Understanding Gravel Beach Dynamics by Measurmg and Modelling Annual, Seasonal and Storm-Induced Morphological Response

The aim of this study is to gain a better understanding of the medium-term (weeks to years) morphological evolution of gravel beaches. In particular, this project aims to better understand and parameterise gravel beach/barrier evolution processes at storm, seasonal and annual time-scales, For this purpose, the morphological variabihty of the Slapton Sands macrotidal gravel beach/barrier (Devon, UK) was measured over a 3-year period using an Argus video-system, and carrying fortnightly and event-driven RTK-GPS field surveys. Qoncurrent nearshore hydrodynamic measurements were collected by a directional wave buoy located 2km off Slapton Sands, and tidal data were obtained from a tidal model specifically developed for the field site. The combination of the morphological and hydrodynamic measurements provided a unique dataset to achieve the central objective of this study. In the past few years, storms have caused significant barrier crest erosion at Slapton Sands. This study showed that the morphological evolution of the barrier was sensitive to the wave direction. The storms affecting the barrier are either southerly swell or easterly wind wave driven. Easterly storms induce supratidal erosion and intertidal accretion, and a significant gain in overall beach volume. During the intervening calm periods, the beach volume increases, and most of the sediment input is used to construct a berm feature in the supratidal zone. Therefore, the beach gains sediment during cross-shore dominance induced by a net onshore sediment transport. In contrast, the beach loses sediment during longshore dominance, which occurs during southerly storms that cause accretion of the supratidal zone and erosion of the intertidal beach. The higher frequency of one storm type over the other results in an imbalance between the direction of the net alongshore sediment transport, which can result in beach rotation. Thus, the relative dominance of the two opposing storm types has sttong control on the morphological evolution and stability of the barrier. The Argus video-system was found an appropriate technique to obtain shoreUne measurements on a gravel beach, with an associated average root mean square error (RMSE) of less than 2m over 2.5 km of beach, and equivalent to deviations of 5% and 10% of the measured maximum and mean shoreline variability. The video-system was also used to obtain wave height estimates with typical errors of ± 0 15 m. Video-derived shorelines were used to study wave-shoreline dynamics, and obtain shoreHne predictions from known forcing through the application of the Canomcal Correlation Analysis data-driven statistical technique. The CCA-derived shorelines accounted for a RMSE of less than 3 m over 16 weeks of predictions, equivalent to 15% of the mean shoreline variability. Thus, the CCA was found of practical use to obtain shorelines within the range of vanability of the input data, and provide assessment on shoreKne evolution in response to a projected wave climate. Wave modelling undertaken using the MIKE 21 SW flexible mesh spectral model revealed that easterly and southerly storms generate opposing sediment transport patterns along the barrier, northward transport occurs during southerly storms, and southwards transport during easterly storms. Wave modelling outputs were used to calculate longshore sediment transport (LST) rates from three equations previously applied to gravel. The CERC (1984) equation, not being site-specific, provided reasonable results using a k value of 0.054, which is an order of magnitude lower that the defined for sandy beaches, and agrees with the originally defined for a grain size of 20 mm. This project has assessed, for the first time, the performance of the XBeach morphodynamic model to simulate storm-induced cross-shore morphological change on a gravel beach/barrier system. The morphological response of the beach was best modelled using a drag coefficient CD of 0.007, and a hydrauHc conductivity KofO 05 ms"^. Model results highlight the relevance of considering groundwater effects when modelling gravel beach dynamics, in order to be able to account for an accurate morphological change. Moreover, model outputs indicated the importance of the position of the lagoon water table m relation to the sea level, which determines the amount of beachface morphological vanability, and barrier stability. At present, the model is unable of reproducing the formation of a berm, thus, beach recovery conditions cannot be modelled. This is attnbuted to the fact that XBeach models long group waves rather than individual waves, which fails to account for the sediment transport of individual swashes capable of reaching the berm crest, and accreting it. In contrast, the model can effectively provide good estimates of post-storm profiles, and it can determine the threshold for overwash occurrence across a gravel beach/barrier