Ce travail de thèse est une étude multi-échelles de temps de la morphodynamique d'une île barrière soumise à des conditions de forçages paroxysmaux de mousson, de tempêtes tropicales et de typhons. Cette étude se concentre dans un premier temps, sur un large travail de terrain effectué dans le cadre du projet KUN-SHEN. Sept mois de mesures (de novembre 2011 à janvier 2012 et de mai à septembre 2012) ont permis l'acquisition d'un jeu de données inédit dans la littérature, exhaustif et de haute résolution temporelle et spatiale. Des mesures hydrodynamiques du large jusqu’à la zone de jet de rive ont été acquises en parallèle d'un suivi topographique de la plage émergée. Les mesures de vagues ont été acquises à partir d'une bouée au large, de deux houlographes-courantomètres en avant-côte (mesures en continu à 2 Hz) ainsi que de pressiomètres enfouis dans la plage sub-aérienne (mesures en continu à 5 Hz). Les levés topo-bathymétriques ont été réalisés au D-GPS (résolution centimétrique) une fois par semaine pendant la saison de mousson et juste après et avant chaque événements de tempêtes extrêmes pendant la saison des typhons. Le travail d'analyse cible ensuite la morphodynamique de la zone de jet de rive et de la plage émergée sur un panel d’échelles spatio-temporelles allant de l'instantanée à annuelle. A l’échelle instantanée, ce sont les variations d’élévation de la surface libre de l'eau et du lit sableux (quelques secondes) qui sont observées au cours de chacune des phases (montant, apex, tombant) de la tempête tropicale Talim (juin 2012, Hs = 10.34 m et Ts = 14.6 s). Pour le même événement de tempête, la réponse morphologique de toute la plage émergée est ensuite décrite et quantifiée dans le détail à l’échelle événementielle (quelques jours). Les bilans sédimentaires de chacune des saisons sont ensuite quantifiés dans le but de caractériser la dynamique saisonnière (quelques mois) à annuelle d'une barrière sableuse soumise à deux types de forçages extrêmes différents (mousson/typhons). L'impact sur le front de plage, des groupes de tempêtes faibles à modérées pendant l'hiver est ainsi comparé à l'impact des tempêtes extrêmes pendant l'été. On souligne finalement 1) l'importance d'un jeu de mesures in-situ de bonne qualité dans un travail d'analyse morphodynamique, 2) l'essentialité de l’emboîtement des échelles spatio-temporelle ainsi que 3) le rôle du profil morphologique héritée dans la réponse morphologique d'une plage émergée lorsqu'elle est soumise à des conditions de forçages de vent et de vague extrêmes. / This work focuses on a multi-scale analysis of a sand barrier morphodynamics exposed to paroxysmal forcings of monsoon, tropical storms, and typhoons. The study provided a large dataset of in-situ measures acquired in the framework of the KUNSHEN project. Seven months of field work (from November 2011 to January 2012 and from May to September 2012) provided a new, exhaustive and hight resolution measures. Measures of hydrodynamic conditions from offshore to the swash zone are acquired as well as high frequency topographic surveys of the emerged beach. Waves monitoring are from an offshore buoy, two current profilers in the shoreface (continuous measures at 2 Hz) and pressure sensors buried in the sub-aerial beach (continuous measures at 5 Hz). The topo-bathymetric surveys were acquired using D-GPS (centimetric resolution) one time a week during winter season of monsoon and just before and after each extreme storm during summer season of typhoons.Then, analysis focus in the morphodynamics of the swash zone and the emerged beach on a range of scales from instantaneous to one year. At the instantaneous scale, we analyze variations of the water surface fluctuation and sand bed elevation (seconds) during each phases (rising, apex and falling) of the Talim tropical storm (June 2012, Hs = 10.34 m and Ts = 14.6 s). Concerning the same storm, we characterize the morphologic changes of the whole emerged beach at the evenemential scale (days). Sand balances of each seasons are quantified in order to characterize the dynamics from seasonal scale (months) to annual scale of a sand barrier driven by two types of extreme forcings (monsoon/typhoons). On the beach-front, impact of groups of winter weak storms is compared to the impact of extremes tropical storms and typhoons. To conclude, we highlight 1) importance of in-situ measures in morphodynamic studies, 2) the major role of scales stacking and 3) the role of herited beach profile in the morphologic responses of the emerged beach driven by extremes forcings of wind and wave.

