A morphodynamical linear stability analysis is used to predict the natural development of crescentic bed patterns and rip channels. The purpose is to investigate whether this technique, which is useful for understanding the physics of emerging bed-forms, can be used to make quantitative predictions in the field, which may then be of use for coastal engineers. To this end a morphodynamical linear stability model (Morfo60) is used to describe the development of crescentic bed patterns at the coast at the USACE Field Research Facility in Duck, North Carolina, USA. Wave, tide and alongshore-averaged bathymetry data recorded at Duck over a 2 month period in 1998 are used to model the development of these morphodynamical patterns at an open coast. The model predictions are compared with field observations by van Enckevort et al. (2004) made at Duck using the Argus imaging technique, over the same 2 month period. Field observations and model predictions show similar length scales of the crescentic bed patterns. Immediately after a storm, large length scales of around 500 to 800 m are predicted and observed, which in a couple of days decrease to around 150 to 400 m, until the next storm occurs. The model predictions show more fluctuations in the predicted length scale than those observed in the field. These fluctuations are due to variations in the wave conditions and tidal level and the lack of pre-existing bed patterns in a linear stability analysis. An algorithm is developed to identify the more physically significant model predictions based on large growth rates and consistency in length scales. The presented algorithm, referred to as a Physically Significant Development (PSD) algorithm, is able systematically to identify the more physically representative model results. These compare better with field observations, as shown by the good agreement between predicted and observed crescentic bed pattern length scales. Accordingly, the conclusion is that linear stability analysis in combination with an appropriate physically based significant development algorithm may be of use to coastal engineers.