Retrieval of snow water equivalent over boreal forest in Canada from passive microwave measurements using the HUT snow emission model

Abstract

The objective of this research is to validate the Helsinki University of Technology (HUT) snow emission model and evaluate it's efficiency for snow water equivalent (SWE) estimation in a Canadian boreal forest environment. The model validation over the BOREAS study area shows a correlation (R/sup 2//spl ges/0.4) between the errors in the simulated brightness temperatures and the snow grain size, which controls the snow cover extinction behavior in the HUT model. Therefore, we propose a different empirical relationship for the extinction coefficient of the snow cover for the environment considered. Better predictions of brightness temperatures at 18 and 37 GHz in both polarizations were obtained using this relation. The accuracy of SWE estimated by inverse modeling based on minimization techniques was also improved, with a mean retrieval error on SWE of less than 15 mm and negligible bias. In comparison, conventional empirical algorithms, based on brightness temperature gradients, tend to underestimate SWE in boreal conditions with a mean retrieval error in the order of 20 to 30 mm.