Variations in Os- and Mo-isotope compositions and trace element abundances across the Permo-Triassic boundary, Meishan, China: Proxy evidence for large-scale oceanic anoxia?

We have carried out Os-isotope analyses, and major and trace element determinations, on samples from the Permo-Triassic section at Meishan B sections, adjacent to the GSSP, China, to assess environmental change at the P-Tr boundary (Bed 27) and the underlying main extinction horizon (Bed 25). Major and trace element abundances were determined by XRF on 38 samples from a 2.3-metre interval of limestone, shale and dolomitic marl from Bed 24a to Bed 32. Re, Os and Mo abundances, and Os-isotope compositions, were determined on a subset of 26 samples, predominantly dark-coloured mudrocks; new Mo-isotope data will be presented. Initial 187Os/188Os(t=250Ma) ranges from 0.3 to 2, with no obvious pattern of change up-section. The data thus provide no unambiguous evidence at Meishan for major perturbation of seawater 187Os/188Os in this region of Palaeotethys, at least at the scale of the sampling. This is unlike early Jurassic and end-Cretaceous boundary sections, and mid-Cretaceous OAE intervals, which record substantial shifts in seawater 187Os/188Os that clearly reflect the influence of LIP emplacement. Major changes in elemental abundances occur between Beds 24 and Bed 27, accompanying the documented excursion in d13C. Redox- and biologically-sensitive elements such as Cu, Ni, P, V and Zn show strong fluctuations in abundance throughout Bed 24, even when abundances are normalised using Al2O3 to minimise the effects of carbonate dilution. In the lower part of Bed 27, however, the Al2O3-normalised concentrations of these elements decrease by an order of magnitude, and remain consistently low in the overlying Triassic marls. The relative decreases in Re and Os abundances throughout Bed 27 are even more substantial. Whilst these decreases are partly an artefact of Al-normalisation, the changes are accompanied by large increases in ratios involving lithogenous elements such as REE, Nb, Zr, Th and Ti. The data thus record a dramatic change in the marine depositional environment at Meishan, particularly within Bed 27, although levels of redox-sensitive elements were also very low within the upper part of Bed 24. A possible cause of these changes may have been an increase in the flux of terrigenous material, but this is not reflected by any obvious changes in 187Os/188O. Alternatively, these observations may reflect the occurrence of widespread oceanic anoxia in distal but contiguous water masses that caused a sudden reduction in the abundance of redox-sensitive elements in seawater. If the Meishan seas were fully open with Paleotethys, then seawater anoxia was global, as suggested by other studies. These results, combined with new Mo-isotope data, will shed further light on the phenomenon of marine anoxia in the vicinity of the P-Tr boundary and its relationship to the largest mass extinction of the Phanerozoic.