Paleoclimate of Antarctica reconstructed from clast weathering rind analysis

https://doi.org/10.1016/j.palaeo.2016.01.026Get rights and content

Highlights

  • Clast rind analysis

  • Mineralogy of clast weathered zones (CWZ)

  • Correlation of CWZ to Paleosol horizons

  • Record of Antarctic temperate/polar ice transition

Abstract

Previous analysis of extractable Fe/Al in paleosols near New Mountain, Antarctica, suggests a pre-Middle Miocene, possibly Late Oligocene–Early Miocene age for paleosol 831 emplaced during the alpine event prior to the growth of the Inland Ice Sheet. Recent analysis of weathering zones in clasts with encrusted Fe/salts from the pebble pavement overlying the 831 paleosol reveals a succession of three weathering zones in a sandstone clast that record a paleoclimate transitioning from warm/wet (temperate) to cold/dry (polar). The first zone corresponds to an association of quartz cemented with berthierine (smectite–serpentine) and illite clay minerals. The second zone transitional from weathered clast to Na-encrusted rind contains smaller amounts of berthierine and illite formed in grains of partially dissolved quartz with minor salt content. Zone three corresponds to an ~ 3 mm (3 × 106 nm) thick mass of porous nitrate and gypsum. The contact between zones 2 and 3 appears to be correlative with the transition from temperate to polar ice which is considered to have occurred prior to or coincident with the Middle Miocene Climatic Optimum.

Introduction

Antarctic paleosol profiles have been studied (Campbell and Claridge, 1987, Retallack and Krull, 1999, Retallack et al., 2007, Bockheim, 2007, Bockheim, 2013, Bockheim and Ackert, 2007, Mahaney, 2015, Mahaney et al., 2001a, Mahaney, 2015; NRCS, http://soils.usda.gov/) in some detail whilst the overlying pebble pavement, complete with Fe/Na encrusted rinds, has been neglected. While the profiles reveal important information on weathering of polar desert sediment, even multiple weathering stages tied to oscillations of outlet glaciers, the pebble pavement is seen to contain in microcosm a longer and more detailed weathering record of the common lithologies — sandstone of the Beacon Supergroup and Ferrar Dolerite. While granite transported from the Antarctic craton and deposited in moraines forming the Middle Miocene Climatic Optimum (MMCO; Warny et al., 2009), it appears that sandstone and dolerite compose the dominant lithologies in older moraines belonging to the earlier alpine event.

Clast rinds have long been used for relative dating of glacial deposits (Birkeland, 1973, Birkeland, 1999, Mahaney, 1978, Mahaney, 1990, Colman and Pierce, 1981), for analysis of clast weathering processes (Sak et al., 2004), and for determination of the bio-influence in clast weathering (Jackson and Keller, 1970). Recently, clast sequences have been probed with high-resolution imaging instrumentation to assess biomineralization (Mahaney et al., 2013a, primary to secondary mineral genesis (Mahaney et al., 2012a), rinds on Mars as possible environmental niches to locate extant or fossil microbes (Mahaney et al., 2012b), and as inventory archives of cosmic impacts/airbursts (Mahaney et al., 2013b, Mahaney and Keiser, 2013). Some rinds have been shown to contain weathering zones that mimic horizons in local paleosols, the resident changes in weathering strength over time registered as an archive in both surface clasts and in the underlying sedimentary substrate (Mahaney et al., 2013b). Still, some rinds are known to become armored thus closing the interface with the atmosphere and biosphere (Mahaney et al., 2012a), with examples placing clasts in weathering lockdown. The prime importance of all these investigations is the wealth of paleoenvironmental data that is archived in weathering rinds, which, when sectioned and analyzed, reveal the slow consumption of primary minerals into myriad organic and inorganic compounds revealing in the process stages of weathering that often correlate with horizons in paleosols (Mahaney et al., 2013b).

