Fig. 5
Link between the 400-kyr phase modulation and the global ocean carbon cycle. a DF ΔTsite (red) and the composite atmospheric CO2 record32 (green). b The difference between the normalized obliquity component of ΔTsite and AMI (same as in Fig. 4g). c The intermediate-to-deep δ13C gradient of benthic foraminifera (Δδ13C), a proxy for deep-water ventilation in the Southern Ocean49. Δδ13C represents the difference in δ13C between two cores, specifically those collected at ODP site 1088 (ca. 2100 m water depth) and site 1090 (ca. 3702 m water depth) in the South Atlantic Ocean. Dotted red line indicates the Δδ13C obtained using C. kullenbergi data at site 1090 and solid red line indicates C. wuellerstorfi data only. d A measure of overturning strength based on δ13C gradient50 between the North Atlantic (ODP site 607)91 and the tropical Pacific Ocean (ODP site 846)92. Small gradients (close to zero) imply drastically reduced ventilation of the Atlantic or enhanced ventilation in the Pacific50. e Sr/Ca of coccolithophores as a proxy for the rate of coccolithophore production in southwestern Indian Ocean (purple line, MD962077) and in the western equatorial Pacific (blue line, V28-239)45. f Mass accumulation rate (MAR) of iron in the subantarctic Atlantic Ocean (ODP site 1090)48. Dashed-dotted black line indicates the long-term trend obtained using a Gaussian filter around 0 kyr−1 with a 0.0045 kyr−1 bandwidth (i.e with periodicity < 222 kyr removed). g Orbital eccentricity (black) and precession (yellow)