O presente estudo modificou o modelo parabólico da forma em planta de equilíbrio de Medellín et al., 2009, proposto para praias de enseada com desembocaduras, incorporando um parâmetro relacionado à dinâmica costeira, que seria o volume de areia necessário para formar a praia emersa próxima à desembocadura. Foram escolhidas 27 desembocaduras na costa brasileira e obtidos parâmetros da dinâmica costeira e estuarina, através de imagens do Googleearth e dados de ondas e marés astronômicas dos programas SMC-TOOLS e WXTide32, respectivamente. O modelo parabólico foi aplicado e não se ajustou adequadamente, com diferenças com relação à linha de costa de 20-40m até 80m. Logo, foi realizada sua adaptação à costa brasileira, apresentando uma moderada a baixa dispersão dos coeficientes A e B de 77 % e 64%, respectivamente. Sua validação apresentou ajuste bom a satisfatório, com diferenças médias de 5m até máximas de 15m com relação à linha de costa. Posteriormente, foi realizada a modificação do modelo parabólico que minimizou a dispersão dos coeficientes A e B com um ajuste de 98 % e 81%, respectivamente. Sua validação apresentou um ajuste excelente a bom, com diferenças médias de 3-5m até máximas de 10m com relação à linha de costa. / This study modified the equilibrium plan form formulations proposed by Medellín et al., (2009) to embayments beachs with inlets, by incorporating a parameter related to coastal dynamics, that would be the volume of sand required to build the aerial beach close to the inlet. For this study 27 embayments beaches with inlets were chosen on the Brazilian coast and obtained parameters of dynamics coastal and estuarine, through GoogleEarth images and waves and astronomical tide datas series by the SMC-TOOLS and WXTide32 programs, respectively. The Medellín parabolic model was applied and did not fit properly, with differences regarding the shoreline of 20-40m up to maximum of 80m. Therefore, their adaptation to the Brazilian coast was carried out, which showed a moderate to low dispersion of the coefficients A and B of 77% and 64%, respectively. Their validation showed good to satisfactory fit, with mean differences regarding the shoreline of 5m up to maximum of 15m. Subsequently, the modification of the parabolic model was carried out which minimized the dispersion of the coefficients A and B with a fit of 98% and 81%, respectively. Their validation showed an excellent to good fit, with mean differences regarding the shoreline of 3-5m up to maximum of 10m.

O presente estudo modificou o modelo parabólico da forma em planta de equilíbrio de Medellín et al., 2009, proposto para praias de enseada com desembocaduras, incorporando um parâmetro relacionado à dinâmica costeira, que seria o volume de areia necessário para formar a praia emersa próxima à desembocadura. Foram escolhidas 27 desembocaduras na costa brasileira e obtidos parâmetros da dinâmica costeira e estuarina, através de imagens do Googleearth e dados de ondas e marés astronômicas dos programas SMC-TOOLS e WXTide32, respectivamente. O modelo parabólico foi aplicado e não se ajustou adequadamente, com diferenças com relação à linha de costa de 20-40m até 80m. Logo, foi realizada sua adaptação à costa brasileira, apresentando uma moderada a baixa dispersão dos coeficientes A e B de 77 % e 64%, respectivamente. Sua validação apresentou ajuste bom a satisfatório, com diferenças médias de 5m até máximas de 15m com relação à linha de costa. Posteriormente, foi realizada a modificação do modelo parabólico que minimizou a dispersão dos coeficientes A e B com um ajuste de 98 % e 81%, respectivamente. Sua validação apresentou um ajuste excelente a bom, com diferenças médias de 3-5m até máximas de 10m com relação à linha de costa. / This study modified the equilibrium plan form formulations proposed by Medellín et al., (2009) to embayments beachs with inlets, by incorporating a parameter related to coastal dynamics, that would be the volume of sand required to build the aerial beach close to the inlet. For this study 27 embayments beaches with inlets were chosen on the Brazilian coast and obtained parameters of dynamics coastal and estuarine, through GoogleEarth images and waves and astronomical tide datas series by the SMC-TOOLS and WXTide32 programs, respectively. The Medellín parabolic model was applied and did not fit properly, with differences regarding the shoreline of 20-40m up to maximum of 80m. Therefore, their adaptation to the Brazilian coast was carried out, which showed a moderate to low dispersion of the coefficients A and B of 77% and 64%, respectively. Their validation showed good to satisfactory fit, with mean differences regarding the shoreline of 5m up to maximum of 15m. Subsequently, the modification of the parabolic model was carried out which minimized the dispersion of the coefficients A and B with a fit of 98% and 81%, respectively. Their validation showed an excellent to good fit, with mean differences regarding the shoreline of 3-5m up to maximum of 10m.