Abstract:
Near estuaries and harbours, submerged shoals and defence structures impact the exposure to overtopping. These
features may be accounted for in two-dimensional horizontal (2DH) numerical models, either based on phase resolving or phase-average solvers for wave propagation. In between, surfbeat solvers, such as in XBeach,
combine an affordable computational cost with the ability to generate and propagate the longer infragravity (IG)
waves. However, surfzone wave characteristics and the overtopping exposure modelled with XBeach are sensi tive to settings such as the shape of the forcing wave spectra and the numerical scheme for wave propagation.
The present paper explores this sensitivity and assesses the performance of different inundation models built with
the 2DH surfbeat solver of XBeach. These models were forced with downscaled water levels and directional wave
spectra and the results fuelled a discussion bounded by data collected downdrift of the entrance to the harbour of
Figueira da Foz (Portugal). The original second-order upwind scheme, which propagates short-waves with a
lower numerical diffusion improved the model performance in terms of long-wave height, and an unconventional
breaking criterion better represented the cross-shore distribution of short-wave height near the shore. A cali brated model was validated through the hindcast of an overtopping event observed under moderate swell
forcing, and was used to map the overtopping exposure during a hypothetical combination of an energetic swell
with a water level having a return period of ~70 years. Compared to the default wave spectra shape and model
settings, using an appropriate representation of the short-wave directional spectrum at the open boundary was
necessary to reproduce the observed overtopping extent. Refining the cross-shore resolution of the model helped
to better represent the observed inundation extents, as also did the reduction of the friction coefficient. Addi tional phase-resolving simulations in 1DH overestimated IG wave energy and produced higher and more frequent
overtopping discharges. The differences with the calibrated 2DH surfbeat model increased with the proximity of
the inlet and with short-wave height and angle of incidence. Overall, required calibration steps were provided.
They aim at making 2DH XBeach surfbeat a credible tool for 1) predicting short and IG wave characteristics up to
the shoreline, as well as for 2) providing first estimates of exposure to overtopping in areas with shallow and
alongshore-irregular morphologies.
