For generations, coastal communities have prepared for floods by relying on historical records. Engineers would examine past data, identify the highest water levels ever recorded, and construct defenses based on those benchmarks.
Throughout much of the 20th century, this strategy was considered the bedrock of coastal risk management.
A new study investigated whether historical flood records for coastlines worldwide are still accurate. This metric has been changing since the 1960s, and weather isn’t the primary driver.
The research was led by Sönke Dangendorf at Tulane University, alongside an international team of researchers. Their findings were published in the journal Nature Climate Change.
The scientists analyzed over a century’s worth of water level data collected by tide gauges—devices that record wave heights hourly—from 130 coastal sites globally. They then correlated this data with climate models to pinpoint the causes of the changes. This produced a global picture of how frequently extreme floods are happening now compared to 1900.
The headline figure is striking. A flood that, in 1900, would have been expected once per century on an average coast now occurs roughly 12 times more often. On nearly half the studied sites, what was once a century-flood event is now happening at least once a decade.
Extreme sea levels occur when high tides, storm surges, and the continuously rising ocean baseline combine. As this baseline is elevated, even a weaker storm can push water levels to heights that previously required a much more severe event.
The difficulty lies in disentangling human influence from natural fluctuations. Oceans have their own rhythms—long-term shifts in currents and winds that have always pushed coastlines up and down.
Careful statistical work is required to separate these natural cycles from warming driven by the burning of fossil fuels. The team ran climate simulations with and without human influence and compared the outcomes to actual tide gauge readings. This allowed them to identify the culprits.
In the early 20th century, natural factors dominated sea level changes, but since the 1960s, human impact has steadily grown. By setting aside daily tides and storms and focusing on the relentless rise in sea level, the researchers calculated that human-caused warming has approximately quadrupled the likelihood of a 100-year flood occurring.
Natural climate variations still shape regional patterns, but along most coastlines, they have taken a backseat to human influence.
Global averages mask vast differences from one shoreline to another. Some locations are far worse off, and the reasons are often intensely local. The land itself can be sinking even as sea levels rise.
Manila, the capital of the Philippines, is an extreme example. Decades of groundwater pumping have caused land subsidence, increasing the frequency of extreme flooding by more than 300 times. A small rise in ocean levels was all that was needed for disaster to become commonplace.
Elsewhere, the same story unfolds in milder fashion. In Sandy Hook, New Jersey, a flood expected once a century in 1900 was occurring about once every 16 years by 2005.
Wellington, New Zealand’s capital, has gone from a once-a-century threat to one that happens twice a year.
Scientists have known for years that sea levels are rising and that human activity is the cause. They also knew, broadly, that rising sea levels lead to more coastal flooding. What no one had measured on this scale was how much of the actual increase in flood magnitudes is attributable to human-caused sea level rise.
The answer comes not from computer modeling looking into the future, but from real historical data. This is where this work makes its contribution. It provides direct evidence—not based on projections, but on actual data from the past century—that human-caused warming has already altered the danger of coastal flooding.
Previous studies relied heavily on models; this one grounds its findings in tide gauge data. The study’s main contribution is quantifying the human role. The team found that human-caused sea level rise has made extreme coastal floods about four times more likely.
The practical snag is that flood defenses are planned using historical data. When engineers design a seawall or an insurer calculates the cost of a policy, they refer to the “century flood”—a flood level that past records suggest should happen once in 100 years.
This benchmark operates on the assumption that the past is a reliable guide to the future. This study shows that, often, it is not. If a century flood is now happening every ten to twenty years, infrastructure built to older standards is quietly becoming inadequate.
Risk maps based on 20th-century data underestimate the true danger coastal communities face today. Dangendorf points to New Orleans as an example of a place that has responded. The city rebuilt one of the world’s most advanced flood defenses after Hurricane Katrina.
The study’s findings explain why diligent maintenance and forward-thinking planning keeps these defenses effective in a changing environment, Dangendorf noted.
New Orleans is thus both a testament to the effectiveness of past investment and a model for other coastal cities facing the same imperative: continuous flood protection.
The benchmark is not static, nor are the defenses built upon it. The most obvious takeaway is that the historical baselines upon which coastal planners have relied are no longer sufficient.
Flood frequency, treated as a fixed geographical fact, has been shifting over decades, and this work shows how dramatically and, crucially, why. It is the “why” that gives planners a handle for action.
Since this rise is now firmly linked to human-caused warming, its dynamics are no longer a mystery communities can simply wait out. The science of establishing causes for observed changes—the science of attributing specific drivers—needs to be integrated into coastal adaptation and risk management, the authors argue.
Coastal defenses, drainage systems, and insurance models for the coming decades must be designed for the ocean that exists today, not the one recorded a century ago. Planning based on old benchmarks is no longer adequate.
For millions living on vulnerable coastlines, these changes could manifest as updated flood maps, revised building codes, and infrastructure designed for more frequent threats. The criteria have changed, and this study makes a compelling case that planning must change with them.
About the Author
