Abstract
Models are used to project coastal wetland distribution under future sea-level rise scenarios to assist decision-making. Model validation and comparison was used to investigate error and uncertainty in the Sea Level Affecting Marshes Model, a readily available model with minimal validation, particularly for wetlands beyond North America. Accurate parameterisation is required to improve the performance of the model, and indeed any spatial model. Consideration of tidal attenuation further enhances model performance, particularly for coastal wetlands located within estuaries along wave-dominated coastlines. The model does not simulate vegetation changes that are known to occur, particularly when sedimentation exceeds rates of sea-level rise resulting in shoreline progradation. Model performance was reasonable over decadal timescales, decreasing as the time-scale of retrospection increased due to compounding of errors. Comparison with other deterministic models showed reasonable agreement by 2100. However, given the uncertainty of the future and the unpredictable nature of coastal wetlands, it is difficult to ascertain which model could be realistic enough to meet its intended purpose. Model validation and comparison are useful for assessing model efficacy and parameterisation, and should be applied before application of any spatially explicit model of coastal wetland response to sea-level rise.