Coupling SNOWPACK-modeled grain size parameters with the HUT snow emission model

摘要

We studied whether the physical snow evolution model SNOWPACK could be used together with the HUT snow emission model to simulate microwave brightness temperatures (TB) of snow cover and to parameterize key a priori variables in the retrieval of snow water equivalent (SWE). We used the extensive in situ measurement data set collected in Sodankyla, Finland, during the Nordic Snow Radar Experiment (NoSREx) campaign in 2009-2013 to model the evolution of snow with SNOWPACK. Resulting snow profiles were validated with manual in situ measurements. Mean agreement scores (for a winter) were 0.85-0.91 for traditional grain size, 0.74-0.75 for optical grain size, 0.65-0.80 for density, and 0.71-0.83 for temperature. Grain sizes modeled with SNOW-PACK were compared to effective grain size retrieved from tower-based microwave radiometer measurements. The bias and RMS error of SNOWPACK optical grain size were -0.03 mm and 0.20 mm, respectively, and those of SNOWPACK traditional grain size were 0.30 mm and 0.33 mm, respectively. SNOWPACK snow profiles were used as input to the HUT snow emission model for calculation of TB, which was compared to microwave radiometer measurements. TB calculated with SNOWPACK optical grain size exhibited lower biases (from -12.5 k to 16.2 K, depending on year and frequency) and RMS errors (from 33 K to 18.5 K) than TB calculated with SNOW-PACK traditional grain size (bias from -42.2 K to -9.9 K, RMS error from 12.0 K to 44.7 K). Grain sizes, temperature, and density modeled with SNOWPACK were used as a priori snow data in the retrieval of SWE from tower based microwave radiometer observations. The lowest overall bias and RMS error were reached when traditional grain size from SNOW-PACK was used, either directly with modelled snow density and temperature ( 33 mm and 58 mm, respectively) or with an effective grain size correction and static snow density and temperature applied (22 mm and 59 mm, respectively). (C) 2016 Published by Elsevier Inc.