GSA Connects 2021 in Portland, Oregon

Paper No. 138-11
Presentation Time: 11:10 AM


PASSEY, Benjamin, PhD1, PACKARD, Natalie R.1, KELSON, Julia R.1, SHELDON, Nathan1, NIEMI, Nathan1, RHODES, Rebekah L.2, HYLAND, Ethan G.2 and CARROLL, Alan3, (1)Department of Earth and Environmental Sciences, University of Michigan, 1100 N University Ave, Ann Arbor, MI 48109-1005, (2)Department of Marine, Earth, and Atmospheric Sciences, North Carolina State University, 2800 Faucette Dr., Rm. 1125 Jordan Hall, NC State University, Raleigh, NC 27695, (3)Department of Geoscience, University of Wisconsin-Madison, 1215 W Dayton St, Madison, WI 53706

The Green River Formation in southwestern Wyoming records the existence of paleolake Gosiute, a massive early Eocene lake that reached areal extents up to ~40,000 km2. During its several million-year history, Lake Gosiute shifted from a largely freshwater lake (overfilled to balance-filled) to a largely saline lake (underfilled), and then back to a balance-filled and overfilled lake. The δ18O values of lacustrine carbonates show pronounced, step-like shifts, with the saline stage ~ 6 to 8 ‰ higher in δ18O than the preceding and ensuing freshwater stages. However, the significance of these shifts is unclear: what was the importance of changes in evaporation versus changes in the isotopic composition of catchment precipitation? What was the δ18O of ‘pristine’ (unevaporated) catchment precipitation during different stages of the lake’s history, and what can this tell us about catchment paleogeography?

The triple oxygen isotope composition (Δ′17O, δ18O) of lake water and lacustrine minerals is sensitive to evaporation, and can be used to estimate the δ18O values of catchment waters prior to evaporation. We report triple oxygen isotope compositions of >30 lacustrine carbonates spanning the Green River formation and stratigraphically adjacent units. Our results show that (1) lake waters were strongly modified by evaporation (ca. + 10‰ in δ18O) during all major stages (overfilled, balance-filled, under-filled); (2) The previously observed 6 to 8 ‰ shift in δ18O largely represents a shift in the δ18O of catchment precipitation and not in the degree of evaporative enrichment; and (3) the δ18O of unevaporated catchment precipitation was ~ –19‰ prior to and following the saline stage, and ~ –12‰ during the freshwater stage. Following previous interpretations, we hypothesize that the saline stage was bracketed by the abandonment and then recapture of a large paloeriver. Using the isotopic mass balance model of Doebbert et al. (2010), a captured river with δ18O ~ –19‰ would shift lake water δ18O by the observed ~ –7‰ if it had an annual discharge of ~ 25 x 109 m3/year, comparable to the modern Missouri River at Bismarck, ND. By analogy with modern δ18O–elevation relationships, a value of –19‰ suggests high elevations (~ ≥ 2500 m) in this region during the early Eocene.

Reference: Doebbert A.C., et al. (2010), GSA Bulletin 122, 236-252.