Award: PLR-1401489

The broadest motivation for this project was to explain why atmospheric CO2 was 80-100 ppm lower during ice ages than it has been over the current interglacial period (prior to the effects of humans) and previous interglacials. In the global surface ocean, organic matter is produced by phytoplankton and sinks to the subsurface, where it is decomposed into inorganic components (i.e., CO2 and the nitrogen nutrient nitrate) and stored in the ocean interior. This transport of C and N is referred to as the biological CO2 pump, as it "pumps" CO2 out of the atmosphere and into the deep ocean. In the mid- and low-latitude ocean, nutrients are scarce and thus completely used so as to drive a highly efficient biological pump. However, in the high nutrient, low chlorophyll (HNLC) zones, high supply rate from subsurface ocean and limitation of phytoplankton growth by a combination of light and iron lead to high concentrations of (unused) surface nutrients, compromising the efficiency of the biological pump and resulting in outgassing of CO2 back to the atmosphere. The Southern Ocean is the biggest HNLC zone, where changes in the completeness of nutrient drawdown have the potential to explain much of the 80-100 ppm change in atmospheric CO2 over glacial/interglacial cycles. The completeness of nutrient drawdown in this region is affected by both the physical circulation of the ocean, which supplies the nutrients from the ocean interior, as well as the physical and biochemical conditions of the surface ocean, which determines the growth of the phytoplankton. To understand how the efficiency of the biological pump change during glacial/interglacial cycles, and what mechanism caused it to change, we need to analyze deep-sea sediment cores to reconstruct past environmental and biological conditions. Diatoms, algal cells that make silica shells, are the dominant phytoplankton in the Southern Ocean. The ratio of the two stable isotopes of nitrogen (the ratio of 15N to 14N, or 15N/14N) of the organic matter within the walls of the diatom shells is a useful tool to reconstruct past surface ocean nutrient conditions. As diatoms grow and reproduce, their 15N/14N reflects the 15N/14N of the N in the water that they are consuming, and the 15N/14N of the N in the water is in turn affected by the completeness of N consumption by the diatoms. Records of diatom shell 15N/14N have been generated previously. Most of these records show a generally elevated 15N/14N during cold glacial intervals, meaning that diatoms consumed a larger fraction of the available N pool, pointing to Southern Ocean changes that yielded a more efficient biological pump. Remarkably, in the Antarctic Zone, the more polar part of the Southern Ocean, the apparent ice age rise in degree of N consumption coincided with a decline in diatom productivity, suggesting a reduction in the circulation-driven nutrient supply to the Antarctic surface ocean. This change is not predicted by global climate models, so that more and better data were called for to test that such a circulation change did indeed occur and, if so, to provide a physical mechanism for the change. Of particular concern, records of diatom shell 15N/14N from other investigators were not that similar across cores from different regions in the Antarctic Zone, with some records actually showing the opposite sense of 15N/14N change over glacial cycles. We generated three new diatom shell 15N/14N records, using improved methods for purifying diatoms from deep sea mud and for separating different diatom species from one another. The resulting data yield a highly coherent picture of N isotopic change in the Antarctic Zone over the last glacial cycle, on opposite sides of the Antarctic continent: the central Pacific sector and the central Indian sector. This gives us much more confidence that the degree of nutrient consumption rose in the Antarctic Zone during past ice ages. Our new records also provide precise information about timing, which had been lacking previously, especially during the ice ages. These findings are critical for diagnosing the physical processes that explain the decline in nutrient supply to the Antarctic Zone during the ice ages. Finally, we have compiled our Southern Ocean data covering the past 10,000 years (the Holocene Period, the current interglacial period during which complex human civilization developed). We find that increased upwelling in the Antarctic Zone can explain the gradual increase of CO2 over the Holocene, before large scale CO2 emissions by humans. This project provided a platform for training a postdoctoral researcher, a graduate student, and multiple undergraduates in environmental and geoscience research methods relating to the polar oceans. Outreach by the PI to middle and high school teachers was conducted through 1-day "Teachers As Scholars" workshops organized by Princeton University?s Program in Teacher Preparation. Last Modified: 07/31/2019 Submitted by: Daniel M Sigman