State of the Land Surface in the Arctic System Reanalysis
Andrew G. Slater1, David N. Kindig2, Mark C. Serreze3
1National Snow and Ice Data Center, University of Colorado, Campus Box 449, Boulder, CO, 80309-0449, USA, Phone 303-735-5358, aslater [at] kryos [dot] colorado [dot] edu
2National Snow and Ice Data Center, University of Colorado, Campus Box 449, Boulder, CO, 80309-0449, USA, kindig [at] kryos [dot] colorado [dot] edu
3National Snow and Ice Data Center, University of Colorado, Campus Box 449, Boulder, CO, 80309-0449, USA, serreze [at] kryos [dot] colorado [dot] edu
The Arctic System Reanalysis (ASR) is an Arctic specific reanalysis, conducted at a high spatial resolution (15km or less) for the 2000–2010 period. The Weather Research and Forecasting (WRF) model, with a polar optimized physics selection, will form the basis of the ASR. WRF will use the Noah land surface model. While the atmosphere will have the benefit of 3D-Var assimilation, the land surface is presently planned to run with minimal assimilation. However, land surface model improvements pertinent to the Arctic have been made, for example to the soil lower boundary, conditions have been developed to better represent permafrost. Additionally we have examined portions of the land surface shortwave radiative budget with particular emphasis on albedo. Using station data from multiple locations across Alaska we assess the incoming flux at the surface over the period August 2005–2006. Remote sensing data from the MODIS instruments is used to compare model results to snow cover fraction, snow albedo and most importantly mean terrestrial albedo. Results suggest an autumnal bias of excess modeled snow cover and albedo. In the spring, the model often provides a terrestrial albedo that is within the span of MODIS pixels contained by a gridbox however compared to the observed mean/median, the model often displays a systematic error.