Mid-Late Cretaceous Nd Isotopic Composition of Southern Ocean Deep Waters

Little is known about ocean circulation during the middle to Late Cretaceous greenhouse world. It is widely recognized that during the majority of the Cretaceous, waters at the seafloor were well oxygenated implying consistent ventilation of deep waters. However the mode of deep ocean ventilation is not known and significant debate continues as to whether deep-water circulation operated in a mode similar to today (high latitude convection), or whether warm saline waters convected in low latitudes. These two circulation patterns would have different effects on the planetary heat budget. Thus identifying the location of deep-water formation is crucial to understanding oceanic heat transport during the Late Cretaceous greenhouse world. Here we use Nd isotopes to define the composition of surface, intermediate, and deep-waters as a first step toward identifying regions of water mass formation. We present new Nd isotopic records (age-corrected 143Nd/144Nd expressed as ɛNd(t)) and Sr records of fish teeth and bones from seven Deep Sea Drilling Project (DSDP) and Ocean Drilling Project (ODP) holes (361, 511, 530A, 690C, 763B, 765C, 766A) within the Atlantic and Indian sectors of the Southern Ocean spanning the mid to late Cretaceous (Cenomanian-Campanian). South Atlantic sector (Sites 361 and 511) ɛNd(t) values gradually decrease from ~ -4 during the Cenomanian to -7 in the Campanian. Hole 530A, located north of Walvis Ridge, likely reflects a separate water mass from that in the Atlantic sector of the Southern Ocean, as the ɛNd(t) values exhibit a different trend (increasing from -7 during the Cenomanian to -2 during the Santonian and then decreasing to -11 by the Campanian). Sites 763, 766, and 765 form a depth transect along the Exmouth Plateau at paleo-depths of 1000, 3000, and 4000 meters and these data display a similar trend as 530A, becoming more radiogenic during the late Turonian (ɛNd(t) ~ -8 in 763B, ~ -5 in 766A and 765C) before returning to slightly more unradiogenic values in the Campanian (ɛNd(t) ~ -11.5 in 763B, ~ -10 in 766A and 765C) as they had in the Cenomanian (ɛNd(t) ~ -10.5 in 763B, ~ -8 in 766A and 765C). The similar timing and trends between the shallower 763B and the deeper sites suggests that a single water mass bathed the entire depth range, with the shallower depths influenced by mixing with more unradiogenic surface waters. The similarity between Site 530 and the Exmouth transect is likely coincidence rather than circulation driven due to the numerous sills isolating Site 530 from the Indian sector of the Southern Ocean, implying numerous regional/local sites of convection/ventilation rather than basin wide circulation. The more radiogenic values and different ɛNd(t) trends at the geographically intervening Sites 511 and 361 corroborate this assertion.