Evaluating CM-SAF solar radiation CLARA-A1 and CLARA-A2 datasets in Scandinavia. (August 2018)

• Record Type: Journal Article
• Authors: Babar, Bilal; Graversen, Rune; Boström, Tobias
• Is Part Of: Solar energy. Volume 170(2018)
• Page Start: 76
• Page End: 85
• Publication Date: 2018-08
• Subjects: CLARA A1 and A2 -- Scandinavia -- ECMWF -- Arctic -- Solar radiation estimation -- Polar orbiting satellites; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2018.05.009
• Languages: English
• ISSNs: 0038-092X

Abstract

Highlights: CM SAF databases of shortwave downward radiation are analyzed for high latitude locations. The latest dataset is compared with the previous version. Snow and cloud differentiation is difficult, complex topography makes this task even more difficult. Missing values exist in the new dataset due to of lack of differentiation between clouds and snow covers. The newer dataset has more values included but with large errors. CLARA-A2 comes closer to measured yearly energy estimates because it has more data points than CLARA-A1. Abstract: Estimating/retrieving solar radiation through satellite-based remote sensing provides larger spatial coverage compared to other methods. Accurate estimates of incoming solar radiation is important when planning new solar energy installations. In addition, these estimates are also used in climate studies. Geostationary satellites are ideal for estimating solar radiation but cannot be used for high latitudes because of an unfavourable viewing angle; however, polar-orbiting satellites provide an alternative. CL oud, A lbedo RA diation edition 2 (CLARA-A2) is the latest retrieval product of cloud properties, surface albedo and surface solar radiation by Satellite Application Facility on Climate Monitoring (CM-SAF) based on Advance Very High Resolution Radiometer (AVHRR) observations from polar orbiting satellites. This data set covers the whole earth and provides daily and monthly averages. In this study, we have evaluated the CLARA-A2 data set and the previous version CLARA-A1 to in-situ high-quality observations from specific locations in Scandinavia, with a focus on solar radiation at high latitudes. The results show that both datasets perform within the target accuracies of CM-SAF, although the new data points, which were previously not available in CLARA-A1 due to snow-cover and cloud differentiation, have high deviations. Nevertheless, yearly average energy estimates are more accurate in CLARA-A2 because of these new points. For Swedish locations, mean absolute deviation (MAD) of 8.1 W m −2 and 8.7 W m −2 for CLARA-A1 and A2 respectively were calculated for updated values. Similarly, for Norwegian locations MAD of 8 W m −2 and 8.9 W m −2 were calculated for CLARA-A1 and A2. Overall, for all locations MAD lies at 8.1 W m −2 and 8.8 W m −2 for CLARA-A1 and A2, respectively. CLARA A2 has more temporal data points than CLARA A1, however, the MAD of the new data points that were not available in CLARA-A1 are 15.2 W m −2 and 17.7 W m −2 for Swedish and Norwegian sites, respectively.