Improved ICESat-2 Sea Ice Thickness by Constraining the NASA Eulerian Snow On Sea Ice Model with CloudSat-Derived Snowfall and Operation IceBridge Snow Depth

Abstract:

Estimates of snow depth are a key, but poorly observationally-constrained, input for deriving sea ice thickness from altimetry measurements. Snow-on-sea-ice models such as the NASA Eulerian Snow On Sea Ice Model (NESOSIM) can be used to produce basin-wide snow depth that is required for sea ice thickness estimates. However, these models include parameters and external inputs that remain highly uncertain.

NESOSIM version 1.1 is a 2-layer scheme capturing simple representations of accumulation, wind packing, blowing snow loss to leads, and redistribution due to sea ice convergence and divergence. Relative to version 1.0, NESOSIM 1.1 features a more extended model domain, finer resolution, and an option of scaling reanalysis snowfall input to be consistent with snowfall retrieved from CloudSat satellite radar reflectivity measurements. This scaling helps reconcile some of the inter-product discrepancies in reanalysis snowfall input to NESOSIM.

The new version of NESOSIM, the rescaled snowfall input, and the availability of 2010-2019 field seasons of Operation IceBridge airborne radar snow depths provide an opportunity to re-calibrate NESOSIM for the next releases of sea-ice thickness products derived from NASA ICESat-2 lidar altimetry. We examine the impact of this re-calibration on sea ice thickness estimates. The impact of estimated uncertainties in snowfall and snow depth on derived sea ice thickness estimates is also assessed.