A framework for attenuation method selection evaluated with ice-penetrating radar data at South Pole Lake

Benjamin H Hills,Knut Christianson,Nicholas Holschuh

doi:10.1017/aog.2020.32

Abstract

AbstractAll radar power interpretations require a correction for attenuative losses. Moreover, radar attenuation is a proxy for ice-column properties, such as temperature and chemistry. Prior studies use either paired thermodynamic and conductivity models or the radar data themselves to calculate attenuation, but there is no standard method to do so; and, before now, there has been no robust methodological comparison. Here, we develop a framework meant to guide the implementation of empirical attenuation methods based on survey design and regional glaciological conditions. We divide the methods into the three main groups: (1) those that infer attenuation from a single reflector across many traces; (2) those that infer attenuation from multiple reflectors within one trace; and (3) those that infer attenuation by contrasting the measured power from primary and secondary reflections. To assess our framework, we introduce a new ground-based radar survey from South Pole Lake, comparing selected empirical methods to the expected attenuation from a temperature- and chemistry-dependent Arrhenius model. Based on the small surveyed area, lack of a sufficient calibration surface and low reflector relief, the attenuation methods that use multiple reflectors are most suitable at South Pole Lake.

Highlights

The measured power of a reflected radar wave preserves information from both reflector properties and path effects
Even though we observe some discrepancy between the attenuation methods applied here (Fig. 6), after discarding the results which do not meet our uncertainty criteria, the calculated attenuation rates approximately agree with what is expected based on the predicted attenuation from an Arrhenius model (MacGregor and others, 2007) (Fig. 6)
We highlighted the site and ice-penetrating radar survey characteristics that affect attenuation estimation, and used those to develop a framework for method selection

Summary

Introduction

The measured power of a reflected radar wave preserves information from both reflector properties and path effects. For ice-penetrating radar, returning waveforms provide insight into the nature of subsurface interfaces as well as the ice column. Attenuation is generally the strongest of the path effects acting in polar ice that is well below the pressure-melting point. Radio-frequency attenuation is mostly due to the dielectric relaxations that occur due to lattice defects (Fletcher, 1970). The attenuation rate, or the energy absorption as a function of travel path length, depends on the defect concentration and ice temperature (Corr and others, 1993; MacGregor and others, 2007; Stillman and others, 2013). Empiricallyderived attenuation measurements in ice generally have high uncertainty, inhibiting comprehensive, ice-sheet scale, radar power interpretation (Jordan and others, 2016)

Methods

Discussion

Conclusion