• • 대부분의 기존 연구가 전 세계 해안 해수면 높이를 평균 0.3m 정도 과소평가했을 가능성이 있다는 주장이 5일 '네이처'에 발표됩니다. 개발도상국 일부 지역에서는 실제 해수면 높이가 기존 추정치보다 최대 1m 더 높을 수도 있다는 뜻입니다.
    • 논문 제목: 'Sea level much higher than assumed in most coastal hazard assessments' (Nature)
    • (엠바고 5일(목) 01시 KST 해제)
  • 네덜란드 연구팀은 2009년부터 2025년까지 발표된 해안 노출 및 위험 영향 평가에 관한 논문 385편을 분석하고, 메타 분석을 통해 일반적으로 가정되는 해안 해수면과 실제 측정된 해수면 간의 차이를 계산했습니다.
  • 그 결과, 전체 연구의 90% 이상이 실제 측정 해수면이 아닌 중력 모델(지오이드)에 기반한 가정 해수면에 의존한 것으로 나타났습니다.
  • 해수면은 사용된 지오이드 모델에 따라 평균 0.24~0.27m 정도 과소평가됐으며, 일부 연구에서는 최대 5.5~7.6m의 차이가 발생했음을 확인했습니다.
  • 해수면 상승 및 기타 해안 위험 영향 평가 역시 평균적으로 실제 해수면보다 낮은 해안 해수면을 사용해 계산된 것으로 나타났습니다.
    • 결과적으로 해안 노출을 과소평가했다고 연구팀은 주장했습니다.
  • 연구팀은 기존 값을 재평가하고, 향후 해안 재해 평가에서 해수면 높이 측정값과 해안 고도를 정확하게 결합해 기후 변화가 해안 지역에 미치는 영향을 보다 정확하게 파악해야 한다고 강조했습니다.

*2026.3.4.

Comment author (incl affiliation): Bill Hare, CEO, Climate Analytics
Contact details: press@climateanalytics.org
Conflicts of interest:
Available for interviews in English. Based in Berlin, CET

Statement:

“This is a groundbreaking piece of science that has clearly taken some years to put together and is extremely robust. This paper brings some profound bad news to the assessment of sea level rise and coastal impacts. It essentially shows that most previous assessments of the impacts of sea level rise based on satellite data have likely systematically underestimated exposure to severe risks as well as the rate at which these risks will increase in the future.

“Very substantial progress has been made in projecting sea level rise in recent years, most of which has shown that sea level rise is likely to be higher than previously thought. Previous research was already showing we are at major risk of sea level rise of up to one metre by 2100. This new science shows that the projected impacts from SLR are likely to be much higher than previously thought. And extra risk of damage is much greater than previously forecast, particularly in the most vulnerable countries, especially small island states, low lying developing countries and the large archipelago of Southeast Asia.”

Results

“The methods and results in this paper appear quite robust and challenge sea level rise impact assessments based on satellite data. The geoid method that has been used historically worked very well in the global north where most research has originated. But few were aware that in other parts of the world, this method has been producing very large discrepancies.”

Sea level rise

“Based on the evaluation made from literature, it is clear that the impacts of this underestimation of satellite-based sea level rise impact assessments are very large and very significant. This is particularly true for the developing countries where the gap between sea levels estimated by land or sea is the largest. We have always known that sea level rise is a relative phenomenon – sea levels do not rise evenly across the world and in some places will go down.  Properly estimating sea level rise impacts requires taking into account the position on the Earth, and whether or not land is rising or subsiding – often due to human effects.

“The results here could add a staggering new dimension to sea level rise impact assessments – particularly given that many impact assessments cited by the authors as underestimations have been used to guide coastal impact, vulnerability and adaptation risks and measures. The two regions where the gap in estimation is the largest, Southeast Asia and the Pacific, are home to many Small Island Developing States (SIDS) and Least Developed Countries (LDCs), which were already the most vulnerable to the impacts from sea level rise and other climate impacts. The paper identifies low-lying atoll countries in the Pacific as places where the assumed sea levels are lower than actual measurements (in two out of three scenarios). Under these new circumstances, they will face increased vulnerability and likely higher costs to deal with adaptation and minimise the impacts of disasters. The urgency to mitigate emissions to minimise sea level rise impact and improve access to finance for adaptation and loss and damage is ever clear.

