Solar power plants as their name implies use solar energy as fuel. It is therefore important that accurate long term solar energy datasets be available for estimating production. Depending on the technology either the direct or the total solar radiation is the variable of importance. In renewable energy parlance those two quantities are defined as Direct Normal Irradiance (DNI) and Global Horizontal Irradiance (GHI). DNI estimates are required for Concentrated Solar Power (CSP) technology while GHI estimates are important for Photovoltaics (PV). While ground based measurements provide the most accurate data it is not logistically viable to make multi-year measurements at all locations of prospective interest. The alternative solution is to retrieve GHI and DNI from geostationary satellites because of the availability of continuous long-term coverage. Satellite estimates of surface radiation have inherent uncertainties including the effect of bright surfaces, broken clouds, multi-layer clouds and aerosols. Again all locations are not suitable for solar development because of various issues including land usage, terrain and availability of transmission lines.
The National Renewable Energy Laboratory (NREL) has created multi-year maps of solar resources that are part of the National Solar Radiation Database (NSRDB) (Figure 1). The NSRDB datasets are widely used by industry for power production estimates using various models including NREL's Solar Advisor Model (SAM). NREL additionally creates resource datasets for other countries around the globe. We will present a description of what is involved in the creation of such datasets and what issues specific issues have to be considered for making these datasets useful for renewable energy purpose.
Figure 1: A map of average daily DNI resource for the United States.