Understanding the impact of offshore wind farms on marine fauna is vital for sustainable development of this renewable energy resource. This paper presents an application to real data of a new simulation-based impact assessment method that was developed using artificial data. The method simulates surveys of seabird counts at post-construction survey locations using knowledge obtained from the undisturbed pre-construction phase. Next, using hypothesis-testing it is investigated whether the actually collected post-construction counts are statistically different from the set of simulated surveys of the undisturbed situation. We investigate the applicability of this method in a real case using a dataset that was collected to assess the impact of an offshore wind farm in the North Sea on the seabird species Guillemots (Uria aalge). It appeared that several elaborations of the method were needed to accommodate the properties of the dataset at hand. These elaborations included the formulation of an appropriate deterministic model to accommodate for zero-inflation in the species data, transformations of the residual data to be appropriate for the stochastic modelling part, and accommodation of directional spatial correlation in the residuals.