Solar radiation budget derived by integrating ground-based and satellite observations with a Monte Carlo radiation model

Estimates of the global radiation budget from globally integrated data have been calculated using the MACR ( Monte Carlo Aerosol-Cloud-Radiation) radiative transfer model. The model was first run at daily and monthly frequencies for selected stations to validate against ground-based BSRN (Baseline Surface Radiation Network) network and satellite based CERES (Clouds and the Earth's Radiant Energy System) measurements. This study largely relies on observational inputs for estimating the fluxes at the surface and the top of the atmosphere (TOA). For the model inputs, the quality assured level 2.0 data from the AERONET (Aerosol Robotic Network) derived aerosol parameters (aerosol optical depth, single scattering albedo, and asymmetry factor), radiosonde retrieved column water vapor amount, and the TOMS (Total Ozone Mapping Spectrometer ) total ozone amount are used. The cloud fraction and cloud optical depth from the ISCCP (International Satellite Cloud Climatology Project)-D1 3 hour cloud data are used for the cloudy sky flux calculation. The comparison of the model results with diurnal mean surface solar radiations derived from the surface measurements indicates that the mean bias (MACR minus Observed) is around +5 W/m2 without clouds with an RMS error of +7 W/m2. The inclusion of clouds increases the bias error to only 7 W/m2 , but the RMS error is as high as 22 W/m2. At the TOA, the mean bias when compared with CERES data is less than 1 W/m2. The overall uncertainties of the clear sky calculation at the surface arising from variabilities in the input parameters are in the range of 5-6 W/m2 as derived from sensitivity tests. The larger uncertainties under cloudy sky conditions mainly come from the uncertainties in cloud optical depth and cloud fraction as well as additional errors of interpolation from grid to station values. We ran the MACR model globally with the 2000-2002 averaged input values, and then compared the outputs with the CERES data (2000-2002 average) and with the Earth Radiation Budget Experiment (ERBE) data (1985-1989 average). The global annual mean values show that the MACR derived annual mean TOA fluxes are within 1-4 W/m2 of satellite measurements under both clear and cloudy sky conditions. The MACR global mean fluxes show better agreement with the CERES than the ERBE under both clear and cloudy sky conditions (±1.0 W/m2 vs. 4.0 W/m2). The present study suggests that the planetary albedo is 0.29 while without clouds it is only 0.15. The global mean TOA cloud forcing is -47.7 W/m2, compared with the satellite estimates of -47.9 W/m2 (ERBE) and -48.2 W/m2 (CERES). The clear sky atmospheric absorption is 71 W/m2 and the surface absorption is 221 W/m2. Clouds enhance atmospheric absorption from 71 W/m2 to 79 W/m2 and decrease surface solar absorption from 221 W/m2 to 165 W/m2. The global mean clear sky aerosol radiative forcing at the TOA and the surface are -5.0 and -9.1 W/m2, respectively and are -2.6 W/m2 (at TOA) and -6.2 W/m2 (at the surface) in the presence of clouds.