Atlantic overturning responses to obliquity and precession over the last 3 Myr

Abstract

This study analyzes 39 Atlantic and seven Pacific benthic 13 C records to characterize obliquity and precession responses in Atlantic overturning since 3 Ma. Regional benthic 13 C stacks are also analyzed. A major transition in orbital responses is observed at 1.5-1.6 Ma coincident with the first glacial shoaling of Northern Component Water. Since ∼1.5 Ma, the phases of Atlantic benthic 13 C records from 2300 to 4000 m depth lag maximum obliquity by 59 • (6.7 kyr) and June perihelion precession forcing by 133 • (8.5 kyr). Comparison with North Atlantic sea surface temperature suggests that these orbital responses (particularly precession) in middle deep Atlantic 13 C are associated with changes in ocean heat transport and overturning rates. The mid-Pleistocene transition had little effect on the obliquity and precession phases of benthic 13 C but did result in a ∼50% decrease in the obliquity power of middle deep Atlantic 13 C at 0.6 Ma.

Mediterranean Outflow Water variability during the Early Pleistocene

“…4a). In this regard, and Lisiecki (2014) suggested the lack of increased overturning circulation in the deep water during this time interval. At a first glance this stands in contrast to the proposed increased overturning circulation postulated by Bell et al (2015) in relation to changes in the North Atlantic surface water mass trajectory.…”

Deglacial carbon cycle changes observed in a compilation of 127 benthic <i>δ</i><sup>13</sup>C time series (20–6 ka)

“…On orbital timescales, changes in the intermediate-todeep vertical δ 13 C gradient closely match atmospheric CO 2 , with weaker vertical δ 13 C gradients corresponding to higher CO 2 levels Flower et al, 2000;Hodell et al, 2003b;Köhler et al, 2010;Lisiecki, 2010). This relationship suggests that many of the processes affecting CO 2 also alter the vertical δ 13 C gradient.…”

“…S1, Table A2). Additionally, we evaluate an alternate gradient, δ 13 C (INA/2)−DP , defined as the difference between half the intermediate North Atlantic stack and the deep Pacific stack; Lisiecki (2010) found that this gradient optimized correlation with CO 2 from 0-800 ka.…”

“…Changes in the vertical δ 13 C gradient should closely resemble time series of atmospheric CO 2 if the deglacial CO 2 increase is caused by a decrease in deep ocean carbon storage. This hypothesis is supported by findings on orbital timescales using a smaller number of sites Flower et al, 2000;Hodell et al, 2003b;Lisiecki, 2010), but the link between the vertical δ 13 C gradient and CO 2 has not yet been evaluated at millennial timescales or using a global data compilation. Observing such a link would improve our understanding of deglacial atmospheric CO 2 increase and, furthermore, demonstrate that the data compilation presented here has adequate spatial and temporal resolution with sufficiently precise age models to reconstruct millennial-scale changes in the global carbon cycle.…”

“…Averages of benthic foraminiferal δ 13 C time series, called stacks, can improve the signal-to-noise ratio of regional or global seawater changes (e.g., Lisiecki et al, 2008;. Global mean benthic δ 13 C change is likely caused by changes in terrestrial organic carbon storage (Shackleton, 1977;Curry et al, 1988;Duplessy et al, 1988;Ciais et al, 2012;Peterson et al, 2014), while vertical δ 13 C gradients may record changes in deep ocean carbon storage and atmospheric CO 2 Flower et al, 2000;Hodell et al, 2003b;Lisiecki, 2010). The vertical δ 13 C gradient between the surface (high δ 13 C) and deep ocean (low δ 13 C) primarily results from the accumulation of low-δ 13 Crespired organic carbon in deep water, which temporarily sequesters it from the atmosphere.…”