Lead: New research published in Nature indicates that global coastal sea levels have been systematically underestimated because many studies relied on land-elevation references tied to global geoid models rather than direct, local sea-level measurements. The analysis of 385 peer-reviewed papers (2009–2025) shows average ocean levels are about 30cm higher than previously reported, with parts of south-east Asia and the Indo-Pacific possibly 100–150cm higher. The authors warn this recalibration could bring forward and amplify the projected impacts of warming seas on coastal settlements and populations.
Key takeaways
- The study reviewed 385 peer-reviewed papers published between 2009 and 2025 and found systemic measurement choices that bias coastal elevations.
- Globally, coastal sea levels are on average about 30cm higher than commonly assumed, with regional highs of 100–150cm in parts of south-east Asia and the Indo-Pacific.
- Common geoid-based approaches underestimate sea levels by roughly 24–27cm on average, though local discrepancies can reach 550–760cm in extreme cases.
- Following a relative sea level rise of 1 metre, an estimated 37% more coastal area would lie below sea level, potentially affecting up to 132 million people.
- The study’s dataset integrates up-to-date sea-level measurements with coastal elevation data and is offered to help recalibrate coastal hazard assessments.
- Authors identify an “interdisciplinary blind spot” where oceanographic and terrestrial elevation methods have been mixed without reconciling their differences.
Background
Sea-level projections and coastal-vulnerability assessments depend on two key inputs: measurements of the water surface and measurements of land elevation. Traditionally, many studies have used land-elevation products referenced to global geoid models — mathematical representations of mean sea level determined by Earth gravity and rotation — rather than local tide-gauge or satellite altimetry data tied to actual coastal water levels. That methodological choice can introduce a consistent offset between the assumed baseline elevation of coasts and the real, measured sea-surface height.
Since the Intergovernmental Panel on Climate Change (IPCC) provides policy-relevant projections — citing a likely global rise by 2100 of about 28–100cm under a range of scenarios — accurate baseline elevation is critical to translate future sea-level change into impacts on land, infrastructure and populations. Coastal regions with low relief, densely populated deltas and small islands are particularly sensitive to small baseline shifts, which can change exposure estimates significantly.
Main event
The Nature study led by Dr Philip Minderhoud (Wageningen University) and Katharina Seeger assessed the methods and reference frames used in 385 coastal-elevation and hazard studies between 2009 and 2025. They report that over 90% of those studies used land elevation products tied to geoid references rather than direct coastal water-level observations. By recalculating coastal elevations using local sea-level measurements and updated datasets, the authors found a global average upward revision of around 30cm.
Regionally, the recalibrated estimates indicate much larger upward adjustments in some parts of the global south: parts of south-east Asia and across the Indo-Pacific could be 1 to 1.5 metres higher than previous baselines indicated. The paper attributes these differences to factors not captured by geoid-only approaches, including prevailing winds, ocean currents, seasonal sea-surface slopes, and gradients in temperature and salinity that alter local sea surface height.
The authors quantify the practical implications: under a scenario where relative sea level rises by 1 metre, the area below sea level expands by 37% compared with prior baseline assumptions, translating into as many as 132 million additional people exposed to inundation risk. The paper includes a ready-to-use global coastal elevation dataset paired with measured sea-level values and recommends revising coastal hazard methodologies to use those integrated data.
Analysis & implications
The research has immediate implications for the accuracy of coastal flood mapping, adaptation planning and economic loss estimates. If baseline sea levels are higher than assumed, many existing exposure assessments may undercount at-risk land, asset values and population. This matters for infrastructure planning, insurance pricing and the prioritisation of adaptation funding, especially in low-lying and densely populated coastal regions.
Policy frameworks that use percentile-based thresholds (for example, area flooded under a 1-in-100-year event) will shift if the baseline is raised; thresholds will be reached sooner and more frequently. For vulnerable island states and delta regions, an upward revision in baseline sea level could accelerate timelines for relocation, flood defences and other costly adaptation measures, complicating planning and finance.
International assessments such as IPCC reports aggregate many regional studies. The authors warn that an “interdisciplinary blind spot” in the literature — the routine mixing of geoid-referenced land heights with sea-level projections without local correction — may propagate systematic bias into high-profile syntheses. If confirmed and widely adopted, the study’s dataset and recommendations could prompt the re-evaluation of numerous regional hazard studies and national adaptation plans.
Comparison & data
| Metric | Previously assumed | Revised / study finding |
|---|---|---|
| Global average coastal offset | 0 cm (baseline) | +30 cm |
| Regional peak adjustments (SE Asia, Indo-Pacific) | 0–20 cm typical | +100–150 cm |
| Geoid-model average underestimate | — | ≈24–27 cm |
| Maximum local discrepancies reported | — | ≈550–760 cm |
| IPCC projected rise by 2100 | 28–100 cm (scenario dependent) | Context: baseline shifts make impacts occur sooner |
| Area below sea level after 1 m rise | Baseline estimate | +37% area; up to 132 million additional people exposed |
The table summarises the principal numerical findings from the study and places them alongside the IPCC’s 2100 envelope. The authors stress that the magnitude of local discrepancies varies widely by location and by the elevation product used; the largest differences are driven by highly localised combinations of topography, oceanographic forcing and data gaps in land-elevation models.
Reactions & quotes
“In reality, sea level is influenced by additional factors such as winds, ocean currents, seawater temperature and salinity,”
Dr Philip Minderhoud, Wageningen University (study lead)
Context: Minderhoud highlighted that geoid-based baselines do not capture dynamic and thermosteric effects that change local sea-surface height. The study authors argue that failing to account for these factors leads to systematic underestimation of coastal exposure.
“If sea level is higher for your particular island or coastal city than was previously assumed, the impacts from sea level rise will happen sooner than projected before,”
Dr Philip Minderhoud, Wageningen University
Context: The point underscores how a revised baseline shortens adaptation timescales for at-risk communities and could require acceleration of planned interventions.
“An interdisciplinary blind spot”
Study authors (summary phrasing)
Context: The authors use this phrase to describe routine methodological mismatches between terrestrial elevation products and oceanographic sea-surface observations, and call for tighter cross-discipline calibration.
Unconfirmed
- Extent to which every high-profile IPCC-cited regional study is biased by geoid-referenced elevations is not fully enumerated; the authors note many references but not an exhaustive list tied to IPCC chapters.
- Some extreme local discrepancy figures (550–760cm) appear as outliers; the precise locations and causes of the largest mismatches require further local verification and peer review.
- How quickly national planning agencies will adopt the study’s dataset and revise official hazard maps is uncertain and will depend on institutional capacity and data access.
Bottom line
The study presents robust evidence that many coastal elevation baselines used in the scientific literature and applied assessments have been biased low because they rely on geoid-referenced land heights rather than local sea-level measurements. That bias implies that hazard maps, exposure counts and timelines for impacts could be systematically optimistic, particularly in densely populated, low-lying regions of the global south.
Practical follow-up steps include independent verification of large regional discrepancies, the incorporation of the study’s integrated coastal elevation dataset into national hazard assessments, and a methodical review of past exposure studies cited in major assessments. Policymakers, planners and insurers should treat baseline recalibration as a priority to ensure adaptation investments and timelines reflect the true starting point for sea-level impacts.