Two possible source regions for central Greenland last glacial dust

Abstract

Dust in Greenland ice cores is used to reconstruct the activity of dust‐emitting regions and atmospheric circulation. However, the source of dust material to Greenland over the last glacial period is the subject of considerable uncertainty. Here we use new clay mineral and <10 µm Sr–Nd isotopic data from a range of Northern Hemisphere loess deposits in possible source regions alongside existing isotopic data to show that these methods cannot discriminate between two competing hypothetical origins for Greenland dust: an East Asian and/or central European source. In contrast, Hf isotopes (<10 µm fraction) of loess samples show considerable differences between the potential source regions. We attribute this to a first‐order clay mineralogy dependence of Hf isotopic signatures in the finest silt/clay fractions, due to absence of zircons. As zircons would also be absent in Greenland dust, this provides a new way to discriminate between hypotheses for Greenland dust sources.

Sr‐Nd‐Hf Isotopic Analysis of <10 mg Dust Samples: Implications for Ice Core Dust Source Fingerprinting

“…After intense stirring, the samples were allowed to settle and the remaining liquid was decanted by syringe after 55–57 min, and dried on a hotplate at 50° C leaving the fine particles behind. Grain size distributions of these fine separates are dominated by 1–10 μm particles with peaks at 3–4 μm (see supporting information in Újvári et al, ). To investigate the effects of different acid treatments on measured isotopic ratios, the sample size was reduced (∼20 g) and various sample‐to‐acid ratios were employed.…”

“…Also, potential source area (PSA) samples such as loess sediments (including the finer fractions) geochemically representing a larger region of the upper continental crust partly consist of secondary carbonates formed during pedogenic processes subsequent to dust deposition (Barta, ; Becze‐Deák et al, ). These minerals also have Sr isotopic compositions different from the dust parent material, therefore the strategy of comparing the carbonate‐free, aluminosilicate fractions is usually adopted (Biscaye et al, ; Bory et al, ; Delmonte et al, ; Sugden et al, ; Svensson et al, ; Újvári et al, ). Selective dissolution of calcite in loess was achieved using weak (5% or 0.5 mol/L) acetic acid at room (or slightly elevated) temperature (Chen et al, ; Újvári et al, ; Yang et al, ).…”

“…To investigate the effects of different acid treatments on measured isotopic ratios, the sample size was reduced (∼20 g) and various sample‐to‐acid ratios were employed. This different sample treatment was then compared to the scheme outlined in Újvári et al (), which used larger samples (150–200 g) for the acid dissolution.…”

“…Since the most powerful dust fingerprinting methods, such as REE composition (Ferrat et al, ; Gallet et al, ; Zdanowicz et al, ) and Sr‐Nd‐Pb isotopic analyses (e.g., Grousset & Biscaye, ), are destructive, there is a clear need to establish sequential separation techniques of Sr, Nd, Pb and other REEs to retrieve the most information from small (5–10 mg) dust samples recovered from ice cores. Although Hf isotopes have recently been added as a tool of aerosol/dust source discrimination (Aarons et al, ; Aciego et al, ; Blakowski et al, ; Lupker et al, ; Pettke et al, ; Pourmand et al, ; Újvári et al, ; Zhao et al, ), precise Hf isotopic measurements of small (<10 mg) dust samples are still challenging due to their minute amounts of Hf (on the order of 1–10 ng); these analyses are often compromised by problems that can arise during ion exchange chemistry. Such problems may include high Lu and Yb blanks causing inaccurate interference corrections on 176 Hf (Münker et al, ), insufficient Ti removal leading to reduced Hf transmission in the mass spectrometer (Blichert‐Toft et al, ; Wimpenny et al, ), and high Zr/Hf producing bias on 176 Hf/ 177 Hf ratios (Peters et al, ).…”

“…International rock reference materials AGV‐1 and BCR‐2 (10 mg aliquots) were used to demonstrate that the experimental setup allow for valid 176 Lu and 176 Yb corrections, and in turn results in reliable 176 Hf/ 177 Hf (and also 87 Sr/ 86 Sr, 143 Nd/ 144 Nd) ratios. Along with the standards 50, 10, and 5 mg aliquots of the fine (<10 μm) fractions of three wind‐blown loess sediment samples from Central Europe, China and the US were processed to (1) demonstrate that reproducible Sr, Nd and Hf isotopic ratios can be obtained from small size (<10 mg) dust samples with the applied procedure by comparing these ratios to those acquired on larger samples (Újvári et al, ), (2) to reveal intra‐ and inter‐sample variabilities in isotopic signatures, (3) to test acid pretreatment effects on the isotopic ratios, and (4) to gain insight into the uncertainties of Hf isotopic ratios of small volume dust samples and their suitability for distinguishing dust sources. Furthermore, Monte Carlo simulations were performed to model zircon depletion effects for two grain size fractions (<10 and <2 μm) of a hypothetical 5 mg dust sample.…”