Characterization of a sulfur-rich Arctic spring site and field analog to Europa using hyperspectral data

A unique glacial spring system exists at Borup Fiord pass, in the Canadian High Arctic, emerging from a glacial surface and depositing elemental sulfur, gypsum and calcite across a portion of the glacier. The presence of sulfur springs associated with glacial ice is extremely rare in a terrestrial context, and the resulting deposits may provide a field analog to non-ice materials on the surface of Europa. Spectral characterization of the supraglacial deposits in the visible-near infrared (VIS-NIR) range, 0.4–2.5 µm, was carried out using reflectance spectra collected in situ using a field spectrometer during the 2006 field season. These spectra, while dominated by melting snow, ice, and sulfur, show that some absorption features of the sulfates are shifted in wavelength with respect to library spectra due to the effects of mixing or temperature. Absorption features of calcite are largely absent, potentially due to mineral partitioning effects within the deposits. Investigations into changes in mineralogy within the deposits over the course of the active season using data collected by the Hyperion hyperspectral visible/infrared spectrometer aboard the Earth Observing 1 spacecraft (EO-1) were limited by low signal-to-noise (SNR) ratios in the data, although they indicate that sulfur is remaining stable. This is confirmed by seasonal data on the extent of the deposits, obtained using a classification algorithm running onboard the satellite, which continued to detect the presence of sulfur until snow obscured the site. Ground truth for the observations is provided by mineralogical analyses obtained by X-ray diffraction (XRD) measurements and laboratory reflectance spectra from 0.2–25 µm obtained of samples returned from the site in 2006. We show that while sulfur, the main constituent of the deposits, is well represented in Hyperion data, minor constituents such as calcite and gypsum can be partially or entirely masked in the data. In spite of these effects, autonomous detection methods can be utilized to monitor the generation and extent of the deposits, whose spectral properties show similarities with those of Europaʹs non-ice materials.