Mesoscale and submesoscale dynamics and their role on the transport and mixing of ocean biogeochemical tracers have been investigated widely in recent literatures. In general, large circulations and mesoscale eddies control the absolute dispersion of tracers horizontally, while submesoscale motions affect horizontal mixing and, with their strong vertical velocities, control upwelling or subduction processes in the boundary layers, at the ocean surface and near its bottom.

The exponential growth of climate data combined with advances in machine learningoffers new opportunities to understand the climate system and its response to external forcings. This thesis explores and proposes data mining frameworks to reduce the complexityof spatiotemporal climate fields and facilitate analysis and interpretation.As complex as it appears, the dynamics of the climate system is dominated by spatiotemporal patterns and the identification of these patterns and their linkages offers a useful framework for dimensionality reduction.