How Consistent are Sea Surface Temperature Estimates from Different Proxies? An Assessment of the Alkenone, Mg/Ca, and Faunal Paleothermometers Using Records from the Plio-Pleistocene

Abstract:

A variety of different paleothermometers have been used to estimate past sea surface temperature (SST). However, a limited number of studies have explored the consistency of the temperature estimates produced by these proxies or assessed the cross proxy reproducibility of the major results gleaned from these records. Here, incorporating novel and previously published records from a variety of different sites (ODP Site 662 in the Eastern Equatorial Atlantic, DSDP Site 607 and Site 609 in the North Atlantic, and ODP Site 1125 in the Southwest Pacific), we assess the consistency of the long-term trends and orbital signatures of U^k’₃₇ SST records, relative to faunal and Mg/Ca estimates from the same sites. Exploring the similarity of companion SST datasets generated for different time windows within the Plio-Pleistocene at several different sites, we find that while there are structural differences between records, they produce very similar long-term trends and produce absolute temperature estimates that are generally within error of each other. Furthermore, companion records successfully capture the same dominant milankovtich period and produce phase estimates relative to benthic oxygen isotope estimates that are within error of each other. While there are some differences in the strength, temporal persistence, and presence/absence of the non-dominant milankovitch periodicies and some intervals in which one proxy record is coherent with δ¹⁸O while another is not, our results suggest that the primary results gleaned from SST proxy records remain intact regardless of which SST proxy is used.