Résumé
How will mangrove forests cope with fast-changing tropical coastal environments under global change? How will coastal ocean processes impact thousands of kilometres of mangrove shorelines where millions of people live? So far, these questions have been insufficiently addressed due probably to the complexity and diversity of interwoven sedimentary, hydrological and ecological processes involved on any mangrove coast. Evidence of changes on several open mangrove coasts is, however, highlighted by decades of aerial and satellite images. Besides, time series of ocean data are now provided worldwide by different international services such as the E.U. Copernicus Marine Service. The French Guiana (FG) coast of South America offers foundations for exploratory semi-empirical models of the impacts of hydro-sedimentary processes on mangrove coasts. The 320-km long mangrove shoreline is submitted to marked, often extremely rapid, alternating erosional or accretional phases generated by the alongshore northwestward migration of giant mud banks. Experimental studies have been carried out for decades to show and explain processes involved in the permanent transformation of the FG coast and its consequences for biodiversity and socioeconomic activities. Studies suggest a leading forcing role of ocean wave and current regimes on mud bank migration, erosion and accretion, the combination inducing exceptional rates of mangrove landward retreat or seaward expansion that can attain up to 500 m per year. Here, we present a modelling approach, named MANGCOAST, designed to simulate mangrove shoreline landward retreat and seaward advance, as close as possible to observations of those mangrove shorelines visually and annually delineated in satellite images acquired since 2013. Time series of wave and current data associated with data on the extents, shapes and locations of mud banks constitute input data. Our modelling approach is based on interaction graphs describing relations between three entities (mangrove, ocean, mud bank) and implemented using Ocelet, a domain-specific language. Two equations were proposed to consider: (1) the role of the subtidal part of the mud bank in the dissipation of ocean waves and currents, and (2) the ability of mangroves to expand over new consolidated mud substrates. Five coefficients are used to weigh the respective influences of waves, currents, mudbank extent and mangrove expansion. Their final values are adjusted from an iterative optimization process searching for convergence of the simulated mangrove shoreline with the observed ones. This model was run on different sectors of the FG coast for the 2013-2022 period to deliver a set of coefficients for each sector. First, the geographical variability of coefficient values is discussed as a function of local coastal geomorphology and orientation. Second, the particular seasonal effect of waves and currents on mangrove shoreline fluctuations is tested by embedding final coefficient values in the equations and by applying daily-computed trends of wave and current data to different mud bank configurations. We put particular emphasis on examining the role of the ocean wave regime during the high wave-energy season from December to April. Third, we explain how indices of coastal vulnerability can be built and could contribute to the Guyana Coastal Observatory (https://observatoire-littoral-guyane.fr/). [Copyright Author(s) 2023. CC Attribution 4.0 License: https://creativecommons.org/licenses/by/4.0/legalcode]