Deep-water formation in the Labrador Sea is simulated with the Finite-Element Sea-Ice Ocean Model (FESOM) in a regionally focused, but global covered model setup. The model has a regional resolution of up to 7km. Our simulations cover the time period 1958-2007. We evaluate the capability of the model setup to reproduce a realistic deep water formation in the Labrador Sea. Two classes of Labrador Sea Water (LSW) are analysed and compared to LSW layer thicknesses derived from observations in the formation region for the time interval 1988-2007. It is shown that the model is able to reproduce four phases in the temporal evolution of the potential density, temperature and salinity, since the late 1980s, which are known in observational data. These four phases are characterized by a significantly different LSW formation. The first phase is characterized in the model by a fast increase in the the convection depth of up to 2000m, accompanied by an increased Spring production of deep Labrador Sea Water (dLSW). In the second phase, the dLSW layer thickness remains on a high level for several years, while the third phase features a gradual decrease in the deep ventilation and the renewal of the deep ocean layers. The fourth phase features an almost constant dLSW layer thickness on a reduced level. By applying a Composite Map Analysis between an index of dLSW and sea level pressure over the entire simulation period from 1958-2007, it is shown that a pattern which resembles the structure of the North Atlantic Oscillation (NAO) is one of the main triggers for the variability of LSW formation. Our model results indicate that the process of dLSW formation can act as a low-pass filter to the atmospheric forcing, so that only persistent NAO events correlate with the dLSW index. Based on composite maps of the thermal and haline contributions to the surface density flux we can prove that the central Labrador Sea in the model is dominated by the thermal contributions of the surface density flux, while the haline contributions are limited to the branch of the Labrador Sea boundary current system (LSBCS), where they are dominated from the haline contributions of sea ice melting and formation. Our model results feature a shielding of the central Labrador Sea from the haline contributions by the LSBCS, which only allows a minor haline interaction with the central Labrador Sea by lateral mixing. Based on the comparison of the simulated and measured LSW layer thicknesses as well as vertical profiles of potential density, temperature and salinity we show that the FESOM model is a suitable tool to reproduce the regional dynamics of the LSW formation in a global covered context.