UC Davis

The removal of exchangeable oxygen from diatom opal prior to δ18Odiatom analysis is a crucial first step before analyzing frustule oxygen isotopes for paleoceanographic applications. We present the results of experiments that quantify the temperature-dependent reactivity of biogenic silica with water under laboratory conditions. We demonstrate that controlled exchange between rinse water and diatom opal at room temperature results in predictable alteration of δ18Odiatom, after vacuum dehydroxylation. Diatom samples equilibrated with an 18O-enriched δ18Oequil. water solution of +94.4‰ at ~21 °C for 70 hr prior to dehydroxylation yield δ18Odiatom data that can be directly interpreted with existing empirical δ18Odiatom versus temperature relationships. We compare sediment trap-based δ18Odiatom temperature data with modern δ18Oforam temperatures. Finally, we present an ~220 year record of δ18Odiatom data from an eastern Pacific Guaymas Basin boxcore, analyzed using the microfluorination technique, that indicates diatom chlorophyll maximum (Chlmax) temperatures have shifted from a period during the late eighteenth and nineteenth centuries that was 8 °C cooler during the fall bloom relative to mean annual alkenone Uk′37 sea surface temperature, to the late twentieth century when we observe no difference between alkenone sea surface temperature and diatom Chlmax. These data suggest a twentieth century seasonal shift in the timing of fall upwelling and Guaymas Basin stratification breakdown that could be due to reduced upwelling efficiency and/or increased summer mixed layer thermal stratification.