ASSESSING CRITICAL DRIVERS OF THE ORDOVICIAN GREENHOUSE-ICEHOUSE TRANSITION WITH PAIRED STRONTIUM AND NEODYMIUM ISOTOPE STRATIGRAPHY

Modeling the control of global carbon fluxes on atmospheric carbon dioxide (CO₂) levels through time in accordance with oxygen isotope paleotemperature curves is one of the leading approaches used to understand the origin and timing of the Ordovician greenhouse-icehouse transition. Primary fluxes considered in carbon cycle models are the release of carbon dioxide by volcanic and metamorphic degassing, drawdown by chemical weathering of calcium silicate minerals, and to a lesser extent burial and oxidation of marine organic matter.

A modeling study by Young et al. (2009) utilized the seawater Sr isotope curve to drive enhanced non-radiogenic basaltic weathering and a mid-Ordovician drawdown of atmospheric CO₂, albeit with the assumption that volcanic degassing balanced this flux for most of the Late Ordovician to match the temperature curve of Trotter et al (2008). More recently, McKenzie et al. (2016) proposed that a lowering of the degassing flux was the principal cause of the Ordovician greenhouse-icehouse transition; in contrast, Swanson-Hysell and Macdonald (2017) presented a Laurentian paleogeographic reconstruction which supports the basaltic weathering hypothesis by placing the paleocontinent and Taconic arc in the humid tropics. Collectively, these studies raise questions about the importance of silicate weathering in driving Ordovician atmospheric CO₂ and climate.

Our current effort examines the possibility that the decrease in mid-Ordovician Sr isotope ratios was actually caused by a flux of non-radiogenic Sr from mid-ocean ridge hydrothermal alteration. We measured paired Sr (⁸⁷Sr/⁸⁶Sr) and Nd (¹⁴³Nd/¹⁴⁴Nd) isotopes in marine carbonates in the Appalachian region because there is no appreciable hydrothermal flux of Nd and therefore changes in Nd isotopes should signify changes in weathering sources. Our results show a fall in Nd of 10 epsilon units that is coeval with the mid-Ordovician ⁸⁷Sr/⁸⁶Sr drop of 0.0005, which is consistent with enhanced basaltic weathering in the Appalachian region that was large enough to affect the global Sr cycle. Ongoing and future studies will construct bulk rock and conodont Sr and Nd curves for other sections in the Appalachian Basin and beyond (e.g. Nevada, Oklahoma) to constrain the timing of shifts in these signals to clarify the role of hydrothermal Sr input.