Background

Project Description

Available Data

Validation

Known Problems

Data Access

References

Gallery

Links

Contacts

Specialized Data Sets

Back to Front Page

Validation

The algorithm used to produce current information  has been evaluated within the limits of the quality and scope of available ground truth, and satellite inputs. Version 1.1 of the algorithm as implemented at NASA/ LaRC at 2.5 degree grids with ISCCP C1 data, has been evaluated against ground truth as available from the Global Energy Budget Archive (GEBA) for about 200 stations (Whitloch et al. 1995, Ohmura et al. 1995) for 1986. Emphasis was put on 1986  which was selected as a validation year for both the GEWEX/SRB project and Clouds and the Earth's Radiant Energy System (CERES) data subsystem (Wielicki et al., 1995). In support of these activities, quality controlled ground based observations have been prepared for validation, under the GEWEX/SRB activity at the NASA Langley Research Center (LaRC), Hampton, VA. They are available from the Global Energy Budget Archive (GEBA), housed at the Swiss Federal Institute of Technology, Zurich, Switzerland (Ohmura and Gilgen, 1993), and augmended at LaRC. Version 2.1 of the algorithm was implemented at the University of Maryland with the improved ISCCP D1 data, and two sources of water vapor. One  from TOVS, as appended to the ISCCP D1 satellite observations. The new source of information on water vapor comes from Version 1 of the Goddard Earth Observing System (GEOS) Data Assimilation System (DAS) developed by the Data Assimilation Office (DAO) of the NASA Goddard Space Flight Center and preprocessed at the NASA Langley Research Center under the GEWEX/SRB project . 

Results have been tested against an expended network of about 500 stations for 1986, at monthly time scales, and are presented in Figure 1.  Ability to perform validation at different scales, might help to determine the limits of attainable accuracy using satellite observations. As of now, methodologies were implemented with satellite inputs of different temporal resolution. Darnell et al. (1992) and Staylor (see Whitlock et al., 1995), use daily average radiance to compute the daily mean surface radiation form the ISCCP C1 data. They use ERBE data, to infer surface albedo. Bishop and Rossow (1992) use monthly mean values of ISCCP C1 data, and Li et al. (1993) use monthly mean ERBE data, to infer directly surface net shortwave flux. To derive the results presented at this Web site, three hourly satellite observations were used as input, and the derived fluxes were averaged, to obtain values at lower temporal resolutions. Of consequence to the evaluation of the derived radiative fluxes against ground observations are issues related to the compatibility of the satellite observations and the ground truth. An example is presented in Figure 2, where results of evaluation against ground truth for the entire year of 1986 are presented for both the D1 and DX scales, separately for Europe and Africa. Information on surface radiative fluxes at high spatial resolution is needed to address climate issues on the meso-scale, and to bridge the scale gap with future EOS products (e.g., the CERES product will be produced at one degree resolution). For validation activities, data from the Global Energy Budget Archive (GEBA), housed at the Swiss Federal Institute of Technology, Zurich, Switzerland (Ohmura and Gilgen, 1993), and amended at the NASA Langley Research Center (LaRC), Hampton, VA, are used.

Till now, global scale estimates of surface radiation budgets have been obtained only from low resolution (temporal and/or spatial) satellite observations, available at best, on 2.5 degree grids, such as the ISCCP C1 or C2 (Whitlock et al. 1995) and ISCCP D1 data (Pinker et al. 2000), ERBE (Barkstrom, 1984), or NIMBUS 7 (Kyle et al., 1993). Estimates of downwelling shortwave fluxes have been validated on such scales, frequently, using observations from one ground observing stations only.