Computer modelling serves a number of purposes and plays an important role in scientific research. Examples are interpolation between or extrapolation from existing observations or the dynamics of a system and how it is influenced by changes to the parameters or the boundary conditions.
Things should be made as simple as possible, but not any simpler than that, to paraphrase Albert Einstein. Our approach to modelling involves these few, but fundamental elements, essential when establishing a sound, reliable model:
When developing a model to be used by others, collaborators or clients, we furthermore value these properties:
In contrast, large computer models that build on generalized descriptions of the governing equations, including Newton's second law, may be hard to have confidence in. Such models are often assumed to be valid simply if model results agree with observations. However, since these models usually include a large number of parameters, they may easily be fiddled with to produce almost any opportune result.
A prominent example of what we do and offer is the Ringkøbing Fjord model. This was developed in order to find the optimal management of the sluice in Hvide Sande with respect to water level, salinity and stratification, having an impact on, among other things, the risk of both oxygen depletion in the fjord and flooding of low-lying areas around the fjord. The model was implemented to run in a widely available spreadsheet program, such as Microsoft Excel, and allowed the user to determine the impact of a given sluice usage on her or his own.
Subsequently the model was modified for Nissum Fjord, a more complex, three-basin fjord, which is subject to similar problems. The Ringkøbing Fjord model is described in Nielsen et al., Estuarine, Coastal and Shelf Science, vol. 63, pp. 235-248, 2005.