The Efficiency of the Biological Pump in the Southern Ocean over the Past 40,000 Years As Evidenced from Deep-Sea Corals

Abstract:

The efficiency of the biological pump (E_BP) in the Southern Ocean (i.e. the completeness of nutrient consumption in its surface waters) is a critical modulator of the atmospheric CO₂ concentration (pCO₂), but how it has varied in the past is uncertain. The nitrogen isotopic composition (d15N) of organic matter is a promising tool in reconstructing E_BP in the past Southern Ocean because phytoplankton preferentially consume ¹⁴N-nitrate relative to ¹⁵N-nitrate, resulting in a correlation between the d15N of sinking organic matter and the fraction of the nitrate supply consumed in the surface ocean (E_BP). Here we present the first set of Southern Ocean d15N records from deep-sea fossil corals, which consume organic matter that derives from the sinking flux. The corals are from the Drake Passage, with collection sites in both the Antarctic and Subantarctic Zones (AZ, SAZ), and the SAZ south of Tasmania. While previous work has shown that the productivity changes from the Last Glacial Maximum (LGM) to the Holocene were opposite in the SAZ and AZ (LGM productivity was higher in the SAZ but lower in the AZ), all of our three regions show that the d15N during the LGM/deglaciation are ~4‰ higher than the Holocene, indicating a much more efficient biological pump across the entire glacial Southern Ocean. Greatest inter-coral d15N variability appears to occur during the deglaciation (10-15 ka) than during the Holocene and glacial period, and the data generated so far on intra-coral d15N variation (multiple measurements within individual coral septa) also show a maximum in variability during the deglaciation. This relatively high inter- and intra-coral d15N variability during the deglaciation suggests fluctuations in E_BP on annual/decadal timescales. Despite the short-timescale variation, the coral data in hand suggest that the overall deglacial decrease in d15N began at ~15 ka, later than expected if the data recorded a whole-Southern Ocean E_BP decline that led to the first rise in atmospheric pCO₂.