Intertidal beach slope predictions compared to field data

This paper presents a test of a very simple model for predicting beach slop e changes. The model assumes that these changes are a function of both the incident wave conditions and the beach slope itself. Following other studie s. we hypothesized that the beach slope evolves towards an equilibrium valu e that depends nonlinearly on wave steepness (H/L). The rate of beach slope response is assumed to depend on both the degree of slope disequilibrium a nd on the incident wave energy. The model was tested against daily beach sl ope observations derived from digital images of the nearshore zone. Approxi mately, 10(4) images were analyzed over eight, mostly consecutive, month-lo ng periods along a 2 km length of beach. The slope change model was calibra ted by fitting it to the daily differences in the alongshore-averaged slope s, which were obtained from a. 500 m (alongshore) subset of the observation s. An equilibrium slope prediction proportional to the wave steepness (H/L) raised to the -1st to -2nd power performed best compared to several altern ative models. The response rate of beach slope changes depended on the wave height, raised to the 3rd or 4th power. A characteristic response time for the system was found to be 1-2 days. The calibrated (i.e. hindcast) model explained 30-40% of the observed slope change variance, indicating that the model was consistent with the data. However, when the model was used to pr edict the evolution of the beach slept: time series (i.e. to forecast), the prediction error variance was equal to or only slightly lower than the obs erved temporal variability in the slopes. The present model is sufficiently accurate to characterize beach slope dynamics, but its predictive capabili ty would not outperform a model that predicts a constant, mean slope. (C) 2 001 Elsevier Science B.V. All rights reserved.