Quaternary ice age cycles are characterized by fluctuations in atmospheric CO2, global temperature changes, ice volume changes, global vegetation changes, and physical and chemical changes in the ocean, including changes in deep ocean ventilation and circulation patterns. The ocean is the largest carbon reservoir and is likely a sink for atmospheric CO2 during glacials, but no one hypothesis can account for all the observed CO2 changes. A better understanding of global mean delta13C changes through time can help constrain the roles of vegetation and circulation changes on atmospheric CO2. By amassing a benthic delta 13C dataset covering as many glacial cycles as geologic and coring constraints allow, we can systematically improve estimates of global mean delta 13C, water mass boundary movements, deep ocean carbon storage changes, and past changes in ocean circulation. Additionally, deglacial ventilation of respired carbon represents an important feedback in the climate system that impacts the carbon cycle, global temperatures, sea level, etc. Therefore, reconstructing deglacial changes in benthic delta13C can improve estimates of changes in deep ocean carbon storage and the global carbon cycle.

At the Last Glacial Maximum (LGM, 23-19 ka), most of the carbon released by the terrestrial biosphere was stored in the ocean, where the light isotopic signature of terrestrial carbon is observed as a 0.32--0.7‰ depletion in mean benthic foraminiferal delta13C values relative to the Late Holocene (6--0 ka). We present a more precise estimate of glacial-interglacial delta 13C change of marine dissolved inorganic carbon using benthic Cibicidoides wuellerstorfi delta13C records from 480 core sites (more than 3 times as many sites as previous studies) from 0.5-5 km depth which yields a mean delta13C decrease of 0.38 +/- 0.08‰ (2?). Estimating large uncertainty ranges for delta13C change in the top 0.5 km, below 5 km, and in the Southern Ocean, we calculate a whole-ocean change of 0.34 +/- 0.19‰. This implies a terrestrial carbon change that is consistent with recent vegetation model estimates of 330--694 Gt C. Additionally, we find that a well-constrained surface ocean delta 13C change is essential for narrowing the uncertainty range of estimated whole-ocean delta13C change.

Here we estimate the mean ocean delta13C in 1-kyr time steps across the deglaciation (20-6 ka) by placing 118 benthic delta 13C records on regional age models, and averaging these nine regional delta 13C stacks. Additionally we volume-weight intermediate and deep delta 13C stacks, and a global ocean (0.75-5 km) mean delta13C stack. We perform empirical orthogonal function (EOF) on 118 benthic delta 13C records to identify the principle modes of variability in benthic delta 13C. We find the first component of the EOF explains 39% of the variance in benthic delta13C, and describes the long-term deglacial trend in the reduction in deep ocean carbon storage related to the deglacial increase in atmospheric CO2. The second component in benthic delta 13C explains 20% of the variance, and resembles proxies of changes in the strength of North Atlantic deep water formed and millennial scale (1000's of years) variability in North Atlantic iceberg calving event-sourced fresh water. This suggests a fifth of the variability in benthic delta 13C is due to rapid changes in the formation of deep water in the North Atlantic. A simulated mixing experiment with Atlantic delta 13C records from deep water formation regions indicates that the secondary signal in benthic delta13C is most consistent with a millennial-scale reduction in percent North Atlantic Deep Water at 2-4 km in the North Atlantic.

Similar to PC1, the volume-weighted Deep delta13C stack minus Intermediate delta13C stack tracks deep ocean carbon storage and covaries with atmospheric fn. Our volume-weighted global mean delta 13C stack suggests the terrestrial biosphere expanded from 19-6 ka while atmospheric CO2 reached its interglacial value around 8 ka. This 4D delta13C data set is suitable for model-data comparisons and time-stepping 3D visualizations.

Can 3D visualizations improve undergraduate students' understanding of fundamental Earth science concepts like plate tectonics? Our simple 3D Atlantic seafloor movie helps non-STEM students more than STEM students understand what the seafloor looks like and how ocean basins are formed. Simple 3D visualizations potentially benefit novice science students and closes the gap between STEM and non-STEM students' scores.