Optimal Estimation of Snow and Ice Surface Parameters from Imaging Spectroscopy Measurements

Abstract:

Imaging spectroscopy can leverage the understanding of snow and ice melt processes, which are directly related to a decrease in surface reflectance mainly caused by accumulation of liquid water and light-absorbing particles (LAP). We present a new method to retrieve grain size, liquid water fraction as well as LAP mass mixing ratios from imaging spectroscopy measurements. It is based on a simultaneous retrieval of atmospheric and surface parameters using optimal estimation (OE), which incorporates prior knowledge and measurement noise as well as model uncertainties. We extend a comprehensive library of reflectance spectra representing prior knowledge of the surface state by assigning grain size, liquid water fraction and LAP mass mixing ratios as additional parameters to each spectrum. As for LAP, we focus on snow and ice algae but also include black carbon and mineral dust representing inorganic particles. To build the spectral library we use the snow radiative transfer model BioSNICAR-GO. In contrast to previous algorithms, our method introduces a new approach to retrieve snow and ice surface parameters simultaneously to the atmospheric correction procedure and to quantify associated uncertainties. A sensitivity analysis based on simulated EnMAP spectra indicates an overall good performance of the algorithm. All parameters are retrieved with an R² of more than 0.75 and less than 0.8% posterior uncertainty. An additional validation of the extended surface model with field observations of algae mass mixing ratios and surface reflectance from the Greenland Ice Sheet yields an R² of 0.55 for glacier algae and less than 3% residual for the retrieved surface reflectance. Finally, we evaluate retrievals of grain size, liquid water fraction and LAP mass mixing ratios with a case study of an AVIRIS-NG acquisition from the Greenland Ice Sheet. The demonstrated potential is well suited for upcoming orbital imaging spectroscopy missions such as SBG, EnMAP and CHIME.