Study of land surface temperature and spectral emissivity using multi-sensor satellite data |
| |
Authors: | P K Srivastava T J Majumdar Amit K Bhattacharya |
| |
Institution: | (1) Graduate School of Environmental Studies, Nagoya University, Nagoya, Japan;(2) Department of Earth Sciences and Earth Dynamic System Research Center, National Cheng Kung University, Tainan, Taiwan |
| |
Abstract: | In this study, an attempt has been made to estimate land surface temperatures (LST) and spectral emissivities over a hard
rock terrain using multi-sensor satellite data. The study area, of about 6000 km2, is a part of Singhbhum-Orissa craton situated in the eastern part of India. TIR data from ASTER, MODIS and Landsat ETM+
have been used in the present study. Telatemp Model AG-42D Portable Infrared Thermometer was used for ground measurements
to validate the results derived from satellite (MODIS/ASTER) data. LSTs derived using Landsat ETM+ data of two different dates
have been compared with the satellite data (ASTER and MODIS) of those two dates. Various techniques, viz., temperature and
emissivity separation (TES) algorithm, gray body adjustment approach in TES algorithm, Split-Window algorithms and Single
Channel algorithm along with NDVI based emissivity approach have been used. LSTs derived from bands 31 and 32 of MODIS data
using Split-Window algorithms with higher viewing angle (50°) (LST1 and LST2) are found to have closer agreement with ground
temperature measurements (ground LST) over waterbody, Dalma forest and Simlipal forest, than that derived from ASTER data
(TES with AST 13). However, over agriculture land, there is some uncertainty and difference between the measured and the estimated
LSTs for both validation dates for all the derived LSTs. LST obtained using Single Channel algorithm with NDVI based emissivity
method in channel 13 of ASTER data has yielded closer agreement with ground measurements recorded over vegetation and mixed
lands of low spectral contrast. LST results obtained with TIR band 6 of Landsat ETM+ using Single Channel algorithm show close
agreement over Dalma forest, Simlipal forest and waterbody with LSTs obtained using MODIS and ASTER data for a different date.
Comparison of LSTs shows good agreement with ground measurements in thermally homogeneous area. However, results in agriculture
area with less homogeneity show difference of LST up to 2°C. The results of the present study indicate that continuous monitoring
of LST and emissivity can be undertaken with the aid of multi-sensor satellite data over a thermally homogeneous region. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|