During berthing operations vessels use their bow thruster(s) to improve their manoeuvrability, making them less dependent on the assistance of tugboats. The jet from a bow thruster reflects on the quay wall. It is directed towards the bottom where it reflects, causing high flow velocities over the bed. This may scour the nearby bed when it is left unprotected, leading to instability of the quay wall. Over the years, the shipping industry has been developing continuously, characterized primarily by the upscaling in size of inland- and sea-going vessels. As a result, vessels have more power and larger thruster diameters leading to higher hydraulic loads on quay walls and bed protections of berthing facilities. The most common type of bed protection is rip-rap (partially) penetrated with concrete. However, due to the complex flow field of the reflected jet, the decay profile of the near-bed flow velocities is unknown. This results in uncertainties in the design of bed protections and the required width of these protections that must be penetrated with concrete. In this research, the decay of the near-bed flow velocity in perpendicular direction to the quay wall induced by a 4-channel bow thruster is researched. The eventual goal is to provide a better indication to what extent the bed protection must be penetrated with concrete. Field measurements have been conducted in the North Sea Port of Gent with the Somtrans XXV, one of the largest inland vessels in the Netherlands. The flow velocities near the bed, induced by the bow thruster, have been measured with Acoustic Doppler Velocimeters (ADV), Acoustic Doppler Current Profilers (ADCP) and Ott meters (Ott). The results from the flow velocity measurements have been analysed on three main parameters: influence of the applied bow thruster power, the distance between the measurement instrument frame and the bow thruster and quay wall clearance. The highest flow velocities are measured near the quay wall in the order of 1 m/s reaching up to a maximum of 1.8 m/s. Further away from the quay, the flow rapidly declines towards a more constant level of approximately 0.3-0.4 m/s. To define the maximum load on the bed, the mean flow velocity plus three times the standard deviation is used resulting in a maximum load ranging between 1.6-2.8 times the mean horizontal flow velocity. The Dutch and German guidelines for determining the near-bed flow velocities generally overestimate the measurement results. In addition, the dependency of the Dutch method on the total travelled distance by the jet, based on the sum of the quay wall clearance, the height of the bow thruster above the bed and the distance x from the quay wall, are not reflected in the measurement results. It is recommended that this extensive and unique data set acquired through the field measurements in Gent is further used to analyse the flow field of a reflected jet on a vertical quay wall by validating numerical and scale models. Combining these three different methodologies will contribute to a better understanding of this phenomenon with the eventual goal of optimizing the design of bed protections.