Improving the Accuracy Satellite Based Solar Resource for use in NREL's National Solar Radiation Database

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Monday, 3 February 2014: 4:00 PM
Room C114 (The Georgia World Congress Center )
Manajit Sengupta, National Renewable Energy Laboratory, Golden, CO; and P. Gotseff, A. Habte, C. Molling, and A. K. Heidinger

Models to compute Global Horizontal Irradiance (GHI) and Direct Normal Irradiance (DNI) have been in development over the last 3 decades. These models can be classified as empirical or physical based on the approach. Empirical model relate ground based observations with satellite measurements and use these relations to compute surface radiation. Physical models consider the radiation received at the satellite and create retrievals to estimate surface radiation. While empirical methods have been traditionally used for computing surface radiation for the solar energy industry the advent of faster computing has made operational physical models viable. The The Global Solar Insolation Project (GSIP) is an operational physical model from NOAA that computes GHI using the visible and infrared channel measurements from the GOES satellites. GSIP uses a two-stage scheme that first retrieves cloud properties and uses those properties in the Satellite Algorithm for Surface Radiation Budget (SASRAB) model to calculate surface radiation at approximately 12 km. NREL, University of Wisconsin and NOAA have recently collaborated to adapt GSIP to create a 4 km GHI and DNI product every 30 minutes. The product initially generated the cloud properties from the AVHRR Pathfinder Atmospheres-Extended (PATMOS-x) algorithms while the global horizontal radiation (GHI) and diffuse horizontal radiation (DHI) were calculated using SASRAB. Comparison with ground sites resulted in the finding that the satellite based surface radiation suffered from a significant low-bias. As SASRAB created this low bias we tested three new radiative transfer algorithms that required aerosol optical depth and precipitable water vapor as inputs. These experiments showed significant improvement in the clear sky radiation. An example of the improvement in the clear sky radiation is shown in Figure 1 for GHI for the Desert Rock, NV NOAA SURFRAD site. These improvements and NREL's plans to create a new National Solar Radiation Database (NSRDB) will be presented.