Application of high-resolution thermal infrared remote sensing to assess land surface temperature and emissivity in different natural environments

Airborne thermal infrared image data at high spatial resolution (2-4 m) acquired with the Airborne Hyperspectral Scanner (AHS) over Barrax (Spain), Görmin (Germany) and Cabauw, Loobos and Speulderbos (The Netherlands) were used to study the thermal signatures of land cover types. The processing of the airborne imagery includes atmospheric correction of the AHS thermal infrared (TIR) bands and the retrieval of different biogeophysical variables such as Land Surface Temperature (LST), Land Surface Emissivity (LSE) and Evapotranspiration (ET). The AHS instruments consists of 80 spectral bands which cover the visible and near infrared (VNIR), short wave infrared (SWIR), mid-infrared (MIR) and thermal infrared (TIR) spectral range. This paper is mainly focused on the 10 AHS TIR bands (71 – 80), located in the region between 8 and 13µm, which are used for LST, LSE and ET. The AHS imagery used in this study was acquired in the framework of three different campaigns performed in different natural environments and supported by the European Space Agency (ESA): SEN2FLEX (June and July 2005), AGRISAR (June and July 2006) and EAGLE (July 2006). LST has been retrieved from AHS TIR data using Two-Channel (TC) and TES algorithms, taking into account different band configurations. LSE has been estimated using a simplified approach with the NDVI and also with the TES algorithm. LST and LSE products have been used to retrieve the daily ET applying the simplified SEBI model. The results have been tested against measurements collected in situ, showing errors lower than 2 K for temperature and errors around 1 mm/day for evapotranspiration.