Identification of atmospheric influences on the estimation of snow water equivalent from AMSR-E measurements

Radiometric measurements near 19 and 37 GHz have been used for estimation of snow water equivalent (SWE) for many years. Most conventional SWE retrieval algorithms depend on the difference between the brightness temperatures (Tbʹs) at these frequencies. The effect of atmospheric absorption is generally assumed to be insignificant, and thus often not taken into account in such estimation. In this paper this effect is closely examined with the aid of AMSR-E and radiosonde data sets over two widely separated regions in the continental U.S.A. Results of the analysis show that even under a clear sky the atmospheric absorption could account for as much as ∼ 25–50% to the estimation of SWE. For example, the AMSR-E estimated SWE of 10 cm would become about 13.6 cm when measured at the ground level under the same atmospheric condition; the estimation based on surface emission alone (i.e., no atmosphere) would be about 15.2 cm. There is some regional dependence of this atmospheric absorption effect, but the effect of seasonal variation is negligibly small. Under cloudy conditions, the impact of liquid cloud absorption is significant and it appears necessary to perform either cloud screening or quantify the cloud effects on SWE estimation from 19 to 37 GHz radiometric measurements at ground level or satellite altitudes.