Prediction of cross-shore beach profile evolution using a diffusion type model

A medium term morphodynamic prediction model for beach profiles, based on a 1-D diffusion formulation, is demonstrated here. The model combines an inverse methodology with a data-driven approach to derive unknown key parameters in the model governing equation. The field site used to demonstrate the model is Milford-on-Sea beach located within the Christchurch Bay beach system on the south coast of the UK, where historic measurements of cross-shore beach profile surveys and incident waves have been recorded over two decades. Despite the simplicity of the modelling approach, the model gives encouraging predictions of cross-shore beach profile changes at Milford-on-Sea. The predictive ability of the model is tested by forecasting measured beach changes on the basis of parameter calibrations performed on an independent set of measurements. Comparisons are also made against the results of a purely data-driven technique. In both cases the new method shows measureable improvements.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► Medium term morphodynamic is very important for beach management. ► Simple beach profile models provide useful information of medium term beach change. ► A novel beach profile model was developed using a hybrid approach. ► The model gives successful predictions of medium term beach profile change. ► The performs better than a direct data-driven approach.