Global application of a regional frequency analysis to extreme sea levels

Coastal floods, driven by storm surges, waves and tides, can have devastating impacts on communities worldwide. This research paper, led by Fathom’s Senior Coastal Model Developer Tom Collings, develops an innovative approach for estimating global coastal extreme sea levels.

There is an urgent need for accurate assessments of coastal flood risk around the world, but estimating coastal extreme water levels on a global scale is a difficult and complicated task. Tide gauge records cover only a small proportion of the world’s coastline, leaving large areas ungauged and in the areas which are gauged, the records are often short and incomplete. Global tide, surge and wave models have been developed to help fill the gaps in observations but these have relatively coarse resolution at the coastline.

To simulate coastal flooding using hydrodynamic inundation models, scientists need high-resolution boundary conditions of total water levels at regular intervals along the coastline. To obtain these from publicly available datasets requires combining and downscaling the available data. 

To address this, the research team used a novel approach involving regional frequency analysis for the first time to estimate extreme sea level across the entire global coastline. 

The research: A new approach to estimating extremes

Regional frequency analysis offers several advantages over traditional methods of estimating extremes, particularly in areas where high-quality observational data is scarce. 

Combining several observed and hindcast global datasets, this study builds upon a regional methodology developed by William Sweet from the US’s National Oceanic and Atmospheric Administration (NOAA) to fill in data gaps. 

The team used a regional frequency analysis approach to synthesise tide, surge and wave data into an estimate of extreme sea levels at any point along the global coastline. With this approach, individual records are normalised by tidal range so that they can be merged with nearby records. The merged regional records are then used to characterise the extreme water levels within the area. 

Finally, the regional data is adjusted back to local conditions using tidal range estimates, giving an estimate of extreme sea levels for specific locations along the coast.

Key findings: Quantifying coastal flood risk

The authors show that using this regional frequency analysis approach can help improve upon estimates of water levels driven by rare extreme events, such as tropical cyclones. 

The high-resolution (approximately 1km) output data is a useful tool for assessing flood risk along the entire global coastline – it was subsequently used by Wing et al 2024 as boundary conditions for coastal flood hazard mapping.

In conclusion, the study is a valuable resource for quantifying the coastal flood risk globally. It offers an innovative methodology that can help communities worldwide build resilience to and mitigate against coastal flooding.