Beaches and coastal environments are dynamic, constantly shaped and reshaped by natural processes and anthropogenic modifications. The morphodynamics and influence of natural and anthropogenic factors of two different coasts at various temporal and spatial scales are discussed. To quantify the performance of several beach nourishment projects at annual temporal and kilometer spatial scales on three adjacent microtidal low-wave energy barrier islands in west-central Florida, a total of 5,200 beach and nearshore-profiles spaced at 300 m were surveyed monthly to bi-monthly from 2006-2010. Beach nourishment performance is most significantly influenced by the interruption of longshore sediment transport by complex tidal-inlet processes. More specifically, the tidal-inlet processes influencing adjacent beach nourishment performance includes longshore transport interruption resulting from divergence induced by wave refraction over an ebb-tidal shoal, flood-tidal currents along the beach, and total littoral blockage by structured inlets. A morphologic indicator of a large longshore transport gradient within the study area is the absence of a nearshore sandbar. These non-barred beaches are characterized by persistent shoreline erosion and were almost exclusively located in areas with a large longshore transport gradient. The more typical beach state along the three barrier islands was one exhibiting a migratory bar and relatively stable shoreline. The presence of a sandbar indicates the dominance of cross-shore processes, with onshore migration during calm wave conditions and offshore migration during energetic wave conditions. The onshore and offshore migration of the sandbar is closely related to non-stormy summer and stormy winter seasonal beach changes, respectively. The morphodynamics of a mixed sand and gravel beach in Delaware were investigated based on 740 beach profiles surveyed almost monthly from 2009 to 2011, 60 sediment cores, and 550 surface sediment samples collected at various alongshore and cross-shore transects. Inter-seasonal temporal scales of storm-induced beach changes and post-storm recovery were examined based on a hurricane, a typical energetic winter storm, and an extremely energetic storm resulting from the rare collision of a hurricane and winter storm ("Nor'Ida") occurring within a 3-month period in 2009. The mixed sand and gravel beaches in Delaware are characterized by monotonically increasing water depths lacking a sandbar under all wave conditions. A distinctive beach cycle was identified consisting of a built-up berm profile and depleted nearly-planar storm profile, with a time-scale related to the frequency and intensity of storm impact and duration of intra-storm recovery instead of simple seasonality. The sedimentological characteristics of the storm deposit associated with Nor'Ida demonstrated substantial cross-shore variation ranging from sandy-gravel and gravelly-sand within the storm swash zone (near the pre-storm dune edge) to well-sorted medium to coarse sand seaward of the storm swash zone, suggesting that storm deposits along mixed beaches demonstrate a variety of sedimentological characteristics. A new dynamic beach cycle model is proposed for the non-barred mixed sand and gravel beach with temporal variability controlled by storm occurrence and inter-storm duration.

Une approche modélisante du potentiel TSL le long des plages du golfe du Lion est proposée sur la base du modèle d’ingénierie côtière Unibest-LT (Deltares). Sont exploités en entrée de modèle des données de houle de 1979 à 2010 (base de données ANEMOC-2 produite par le Cerema à partir d’une approche modélisante également) conjointement à des données topo-bathymétriques haute résolution LiDAR (base de données Litto3D produite conjointement par le SHOM et l’IGN). Le potentiel de TSL annuel est calculé pour 157 profils topo-bathymétriques couvrant l’ensemble des ∿250 km de littoral étudié dans ce travail de thèse.En premier lieu, le potentiel de TSL annuel résultant est d’abord estimé sur la base d’une année de houle type, établie à partir de 31 ans de houle horaire entre 1979 et 2010, puis pour chaque année entre 1979 et 2010. La différence observée avec le potentiel de TSL annuel résultat « type » permet de discrétiser les années en trois situations : anomalie positive, anomalie négative ou proche de la normale. Par ailleurs, un lien est clairement établi entre la variabilité interannuelle identifiée dans les simulations et les tempêtes. En outre, cette analyse interannuelle menée à l’échelle des plages du golfe du Lion amène à affiner les limites des cellules de dérive littorale potentielle, pour lesquelles trois scénarios d’organisation sont établis : habituel, occasionnel, exceptionnel.Enfin, le lien entre potentiel de TSL annuel résultant et variations du rivage est testé, conformément au large corpus bibliographique qui existe à ce sujet en ingénierie côtière. Cependant, les résultats s’avèrent décevants : une remise en cause de la théorie sous-jacente est proposée. / Wave data from 1979 to 2010 derived from a large-scale modelling database (ANEMOC-2) were used together with high resolution topo-bathymetric LiDAR data as inputs within the coastal engineering model Unibest-LT (Uniform Beach Sediment - Longshore Transport). The spatial and temporal coverage of these data offers a unique opportunity to carry out a comprehensive study of potential longshore transport intensity and direction.Over the 250 km-long stretch of coast covering the study area, the longshore drift directions computed with the 30-year mean wave climate closely match those of previous findings based on experimental geomorphological methods.Potential LST rates are then computed for each year individually over the 1979-2010 period. Deviations from the 30-year mean LST rates are used to identify 3 cases: positive anomalies, negative anomalies and close to normal. Storm contributions to longshore transport are brought into light: inter-annual variability in LST rates is strongly correlated to >4 m waves occurences.The relationship between LST rates gradients alongshore and shoreline changes is also investigated : despite the significant amout of papers on that subject, results showned in this thesis appears to be disappointing.In the light of these findings, coastal drift cell boundaries are defined taking into account the 30-year mean potential LST rates and the inter-annual variability. This thesis thus highlights the good potential of longshore transport modelling in yielding coherent results that are essential from a shoreline management perspective for future coastal sustainability.

The variability of the nearshore wave climate is investigated via the analysis of over 10 years of Acoustic Doppler Current Profiler (ADCP) data from a gauge deployed at Melbourne Beach, FL. Examples of large yearly variability in the significant wave height, peak period, mean direction and energy distribution are found in the data. Estimates of the averaged spectra for the entire record show that the average wave energy is distributed almost symmetrically with the peak being close to shore-normal. It was expected that the peak would be shifted towards the north of shore-normal considering net north to south longshore sediment transport at this location. Further analysis of the directional spectra partitioned into three directional windows reveals that waves from the southeast (avg. Hmo = 0.78 m) are less energetic than those from the northeast (avg. Hmo = 0.87 m), but they arrive from the south 53% more often. Additionally, energy-based significant wave height (Hmo), peak period (Tp) and mean period (Tmean) distributions are studied and modeled with notable success. Radiation stress (Sxy) estimates are computed using both rigorous integration as well as parameter-based approximations. These two estimates are correlated but the parameter-based approximation over predicts Sxy by 42%, because this method assigns all the wave energy into one direction (Ruessink et al., 2001). Finally, it is shown by the Sxy total average that the net longshore forcing at this location is indeed north to south, but yearly and seasonal variability were quite high. The results indicate that short-term wave records may not provide accurate information for planning purposes. For example, if only 3 months of data were collected at this site, there would be a 33% chance that the mean longshore forcing would be erroneously directed from south to north.