Combining Solar Irradiance Measurements and Various Satellite-Derived Products to a Site-Specific Best Estimage

Abstract

Energy yields of concentrating solar thermal power plants depend strongly on the availability of beam irradiance. Direct solar irradiance is highly variable in space and time and thus, requires in-depth analysis of measurements with high quality, well tracked calibrations and approved quality control to achieve reliable results. Strong vari-ability from year to year makes it necessary to combine precision data, available from limited measurement peri-ods with long-term data, which should cover at least 10 years. For most places this may only be reached by ap-plication of historic satellite data. For sites around the Mediterranean region, Africa and the Mideast solar prod-ucts based on data from the Meteosat First Generation so far have been used. Since mid 2006 this satellite gen-eration is out of service and replaced by Meteosat Second Generation. Therefore, recent measurements can not be inter-compared to the satellite-series, which is used to retrieve long-term time-series. In this paper, a satellite-retrieval for beam irradiance from the new generation is introduced and inter-compared with data derived from the first generation satellite for an over-lapping 2 year time period. Applying more than one satellite data set has the further advantage to incorporate several independent data sets and thus strengthens credibility of the final result. A new procedure which combines the value of independent data sets is introduced: A quality-weighted average is produced depending on the individual uncertainties of each data set. By the help of long-term data the most representative climatological values are determined for the final best estimate, which is required for bank-able simulations of energy yields.