Ice shelf basal melt rates from a high-resolution digital elevation model (DEM) record for Pine Island Glacier, Antarctica

David E Shean,Etienne Berthier,Pierre Dutrieux,Benjamin E Smith,Ian R Joughin

doi:10.5194/tc-13-2633-2019

Abstract

Abstract. Ocean-induced basal melting is responsible for much of the Amundsen Sea Embayment ice loss in recent decades, but the total magnitude and spatiotemporal evolution of this melt is poorly constrained. To address this problem, we generated a record of high-resolution digital elevation models (DEMs) for Pine Island Glacier (PIG) using commercial sub-meter satellite stereo imagery and integrated additional 2002–2015 DEM and altimetry data. We implemented a Lagrangian elevation change (Dh∕Dt) framework to estimate ice shelf basal melt rates at 32–256 m resolution. We describe this methodology and consider basal melt rates and elevation change over the PIG ice shelf and lower catchment from 2008 to 2015. We document the evolution of Eulerian elevation change (dh∕dt) and upstream propagation of thinning signals following the end of rapid grounding line retreat around 2010. Mean full-shelf basal melt rates for the 2008–2015 period were ∼82–93 Gt yr−1, with ∼200–250 m yr−1 basal melt rates within large channels near the grounding line, ∼10–30 m yr−1 over the main shelf, and ∼0–10 m yr−1 over the North shelf and South shelf, with the notable exception of a small area with rates of ∼50–100 m yr−1 near the grounding line of a fast-flowing tributary on the South shelf. The observed basal melt rates show excellent agreement with, and provide context for, in situ basal melt-rate observations. We also document the relative melt rates for kilometer-scale basal channels and keels at different locations on the ice shelf and consider implications for ocean circulation and heat content. These methods and results offer new indirect observations of ice–ocean interaction and constraints on the processes driving sub-shelf melting beneath vulnerable ice shelves in West Antarctica.

Highlights

The Amundsen Sea Embayment (ASE) of the West Antarctic Ice Sheet (WAIS, Fig. 1) has experienced significant acceleration, thinning, and grounding line retreat since at least the 1970s (Joughin et al, 2003; Konrad et al, 2018; Mouginot et al, 2014; Rignot et al, 2014; Rignot, 1998)
We evaluated five different bed datasets for Pine Island Glacier (PIG) (Fig. S1), including Bedmap2 (Fretwell et al, 2013), an aerogravity inversion constrained by Autosub bathymetric data (De Rydt et al, 2014; Dutrieux et al, 2014b), an aerogravity and Autosub inversion constrained by active-source seismic surveys (Muto et al, 2016), a mass-conserving bed embedded in Bedmap2 (Morlighem et al, 2011), and the CReSIS L3 gridded Multichannel Coherent Radar Depth Sounder (MCoRDS) ice thickness product from 2009–2010 airborne radio echo sounding
We described basal melt-rate calculations for a single DEM acquired at time ti (DEMi)–DEM acquired at time tj (DEMj) combination with sufficient overlap and a ti − tj time interval that falls within the chosen Dt range (∼ 2 years), which represents only one of many potential valid DEMi–DEMj combinations that can be formed from the full set of digital elevation models (DEMs) over the PIG ice shelf

Summary

Introduction

The Amundsen Sea Embayment (ASE) of the West Antarctic Ice Sheet (WAIS, Fig. 1) has experienced significant acceleration, thinning, and grounding line retreat since at least the 1970s (Joughin et al, 2003; Konrad et al, 2018; Mouginot et al, 2014; Rignot et al, 2014; Rignot, 1998) During this period, regional mass loss increased to present-day estimates of ∼ 100–120 Gt yr−1 (Medley et al, 2014; Sutterley et al, 2014; Velicogna et al, 2014). Shean et al.: Ice shelf basal melt rates from a high-resolution DEM record bility, which could lead to ∼ 3.3 m of global sea level rise (Bamber et al, 2009)

Results

Discussion

Conclusion