Sensitivity of Estuaries to Compound Flooding

Abstract Fluvial and surge-tide extremes can occur synchronously resulting in compound flooding in estuaries, greatly intensifying the hazard. This flood risk has the potential to increase in the future as the frequency, phasing and/or intensity of these drivers change. Improved understanding of how extreme fluvial discharge and surge-tides interact will help inform future flood mitigation methodology. In this paper, therefore, we resolve for the first time intra-estuary sensitivities to fluvial and surge-tide extremes, for two contrasting UK estuaries (Humber and Dyfi). Model simulations at hyper-spatial resolution (< 50 m) using a 2D hydrodynamic model predicted compound flooding hazards based on: (1) present-day extreme events (worst on record); (2) present-day extreme events with shifted timings of the drivers to maximise flooding; and (3) modified drivers representing projected climate change. We found that in a small estuary with short-duration, high-intensity fluvial inputs (Dyfi), flood extent is sensitive to the relative timing of the fluvial and surge-tide drivers. In contrast, the relative timing of these drivers did not affect flooding in a larger estuary with a slower fluvial response to rainfall (Humber). In the Humber, extreme fluvial inputs during a compound hazard actually reduced maximum water depths in the outer estuary, compared with a surge-tide-only event. Projected future changes in these drivers by 2100 will increase compound flooding hazards: simulated sea-level rise scenarios predicted substantial and widespread flooding in both estuaries. However, projected increases in surge-tide behaved differently to sea-level rise of the same magnitude, resulting in a greater seawater influx and more flooding. Increased fluvial volumes were the weakest driver of estuarine flooding. In this paper we show how these interactions are complex and how the hydrodynamics vary considerably between different estuaries and sites within estuaries, making it difficult to generalise, use probabilistic or use 1D approaches for assessing compound flooding hazards. Hence, we contribute new knowledge and methods for catchment-to-coast impact modelling used for flood mitigation strategies.