Evaluation of Four Reanalysis Surface Albedo Data Sets in Arctic Using a Satellite Product

Abstract

Surface albedo has been widely used in studying energy budgets and climate dynamics in the Arctic region. Previous efforts have focused on using reanalysis albedo data, but their uncertainties remain unknown. In this letter, we evaluated four popularly used reanalysis surface albedo products, namely, the European Centre for Medium-Range Weather Forecasts Interim Reanalysis (ERA-Interim), the Modern-Era Retrospective Analysis for Research and Applications (MERRA), the National Centers for Environmental Prediction Climate Forecast System Reanalysis (CFSR), and the Japanese 55-Year Reanalysis (JRA-55), over the Arctic Ocean using satellite-retrieved product (CLARA-SAL) from 1982 to 2009. Owing to the flawed parameterization scheme or problematic model inputs, reanalysis products are unable to capture both the interannual variation and long-term reduction of surface albedo in the Arctic. This results in a large bias in the decline of shortwave radiative forcing at both surface (from-11.74 to-38.25 W m-2) and top of atmosphere (from-5.35 to-20.19 W m-2). The most significant underestimation occurred in the melt season and after sea-ice melting acceleration started since 1996, in the central Arctic Basin north of 80° N, which is likely due to the failure in simulating the influence of thinning ice and decreasing snow depth. The JRA-55 albedo product outperformed the other three products, which is likely due to the employment of observed sea-ice concentration on the parameterization scheme. On the other hand, the other three reanalysis products, namely, ERA-Interim, MERRA, and CFSR, are unable to effectively track the interannual variation of surface albedo and significantly underestimate (from-0.016 to-0.021 relative to-0.048, by one-third to half) the decreasing surface albedo.