The relation between salt marsh accretion and flooding regime was quantified by statistical analysis of a unique dataset of accretion measurements using sedimentation-erosion bars, on three barrier islands in the Dutch Wadden Sea over a period of c. 15 years. On one of the islands, natural gas extraction caused deep soil subsidence, which resulted in gradually increasing flooding frequency, duration, and depth, and can thus be seen as a proxy for sea-level rise. Special attention was paid to effects of small-scale variation e.g., in distance to tidal creeks or marsh edges, elevation of the marsh surface, and presence of livestock. Overall mean accretion rate was 0.44 ± 0.0005 cm year −1 , which significantly exceeded the local rate of sea-level rise of 0.25 ± 0.009 cm year −1 . A multiple regression approach was used to detect the combined effect of flooding regime and the local environment. The most important flooding-related factors that enhance accretion are mean water depth during flooding and overall mean water depth, but local accretion strongly decreases with increasing distance to the nearest creek or to the salt marsh edge. Mean water depth during flooding can be seen as an indicator for storm intensity, while overall mean water depth is a better indicator for storm frequency. The regression parameters were used to run a simple model simulating the effect of various sea-level scenarios on accretion and show that, even under extreme scenarios of sea-level rise, these salt marshes can probably persist for the next 100 years, although the higher parts may experience more frequent inundation.