As with coatings on sand fractions in soils/paleosols or on bedrock (Mahaney, 2002, Dixon et al., 2002, Gordon and Dorn, 2005), clast rinds likewise reveal a similar residential corpus of natural and anthropogenic activity that can be correlated with one another. Sand coatings down-section in paleosols may correlate with clast rinds in deposit surfaces providing parallel datasets that reinforce each other with regard to interpretations gleaned from parallel weathering trends. As indicated here, clast rind history of site 831 in the New Mountain area, Antarctica, is seen to contain a somewhat more chemically active and perhaps more longer-lived record compared with the underlying paleosol (Site 831, referenced from Mahaney et al., 2001a). Research into the pebble pavement beds of these Antarctic paleosols is designed to answer several seminal questions related to sediment origin and age. First, to what degree did most of the matrix sediment deposited with the coarse clastic load either suffer deflation or respond to the warm/wet temperate climate to find residence in the profile below? Second, does proximity of the pebble pavement to the surface subaerial environment lead to selectively greater presence of water to insure hydrolysis and slow growth of clay minerals compared with the drier underlying paleosol? Is the presence of berthierine with kaolinite layers in the weathered clast and its absence in the paleosol suggestive of a more active weathering environment in the pebble bed? Is the nitratine (NaNO3) ~ 3 mm thick salt bed adhered to the upper clast rind conclusive evidence of a rapid transition to cold/dry polar climate either during the MMCO or before?

Analysis by XRF of an ~ 8 mm × ~ 8 mm section of a clast from the Ant-831 pebble bed would appear to answer many of the above mentioned questions.

Section snippets

Regional geology

The Transantarctic Mountains, including the Taylor Glacier–New Mountain area (Fig. 1A location relative to McMurdo station; Fig. 1B location relative to the Dry Valleys), consist of sedimentary layers overlying a basement of granites and gneisses. The sedimentary layers include the Beacon Supergroup (Stewart, 1934, Shaw, 1962) consisting mainly of arkosic sandstone locally intruded by doleritic dikes and sills (Ferrar dolerite). This mix of sandstone, dolerite, granite and gneiss comprise the

Field sedimentary analysis

The 831 section detailed here was selected and excavated in a moraine ridge deposited by alpine ice above the embayment below where sites 828 and 829 are located (see Mahaney et al., 2014b for location of sites). The site, excavated by hand, was sectioned to expose in situ sediment. The paleosol horizon descriptions reported here (Fig. 2) are genetic following horizon designations used previously by Mahaney et al. (2009) adopted from the NSSC (1995) and the Canadian Soil Classification (CSSC,

Pebble pavement

The 831-paleosol pebble pavement is ~ 3 cm in thickness, similar in most respects to pebble pavement thicknesses in younger profiles dating from the MMCO (Mahaney et al., 2001a). Beyond horizon thickness, the two sets of younger pavements (827, 828 and 829) vs the older 831 bed, differ with clast color expression, thickness of rind development, and where present, thickness of Na-salt encrustation. Surface pebble colors (Oyama and Takehara, 1970) in the younger clast set range from 10YR 4 and 5

Discussion

Weathering rind analysis in the Antarctic has been a neglected source of paleoenvironmental information over the time that research has been carried out in earnest, that is, the last 5 decades or so. Because weathered regolith almost always classifies as of paleosol – not soil – vintage, the vast majority of pedons are classed as forming under more than one climate, either before during or following the MMCO of ~ 15 Ma (Graham et al., 2002), and most but not all, contain pebble pavements. Some

Conclusions

The Beacon Supergroup sand from the Ant-831 pebble pavement provides an ~ 8 × ~ 8 mm image of clast and Na encrustation, with XRF analysis recording chemical changes from deep within the clast to its outer weathered edge transitioning across a narrow contact upward into the salt body. The described weathering zones 1 and 2 record weathering processes, presumably largely oxidation, hydrolysis, and hydration operating since at least the Early Miocene, and quite possibly reaching into the Late

Acknowledgments

This research was funded by Quaternary Surveys, Toronto and the New Zealand Antarctic Programme. We thank Event K-105, Antarctic NZ Programme for support. We gratefully acknowledge reviews by Dustin Sweet (Texas Tech University, Lubbock, Texas) and one anonymous reviewer.

References (56)

  • G.J. Retallack et al.

    Return to Coalsack Bluff and the Permian–Triassic boundary in Antarctica

    Glob. Planet. Chang.

    (2007)
  • P.B. Sak et al.

    Rates of weathering rind formation on Costa Rican basalt

    Geochim. Cosmochim. Acta

    (2004)
  • D.D. Wynn-Williams et al.

