Ocean velocities as inferred from Argo floatsmethodology and applications

Laburpena

The velocity fields inferred from the trajectories of Argo floats have proven to be a valuable tool for observing the major ocean currents. The velocity fields can be used to examine statistical properties at regional and global scales, and to describe regional and global circulation patterns at temporal scales that range from months to years. The first application has been the statistical study of ocean velocities, in particular the analysis of the Gaussian character of the velocity PDFs. The Argo-inferred velocities show to be adequate for this analysis as the inferred PDFs are comparable to previous results in the literature. We observe a deviation of the PDFs from Gaussianity, but yet we find no conclusive arguments to state that these PDFs are not Gaussian at a local scale. Our results emphasize the dependence of the resulting PDFs on the selection of both temporal and spatial sampling intervals which likely depend on the regional dynamics. In order to integrate relatively large temporal and spatial regions, a proper local normalization is previously required. This is a delicate process which, if incorrectly done, may lead to residual lateral and temporal inhomogeneities. We end up developing a simple model that shows to be quite useful at reproducing the velocity PDFs. The second application is an exhaustive study of the circulation of the equatorial Atlantic Ocean at several depths. It shows the great potential of the methodology, both to obtain the annual-mean ocean currents at several depths and to unveil the seasonal variations at the sea surface and 1000 dbar. In particular, we calculate monthly velocity fields which are quite different from the annual averaged fields. We have examined the intricate coupling of the surface winds and the surface currents. The intensity of the surface easterlies is associated to the annual latitudinal displacement of the ITCZ, in most of the Atlantic taking place between about 2ºN and 10ºN. A harmonic analysis also helps to understand the seasonal cycles both at surface and depth. The way the NECC grows in intensity across the Atlantic, from east to west, and the novel view of the NBC retroflection as the westernmost path for the recirculating nSEC, are two remarkable outputs. The five jets that appear in the annual-mean velocity field at 1000 dbar are the result of the composition of three alternating jets that reverse through the year while changing their latitudinal position. Two extra-equatorial jets, the NIEC and the SIEC, develop on both sides and flow opposite to the EIC. We show that some of these inversions are likely the result of westward propagating waves. The third, and final, application dealt with the surface and intermediate currents around Australia. The velocity data is complemented with the CTD data contained in the profiles and a novel method is developed to estimate the water transport in the intermediate layer. Our results confirm the existence of westward transport south of Tasmania and along the Great Australian Bight, from the Pacific Ocean to the Indian Ocean, through the Tasman Leakage. We obtain the first actual measurements of water transport, in rough agreement with previous model estimates: 3.8 ± 1.3 Sv with 39% contribution through eddy transport. We also show the existence of substantial mesoscale variability. This represents a confirmation of the existence of the Southern Ocean Supergyre, previously suggested by models and hypothesized from oceanographic cruise data. The dissertation is complemented with several appendices. In appendix A we propose that ocean velocities have a Gaussian distribution, which maximizes entropy. Appendix B presents some thoughts on the scientific method, which arose during the investigation process. Appendix C presents the basics of the harmonic analysis used in chapter 4. Finally, appendix D is found in a complementary CD, which contains a version of the basic scripts used during the research.