Evaluation of Icebridge Snow Radar Measurements over Sea Ice in the Canadian Arctic

Abstract:

Recent efforts to retrieve snow depth on sea ice using the Operation IceBridge (OIB) snow radar have identified uncertainties related to the vertical heterogeneity of snow, ice deformation, and radar side lobes (e.g. Farrell, et. Al., 2012, Kurtz et. Al., 2013, Kwok and Maksym, 2014). To characterize and evaluate snow depth retrieval uncertainties as related to snow physical properties, an OIB mission was flown near Eureka, Nunavut (79°59’20”N, 85°56’27”W) within the Canadian Arctic archipelago as part of the 2014 OIB Arctic campaign. A series of 12 parallel flight lines covered a narrow swath of first year sea ice approximately 50 km in length. Immediately following the OIB mission, an intensive 10-day field campaign was completed to characterize snow and ice properties within the footprint of the OIB snow radar at multiple scales. Measurements were divided between two observation areas: (1) a primary sampling transect along the length of the flights to characterize horizontal variability in bulk snow properties and (2) a set of intensive grids (250 m x 250) to evaluate variations in snow properties sub-grid to OIB products. As part of each experiment, standard sampling methods were used to collect geo-located snow depth and snow pit measurements (stratigraphy, density, grain size, and salinity). More than 30,000 geo-located snow depth measurements were collected along the primary transect with 94% located within the snow radar footprint. The substantial volume of field measurements coincident with OIB snow radar observations provides an excellent opportunity to evaluate and advance the retrieval of radar-derived snow depth over sea ice. In this study, we present statistical analysis of the observed radar signal and measured snow properties at multiple scales to address previously identified ambiguities in the interpretation of the radar returns.