The role of an Arctic ice shelf in the climate of the MIS 6 glacial maximum (140 ka)

https://doi.org/10.1016/j.quascirev.2010.06.023Get rights and content

Abstract

During the last decade, Arctic icebreaker and nuclear submarine expeditions have revealed large-scale Pleistocene glacial erosion on the Lomonosov Ridge, Chukchi Borderland and along the Northern Alaskan margin indicating that the glacial Arctic Ocean hosted large Antarctic-style ice shelves. Dating of sediment cores indicates that the most extensive and deepest ice grounding occurred during Marine Isotope Stage (MIS) 6. The precise extents of Pleistocene ice shelves in the Arctic Ocean are unknown but seem comparable to present existing Antarctic ice shelves. How would an Antarctic-style ice shelf in the MIS 6 Arctic Ocean influence the Northern Hemisphere climate? Could it have impacted on the surface mass balance (SMB) of the MIS 6 Eurasian ice sheet and contributed to its large southward extent? We use an Atmospheric General Circulation Model (AGCM) to investigate the climatic impacts of both a limited MIS 6 ice shelf covering portions of the Canada Basin and a fully ice shelf covered Arctic Ocean. The AGCM results show that both ice shelves cause a temperature cooling of about 3 °C over the Arctic Ocean mainly due to the combined effect of ice elevation and isolation from the underlying ocean heat fluxes stopping the snow cover from melting during summer. The calculated SMB of the ice shelves are positive. The ice front horizontal velocity of the Canada Basin ice shelf is estimated to ≈ 1 km yr−1 which is comparable to the recent measurements of the Ross ice shelf, Antarctica. The existence of a large continuous ice shelf covering the entire Arctic Ocean would imply a mean annual velocity of icebergs of ≈12 km yr−1 through the Fram Strait. Our modeling results show that both ice shelf configurations could be viable under the MIS 6 climatic conditions. However, the cooling caused by these ice shelves only affects the Arctic margins of the continental ice sheets and is not strong enough to significantly influence the surface mass balance of the entire MIS 6 Eurasian ice sheet.

Introduction

The Antarctic continent is covered by a complex system of glacier ice masses grounded both above and below sea level. This continental ice sheet system flows into the ocean at several places to form large floating ice shelves. The largest of these are the Ross and Ronne–Filchner Ice Shelves that together cover more than 900,000 km2 and bound the marine-based West Antarctic Ice Sheet (British Antarctic Survey, 2005). The Arctic Ocean, on the other hand, hosts no similar ice shelves today. Only relatively small and thin ice shelves along the Northern Ellesmere Island exist today (Jeffries, 1992, Jeffries, 2002). However, during the last decade of Arctic icebreaker and nuclear submarine expeditions, large-scale glacial erosion and glaciogenic bedforms have been mapped on the central Arctic Ocean seafloor in water depths down to approximately 1000 m below present sea level (e.g. Jakobsson, 1999, Polyak et al., 2001). Some of the mapped glaciogenic features on the Lomonosov Ridge, Chukchi Borderland and along the Northern Alaskan margin indicate that the glacial Arctic Ocean hosted large Antarctic-style ice shelves (Polyak et al., 2001, Jakobsson et al., 2005, Jakobsson et al., 2008b, Engels et al., 2008). Several decades before this geophysical evidence emerged, Mercer (1970) proposed that there may have existed Antarctic-style ice shelves in the Arctic Ocean during the Pleistocene glacial periods. His idea was further developed by Hughes et al. (1977) who suggested a 1-km thick ice shelf covering the entire Arctic Ocean and forming a critical part of a huge ice sheet, including the Laurentide and Eurasian ice sheets, which behaved as one dynamic system during the Last Glacial Maximum (LGM, ≈21 ka).

