Anisotropy of Land Surface Skin Temperature Derived from Satellite Data

Abstract

The land skin temperature, an important feature for agricultural monitoring, convective processes, and the earth?s radiation budget, is monitored from limited-view satellite imagers. The angular dependence of this parameter is examined using simultaneous views of clear areas from up to three geostationary satellites. Daytime temperatures from different satellites differed by up to 6 K and varied as a function of the time of day. Larger differences are expected to occur but were not measured because of limited viewing angles. These differences suggest that biases may occur in both the magnitude and phase of the diurnal cycle of skin temperature and its mean value whenever geostationary satellite data are used to determine skin temperature. The temperature differences were found over both flat and mountainous regions with some slight dependence on vegetation. The timing and magnitude of the temperature differences provide some initial validation for relatively complex model calculations of skin temperature variability. The temperature differences are strongly correlated with terrain and the anisotropy of reflected solar radiation for typical land surfaces. These strong dependencies suggest the possibility for the development of a simple empirical approach for characterizing the temperature anisotropy. Additional research using a much greater range of viewing angles is required to confirm the potential of the suggested empirical approach.