Since we produced our first comprehensive diagnosis of radiative fluxes at the top of the atmosphere (TOA) and surface in the early 90's (Zhang et al., 1995), the NASA GISS radiation model has undergone substantial revision. Also, the main input data for our flux calculation, new improved ISCCP D series data (D1 and DX), has become available and is currently being produced.

The improvements or new features of our current radiative transfer model include: (1) the spectral resolution of the correlated k-distribution method increased to 33 k's from 25 k's; (2) better water vapor continuum parameterization for LW calculation; (3) treatment of ice cloud scattering with non-spherical particle shape; (4) better size and solar zenith angle dependence in SW scattering calculation for aerosols and clouds; (5) better ocean albedo; (6) correction of land surface reflectance from D1 due to aerosols; (7) introduction of non-unit emissivity for surface skin temperature correction and LW flux calculation; (8) parameterization of inhomogeneous cloud effects; (9) introduction of cloud particle size climatology; (10) updated aerosol climatology; (11) better cloud vertical structure climatology and (12) introduction of climatology of diurnal variation to temperature profile of TOVS that is used in our standard flux production.

We have tested the new radiation code using the current ISCCP D1 data as the main input for 1986. Compared with ERBE, the new results have improved TOA global mean fluxes by several Watts/m2 against our old code with ISCCP C1 data (Rossow and Zhang, 1995). This reduces our TOA uncertainties to about 5 W/m2 from about 10 W/m2. The new code results decrease clear/full-sky surface downwelling SW by about 4.4 and 2.7 W/m2, respectively, primarily because of an increase of the atmospheric absorption. The surface downwelling LW has also decreased by about 4 W/m2 .These changes have generally improved our surface flux results. Our conclusion is that, if all the input datasets have reasonably high quality, it is possible to reduce the uncertainties of TOA and surface fluxes to about 5 W/m2 and below 15 to 20 W/m2, respectively.

With all these improvements, we are currently producing global radiative fluxes at TOA and surface, and also vertical profiles for all the whole ISCCP time period (July 1983 through and beyond the current year) with a spatial resolution of 2.5° equatorial equal area and a temporal resolution of 3 hours. In addition, many DX (30 km nominal spatial resolution and same temporal resolution as D1) level flux products will be produced to meet the need of various field experiments, e.g., FIRE, ARMIOPs, SHEBA. Along with all the production, further research is being planned.

REFERENCES.

Zhang, Y.-C., W.B. Rossow and A.A. Lacis, 1995, Calculation of surface and top of atmosphere radiative fluxes from physical quantities based on ISCP data sets: 1. Method and sensitivity to input data uncertainties, J. Geophys. R., 100, 1149-1165.

Rossow, W. B. And Y.-C. Zhang, 1995, Calculation of Surface and Top of Atmospheric Radiative Fluxes from Physical Quantities Based on ISCCP Data Sets: 2. Validation and First Results, J. Geophys. R., 100, 1167-1197.