On the need for uncertainty consideration in coastal erosion projections to support adaptation planning

Abstract:

Coastal erosion is one of the main threats to coastal areas worldwide. This impact can occur at different time scales, including permanent loss of beach surface due to mean sea-level rise, and episodic extreme retreat events due to the combined effect of waves, storm surges and tides, which due to the rise in mean sea-level will become more frequent.

These coastal erosion drivers will be affected by climate change, and their associated uncertainty grows towards the end of the century and beyond. Their uncertainty arises from multiple sources, including emissions scenarios, climate models and forcing variables and their bias, downscaling techniques, and impact models, cascading through the top-down process and accumulating in shoreline change estimates. The identification of the aforementioned uncertainty sources is essential to select the most appropriate sampling approach (e.g., ensemble or probabilistic), which would benefit from the highest degree of probabilistic development (Toimil et al., 2020).

In this presentation, and using a pocket beach as an illustration, we analyse uncertainty in coastal erosion projections. We consider using different emissions and concentration scenarios, applying dynamic projections of storm surges and ocean waves (derived from a multi-model ensemble and including bias correction) downscaled to the coastal zone, and performing an ensemble of erosion models that allow obtaining probabilistic estimates of coastal erosion. Additionally, we discuss the importance of uncertainty consideration to make better decisions on adaptation. We also highlight the need for flexible adaptation strategies, as the best way to incorporate uncertainty in adaptation planning is through flexible approaches such as adaptation pathways. Adaptation pathways offer the advantage of identifying when, how and how much to adapt keeping pace with changing conditions, provided systematic monitoring support them.