Study On Land Surface Temperature Retrieving Algorithms Using Thermal Infrared Band Of ETM+ Data In Nanjing

Land surface temperature(LST) is a crucial parameter in indicting thermal energy distribution,the most efficient way to acquire it is through thermal infrared remote sensing images.Thermal infrared remote sensing is an important branch of remote sensing whereas LST retrieving is the premise of it.First of all,mono-window algorithm and single-channel algorithm based on Landsat TM and ETM+ images have been hitherto used in merely several areas so the broad applicability should be tested more deeply.Secondly,the contrast and comparison of the two algorithms have been hardly researched nowadays.So this paper retrieves LSTs of Nanjing using mono-window algorithm and single-channel algorithm and compares the results of these two algorithms.Thirdly,the thermal band of ETM+ consists of high gain and low gain statuses but the comparison of the two statuses in LST retrieving has not been researched by other people.So the paper retrieves LST through the two statuses separately and compares the results.This paper retrieves LSTs using mono-window algorithm and single-channel algorithm on low gain status and high gain status of ETM+ thermal infrared band in June,2000, Nanjing.Several spatial models are built to meet the goal.Based on the LSTs the paper analyzes the following problems.(1)spatial diversity distribution of LST.(2) single-channel LST and mono-window LST's diversity.(3) diversity of LSTs retrieved from these two algorithms and brightness temperature(4) diversity of low gain LST and high gain LST.The results show that:(1) LSTs of different underlying surface types are of great diversity.Thermal attribute of underlying surface,soil moisture,vegetation coverage and anthropogenic heat are the internal causes of these diversities.(2) mono-window LST and single-channel LST are close to each other,but the later is a bit higher than the former about 1.3℃and these two LSTs are both higher than brightness temperature,for the mono-window algorithm,it is about 1.7℃and for the single-channel algorithm,about 5.7℃.(3) Low gain status LST and high gain status LST are of a delicate difference,the former are slightly higher than the later.So the result illustrates that these two statuses are both suitable for Nanjing region LST retrieving.