Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1990

Description: As a primary feature of ocean circulation and a key component of the global carbon cycle, changes in the thermocline must be accounted for if we are to understand the processes involved in Quaternary climatic fluctuation. Toward this goal, this thesis contains studies of the modern and glacial thermoclines at the Bahama Banks and it presents an novel approach to determine sea level based on the flux of 230Th and 231Pa from thermocline waters to the seafloor. In the first chapter, the hydrography of the modern thermocline in Northwest and Northeast Providence Channels, Bahamas, is investigated using CTD data. Potential temperature- salinity relationships demonstrate that the deep waters and most of the thermocline waters in these channels originates in the Sargasso Sea. Cross channel sections of water properties suggest the following: (1) water from the shallow core of the Deep Western Boundary Current (Fine and Molinari, 1988) may circulate along the channel margins, and (2) where the western end of Northwest Providence Channel opens to the Florida Straits, shallow flow is toward the straits in the southern portion of the channel and away from the straits in the northern portion. In the next two chapters, changes in the temperature and nutrient structures of the thermocline from the last glaciation to the recent Holocene are inferred from isotopic variations of the planktonic foraminifera Globigerinoides ruber (212-250μm) and G. sacculifer (300-350μm) and the benthic foraminifera Planulina wuellerstorfi, P. ariminensis, P. foveolata and Cibicidoides pachyderma (〉250μm) in a suite of cores from the margins of Little and Great Bahama Banks. During the last glaciation, δ18O values were from 1.4 to 1.9 per mil greater than during the recent Holocene. Based on the δ18O/sea-level model of Fairbanks (1989), we estimate that the upper 1500 m of the water column was cooler by at least 1°C- the deepest waters were several degrees cooler. The temperature gradient (dT/dz) was steeper and the base of the thermocline appears to have stayed at about the same depth or risen slightly. At all depths in the thermocline, δ13C was greater during the last glaciation than during the recent Holocene by at least 0.1-0.2 per mil and as much as 0.6 per mil in the lower thermocline. While recent Holocene δ13C reaches minimum values in the lower thermocline (the poorly-ventilated oxygen minimum/phosphate maximum layer), this feature was not present during the last glaciation. These data show that the concentrations of nutrients throughout the thermocline were reduced and that there was no oxygen minimum layer, indicating greater, more uniform ventilation of thermocline waters. These results are consistent with our understanding of the physics of thermocline circulation and evidence for hydrographic conditions at the ocean surface during the last glaciation, indicating a direct response of thermocline circulation to changes in climate. Cooler thermocline waters reflect cooler surface ocean temperatures at mid-latitudes where thermocline isopycnal surfaces outcrop. Increased, more uniform ventilation of the glacial thermocline is consistent with both more vigorous glacial winds leading to increased Ekman pumping and all isopycnal surfaces of the thermocline outcropping in the area of Ekman downwelling. Taken together with previous studies of intermediate-depth waters, these data document that the entire upper water column of the North Atlantic was depleted in nutrients during the last glaciation. A final suggestion of the third study is that Mediterranean and southern source waters contributed little to deeper intermediate-depth waters in the North Atlantic. The fourth chapter presents two new approaches to reconstruct the sea-level history based on the fluxes of 230Th and 231Pa to the seafloor. The approaches rely on the fact that fluxes of these nuclides to a site on the seafloor are proportional to the height of the water column above the site. Consequently, a change in sea level causes changes in the 230Th and 231Pa fluxes which, at shallow sites, are large fractions of the total fluxes. Past sea level can be reconstructed using either the record of nuclide accumulation in a single core of sediment, or nuclide concentrations in synchronously deposited sediment samples from cores collected over a range of water depths. Importantly, this record of sea level is both continuous (not just high stands) and independent of the assumptions of constant seawater temperature or uplift rate required by some other approaches.

Description: This work was supported by NSF grant number OCE-8813307 to W. Curry, NSF grant number OCE-8800693 to J. Broda and W. Curry and the Education Office of the MIT/WHOI Joint Program in Oceanography.