Wind-driven water level fluctuations drive marsh edge erosion variability in microtidal coastal bays

Marsh shorelines are retreating rapidly in coastal Louisiana, largely driven by wind waves attacking the marsh edge. The amount of wave power hitting the marsh is a major predictor for marsh retreat rates; however, marsh erodibility (erosion rate per unit of wave power) has a large spatial variability. Identifying the causes of this variability is essential to obtain more reliable predictions and to optimize marsh protection strategies. Here we investigate marsh edge erosion in a small (similar to 3 km(2)) bay within Barataria Bay, LA, USA. Long-term (similar to 140 years) erosion data and short term (similar to 1 year) field measurements show that, for the same wave power, north-facing marsh edges erode twice as fast as south-facing marsh edges. A possible explanation might reside in the peculiar hydrodynamics of coastal Louisiana, where northerly winds are associated with low water levels and southerly winds are associated with high water levels. This causes south-facing shores to experience high water levels when being impacted by waves and north-facing shore to experience low water levels when being impacted by waves, which could subsequently affect marsh edge erosion in three different ways. First, south-facing shores experience a higher frequency of wave overshooting, which limits the ability of waves to cause erosion. Second, north-facing shores experience a higher frequency of waves impacting the highly erodible soil below the root mat, thus undercutting the marsh. Third, south-facing marsh edges have a higher elevation and a higher soil shear strength in the root layer (0-20 cm depth), likely because these shores receive more sediment during wave events. These three processes were combined into a single empirical correction to represent effective marsh erodibility and the correction was used in a 2D model of marsh edge retreat. The model accurately predicts marsh edge erosion and can be used to determine whether historical marsh loss was due to edge erosion or to other processes, such as ponding or drowning.