Modeling δ18Ow to calculate paleotemperatures from deep time lacustrine carbonates

BOWEN G.J.; IRMIS, RANDALL B.; BENAVENTE, CECILIA ANDREA; MANCUSO, ADRIANA CECILIA

hawai

Congreso; Goldschmidt2020; 2020

Paleotemperature proxy data from lacustrine sediments are fundamental to understanding terrestrial climate in deep time. Because the relationship between the oxygen isotope composition of water (δ18Ow), the oxygen isotope composition of carbonate (δ18Oc), and temperature, are well understood in the modern, δ18Oc has been used to estimate Quaternary lacustrine records paleotemperatures. This could be applied for deep time calculations, being challenging because δ18Ow can be affected by geographic and climatic factors which control precipitation isotope ratios as well as evaporation of lake water. We have faced these challenges in interpreting records from Anisian carbonate-rich paleolakes from the Cuyana Rift Basin, central-west Argentina. A multiproxy paleoenvironmental-paleoclimate reconstruction was obtained for these systems using stable C and O isotope compositions. To infer paleotemperature, we estimated δ18Ow using the geographic relationship of Bowen and Revenaugh (2003), to predict precipitation δ18O, with two end members of paleolatitude calculated from paleopole data and two end members of paleoelevation from the range of modern lake systems within the East Africa Rift. Following Leng and Marshall (2004), this yielded four different paleotemperature estimates for each δ18Oc value. This approach is limited in that: a) is imprecise; b) it assumes a modern geographic distribution of precipitation isotope ratios, and c) it does not address potential impacts of lake water evaporation. To address these issues, we explore a multiproxy data analysis using the joint proxy inversion method of Bowen et al. (2020), and combining δ18O and δ13C data from our lake carbonates with mid-Triassic sea surface temperature proxies. The resulting analysis leverages multiple lines of information to help constrain our regional paleotemperature estimates, providing a promising path forward for developing more accurate and precise continental paleotemperature reconstructions from these and other deep time lake sequences.