For several years the large-scale mining of sand from the Dutch Sector of the North sea is in discussion related to the need of sand for shoreface, beach and dune nourishment and large-scale engineering works at sea (Maasvlakte extension, airport at sea). The mining methods considered, basically fall into two categories: wide, shallow or small, deep mining pits. Presently, most sand mining pits with a limited depth, not deeper than about 2 m, are excavated beyond the 20 m depth contour. Deep mining pits have not yet been made extensively. As morphological models are primarily used to assess the impact of sand mining pits it is essential to have a good insight in the quality of the predictions made with these models. In the present study the Delft3D model is evaluated with measurements from a sand mining pit located some 10 km offshore of the Rotterdam harbour entrance. The evaluation involves a comparison with measured water levels and current velocities in the pit and the surrounding area. The evaluation study has shown that the Delft3D-model is capable of reproducing the measurements with reasonable to good accuracy. However, the agreement did vary in the two periods that were considered. In the first period, around neap tide with relatively high waves and wind, occasionally large deviations between the predictions and measurements were observed. The second period, around spring tide with low waves and wind, was reproduced accurately. In the 2DH-simulations the effect of waves, wind and salinity was limited. The tidal forcing appeared to be dominant at the investigated pit. Comparison with the velocity profiles over the vertical showed that the 3D-model was able to represent the 3D-structure of the currents with good accuracy. The morphodynamic evaluation, based on a 2DH simulation with representative tidal, wind and wave forcing, showed a reasonable agreement with the sedimentation-erosion patterns derived from the bathymetric surveys. As the measured morphological development has considerable uncertainties an unambiguous conclusion regarding the morphological predictive capabilities of the Delft3D could not be drawn. The morphodynamic sensitivity analysis revealed some differences between the results obtained in 2DH and 3D-mode. In general the morphological changes were larger in 3D. At this time we can not assess the quality of these predictions due to the lack of a reliable measured morphological development. The differences between the 2DH and 3D morphodynamic simulations on the considered time scale of one year are limited. However, more research is required to investigate 3D effects, especially on longer time scales.