New Space-Based Perspectives on Blowing Snow Over Arctic Sea Ice

Abstract

Blowing snow over Arctic sea ice is an important but poorly constrained process connecting the cryosphere and atmosphere. Sublimation of blowing snow is a significant removal pathway for snow from the sea ice surface, thereby affecting the surface mass balance of sea ice as well as the surface energy and radiation budgets. While model parameterizations of blowing snow have been developed, the resulting predictions can be highly uncertain due to the scarcity of observations to evaluate them. However, active remote sensing satellite platforms provide a novel and unique opportunity to constrain blowing snow processes on an Arctic-wide scale. The ultimate goal for this dissertation is to better constrain blowing snow occurrence, its properties, and role in the hydrology of snow on Arctic sea ice using satellite observations, thereby providing a framework to evaluate model predictions of sea ice mass balance, polar chemistry, and Arctic climate. This dissertation first evaluates and optimizes an algorithm for detecting blowing snow over Arctic sea ice using observations from the NASA Ice Cloud and land-Elevation Satellite-2 (ICESat-2) satellite. In particular, refinements were made to the algorithm to account for the presence of clouds which could be mis-identified as blowing snow. To conduct this optimization, ICESat-2 orbits coincident with the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) campaign were examined for a six-month period (November 2019 through April 2020). Both ICESat-2 and MOSAiC suggested that blowing snow occurred frequently (17-18%) during the campaign associated with passing storms. This analysis also showed that ICESat-2 inferred blowing snow number and mass concentrations were broadly consistent with in-situ observations made during MOSAiC. ICESat-2 data near MOSAiC suggest that blowing snow sublimation may explain a substantial portion of precipitation mass loss during the campaign, offering important context to the observed snowfall and snow depth. Using the optimized blowing snow detection algorithm, this dissertation then examines ICESat-2 observations across a multi-year period (2018-2023). Retrieved blowing snow occurrence was found to average 20% across the sea ice during the cold season (November through April), with some regions of the Central Arctic reaching as high as 35%. Blowing snow occurrence detected by ICESat-2 shows substantial interannual variability that is related to large-scale climate variability including the Arctic Oscillation (AO). The ICESat-2 observations confirm windspeed plays the strongest role in modulating blowing snow occurrence, height, and optical depth, with all increasing by more than a factor of five across the 4-15 m s⁻¹ range. By combining retrieved ICESat-2 blowing snow properties with reanalysis meteorology this chapter shows that sublimation averages 1.63 ± 0.74 cm snow-water-equivalent (SWE) per cold season, corresponding to a mean precipitation mass loss of 13.6 ± 5.9 %. Predictions from two models of ranging complexity are consistent with these totals (1.66-2.07 cm SWE equaling a 14.1-16.9% precipitation mass loss). Blowing snow sublimation is more than a factor of 3 larger than surface sublimation (0.3-0.5 cm SWE per season), highlighting that it exerts considerable forcing on snowpack evolution during the Arctic cold season. Critically, the findings from this chapter provide the first multi-year satellite-based constraints on blowing snow processes over Arctic sea ice and underscore its importance in the Arctic sea ice snow budget.