    Proximal analysis of regolith habitats and protective biomolecules in site by Raman spectroscopy: overview of terrestrial Antarctic habitats and Mars analogy

    Icarus

    (2000)
  • J.B. Anderson et al.

    Progressive Cenozoic cooling and the demise of Antarctica's Last Refugium

    PNAS

    (2011)
  • R.L. Armstrong

    K–Ar dating: Late Cenozoic McMurdo volcanics and Dry Valley glacial history

    N. Z. J. Geol. Geophys.

    (1978)
  • P.J. Barrett

    History of the Ross Sea region during the deposition of the Beacon Supergroup 400–180 million years ago

    J. R. Soc. N. Z.

    (1981)
  • P.W. Birkeland

    Use of relative age dating methods in a stratigraphic study of rock glacier deposits, Mt. Sopris, Colorado

    Arct. Alp. Res.

    (1973)
  • P.W. Birkeland

    Soils and Geomorphology

    (1999)
  • S. Bo et al.

    The Gamburtsev mountains and the origin and early evolution of the Antarctic Ice Sheet

    Nature

    (2009)
  • J.G. Bockheim

    Soil processes and development rates in the Quartermain Mountains, Upper Taylor Glacier Region, Antarctica

    Geogr. Ann.

    (2007)
  • J.G. Bockheim

    Paleosols in the transantarctic mountains: indicators of environmental change

    Solid Earth Discuss.

    (2013)
  • M. Bradshaw

    The Taylor Group (Beacon Supergroup). The Devonian sediments of Antarctica

    J. Geol. Soc.

    (2013)
  • I.B. Campbell et al.

    Antarctica: Soils, Weathering Processes and Environment

    (1987)
  • Canada Soil Survey Committee (CSSC)

    The Canadian System of Soil Classification

    (1998)
  • S.M. Colman et al.

    Weathering rinds on andesitic and basaltic stones as a Quaternary age indicator, western United States

    US Geol. Surv. Prof. Pap.

    (1981)
  • K.M. Cuffey et al.

    Entrainment at cold glacier beds

    Geology

    (2000)
  • J.C. Dixon et al.

    Weathering rinds and rock coatings from an Arctic alpine environment, Northern Scandinavia

    Geol. Soc. Am. Bull.

    (2002)
  • H.T. Ferrar

    Report on the field geology of the region explored during the Discovery Antarctic Expedition1901–1904. Nat Antarct. Exped. 1901–1904

    Nat. Hist.

    (1907)
  • Cited by (11)

    • Clast rind-paleosol record of the Antarctic early Alpine glaciation

      2021, Polar Science
      Citation Excerpt :

      With clast positions within the pavement and juxtaposition to the upper paleosol beneath, it is possible the bulk of available snowmelt was probably contained within these two mass areas. The pebble pavement previously described by Mahaney et al. (2001) and Mahaney and Schwartz (2016) is important as an integral part of the sections at Aztec and New Mountain (Fig. 2B). Within these stratigraphic sections, the paleosol body yields important physical, mineral, and chemical data that assist in paleoenvironmental reconstruction, but on the physical side the pebble pavement provides even a higher level of information.

    • Evidence for cosmic airburst in the Western Alps archived in Late Glacial paleosols

      2017, Quaternary International
      Citation Excerpt :

      Limited chemical weathering in Gale Crater rivals similar paleosols acting as proxies in Late Oligocene profiles belonging to the early alpine sequence of moraines near New Mountain Antarctica, albeit without A horizons, profiles revealing at best Cox/C/Cu profiles (Mahaney, 2015; Mahaney et al., 2001). While weathering rinds on Earth are reported to carry paleoenvironmental records over Late Neogene time (Mahaney et al., 2012), and even to the Late Paleogene (Mahaney and Schwartz, 2016), on Mars the record extends into the Noachian (Mahaney et al., 2012), and likely Hadean-age-equivalent information that has been all but been destroyed on Earth through subduction and plate-tectonic erosion, except for resilient zircons (e.g., Wilde et al., 2001). Because rinds on Mars are so far limited to deep chemical weathering of meteorites (Mahaney et al., 2012), followed by burial and exhumation, followed, in turn, by slower weathering during the Amazonian, instruments aboard the Opportunity Rover were not sensitive enough to measure successive mineral alteration stages related to the two hypothesized time of weathering.

    View all citing articles on Scopus
    View full text