Sea level rise (SLR) poses a huge and growing risk to vast coastal plains and densely populated river deltas around the world. The latest study however, it reveals an extremely disturbing fact: More than 99 % of the scientific assessments of the impacts and risks associated with sea level rise to date incorrectly work with data on elevation and actual sea level. This systemic failure leads to a massive underestimation of the true threat to coastal areas. The results show that a much larger part of the land and significantly more people are actually at risk of flooding than previously reported.

Elevation models, geoids, and overlooked ocean dynamics Accurately linking elevation data, mainly provided by satellite digital elevation models (DEMs), with local sea level elevation is essential for proper assessment of coastal vulnerability. However, an in-depth analysis of 385 peer-reviewed scientific publications from 2009 to 2025 revealed a huge „blind spot.“. Up to 90 % of them incorrectly assume that sea level is identical to global geoid models (such as EGM96 or EGM2008), completely ignoring actual local measurements.

The geoid is a mathematical model that approximates mean sea level based solely on the Earth's gravity and rotation. However, the actual level of the oceans is dynamic. It is significantly shaped by ocean currents, wind, temperature changes, and water salinity. The difference between the geoid and the real time-averaged ocean level is called the mean dynamic topography (MDT). On a global average, neglecting this difference means a deviation of about 0.24 to 0.27 meters to the detriment of the actual level. The authors found that only an absolute minimum of studies – exactly one, or 0.3 % out of 385 evaluated – correctly and completely documented and converted the height data with respect to the actual dynamics of the seas. Other publications either skipped this key step altogether or performed it incorrectly.

The Global South is paying the price for a lack of data This methodological error is not geographically evenly distributed. While in data-rich regions of the Global North (such as northern and western Europe or the east coast of North America) geoids reflect the true level relatively well, in regions of the Global South the inaccuracies are catastrophic. The largest deviations have been recorded in Southeast Asia (where there are large and extremely densely populated deltas) and in the Pacific region. In these areas, the actual mean sea level can be more than 1 meter higher than indicated by the global geoids.. In remote areas with poor quality gravity data, local differences can grow to several meters.

Compounding the problem is the fact that many studies do not take into account that land itself is often rapidly sinking (subsidence) due to human activity. Instead, they often rely on outdated reference sea levels from historical tide gauges without correction for current conditions, thereby significantly distorting the relative sea level rise (RSLR) rate.

Real impacts: Tens of millions more people at risk When the research team applied the best current global elevation models with correct data on actual local sea level (LDT), they revealed shocking findings. In simulating a hypothetical scenario of a relative sea level rise of 1 meter, they found that The area of land that would be submerged is 31 % to 37 % larger (meaning a total increase of 460,100 to 670,000 km²) than previously estimated.

Even more tragic are the numbers regarding the exposed population. Not 34 to 49 million people would be below sea level, as the incorrect methods suggested, but as many as 77 to 132 million people, representing a staggering increase of 48 to 68 billion years.. In the most critical region – Southeast Asia – the estimates of the flooded area and the population at risk increase by up to 94 % and 96 %, respectively, after the correct recalculation. For example, in the Mekong River Delta in Vietnam alone, the estimate of the affected population increased from the originally reported fractions to up to 5.4 to 10 million people after the correction of the calculations. Global estimates for low-lying coastal zones up to 10 meters (LECZ) have also been revised – after taking into account the actual sea level, up to 0.82 to 1.07 billion people actually live there.

The need for systemic change in science and politics The findings also highlight the extremely risky spread of this erroneous data. The review confirmed that 46 studies suffering from analytical errors were included as reference material in the latest Intergovernmental Panel on Climate Change (IPCC) reports under the AR6 cycle. Since it is the IPCC reports that shape global policies and guide the allocation of funding for coastal adaptation, inaccuracies at this level are critical and can lead to a profound underestimation of climate risks.

Efforts to measure sea level rise with millimeter precision are completely meaningless if the results are distorted by elementary errors in combining with terrain elevation data.. To address this problem, scientific journals need to implement strict guidelines and checklists for aligning vertical reference systems. In addition, the authors suggest that satellite model providers (such as CoastalDEM or DeltaDTM) should start offering ready-made products that are properly aligned to real dynamic sea levels, rather than leaving complex geodetic conversions to end users.

Existing coastal adaptation and protection plans require urgent reassessment in light of these facts, as we may reach critical levels of threat much sooner than many world governments currently anticipate. JRi&CO2AI