Surface exposure geochronology using cosmogenic nuclides : applications in Antarctic glacial geology

Abstract

Cosmogenic 3He, 26A1, and 10Be have been measured in a variety of Antarctic glacial deposits in the McMurdo Sound-Dry Valleys region. The goals of this project were to provide age constraints for Antarctic glacial events, to investigate production mechanisms of 3He, lOBe, and 26 A1 in terrestrial rocks, to constrain the importance of loss of 3He from quartz due to diffusion, and to refine methods of exposure-age dating. Moraines deposited in Arena Valley by the Taylor Glacier, an outlet glacier of the East Antarctic Ice Sheet, have exposure ages from ~120 kyr to 2 myr. 10Be and 3He ages of 122 ± 29 and 134 ±54 kyr, respectively, for the Taylor II moraine are consistent with deposition during isotope stage Se (~ 120 kyr) and with aerial expansions of the East Antarctic Ice Sheet during interglacial periods. Mean 10Be exposure ages for older moraines in the valley are 362 ± 26 kyr (Taylor III), 1.1 ± 0.1 myr (Taylor IVa) and 1.9 ± 0.1 myr (Taylor IVb). Because these older moraines were deposited at most ~200m above the Taylor II limit their ages suggest that major ice sheet advances during the last 2 myr have been broadly similar in magnitude to changes during the last glacial-interglacial cycle. 10Be ages for stratigraphically older drift deposited by the Taylor Glacier allow extension of this conclusion to ~ 3 myr. lOBe measurements in high altitude, pre-Pleistocene glacial deposits in the Dry Valleys preclude rapid uplift of the Transantarctic Mountains (400-1000 m/myr) suggested by controversial biostratigraphic studies of Sirius Group tills. Comparison of measured lOBe concentrations in Sirius Group deposits with those predicted with a model of the effects of uplift on 10Be production suggests minimal uplift of the Transantarctic Mountains over the last 3 myr. 3He, 10Be and 26A1 ages for the "late Wisconsin" Ross Sea Drift, a glacial drift deposited on the coast of McMurdo Sound by the Ross Sea Ice Sheet, range from 8-106 kyr. The age range suggests that this deposit does not, as previous studies suggested, represent a single ice advance in response to lowered sea level at the last glacial maximum. The age range may reflect several ice sheet advances during the last glacial period. Paired measurements of 3He and 10Be in a number of quartz sandstones from the Dry Valleys region show that cosmogenic 3He is not completely retained on time scales of greater than 200 kyr. The 3He and 10Be data, combined with measurements of 3He concentrations in quartz as a function of grain size, suggest effective 3He diffusion coefficients in quartz from ~10-19 to 10-17 cm2 s-1. These values are one to three orders of magnitude greater than previous experimental determinations extrapolated to low temperature. The data also suggest that 3He diffusion rates in quartz do not follow simple volume diffusion trends and are probably sample dependent. Crushing quartz grains in vacuo releases variable concentrations of 3He and 4He and releases helium with surprisingly high isotopic compositions, up to 148 x Ra (Ra = atmospheric 3He/4He ratio: 1.384 x 10-6). The high ratios suggest that crushing releases cosmogenic 3He. As a result, in vacuo crushing cannot be used to correct 3He concentrations in quartz for a magmatic or "non-cosmogenic" component, as is possible in olivine and clinopyroxene phenocrysts in volcanic rocks. Replicate analyses of quartz samples show widely varying 4He concentrations and constant 3He concentrations. These data allow constraint of the nucleogenic 3He/4He production ratio to less than 0.05 ± 0.2 Ra. Corrections to exposure ages for nucleogenic 3He are in most cases < 1 % of total 3He. Depth profiles of 10Be and 26 A1 in two quartz sandstone bedrock cores have exponential attenuation lengths of 145 ± 5 and 145 ± 6 g cm-2 (10Be) and 153 ± 13 and 152 ± 5 g cm-2 (26 A1). The close agreement between the apparent attenuation lengths for the two nuclides supports previous suggestions that the 26A1/10Be ratio does not change with depth and therefore can be used as an independent chronometer. 3He profiles in 0.5-0.7 mm and 1.0-1.3 mm grains in one of the cores have scale lengths of 135 ± 6 and 152 ±7 g cm-2, respectively. In the second core the exponential scale length for 3He (0.5-0.7 mm grains) is significantly higher, 227 ±14 g cm-2. Models incorporating 3He production by muons and neutrons, erosion, and diffusion suggest that, to explain the core data, muon-induced production rates of 3He must be greater than previously believed. Previously overlooked (n,T) or (n,3He) reactions on Si and 0 by relatively low energy (e.g., lO's of MeV) neutrons produced in muon capture reactions are a possible explanation for the discrepancy.