Holocene glacial activity in Barilari Bay, west Antarctic Peninsula, tracked by magnetic mineral assemblages: Linking ice, ocean, and atmosphere

Abstract

We investigate the origin and fate of lithogenic sediments using magnetic mineral assemblages in Barilari Bay, west Antarctic Peninsula (AP) from sediment cores recovered during the Larsen Ice Shelf System, Antarctica (LARISSA) NBP10‐01 cruise. To quantify and reconstruct Holocene changes in covarying magnetic mineral assemblages, we adopt an unsupervised mathematical unmixing strategy and apply it to measurements of magnetic susceptibility as a function of increasing temperature. Comparisons of the unmixed end‐members with magnetic observations of northwestern AP bedrock and the spatial distribution of magnetic mineral assemblages within the fjord, allow us to identify source regions, including signatures for “inner bay,” “outer bay,” and “northwestern AP” sources. We find strong evidence that supports the establishment of a late Holocene ice shelf in the fjord coeval with the Little Ice Age. Additionally, we present new evidence for late Holocene sensitivity to conditions akin to positive mean Southern Annual Mode states for western AP glaciers at their advanced Neoglacial positions.

Late Holocene relative sea levels near Palmer Station, northern Antarctic Peninsula, strongly controlled by late Holocene ice-mass changes

“…Proxy records from the Palmer Deep km south of Palmer Station point to almost 1.3°C warming of sea-surface temperatures between 1.8 ka and 1.6 ka with a similar magnitude cooling between 0.5 ka and 0.1 ka with no fewer than 8 additional but smaller amplitude sea-surface temperature changes since 1.8 ka (Shevenell et al, 2011). Several other proxy records of climatic changes have been retrieved from marine sediments within the Palmer Deep (Leventer et al, 1996;Shevenell and Kennett, 2002) and elsewhere across the western Antarctic Peninsula (Heroy et al, 2008;Allen et al, 2010;Yoon et al, 2010;Reilly et al, 2016;Kim et al, 2018) all pointing to climatic and potentially glacial fluctuations through the late Holocene.…”

The Larsen Ice Shelf System, Antarctica (LARISSA): Polar Systems Bound Together, Changing Fast

“…Similar to the Larsen A embayment, reduced glacier-ice extent and seasonally open water during the early to mid-Holocene is observed on the western Antarctic Peninsula in Barilari Bay, followed by late Holocene expansion of sea-ice cover that reached a maximum during the Little Ice Age (Christ et al, 2015). Outer bay glaciers in their advanced late Holocene positions were also sensitive to conditions akin to positive mean Southern Annular Mode (SAM) states (Reilly et al, 2016). Late Holocene cooling is also recorded in sediments from the tip of the Antarctic Peninsula, where the western and eastern Antarctic Peninsula systems meet (Kyrmanidou et al, 2018).…”

Authigenic Ferrimagnetic Iron Sulfide Preservation Due to Nonsteady State Diagenesis: A Perspective From Perseverance Drift, Northwestern Weddell Sea

“…The relationship between H cr and S‐ratio in Zone 1 is different than the relationship in Zones 2 and 3 (Figures 9c and 9d), which suggests that the magnetic minerals driving S‐ratio and HIRM variability are largely different above the dissolution front versus below—which likely indicate that while in Zone 1 HIRM and S‐ratio reflect the preferential dissolution of detrital (titano)magnetite relative to hematite, in Zones 2 and 3 HIRM and S‐ratio largely reflect the addition of authigenic hexagonal 3C pyrrhotite. The trend of these samples in Zone 3 towards where the sedimentary 3C pyrrhotite nodules of Horng (2018) plot is different from the trends that seem more typical of detrital assemblages, as observed above the dissolution front in Zone 1; in more terrigenous Antarctic Peninsula Fjord glacial marine sediments with less extreme diagenetic conditions (Reilly et al., 2016); in lake sediments that are interpreted to be a mixture of magnetite and hematite (Stober & Thompson, 1979); and the regional bedrock samples (Figure 9).…”

“…Select intervals were measured again in ambient air without an argon atmosphere, to compare magnetic mineral alteration at high temperatures. The argon atmosphere thermomagnetic curves were unmixed using the method of Reilly et al (2016), which adapts the method of Heslop and Roberts (2012) and uses the Simplex Identification via Split Augmented Lagrangian (SISAL) method of Bioucas-Dias (2009). The method was modified to unmix the measured curves rather than the derivative of that curve due to the noisier nature of these weakly magnetic sediments and includes both heating and cooling curves.…”

“…(d) End member temperature dependent χ lf curves (red = heating; blue = cooling), with ±1 σ (shading). (e) End member model selection (vertical red line; τ = 0.7848) using error (left) and shape constraints (right) for a range of regularization parameters (τ; seeReilly et al (2016) for discussion).…”