Comparison of algorithms to retrieve Land Surface Temperature from LANDSAT-7 ETM+ IR data in the Basilicata Ionian band

Land Surface Temperature (LST) is an extremely important parameter that controls the exchange of longwave radiation and sensible heat flux between the Earth’s surface and the atmosphere. Therefore knowledge of LST is essential for a range of issues and themes in Earth sciences central to hydrology, climatology and global environmental change. In particular, it plays a main role in estimating hydrological variables, such as evapotranspiration. However, because of the extreme heterogeneity of most natural land surfaces, LST is a difficult parameter to estimate and to validate. In this study, two models by Qin et al. and Jim´enez-Mu˜noz and Sobrino were applied and compared for the evaluation of the LST on the Basilicata region (Southern Italy). These models were proposed in literature as alternatives to the application of the Radiative Transfer Equation (RTE) in order to overcome some difficulties in obtaining data from radio sounding and in schematizing mass and energy exchange processes in the atmosphere. Two images from Landsat-7 ETM+ (9th August, 1999; 14th June, 2002), covering the whole Basilicata region, were processed to obtain maps of LST. The required meteorological variables, air temperature and relative humidity, global solar radiation and wind speed, were obtained by interpolating data from a network of agro-meteorological stations distributed within the region. The variability of the LSTs retrieved was investigated with respect to different land use types characterized from the CORINE Land Cover map. Then a comparison was made between the LST retrieved by the application of the Qin et al. and the Jim´enez-Mu˜noz and Sobrino models and the in situ measurements of surface temperature taken at ALSIA (Agenzia Lucana di Sviluppo e di Innovazione in Agricoltura) weather stations located in the Ionian band of the Basilicata region. The results show (in agreement with previous works) that the Jim´enez-Mu˜noz and Sobrino model, in this case, is better able to approximate the measured data than the Qin et al. model, also using Landsat-7 ETM+ images and in a different context, such as that of the Lucan Ionian band.