Abstract An integrated field data-modelling approach is employed to investigate relationships between the wavelength of tidal sand waves and four environmental parameters: tidal current amplitude, water depth, tidal ellipticity and median grain size. From echo sounder data at 23 locations on the Dutch continental shelf, the average wavelengths of observed sand waves are determined and compared with the wavelengths obtained with a process-based model. The latter describes the initial formation of these bedforms due to feedbacks between the tidal current and the erodible bed and uses environmental parameters for the 23 locations as input. Good agreement between observed and modelled wavelengths is found if the bottom stress experienced by tidal currents is adequately quantified. Model results show that the wavelength of sand waves increases with increasing water depth, tidal ellipticity and grain size (coarse sand), whilst it decreases with increasing tidal current amplitude and grain size (fine sand). Due to the limited number of stations and the fact that all four parameters change from location to location, the modelled relationships are only partly supported by the field observations.