The precise extent of ice shelves in the Arctic Ocean is unknown, although the geophysical data collected from icebreakers and submarines suggest a more limited extent than that proposed by Hughes et al. (1977) for the LGM. Furthermore, dating of sediment cores collected from glacially eroded areas of the Lomonosov Ridge indicate that the most extensive and deepest ice grounding occurred during Marine Isotope Stage (MIS) 6 and not during LGM (Jakobsson et al., 2001). This implies that the largest ice shelves in the Arctic Ocean coexisted with the largest extent of the Eurasian continental ice sheet, referred to as the Late Saalian with its peak at 140 ka (Svendsen et al., 2004). If an Antarctic-style ice shelf was present in the Arctic Ocean during MIS 6, the question whether it influenced the atmospheric circulation and regional climate arises. Furthermore, could it have impacted on the Surface Mass Balance (SMB) of the Late Saalian Eurasian ice sheet and, thus, play a role in explaining why it grew so large? In order to answer this question we use an Atmospheric General Circulation Model (AGCM) to investigate the climatic impacts of both a limited MIS 6 ice shelf covering portions of the Canada Basin as the recent geophysical mapping indicate (Jakobsson et al., in press) and a fully ice shelf covered Arctic Ocean as proposed by Hughes et al. (1977).

Section snippets

Methods

We have adopted the ice shelf extent in the Canada Basin for our simulation experiments from the ongoing work by Jakobsson et al. (in this issue) involving the compilation of all available geophysical and geological data that constrain the spatial extents of the Pleistocene Arctic Ocean glaciations. However, the precise MIS 6 ice shelf extent in the Canada Basin is not critical for our AGCM experiments as we are generally addressing the potential influence on the atmospheric circulation from an

The Canada Basin ice shelf

The ice shelf occupying the area north of the Canadian Arctic Archipelago and along the Northern Alaskan margin causes a mean annual negative air temperature anomaly of about 3 °C directly over its surface (Fig. 2d). This anomaly is mainly confined to the lower part of the troposphere. The cooling is less than 1 °C above around 2000 m (Fig. 2c). Compared to the recent work by Krinner et al. (in this issue) in which the temperature anomaly induced by sea ice thickness variations is constrained by

Discussion

We addressed the potential climatic impacts of a floating ice shelf in the Canada Basin of the Arctic Ocean because recent marine geophysical and geological data combined suggest that such an ice shelf may have existed some time during MIS 6 (Jakobsson et al., in this issue). Some decades before chirp sonar profiles first revealed extensive ice grounding as deep as 1000 m below present sea level at about 87° N on the Lomonosov Ridge (Jakobsson, 1999), the hypothesis of a 1-km thick ice shelf

Conclusions

This paper is part of a suite of studies on the climatic feedbacks from global and regional factors with the potential of influencing the SMB of the Eurasian ice sheet during the later part of MIS 6, i.e. the late Saalian Ice Sheet at 140 ka (Colleoni et al., 2009a, Colleoni et al., 2009b, Colleoni et al., in press). The main scientific question in focus is why the Late Saalian Ice Sheet grew so much larger than the Weichselian ice sheets. In this present work we have addressed how an ice shelf

Acknowledgments

The authors acknowledge support by the Agence Nationale de la Recherche (project IDEGLACE), the Région Rhône Alpes (programme Explora’Doc), the Ministère des Affaires Étrangères Français, The Bert Bolin Centre for Climate Research (Stockholm University) and the Alexander von Humboldt Foundation for their support The climate simulations were carried out at IDRIS/CNRS and on the Mirage scientific computing plateform in Grenoble (France).

References (42)

  • G. Bigg et al.

    The origin and flux of icebergs released into the Last Glacial Maximum Northern Hemisphere oceans: the impact of ice-sheet topography

    J. Quaternary Sci.

    (2001)
  • British Antarctic Survey

    Antarctic Factsheet Geographical Statistics, British Antarctic Survey

    (2005)
  • CLIMAP

    Seasonal Reconstruction at the Earth’s Surface at the Last Glacial Maximum

    (1981)
  • F. Colleoni et al.

    Sensitivity of the Late Saalian (140 kyrs BP) and LGM (21 kyrs BP) Eurasian ice sheet surface mass balance to vegetation feedbacks

    Geophys. Res. Lett.

