Evaluating the biological pump efficiency of the Last Glacial Maximum ocean using <i>?</i><sup>13</sup>C

Abstract. Although both physical and biological marine changes are required to explain the 100 ppm lower atmospheric pCO2 of Last Glacial Maximum (LGM, ?21 ka) as compared preindustrial (PI) times, their exact contributions debated. Proxies past carbon cycling (such ?13C) document these thus provide constraints for quantifying drivers long-term cycle variability. This modeling study discusses in ocean needed simulate an LGM satisfactory agreement with proxy data, here focusing especially on ?13C. We prepared a PI equilibrium simulation using model NorESM-OC full biogeochemistry (including isotopes ?13C radiocarbon) dynamic sea ice. The modeled LGM–PI differences evaluated against wide range biogeochemical data show key aspects state within uncertainties. However, lack simulated increase regenerated nutrients indicates that additional data. In order examine changes, we explore potential effects different global mean pump efficiencies tracer distributions. Through estimating which efficiency reduces model–proxy biases most, estimate increased from 38 % up 75 (LGM). such may be related circulation incompletely captured by our – stronger isolation Southern Source Water. Finally, even after considering ocean, remaining error exists is 0.07 ‰ larger than 0.19 uncertainty. dynamics