The changing risk of coastal flooding in the United Kingdom

On the night of 31 January 1953, Britain experienced one of the worst natural disasters in its modern history. Strong northerly winds swept over the North Sea, driving an unusually high sea level, known as a storm surge, towards the coast. The surge raised water levels by approximately 2m above the expected high tide, leading to widespread flooding across the UK, the Netherlands, Belgium and Germany. In England alone, 307 people lost their lives and more than 32 000 people were evacuated. Damage to sea defences cost an estimated £50 million, equivalent to over a billion pounds today (Haigh et al., 2017). Since then, the UK has made significant investments in coastal flood defences and early warning systems. The first flood risk management strategy in England was developed for the Thames Estuary. This includes the Thames Barrier, the second largest moveable flood barrier in the world, alongside a network of embankments, walls and other defences along the river (Department for Environment, Food, Figure 2a). However, achieving RCP2.6 is increasingly considered ambitious. The UK Committee for Climate Change recommends that adaptation planning accounts for 2°C–4°C of global warming relative to pre-industrial levels. Under a high-emissions scenario (RCP8.5), we show this risk increases to 8.9% in 2050 and 50.3% in 2100 (Figure 2b). Understanding these risks is essential for informed decision-making and effective coastal adaptation. Coastal flooding poses a serious and evolving threat to the UK, as highlighted by the 1953 storm surge disaster and more recent events such as Storm Xaver. Flood risk arises from a combination of storm surges, tides, wind and waves, rainfall and rising sea levels. While flood defences and forecasting have greatly improved, climate change-driven sea level rise is likely to increase the frequency and severity of extreme coastal flooding during the 21st century, highlighting the importance of continued investment in adaptive flood risk management. The author would like to thank Freya Garry and Thomas Kilsby, who provided helpful comments on draft versions of the manuscript. Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.