Spring snow albedo feedback over northern Eurasia: Comparing in situ measurements with reanalysis products

• Published in: The Cryosphere
• Main Authors: Wegmann, Martin, Dutra, Emanuel, Jacobi, Hans-Werner, Zolina, Olga
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2018
• Subjects: article; Verlagsveröffentlichung; Merra; The Cryosphere
• Online Access: https://doi.org/10.5194/tc-12-1887-2018; https://noa.gwlb.de/receive/cop_mods_00005600; https://noa.gwlb.de/servlets/MCRFileNodeServlet/cop_derivate_00005557/tc-12-1887-2018.pdf; https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018.pdf

This study uses daily observations and modern reanalyses in order to evaluate reanalysis products over northern Eurasia regarding the spring snow albedo feedback (SAF) during the period from 2000 to 2013. We used the state-of-the-art reanalyses from ERA-Interim/Land and the Modern-Era Retrospective Analysis for Research and Applications version 2 (MERRA-2) as well as an experimental set-up of ERA-Interim/Land with prescribed short grass as land cover to enhance the comparability with the station data while underlining the caveats of comparing in situ observations with gridded data. Snow depth statistics derived from daily station data are well reproduced in all three reanalyses. However day-to-day albedo variability is notably higher at the stations than for any reanalysis product. The ERA-Interim grass set-up shows improved performance when representing albedo variability and generates comparable estimates for the snow albedo in spring. We find that modern reanalyses show a physically consistent representation of SAF, with realistic spatial patterns and area-averaged sensitivity estimates. However, station-based SAF values are significantly higher than in the reanalyses, which is mostly driven by the stronger contrast between snow and snow-free albedo. Switching to grass-only vegetation in ERA-Interim/Land increases the SAF values up to the level of station-based estimates. We found no significant trend in the examined 14-year time series of SAF, but interannual changes of about 0.5 % K−1 in both station-based and reanalysis estimates were derived. This interannual variability is primarily dominated by the variability in the snowmelt sensitivity, which is correctly captured in reanalysis products. Although modern reanalyses perform well for snow variables, efforts should be made to improve the representation of dynamic albedo changes.