For the calculation of bottom protection in harbours near quay-walls, analytical formulas are used. The current experiences suggest an overestimation of load on the harbour bottom. Extensive measurements in prototype situations are very expensive. For this reason, the use of a computer model will be far more economical. However, at this moment there is no expertise in the use of numerical models for this type of situation. For this reason, a physical model was made to validate the use of the numerical model. The results of the measurements of Nielsen were not conclusive, because the two models did not match. A sensitivity analysis of the numerical model showed that the model is sensitive to changes in the geometry. Especially the changes to the geometry of the bowthruster caused large differences in the calculation results of the model. For the thruster outflow to be modelled accurately, measurements needed to be done in the physical model. The measurements revealed an unexpected aspect of the flow situation. The low velocity core behind the screw axis was expected to collapse within a certain distance. The measurement results show the core to exist beyond this distance and its contribution to the distribution of the velocities from the bowthruster. Even though this raises new questions, the adaptations to the bowthruster in the numerical model, reproduced the outflow of the physical model. The calculation results of this model were very comparable to the flow situation measured in the physical model. This gives some confidence for the use of the ke-model for modelling bowthruster-induced flows. It also shows the importance of modelling the bowthruster. When the outflow of the bowthruster does not match reality, large deviations are expected.