Measurements from all sites with the exception of San Diego are based on "Version 2 NSF Network Data" (Bernhard et al., 2004), available at http://www.biospherical.com/NSF/Version2. Version 2 data have been corrected for the cosine error of the instruments, have a wavelength uncertainty of less than + .04 nm (+ 1a), and were normalized to a uniform bandwidth of 1 nm FWHM. The expanded standard uncertainty of erythemal irradiance (CIE action spectrum by McKinlay and Diffey, 1987) and spectral irradiance at 400 nm varies between 4.2% and 6.8% (coverage factor 2, equivalent to a confidence level of 95.5%, or 2a-level). Measurements are complemented with calculations of the radiative transfer model UVSPEC/libRadtran (Mayer and Kylling, 2005). Measurements during clear skies agree with model calculations to within + 5% on average. More details on Version 2 data and model spectra from South Pole, McMurdo, Palmer, Barrow, and Summit can be found in publications by Bernhard et al. (2004, 2005, 2006b, 2007, 2008).
Version 2 data for San Diego, including model spectra, are not available as of this writing. Measurements from San Diego are based on the original "Version 0" data release (Booth et al., 1994; Bernhard et al., 2006a). Data were scaled up by 5%, which is the typical difference between Version 0 and 2 for erythemal irradiance observed at the other sites. Version 2 corrections are of less importance for San
Diego because of the smaller solar zenith angles (SZAs) occurring at this site. Nevertheless, San Diego data have an additional uncertainty of about 3%.
Was this article helpful?