The ocean surface boundary layer (OSBL) is one of the largest global regulators of climate. There are two definitions associated with its depth: the mixed layer depth (MLD) and the mixing layer depth (XLD). The former is estimated using vertical profiles of temperature and density, and the latter using measurements of the dissipation rate of turbulent kinetic energy. The relationship between the XLD and MLD has not been extensively studied in the past, since this requires specialized instruments which allow simultaneous measurements of the ocean state variables and turbulence across the full depth of the OSBL. The overarching objective of this thesis is to study the variability of the XLD and MLD under different conditions. To achieve this, different estimates of the MLD and XLD are applied to vertical profiles derived from an autonomous vertical profiler (ASIP) deployed in two different cruises in the North Atlantic. It was found that the MLD criteria differ between themselves when stratification is weaker and the transition between the mixed layer and the pycnocline is not well defined. On the contrary, the XLD criteria had a better agreement between themselves, and a new method was proposed to improve its estimation. Given the close link found between the XLD and the sources of ocean turbulence, a scaling for the XLD was tested dividing the ocean into different mixing regimes according to the ratio between the Monin-Obukhov length, the MLD, and the surface net heat fluxes. The results found using ASIP data were extrapolated to build a climatology of the upper mixing using reanalysis data from ERA5 and ARGO profiles during a period of 10 years (2009-2018). Significant differences between the MLD and the XLD climatology were found during this process.