Past and present ocean dynamics in the western subtropical Atlantic

Abstract

The main physical processes responsible for the past and present Loop Current variability in the Gulf of Mexico and its interconnection with both, the Caribbean Sea and the Florida Straits are investigated in this PhD thesis. The aim is to contribute to the understanding of ocean dynamics in the western part of the North Atlantic across the prominent transition from the Last Glacial Maximum to the Holocene during the last 25 kyr (kilo years). A combined approach using high resolution models, present day observations and paleo-proxies has helped to explore the past and the present spatial and temporal changes of Loop Current dynamics and to understand the relative contributions of different external forcing factors. This PhD thesis consists of three research papers:

The first part of this thesis addresses the influence of atmospheric and internal ocean variability on the Loop Current and the associated Florida Straits transport on interannual to decadal scales. A clear relationship is found between different stages within a ring shedding cycle of the Loop Current in the Gulf of Mexico and transport minima in the Florida Current transport, both in observations and in model simulations. It is demonstrated that transport changes in Florida Straits have a significant influence on the transport variability on monthly to decadal time scales. Differences (and changes) between the ring shedding period and seasonal cycle lead to an interannual to decadal beat frequency, which explains large parts of the variability of the Florida Current transport in the model simulations, even exceeding atmospheric forcing variability on the considered time scales. Although additional trigger events might support the ring shedding process, the Florida Straits transport is influenced mainly by internal dynamics. The second part focuses on the influence of the Loop Current eddy shedding on the heat budget of the Gulf of Mexico at changing sea levels, different wind stress forcings and topographic effects. The model simulations imply that the process of eddy shedding was most likely absent during the Last Glacial Maximum at lowered sea level and modified Yucatan Strait topography. Subsequently, eddy shedding increases gradually across the deglaciation thereby warming the Gulf of Mexico. In support, paleoceanographic proxy data reveal a continuous sea surface temperature increase in the northern Gulf. Although little is known about the glacial atmosphere, a strengthened atmospheric circulation is assumed for the LGM with a shift in the ITCZ position towards the south. As a consequence, glacial wind stress causes enhanced Sverdrup transport within the Subtropical Gyre thus leading to a strengthened Florida Straits and Yucatan Strait through-flow. Eddy shedding decreases the stronger the transport is. Paleoceanographic proxy data are ambiguous in this respect. Although atmospheric circulation was stronger during the Last Glacial Maximum due to the enhanced meridional temperature gradients, paleoceanographic reconstructions reveal both, negative and positive sign in Florida Straits transport for the Last Glacial Maximum.

Finally, in the third part of this thesis the glacial position of the Gulf Stream is discussed. While the lowered sea level by itself does not lead to significant changes in the current system of the North Atlantic, the combination with glacial wind stress forcing does. The changes in the glacial atmospheric circulation leads to the geographical expansion of the northern recirculation gyre towards the south with a subsequent increase in the Ekman pumping within the gyre. Therefore, the line of zero wind stress curl and the Gulf Stream are shifted southwards. The subtropical gyre, hence, is weaker during the LGM. Paleo-evidence is unfortunately scarce mainly due to the highly variable and strong current regime in this region preventing a good conservation of sediment records. Nevertheless, paleoceanographic proxy reconstructions reveal a shift of the polar front due to the large continental ice sheets and the changes in wind supporting the results of this study. Other paleo-observations thus suggests only a slight shift of the present Gulf Stream position during the LGM which contradicts the simulated response to changes in wind forcing. Overall, the separation process off the coast itself is very sensitive to a variety of factors and therefore, a combined effect together with the influence of the atmospheric circulation is also conceivable.