Mercury in marine Ordovician/Silurian boundary sections of South China is sulfide-hosted and non-volcanic in origin

• Published in: Earth and planetary science letters Vol. 511; pp. 130 - 140
• Main Authors: Shen, Jun, Algeo, Thomas J., Chen, Jiubin, Planavsky, Noah J., Feng, Qinglai, Yu, Jianxin, Liu, Jinling
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-04-2019
• Subjects: anoxia; black shale; mass extinction; mercury isotopes; pyrite; volcanism; black shale; mass extinction; mercury isotopes; pyrite; anoxia; volcanism
• ISSN: 0012-821X; 1385-013X
• DOI: 10.1016/j.epsl.2019.01.028

Mercury (Hg) enrichment in stratigraphic successions is now widely used as a proxy for volcanic inputs, often for the purpose of documenting a relationship between large igneous province (LIP) magmatism and ecosystem perturbations. Earlier studies of Hg in Ordovician/Silurian boundary (OSB) sections in South China and Laurentia identified transient spikes in Hg/TOC ratios, on the basis of which a link between volcanism and the Late Ordovician mass extinction (LOME) was claimed. However, Hg enrichments must be tested based on normalization to their main host phase, and Hg/TOC is a suitable proxy only if Hg is mainly complexed by organic matter in the sediment. Here, we demonstrate that Hg in three OSB sections in South China (Qiliao, Yanzhi, and Jiaoye) is overwhelmingly associated with pyrite, as shown by r(Hg–TS) > 0.9 (versus r(Hg–TOC) < 0.1) and by EDS elemental mapping. This association requires that Hg concentrations be normalized to pyrite content as proxied by total sulfur [TS], rather than to total organic carbon [TOC]. The resulting Hg/TS profiles show no significant enrichments at any level within the Upper Ordovician–lower Silurian of the study sections. Also, mercury isotope data show constant mass-independent fractionation (Δ199Hg) values (+0.11 ± 0.03‰) that are inconsistent with volcanic inputs. We therefore infer that previous reports of Hg enrichments in OSB sections were due to the presence of Hg-rich sulfides, and that Hg data from both the present and earlier studies provide no evidence of any volcanic influences on the LOME. The results of the present study highlight the need for caution in applying the Hg proxy for volcanic inputs and the importance of evaluating the main host phase of Hg in paleo-marine sediments.
•First report of pyrite (instead of organic matter) as major host for Hg in Ordovician–Silurian boundary strata.•Absence of Hg/TS (mercury/total sulfur) spikes in three Ordovician–Silurian boundary sections in South China.•Mercury isotope data show constant MIF (Δ199Hg) values that are inconsistent with volcanic inputs.•Evidence against published claims of volcanic Hg inputs and volcanic influences on Late Ordovician Mass Extinction.•Demonstrates need to identify and normalize Hg to its main host phase in marine sediments when used as a volcanic proxy.