The use of airborne radar reflectometry to characterize near-surface snow/firn stratigraphy on Devon Ice Cap, Canadian Arctic: A path to identifying refrozen melt layers

Abstract:

Under present warming conditions, summer surface melt has been observed to intensify and shift towards higher elevations in the accumulation zones of Canadian Arctic ice caps. Consequently, more meltwater percolates into the near surface snow and firn, and refreezes as ice layers. This process can lead to a significant increase in firn densification rates. Knowledge of spatiotemporal variations of the near-surface firn density, especially the distribution of ice layer formation is of great importance when assessing mass change estimates from repeat altimetry measurements.

Here, we present an approach for characterizing the near-surface firn stratigraphy and determining the spatial distribution of refrozen melt layers on Devon Ice Cap, using the surface echo from airborne radio-echo sounding (RES) measurements. The RES surface echo is affected by the upper few meters of snow/firn/ice and thus contains information about the near-surface properties. More specifically, the radar surface return is a combination of a coherent (Pc) and a scattering signal component (Pn). Pc is related to the dielectric constant of the probed surface, whereas Pn is related to the near surface roughness. Hence, different near-surface snow/firn properties can be investigated by analyzing the signal components Pc and Pn and their spatial variability. The Radar Statistical Reconnaissance (RSR) methodology [1] allows the extraction of Pc and Pn from the surface radar return, which then can be used to compute near-surface roughness and firn density estimates. We apply the RSR method to RES data collected on Devon Ice Cap and determine Pc and Pn values. We then compare the results to ground based RES measurements and shallow firn cores (~11 m deep) collected along the airborne RES flight lines. This comparison shows that variations in the scattering coefficient Pn correlate to changes in the pattern of near-surface firn stratigraphy revealed by the ground based RES data and firn cores. Based on these results, we establish an interpretation scheme for RSR results over ice caps and glaciers with complex melting and refreezing firn stratigraphy, which offers a new and fast way of studying the near-surface firn properties, especially to identify refrozen ice layers.

[1] Grima, C. et al. 2013, Planetary and Space Science.