0.58 K,the IB algorithm is lower than that of the MODIS LST data 0.65 K,the correlation coefficient of four algorithm and MODIS LST data can reach more than 0.9,among them,the correlation coefficient between MW algorithm and MODIS LST data was 0.971 1,the highest precision was MW algorithm.