Recovering Surface Temperature and Emissivity from Thermal Infrared Multispectral Data

Abstract

In 1992 Thermal Infrared Multispectral Scanner (TIMS) data were acquired from the NASA C-130 aircraft over the Sahelian region of West Africa as part of the Hydrological and Atmospheric Pilot Experiment in the Sahel (HAPEX). TIMS measures the radiation from the surface modified by the atmosphere in six channels located between 8 mm and 12.5 μm in the thermal infrared. By using a variety of techniques it is possible to extract both the surface temperature and surface emissivity from the areas over which TIMS data were acquired. One such technique was tested with the data acquired during this experiment. Several TIMS images of both the east and west central sites on 2 and 4 September were processed, and the spectral behaviors of different land cover types were determined. These included tiger bush, millet, and fallow grassland sites. There was a 5–10 K difference in the brightness temperature over the six channels when significant bare soil was visible. Channels 1–3 (8.2–9.4 μm) were cooler than the longer wavelength channels (9.6–12.5 μm), which is characteristic of soils rich in quartz. These differences in brightness were converted to emissivity differences using the max–min difference (MMD) method. This method relies on an empirical relationship observed between the range of emissivities and the minimum value for the six TIMS channels. The MMD method was applied iteratively to several entire scenes for the east central site on the two days with the interesting results that Channel 5 showed very little spatial variation in emissivity and the short wavelength channels observed substantial regions with emissivities of about 0.8 or less. There is excellent reproducibility when the same area is seen in different lines on the same day. However, there are differences when the same area is seen on the two days especially for the low emissivity values. Some of these differences may be due to soil moisture differences of 2–3%, which were observed for the two days. The observed surface temperatures were in good agreement with other measures, for example, vegetation temperatures agreed well with the measured air temperatures. Published by Elsevier Science Inc.