Constraining the geometry and volume of the Barents Sea Ice Sheet

Abstract

The ice load configuration of the Barents Sea Ice Sheet (BSIS) over the last glacial cycle is in dispute. The traditional reconstruction, motivated by the observation that paleo‐shoreline emergence increases towards the center of the Barents Sea, places a single dome in the center of the Barents Sea at the Last Glacial Maximum (LGM) that collapses to island‐centered loads during deglaciation. Observations that suggest that ice flowed from the islands into the Barents even at the LGM motivate another reconstruction that places the ice loads over the islands with minimal marine ice. We analyze an ensemble of ice loads that are consistent with the geophysical observations and show that current relative sea level, GPS and gravity measurements do not and cannot distinguish a central dome from an island‐centered BSIS. What is needed are constraints in the central Barents. Improving the gravity data sufficiently will be difficult. However, obtaining even a single GPS uplift rate measurement in the central Barents would resolve the central dome versus island‐centered BSIS geometry question. Uncertainty in the Barents Sea ice load geometry provides a good illustration of statistical methods that we believe will be useful in other areas of glaciology.

GRACE constraints on Earth rheology of the Barents Sea and Fennoscandia

“…The difference in the ocean signal according to the two models is large in the Barents Sea. This uncertainty has not been considered in previous studies of the GIA signal in the region (e.g., Root et al, 2015a;Simon et al, 2018;Kachuck and Cathles, 2018) and thus the errors bars in those studies were provably underestimated. This result has also implications for GRACE studies of non-oceanic mass changes, such as post seismic deformations, in ocean areas (e.g., Han and Simons, 2008;Wang et al, 2012) which possibly have higher uncertainty than previously thought due to errors in the ocean model.…”

“…The interplay between ice deglaciation chronology and Earth rheology makes it difficult to constrain the ice deglaciation chronology in the Barents Sea (Kachuck and Cathles, 2018). Root et al (2015a) used GRACE data to conclude that the SBKIS had less ice than previously thought (5 − 6.3 m of equivalent sea level versus 8.3 m).…”

GRACE constraints on Earth rheology of the Barents Sea and Fennoscandia

“…The Gravity Recovery and Climate Experiment (GRACE) satellite data have been used to constrain numerical models for GIA in North America (Tamisiea et al, 2007;Paulson et al, 2007;van der Wal et al, 2008;Sasgen et al, 2012) and Fennoscandia (Steffen and Denker, 2008;Simon et al, 2018). With longer time series it is now possible to observe weaker GIA signals such as that of the Svalbard-Barents-Kara Ice Sheet (SBKIS) in GRACE gravity data (Root et al, 2015a;Kachuck and Cathles, 2018;Simon et al, 2018). The use of GRACE data is especially relevant in this region as other geodetic observations normally used for GIA studies are only available from the islands surrounding the Barents Sea, in the periphery of the ice sheet that covered the region during the Last Glacial Maximum (LGM).…”

“…Comparing the GRACE-derived gravity disturbance rates with those predicted for different paleo-ice sheet configurations, Root et al (2015a) conclude that the SBKIS contained less ice than previously thought. Kachuck and Cathles (2018) use GRACE data, along with relative sea level (RSL) curves and GPS uplift measurements, to distinguish between two deglaciation histories: one with an ice sheet with a central dome in the Barents Sea and one with the Barents Sea marginally glaciated and domes in the surrounding Arctic islands. They show that the data are inconclusive in this regard.…”

“…Seismic tomography reveals lower seismic velocities in Barents Sea than below Fennoscandia (Levshin et al, 2007;Schaeffer and Lebedev, 2013) but not for all seismic periods and depths. 3D viscosity profiles have been implemented in GIA models for the regions and have been found to affect sea level and uplift rates (Kaufmann and Wu, 1998). However, the difference in properties between Fennoscandia and Barents Sea has not been studied explicitly.…”