Assessing the effects of sediment and tidal level variability on coastal wetland evolution

Abstract

• We include short-term sediment and tidal variability on coastal wetland evolution. • We first simulate 10-year evolution of a mangrove-saltmarsh wetland using field data. • Stormy years promote saltmarsh transgression, but accretion is barely affected. • MC method and synthetic data are used next to assess uncertainty over the century. • Results excluding variability underestimate wetland extent/accretion by up to 25% Assessments of coastal wetland vulnerability to Sea Level Rise (SLR) frequently use simplified representations of tidal levels and sediment input (i.e., average or representative values), disregarding short and long-term variations. The expensive computational cost of using high-resolution time series is one of the main reasons for such simplification. However, observed series contain short-term variations (such as severe storms) and interannual variability induced by climate and ocean phenomena. In this study, we used a previously developed ecogeomorphological model and high-performance computing to carry out simulations with observed and synthetic high-resolution datasets to account for both types of variations. We performed two experiments. In the first experiment we used observed water levels and satellite-derived sediment concentrations from 2002 to 2011 to assess the capability of the model to reproduce in situ measurements of soil elevation change in a mangrove-saltmarsh wetland in SE Australia. Results showed increased wetland extent (particularly in saltmarsh) and soil elevation change in years with important storms, but little effect on the long term trends of soil elevation and wetland evolution. In the second experiment, we implemented a Monte Carlo (MC) approach comprising 100 realizations of 100-year duration using synthetically-generated input data based on 13 years of high-resolution values of water levels and sediment load. Each simulation covered the period between 2000 and 2100 and included increases in SLR following the RCP 8.5 scenario. The MC results show that the variability of inputs generate uncertainties in wetland evolution, with up to 20% of wetland area extent classified as highly uncertain during the second half of the century. In agreement with the first experiment, uncertainty was higher in saltmarsh. In addition, uncertainty in elevation change was lower than in wetland extent. MC results were compared with simulations using representative sinusoidal waves and constant sediment concentration, one using the mean tidal range and the other using a combination of neap, mean and spring tidal ranges. Both simplified formulations underestimated wetland extent (particularly in saltmarsh), but the simulations with the combination of tidal ranges were able to match the soil elevation change results from the MC approach.