    (2009)
  • Colleoni, F., Liakka, J., Krinner, G., Jakobsson, M., Masina, S., Peyaud, V. The Late Saalian surface ocean (140 kyrs...
  • J. Engels et al.

    Sea floor evidence for ice shelf flow across the Alaska–Beaufort margin of the Arctic Ocean

    Earth Surf. Proc. Land

    (2008)
  • H. Fricker et al.

    Distribution of marine ice beneath the Amery ice shelf

    Geophys. Res. Lett.

    (2001)
  • F. Hourdin et al.

    The LMDZ4 general circulation model: climate performance and sensitivity to parametrized physics with emphasis on tropical convection

    Clim. Dynam.

    (2006)
  • T. Hughes et al.

    Was there a late-Wurm Arctic ice sheet?

    Nature

    (1977)
  • S. Jacobs et al.

    Melting of the ice shelves and the mass balance of Antarctica

    J. Glaciol.

    (1992)
  • M. Jakobsson et al.

    Pleistocene stratigraphy and paleoenvironmental variation from Lomonosov Ridge sediments, central Arctic Ocean

    Glob. Planet. Change

    (2001)
  • Cited by (5)

    • An East Siberian ice shelf during the Late Pleistocene glaciations: Numerical reconstructions

      2016, Quaternary Science Reviews
      Citation Excerpt :

      The hypothesis of whether or not MIS 6 Arctic Ocean ice shelves were restricted to the shallow continental shelves or covered the entire ocean is still actively debated. To test if MIS 6 climate conditions are favourable to the maintenance of a large ice shelf, Colleoni et al. (2010a) forced and atmospheric model by prescribing Jakobsson et al. (2010) Canadian ice shelf reconstruction and alternatively a fully ice-covered Arctic Ocean as suggested by Hughes et al. (1977). The simulated surface mass balance was positive, around 0.1–0.2 m/yr, for both ice shelf configurations, implying that a large ice shelf can develop and be maintained over the Arctic Ocean if the basal melting due to ocean heat fluxes and ice flux from the grounded ice sheets are also favourable to its maintenance.

    • Late Quaternary evolution of sediment provenances in the Central Arctic Ocean: Mineral assemblage, trace element composition and Nd and Pb isotope fingerprints of detrital fraction from the Northern Mendeleev Ridge

      2014, Quaternary Science Reviews
      Citation Excerpt :

      In the Eurasian margin, the extent of the Eurasian Ice Sheet is even much less constrained (see discussion in Dowdeswell et al., 2010). Based on recent studies (Colleoni et al., 2010; Stein et al., 2010a; Jakobsson et al., 2010b; Gebhardt et al., 2011) it is accepted that the Eurasian Ice Sheet extended onto Severnaya-Semlya islands and further east in the Laptev Sea. However the presence of an ice sheet in the Indigirka and its region, reaching the Arctic Ocean in the Eastern part of the Laptev sea is much more uncertain (Stein et al., 2010, Fig. 6.41).

    • New insights on Arctic Quaternary climate variability from palaeo-records and numerical modelling

      2010, Quaternary Science Reviews
      Citation Excerpt :

      While several new results indicate that MIS 6 comprised the most extensive Arctic Ocean Quaternary marine ice complex, Strand and Immonen (2010) suggests from studies of quartz sand grain microtextures, in a core from the central Lomonosov, that the Barents-Kara Ice Sheet was thicker and likely more extensive during MIS 4 than during LGM (Fig. 1). Colleoni et al. (2010) use Atmospheric General Circulation Modelling (AGCM) to explore how extensive MIS 6 ice shelves in the Arctic Ocean could have influenced climate and the mass balance of the large contemporary continental ice sheets, in particular the huge Late Saalian Eurasian Ice Sheet. Both the impact from a limited ice shelf in the Amerasian Basin (Jakobsson et al., 2010) and a continuous 1000 m thick ice shelf covering the entire Arctic Ocean (e.g. Hughes et al., 1977) is investigated.

    View full text