Decomposing Shortwave Top-of-Atmosphere and Surface Radiative Flux Variations in Terms of Surface and Atmospheric Contributions

• Published in: Journal of climate 2019-08, Vol.32 (16), p.5003-5019
• Main Authors: Loeb, Norman G., Wang, Hailan, Rose, Fred G., Kato, Seiji, Smith, William L., Sun-Mack, Sunny
• Format: Article
• Language: English
• Subjects: Albedo; Albedo (solar); Albedo variations; Annual variations; Anomalies; Atmosphere; Climate; Clouds; Diagnostic software; Diagnostic systems; Earth Resources And Remote Sensing; Energy balance; Fluctuations; Fluxes; Interannual variations; Satellite observation; Sea ice; Sea ice variations; Variability; Variance analysis; Variation
• ISSN: 0894-8755; 1520-0442
• DOI: 10.1175/JCLI-D-18-0826.1

A diagnostic tool for determining surface and atmospheric contributions to interannual variations in top-of-atmosphere (TOA) reflected shortwave (SW) and net downward SW surface radiative fluxes is introduced. The method requires only upward and downward radiative fluxes at the TOA and surface as input and therefore can readily be applied to both satellite-derived and model-generated radiative fluxes. Observations from the Clouds and the Earth’s Radiant Energy System (CERES) Energy Balanced and Filled (EBAF) Edition 4.0 product show that 81% of the monthly variability in global mean reflected SW TOA flux anomalies is associated with atmospheric variations (mainly clouds), 6% is from surface variations, and 13% is from atmosphere–surface covariability. Over the Arctic Ocean, most of the variability in both reflected SW TOA flux and net downward SW surface flux anomalies is explained by variations in sea ice and cloud fraction alone (r² = 0.94). Compared to CERES, variability in two reanalyses—the ECMWF interim reanalysis (ERA-Interim) and NASA’s Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2)—show large differences in the regional distribution of variance for both the atmospheric and surface contributions to anomalies in net downward SW surface flux. For MERRA-2 the atmospheric contribution is 17% too large compared to CERES while ERA-Interim underestimates the variance by 15%. The difference is mainly due to how cloud variations are represented in the reanalyses. The overall surface contribution in both ERA-Interim and MERRA-2 is smaller than CERES EBAF by 15% for ERA-Interim and 58% for MERRA-2, highlighting limitations of the reanalyses in representing surface albedo variations and their influence on SW radiative fluxes.