Calcium and magnesium isotope systematics in rivers draining the Himalaya-Tibetan-Plateau region: Lithological or fractionation control?

Abstract

In rivers draining the Himalaya-Tibetan-Plateau region, the 26Mg/ 24Mg ratio has a range of 2‰ and the 44Ca/ 42Ca ratio has a range of 0.6‰. The average δ 26 Mg values of tributaries from each of the main lithotectonic units (Tethyan Sedimentary Series (TSS), High Himalayan Crystalline Series (HHCS) and Lesser Himalayan Series (LHS)) are within 2 standard deviation analytical uncertainty (0.14‰). The consistency of average riverine δ 26 Mg values is in contrast to the main rock types (limestone, dolostone and silicate) which range in their average δ 26 Mg values by more than 2‰. Tributaries draining the dolostones of the LHS differ in their δ 44 42 Ca values compared to tributaries from the TSS and HHCS. The chemistry of these river waters is strongly influenced by dolostone (solute Mg/Ca close to unity) and both δ 26 Mg (−1.31‰) and δ 44 42 Ca (0.64‰) values are within analytical uncertainty of the LHS dolostone. These are the most elevated δ 44 42 Ca values in rivers and rock reported so far demonstrating that both riverine and bedrock δ 44 42 Ca values may show greater variability than previously thought. Although rivers draining TSS limestone have the lowest δ 26 Mg and δ 44 42 Ca values at −1.41 and 0.42‰, respectively, both are offset to higher values compared to bedrock TSS limestone. The average δ 26 Mg value of rivers draining mainly silicate rock of the HHCS is −1.25‰, lower by 0.63‰ than the average silicate rock. These differences are consistent with a fractionation of δ 26 Mg values during silicate weathering. Given that the proportion of Mg exported from the Himalaya as solute Mg is small, the difference in 26Mg/ 24Mg ratios between silicate rock and solute Mg reflects the 26Mg/ 24Mg isotopic fractionation factor ( α silicate – dissolved Mg ) between silicate and dissolved Mg during incongruent silicate weathering. The value of α silicate – dissolved Mg of 0.99937 implies that in the TSS, solute Mg is primarily derived from silicate weathering, whereas the source of Ca is overwhelmingly derived from carbonate weathering. The average δ 44 42 Ca value in HHCS rivers is within uncertainty of silicate rock at 0.39‰. The widespread hot springs of the High Himalaya have an average δ 26 Mg value of −0.46‰ and an average δ 44 42 Ca value of 0.5‰, distinct from riverine values for δ 26 Mg but similar to riverine δ 44 42 Ca values. Although rivers draining each major rock type have δ 44 42 Ca and δ 26 Mg values in part inherited from bedrock, there is no correlation with proxies for carbonate or silicate lithology such as Na/Ca ratios, suggesting that Ca and Mg are in part recycled. However, in spite of the vast contrast in vegetation density between the arid Tibetan Plateau and the tropical Lesser Himalaya, the isotopic fractionation factor for Ca and Mg between solute and rocks are not systematically different suggesting that vegetation may only recycle a small amount of Ca and Mg in these catchments. The discrepancy between solute and solid Ca and Mg isotope ratios in these rivers from diverse weathering environments highlight our lack of understanding concerning the origin and subsequent path of Ca and Mg, bound as minerals in rock, and released as cations in rivers. The fractionation of Ca and Mg isotope ratios may prove useful for tracing mechanisms of chemical alteration. Ca isotope ratios of solute riverine Ca show a greater variability than previously acknowledged. The variability of Ca isotope ratios in modern rivers will need to be better quantified and accounted for in future models of global Ca cycling, if past variations in oceanic Ca isotope ratios are to be of use in constraining the past carbon cycle.