Evaluation of the Earth Radiation Budget in NCEP–NCAR Reanalysis with ERBE

Abstract

This study presents an evaluation of the NCEP–NCAR Reanalysis (the reanalysis) by comparing its components of the earth radiation budget to satellite data. Monthly mean clear sky (CS) and total sky of outgoing longwave radiation (OLR), as well as reflected solar radiation (RSW) for 1985 and 1986, are compared to the top-of-the-atmosphere (TOA) measurements from the Earth Radiation Budget Experiment (ERBE). The ERBE-derived data of Staylor and Wilbur are also utilized to validate surface albedo. There are two objectives to this study: (i) to document the general quality of the reanalysis radiation budget, and (ii) to identify some of the general problem areas in the reanalysis global data assimilation system (GDAS). The OLR comparisons show that the global annual mean from the reanalysis is approximately 1.5% higher than that of ERBE. The zonal-average differences are strongly seasonal, which is particularly evident at high latitudes for the CS OLR, and at most latitudes for total-sky OLR. For the geographical distribution, the synoptic patterns from the reanalysis are in good agreement with the observations. Yet many regions in the Tropics and subtropics pose significant systematic biases. Possible causes are from shortcomings in the the cloud/moisture parameterizations of the reanalysis GDAS. The complex topography unresolvable by the T62 model could also be the cause for the biases in tall mountain regions. The global RSW comparisons show that the CS data from the reanalysis is in very good agreement with ERBE, while the total-sky RSW data overestimate ERBE by 12.6 W m⁻² (∼10%) globally. Persistent overestimates of RSW throughout the period indicate that the global energy budget for the reanalysis is not balanced. This result also is consistent with the finding in OLR suggesting that the reanalysis GDAS contains shortcomings in the cloud/moisture parameterizations. Another possibility for the difference in RSW is deficiencies in the GDAS shortwave parameterizations. Over the Sahara Desert, the reanalysis underestimates RSW, and overestimates OLR, both in the clear-sky and total-sky conditions. Comparison with the Staylor and Wilber ERBE-derived surface albedo suggests that GDAS surface albedo in this region should be increased by up to 0.1 (in albedo units). A comparison with the interannual variations of the satellite data for the boreal summer illustrates that the radiation budget data of the reanalysis contains a realistic climate signal.