Instrumental sea surface temperature data only cover the last two centuries. Beyond that, our knowledge of sea surface temperature relies on indirect inference from geochemical indicators in marine sediments known as paleotemperature proxies. Past ocean temperature reconstructions typically rely on a single sediment core to infer climate changes at a study site and across an entire region. Additionally, seawater temperature estimates from different proxies are typically assumed to reflect the mean annual sea surface temperature in global or regional compilations. However, these reconstructions rely on two untested assumptions: that a single core can reliably represent climate variability in the region, and that the different proxies record temperatures from the same season and water depth.
To test these assumptions, PhD student Ru-Yun Tung and Associate Professor Sze Ling Ho from IONTU collaborated with researchers from the Department of Geoscience, NTU, Hokkaido University, the National Museum of Nature and Science in Japan, and the Alfred Wegener Institute in Germany. The team analyzed three commonly used sea surface temperature proxies (foraminiferal Mg/Ca, UK’37, TEX86) from four nearby sediment cores collected within a radius of 10 km from the northern Okinawa Trough by Japanese research vessels Kaiyo and Kairei, the IODP drilling vessel JOIDES Resolution, and a French research vessel Marion Dufresne.
Despite being collected using different coring devices and analyzed in various laboratories, records of the same proxy type from different sediment cores display similar glacial-interglacial temperature trends, thereby confirming the long-held assumption that the first-order pattern in paleotemperature records is representative of the core location despite sediment inhomogeneity. However, there are abrupt jumps in paleotemperature proxy records. Since these abrupt fluctuations did not consistently appear in all sediment cores, they are unlikely to reflect actual climate changes but might instead reflect proxy noise caused by instrumental errors or sedimentary heterogeneity. In addition, a greater similarity between TEX86, which is typically considered an SST indicator, and other intermediate water temperature proxies––compared to SST and thermocline temperature records––suggests that TEX86 may reflect the temperature below the thermocline in the Okinawa Trough region. Overall, our findings provide additional constraints for a quantitative interpretation of paleotemperature proxy records, pushing future paleoclimate reconstructions toward greater accuracy and reliability.
The study was published in JpGU journal Progress in Earth and Planetary Science: Tung, R., Ho, S. L., Kubota, Y., Yamamoto, M., Hefter, J., & Shen, C. (2024). Replicability of paleotemperature records in the northern Okinawa Trough and its implications for paleoceanographic reconstructions. Progress in Earth and Planetary Science, 11(1), 1-23. https://doi.org/10.1186/s40645-024-00664-5
