Dynamic Modelling of Rip Currents for Swimmer Safety on a Wind-Sea–Dominated Mesotidal Beach

Abstract

Sembiring, L.; van Dongeren, A.; Winter, G., and Roelvink, D., 0000. Dynamic modelling of rip currents for swimmer safety on a wind-sea-dominated mesotidal beach. Journal of Coastal Research, 00(0), 000-000. Coconut Creek (Florida), ISSN 0749-0208.We apply the process-based model XBeach to predict bathymetrically controlled rip current flow occurrence, velocities and duration in the wind-sea-dominated environment of Egmond aan Zee, The Netherlands. The model is validated against data obtained during a 5-day field experiment with GPS-equipped floating drifters. Numerical model results show good agreement with data and simulate flow patterns that resemble the offshore flow through the rip channel and subsequent alongshore drift of the drifters. The average relative error of the drifter position is 0.15 and 0.19 for the crossshore and the alongshore position, respectively. Results from the numerical experiments suggest that the rip current intensity is tidally modulated. For the period of analysis, the rips are initiated approximately 5 hours before low tide, reaching a maximum speed during low tide, and becoming inactive 3 hours after low tide. Initiation and duration of the rip current is correlated to a ratio of~0.55 between offshore wave height and water depth over the bar. Further experiments with the numerical model show that at Egmond, drifters that are transported offshore by the rip current are unlikely to circulate back to the shore (with only a 14% return rate). Instead, they are ejected from the surf zone and advected alongshore, dependent on the direction of the tidal current. Without wave group forcing in the model (i.e. with a stationary wave resolving approach), the model's predictability decreases, which mostly affects the prediction of the spatial flow patterns. The rip current durations can still be predicted very well, whereas the rip current magnitude is underestimated. With wind stress forcing, the model's predictive skill increases for the alongshore flow when the wind speed is relatively high, but this is not the case for low wind conditions.