Imaging spectroscopy of albedo and radiative forcing by light-absorbing impurities in mountain snow

Recent studies show that deposition of dust and black carbon to snow and ice accelerates snowmelt and perturbs regional climate and hydrologic cycles. Radiative forcing by aerosols is often neglected in climate and hydrological models in part due to scarcity of observations. Here we describe and validate an algorithm suite (Imaging Spectrometer-Snow Albedo and Radiative Forcing (IS-SnARF)) that provides quantitative retrievals of snow grain size, snow albedo, and radiative forcing by light-absorbing impurities in snow and ice (LAISI) from Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data collected on 15 June 2011 in the Senator Beck Basin Study Area (SBBSA), SW Colorado, USA. Radiative forcing by LAISI is retrieved by the integral of the convolution of spectral irradiance with spectral differences between the spectral albedo (scaled from the observed hemispherical-directional reflectance factor (HDRF)) and modeled clean snow spectral albedo. The modeled surface irradiance at time of acquisition at test sites was 1052 W m(-2) compared to 1048 W m(-2) measured with the field spectroradiometer measurements, a relative difference of 0.4%. HDRF retrievals at snow and bare soil sites had mean errors relative to in situ measurements of -0.4 +/- 0.1% reflectance averaged across the spectrum and root-mean-square errors of 1.5 +/- 0.1%. Comparisons of snow albedo and radiative forcing retrievals from AVIRIS with in situ measurements in SBBSA showed errors of 0.001-0.004 and 2.1 +/- 5.1 W m(-2), respectively. A counterintuitive result was that, in the presence of light absorbing impurities, near-surface snow grain size increased with elevation, whereas we generally expect that at lower elevation the grain size would be larger.