Publications

Nikam, BR; Ibragimov, F; Chouksey, A; Garg, V; Aggarwal, SP (2016). Retrieval of land surface temperature from Landsat 8 TIRS for the command area of Mula irrigation project. ENVIRONMENTAL EARTH SCIENCES, 75(16), 1169.

Abstract
Application of satellite remote sensing in generating various geophysical parameters of land surface has gained tremendous importance in many branches of science and applied research. Out of all geophysical parameters of land surface derived from satellite remote sensing, the land surface temperature (LST) is of prime importance. LST acts as a governing parameter in water and energy exchange between land and atmosphere. It is an essential input to all the numerical weather prediction models, most of the process-based hydrological models and even in irrigation water management activities. Many algorithms for space-based LST retrieval and operational products of LST are available nowadays. However, these products are generally coarser in spatial and/or temporal resolutions. In the sector of irrigation water management, higher spatial resolution of all the geophysical products is prerequisites. Hence, in the present study, LST has been retrieved using two popular algorithms viz. radiative transfer theory (RTT) equation-based method and split-window (SW) algorithm. LST of Mula irrigation project command area has been retrieved using twenty images of Landsat 8 Thermal Infrared Sensor (TIRS) and Operation Land Imager (OLI) for entire Rabi/Winter season. Atmospheric parameters needed for the LST retrieval have been derived using online atmospheric correction tool. The land surface emissivity has been estimated using NDVI threshold technique. The standard daily LST product of MODIS has been used to cross-validate the LST retrieved from both the algorithms. LST retrieved using SW algorithm shows higher correlation coefficient with MODIS LST in entire command area and in agricultural land as well (0.938 and 0.925, respectively) compared to that of LST output of RTT method (0.902 and 0.894, respectively). The relative comparison of the LST products derived from RTT and SW algorithms showed that there is an average difference of +/- 3 degrees K between these two products. However, the difference is very small (-0.8 to 0.5 degrees K) in case of agricultural area in the command. The higher correlation coefficient values between retrieved LST using both the algorithms and MODIS LST products hints toward the higher accuracy of derived LST at higher spatial resolution. These high spatial resolution LST products can further be used for irrigation water management.

DOI:
10.1007/s12665-016-5952-3

ISSN:
1866-6280