An assessment of top-of-atmosphere radiative fluxes in reanalysis data products using NASA CERES broadband radiation observations

Monitoring the top-of-atmosphere (TOA) broadband radiation fields is essential to improve our understanding of the Earth's climate and its variability. The Clouds and the Earth's Radiant Energy System (CERES) instruments on the National Aeronautic and Space Administration (NASA) Terra and Aqua satellite have been gathering these important TOA radiation observations since March 2000 and July 2002, respectively. Based on an improved design of the legacy Earth Radiation Budget Experiment (ERBE) instrument, the CERES instrument produces very accurate and stable climate data set. As of today, the CERES instrument record contains the longest and most comprehensive global broadband radiation budget data set currently available for climate research. The advances in data assimilation technique over the past two decades have also enabled numbers of high quality atmospheric reanalysis data products for climate research. While the dynamic and thermodynamic variables in these data sets are constrained by three dimensional temperature, moisture, and wind observations, the TOA radiation fields in these products are not constrained by actual TOA radiation measurements. Therefore, changes in operational input data sources, instrument sensors, missing data sets, and other data inhomogeneity issues over time can degrade the reliability of the TOA radiation fields. Because these reanalysis data sets are often used to study the energetic of the climate system and its variability with time, it is important to understand the strength and weakness of their TOA radiation fields. The goal of this study is to assess the current state of TOA radiation fields from four state-of-the-art reanalysis data products (ERA-Interim, NASA MERRA, JRA-55, and NOAA CFSR) using CERES TOA broadband radiation observations. Climatological mean at regional, zonal, and global scale will be used to examine the similarities and differences between reanalysis products and CERES observations. Monthly deseasonalized time series will be employed to assess the quality of the interannual variability in these data sets. Trend analysis will be performed to detect drifts and other spurious data issues in the TOA radiation fields. Final recommendations will be provided for improving the quality of TOA radiation fields in the future generation of reanalysis data.