Principal component analysis of Arctic solar irradiance spectra

Abstract

During the First International Satellite Cloud Climatology Regional Experiment Phase III (FIRE III) Arctic Cloud Experiment (ACE) and coincident Surface Heat Budget of the Arctic Ocean (SHEBA) campaign, detailed moderate‐resolution solar spectral measurements were made to study the radiative energy budget of the coupled Arctic Ocean–atmosphere system. The National Aeronautics and Space Administration (NASA) Ames Solar Spectral Flux Radiometers (SSFR) were used to measure moderate‐resolution solar spectral irradiance in the spectral range 380–2200 nm on board the NASA ER‐2 (flying at 21 km) and on the surface at the SHEBA ice camp. Principal Component Analysis (PCA) of SSFR spectra was used to quantify the variability in the data, to determine its causes, to assess our ability to separate cloud and surface effects, and to compare PCA in the Arctic with similar analysis of spectra obtained at a midlatitude continental site. For both the upwelling and downwelling spectra, nearly all of the variance was contained in the first six principal components (PC). The variability in the Arctic downwelling solar irradiance spectra was explained by a few fundamental components including cloud absorption, surface reflectance, molecular scattering, water vapor, and ozone. PCA of the SSFR upwelling spectra separated the influence of the ice surface from that of clouds.