Why the World May Be Underestimating the Reach of Sea Level Rise
New research suggests that flawed assumptions in coastal hazard models could mean millions more people are at risk from rising seas than previously believed.
Breaking the Ice:
A recent study has uncovered a surprising and troubling problem in how scientists estimate the risks of sea level rise. After reviewing 385 coastal hazard studies published between 2009 and 2025, the authors found that more than 99% handled the relationship between sea level and land elevation incorrectly.
At the heart of the issue is a seemingly small but consequential technical assumption. Many studies estimate coastal flooding risk by comparing land elevation data with a simplified model of sea level known as a geoid, which approximates the shape of Earth’s gravity field. But the researchers show that actual measured sea levels are often higher than these models suggest, particularly in parts of the Global South where ocean circulation, temperature, and salinity influence sea-surface height.
The study’s global meta-analysis indicates that with 1 meter of relative sea level rise (RSLR), the area of land projected to fall below sea level could increase by 31–37% more than previously estimated, while the number of people affected could rise by 48–68%, potentially reaching 77–132 million people. In other words, the baseline from which we measure sea level rise may already be higher than many models assume.
Quick Melt:
Coastal hazard assessments shape everything from urban planning and disaster preparedness to international climate finance. If sea level exposure has been underestimated, governments may be underprepared for the scale of future coastal flooding.
The discrepancy arises largely because many studies equate sea level with a geoid surface, effectively assuming that “zero elevation” in global datasets corresponds directly to local sea level. In reality, the ocean surface varies due to currents, wind patterns, temperature gradients, and regional gravitational differences. These variations can create differences of several decimeters to more than a meter in some regions.
When these offsets are corrected using real measurements of sea-surface height, the exposure picture changes dramatically. Southeast Asia, home to some of the world’s most densely populated river deltas, shows particularly large increases in projected vulnerability. In some scenarios, proper sea-level referencing nearly doubles estimates of land and population below sea level in the region.
The study also raises concerns about policy frameworks that rely on existing research syntheses. Some coastal exposure estimates cited in the Intergovernmental Panel on Climate Change (IPCC)’s Sixth Assessment Report may themselves rely on studies affected by these methodological oversights. The authors recommend clearer documentation standards, improved data alignment practices, and the development of datasets that already integrate elevation with measured sea-level information, reducing the likelihood of future errors.
The Thaw:
Don’t Understand Relative Sea Level Rise? AccumulationZone Explains.
While global sea level rise measures how the ocean surface changes globally, RSLR considers the combined effect of ocean height and land elevation at a specific location.
Imagine standing on a beach. If the ocean rises by 30 centimeters, flooding risk increases. But if the land beneath your feet simultaneously sinks by another 30 centimeters (a common phenomenon in sediment-rich deltas) the effective rise relative to the land is 60 centimeters. Many coastal regions already experience this compound effect due to land subsidence, often accelerated by groundwater extraction and urban development.
Accurately estimating these risks requires precise vertical reference systems. If these datasets are not aligned to the same vertical reference frame, the comparison can produce significant errors. This is exactly what the study uncovered: a widespread mismatch between elevation data and actual sea level measurements. According to the analysis, global coastal sea level is on average about 0.24–0.27 meters higher than commonly assumed in hazard assessments, with even larger discrepancies in data-sparse regions.
Many global elevation datasets are referenced to a geoid, a theoretical surface that represents the shape the ocean would take if it were influenced only by Earth’s gravity and rotation. In practice, however, the ocean is far more dynamic. Ocean currents, winds, temperature gradients, and salinity all redistribute water across the planet, subtly raising or lowering the sea surface relative to the geoid.
This difference between the geoid and the real ocean surface is known as mean dynamic topography (MDT). In some regions, MDT can shift the average sea surface upward by several decimeters, or even more than a meter, compared to the geoid reference used in many coastal hazard models.
When hazard assessments assume the geoid represents actual sea level, they effectively start from a baseline that is too low. The result is a systematic underestimation of how much land already sits near or below sea level. The study’s global meta-analysis shows that correcting this mismatch significantly changes exposure estimates. Under a scenario of 1 meter of relative sea-level rise, the number of people projected to live below sea level increases substantially when measured sea level is properly accounted for.
The discrepancy is especially pronounced in the Global South, where many coastal regions lack dense observational networks for gravity measurements and tide gauges. In places like Southeast Asia, home to vast, low-lying river deltas such as the Mekong and the Ganges-Brahmaputra, even small vertical differences can dramatically reshape projections of flood exposure.
Final Thoughts
As rising seas continue to reshape the world’s coastlines, even seemingly small measurement errors can translate into millions more people at risk. The challenge ahead is not just predicting the future of sea level rise, but ensuring we are measuring today’s oceans correctly.