Background and aims – Salt marsh plant communities will be among the first to be exposed to the predicted increase in sea level and to the associated environmental changes. The objectives of this study were to evaluate the influence of three major environmental variables (elevation above sea level, distance from the sea, vegetation age) on vegetation diversity in salt marshes and to predict vegetation changes in the year 2100 according to different scenarios of sea level rise. Methods – Plant communities were sampled in 1257 plots of 1 m ² distributed along transects randomly positioned perpendicular to the shoreline in the Bay of Somme (Picardy, France). Digital elevation model data were used to determine the plot elevation and the distance between the plots and the shoreline. Three centuries of changes in the vegetation cover were reconstructed using historical maps and aerial photographs to estimate the vegetation age. We investigated the relationships between elevation above sea level, distance from the sea, vegetation age and vegetation richness and composition using mixed models. Predictive models of species richness and cover of dominant halophytes were built using the parameter estimates of the previous mixed models and the projections of the explanatory variables in 2100 according to the different sea level scenarios from +0.5 m to +2.5 m. Key results – Mixed models showed that species richness mainly increased with vegetation age. The halophytes exhibited contrasting patterns along elevation and age gradients. Sedimentation rates may counteract the sea level rise until the latter reaches a critical rate that drowns the marsh vegetation. Conclusions – Because the proportions of ancient vegetation will be higher in the bay, mean plant species richness may be higher in predicted communities in 2100 than in recently sampled communities.

ASTER TRIPOLIUM; BAIE DE SOMME; HALOPHYTES; HISTORICAL FACTORS; PLANT COMMUNITY; SALICORNIA; SALT MARSH; SEA LEVEL RISE; SPECIES RICHNESS