New perspectives on the Li isotopic composition of the upper continental crust and its weathering signature

Abstract

Editor: M. Bickle Keywords: Li isotopes loess upper continental crust chemical weatheringLithium isotopes are increasingly used to trace both present-day and past weathering processes at the surface of the Earth, and could potentially be used to evaluate the average degree of past weathering recorded by the upper continental crust (UCC). Yet the previous estimate of average δ 7 Li of the UCC has a rather large uncertainty, hindering the use of Li isotopes for this purpose. New δ 7 Li for desert and periglacial loess deposits (windblown dust) from several parts of the world (Europe, Argentina, China and Tajikistan) demonstrate that the former are more homogeneous than the latter, and may therefore serve as excellent proxies of the average composition of large tracts of the UCC. The Li isotopic compositions and concentrations of desert loess samples are controlled by eolian sorting that can be quantified by a binary mixing between a weathered, fine-grained end-member, dominated by phyllosilicates and having low δ 7 Li, and an unweathered, coarse-grained end-member, that is a mixture of quartz and plagioclase having higher δ 7 Li. We use correlations between insoluble elements (REE, Nd/Hf and Fe 2 O 3 /SiO 2 ), Li concentrations (henceforth referred as [Li]), and δ 7 Li to estimate a new, more precise, average Li isotopic composition and concentration for the UCC: [Li] = 30.5 ± 3.6(2σ ) ppm, and δ 7 Li = +0.6 ± 0.6(2σ ). The δ 7 Li for desert loess deposits is anti-correlated with the chemical index of alteration (CIA). Using this relationship, along with our average δ 7 Li, we infer that (1) the present-day CIA of the average UCC is 61 +4 −2 (2σ ), higher than the common reference value of 53, and (2) the average proportion of chemically weathered components is as high as 37 +17 −10 (2σ )% at the surface of the Earth.

Global climate stabilisation by chemical weathering during the Hirnantian glaciation

“…Therefore, Li inputs must have had an isotope ratio close to 0 ‰. Assuming a modern-like hydrothermal input, this requires that rivers had δ 7 Li values essentially unfractionated from the continental crust (modern value ~0 ‰; Sauzeat et al, 2015). This possibility is supported by δ 7 Li values of ~2 ‰ for the Amazon river (Dellinger et al, 2015), and similarly low values during the peak of the Cenomanian-Turonian hyperthermal (Pogge von Strandmann et al, 2013).…”

“…The δ 7 Li of primary silicate rocks defines a narrow range (continental crust ~0.6 ± 0.6 ‰, basalt ~3-5 ‰; Sauzeat et al, 2015) compared to the high variability in modern rivers (2-44 ‰; Huh et al, 1998;Dellinger et al, 2015;Pogge von Strandmann and Henderson, 2015). Riverine values reflect weathering processes, particularly the extent of preferential uptake of 6 Li into secondary minerals (Dellinger et al, 2015), and therefore reflect "weathering congruency", defined as the ratio of primary rock dissolution (driving rivers to low, rock-like, δ 7 Li = congruent dissolution of rock), to secondary mineral formation (driving rivers to high δ 7 Li; Misra and Froelich, 2012;Pogge von Strandmann and Henderson, 2015).…”

Compositional evolution of the upper continental crust through time, as constrained by ancient glacial diamictites

“…14b). In particular, the diamictite medians for Li, Rb, Y, MREE, HREE, W, and Tl are within uncertainty of the Australian alluvium-based model of Kamber et al (2005), and the median for Li is also similar to that of Teng et al (2004) (35 ppm), as well as a newly proposed average UCC value based on loess analyses (Sauzéat et al, 2015) (30.5 ppm). The diamictite medians for Cr, Co, and Ni, while lower than UCC RG , are higher than the Shaw et al (1967Shaw et al ( , 1976 and Eade and Fahrig (1973) models based on bedrock sampling, but are comparable to the Kamber et al (2005) model for Cr and Ni.…”

“…Most of these studies use either shale, glacially-derived loess, or desert loess to infer the average composition of the UCC, which works well for insoluble elements, but not for soluble elements. Although the cold temperatures and largely physical weathering that produce the glacial loess limits loss of the soluble elements compared with shales, such loess still records a weathering signature (Gallet et al, 1998), and both types of loess are also affected by wind-driven particle sorting that fractionates heavy minerals, which may dominate the budgets of certain elements (e.g., Ti, Zr, Hf, Sn) Liu et al, 1993;Barth et al, 2000;McLennan, 2001;Chauvel et al, 2014;Sauzéat et al, 2015).…”

“…The former approach typically involves the collection and analysis of thousands of samples and has been carried out in the Canadian Shield (Shaw et al, 1967(Shaw et al, , 1976Eade and Fahrig, 1973), Russia Yaroshevsky, 1967, 1976;Borodin, 1998), China (Gao et al, 1998), and Japan (Togashi et al, 2000). The latter approach assumes that natural geological (sedimentary) processes provide a robust average of insoluble elements in the provenance of the sediments, and has been used in many studies aimed at tracking the compositional evolution of the UCC Taylor, 1976, 1977;McLennan et al, 1979McLennan et al, , 1983McLennan, 1985, 1995;Wronkiewicz and Condie, 1987, 1989, 1990Peucker-Ehrenbrink and Jahn, 2001;Kamber et al, 2005;Chauvel et al, 2014;Sauzéat et al, 2015). Most of these studies use either shale, glacially-derived loess, or desert loess to infer the average composition of the UCC, which works well for insoluble elements, but not for soluble elements.…”