Provenance of long-travelled dust determined with ultra-trace-element composition: a pilot study with samples from New Zealand glaciers

• Published in: Earth surface processes and landforms Vol. 30; no. 6; pp. 699 - 716
• Main Authors: Marx, Samuel K., Kamber, Balz S., McGowan, Hamish A.
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01-06-2005; Wiley
• Subjects: Australia; Bgi / Prodig; contamination; dust; Geomorphology; ICP-MS; Methods and techniques; New Zealand; Physical geography; provenance; trace elements; New Zealand; Australia; South Island; Loess; Atmospheric pollution; Aeolian features; Trace-element; Glacier; Aeolian dust; Geochemistry; Aeolian transport; Palaeosol
• ISSN: 0197-9337; 1096-9837
• DOI: 10.1002/esp.1169

We report high‐precision inductively coupled plasma mass spectrometric (ICP‐MS) compositional data for 39 trace elements in a variety of dust deposits, trapped sediments and surface samples from New Zealand and Australia. Dusts collected from the surface of alpine glaciers in the Southern Alps, New Zealand, believed to have undergone long‐distance atmospheric transport from Australia, are recognizable on account of their overabundances of Pb and Cu with respect to typical upper crustal values. Long‐travelled dust from Australia therefore scavenges these and other metals (e.g. Zn, Sb and Cd) from the atmosphere during transport and deposition. Hence, due to anthropogenic pollution, long‐travelled Australian dusts can be recognized by elevated metal contents.
The relative abundance of 25 other elements that are not affected by atmospheric pollution, mineral sorting (Zr and Hf) and weathering[sol ]solubility (alkali and earth alkali elements) reflects the geochemistry of the dust source sediment. As a result, we are able to establish the provenance of dust using ultra‐trace‐element chemistry at regional scale. Comparison of long‐travelled dust chemistry with potential Australian sources shows that fits of variable quality are obtained. We propose that the best fitting potential source chemistry most likely represents the major dust source area. A binary mixing model is used to demonstrate that admixture of small quantities of local dust provides an even better fitting dust chemistry for the long‐travelled dusts. Copyright © 2005 John Wiley & Sons, Ltd.