More people exposed to low-probability coastal flooding than previously expected
Published: 29 June 2021
Using their novel dataset, the researchers found that 192 million people are exposed to a 1-in-1000 year flood, of which 60% can be attributed to tropical cyclones – also called hurricanes or typhoons – alone. Previous studies have underestimated the global exposure to low-probability coastal flooding by 31%. The results of the study were published today in Nature’s Communications Earth & Environment.
Protecting coastal communities
Storm winds can push seawater to extreme levels, causing coastal flooding. This tsunami-like phenomenon is called a storm surge. Storm surges that occur along low-lying, densely populated coastlines can leave devastating societal, economical, and ecological impacts. To gain understanding into the threat imposed by coastal flooding, return periods of storm tides, defined as the water level rises due to the combination of storm surge and tide, must be accurately evaluated. In recent years large-scale hydrodynamic modelling has evolved rapidly at a sufficient accuracy to begin informing policy-making. However, low-probability coastal flooding caused by tropical cyclones is the blind spot of existing datasets. Therefore the researchers created the new global Coastal dataset of Storm Tide Return Periods (COAST-RP).
The researchers used a global dataset (STORM) that contains tracks of tropical cyclones, representing thousands of years of tropical cyclone activity, as input for this study. “By explicitly including low-probability tropical cyclones in hydrodynamic simulations for the first time at a global scale, our new dataset offers a substantial improvement over existing datasets”, says Job Dullaart, researcher at VU Amsterdam (IVM) and first author of the study.
More accurately assessing using the Global Tide & Surge Model
Tides and storm surge are modelled using Deltares’ Global Tide and Storm Surge Model (GTSM). GTSM is developed with the Delft3D Flexible Mesh Suite (Delft3D FM), and uses an unstructured grid to represent coastal areas in more detail than open oceans. The high computational efficiency of GTSM in combination with a novel approach to high performance computing made it possible to run thousands of years of hydrodynamic simulations. “It is great to see that several recent scientific innovations coming together in this new dataset” says Sanne Muis (Deltares / VU Amsterdam and co-author of the paper). “Using the probabilities of extreme sea levels derived in this study we can now improve the mapping of the global risks to coastal flooding.” Validation against regional studies in the United States and Australia shows that the computed return periods of extreme sea levels are very accurate. The COAST-RP dataset can be used to more accurately assess the areas at risk of coastal flooding. Moreover, the dataset provides an improved basis for projection of future risks under sea-level rise. To allow other researchers to make use of the new dataset and support decision-makers, the researchers made COAST-RP freely accessible online.