“Germany and Northwest Europe, according to the authors Figure 3 are amongst the least affected globally. The largest number of people affected are in the Global South.

“Given the significance of what has been discovered there needs to be an urgent program to review, and reanalyse and assessments to determine is coastal adaptation strategies need to be updated and timelines expedited.”

Reporting

“The study argues for the application of new methodologies based on their findings. The authors themselves make clear that this does not mean that the majority of the evaluated studies make errors, or that the IPCC reports that cite these studies are wrong or are containing errors. This is important to bring across in the reporting. Otherwise the study may be misused to disqualify previous insights on SLR or the scientific process as a whole, undermining the basis for climate action.

“This study is part of the scientific process in which the evidence base on climate change, impacts and risks continues to evolve based on new evidence and methodologies. If anything, this new assessment is in line with other new findings in relation to climate change, where risks are increasingly being assessed to occur sooner than previously estimated.

“This is not a matter of disqualifying the science so far, but a further line of evidence that shows that we must err on the side of caution when it comes to climate impacts and act sooner rather than later to curb adverse impacts, of which SLR is one of the most severe.”

Comment author (incl affiliation): Gabriel Mara, Pacific Adaptation and Loss and Damage Analyst, Climate Analytics
Contact details: press@climateanalytics.org
Contact Note: Gabriel is currently in Tuvalu (GMT+12)
Conflicts of interest: Gabriel has previously served as Climate Change Officer for the Fiji Government, supporting national adaptation policy, the NAP process, the development of Fiji’s SOP for Planned Relocation and the operationalisation of the Climate Relocation of Communities Trust Fund. Internationally, he has represented Fiji as an Adaptation Negotiator under the UNFCCC and later served as G77 and China Coordinator for NAPs.

General comment from a Pacific perspective

“This paper finds something quite concerning; many of the projections have underestimated both what we are currently measuring and what we are likely to experience in the future.
“For Small Island Developing States (SIDS), that is far from a discrepancy.
“Because if the projections are conservative, then the policies, infrastructure standards and adaptation plans built on those projections are conservative also. And in our (SIDS) context, conservative does not mean safe. It means SIDS are further behind then currently believed.
“This is serious for Pacific SIDS: as parts of the western tropical and southwest Pacific have historically experienced rates of sea-level rise above the global average due to ocean dynamics. Sea level is not uniform, it shifts with currents, heat and gravity. So while globally the numbers can look one way, locally the lived experience is very different.
“The paper notes (in fig. 4) that compared to other regions, the Pacific shows lower exposure in aggregate terms. But we have to interrogate what “lower exposure” means. If exposure is measured by total population or absolute land area, then yes, small islands will appear smaller in the dataset. But proportionally? Existentially? Losing 1% of land in a large continental country is not the same as losing 1% of a small island state. In our context, that 1% could be an entire village or half a town, a burial ground, a freshwater lens and livelihood for subsistence survival.
“And this is where the implications move past adaptation planning. If projections and threshold triggers are too low, adaptation becomes reactive rather than anticipatory. You build a seawall for yesterday’s sea level. This brings soft and hard limits much earlier, compressing the window between ‘adaptable’ and ‘unavoidable loss’. This leaves residual risk higher than planners assume as countries will face greater losses than projected because hazard baselines are underestimated.
“Even with the levels and projections that are considered standard references, Pacific communities are already facing the impacts of sea-level rise. Right now, coastal erosion and saltwater intrusion is actively occurring in the Pacific. Sacred sites are already being relocated. People are already forced to leave their ancestral lands.
“If sea-level rise is happening faster than what is currently thought, then losses including non-economic losses are also happening faster than our planning and anticipatory frameworks assume or are equipped for. Non-economic loss and damage is rising quietly alongside the water, unavoidable losses will accumulate sooner and at greater scale. That directly affects discussions around Loss and Damage finance, responsibility and equity.
“While the study does not directly assess whether the underestimated projections feed into IPCC assessments, it does highlight something important. Many national policies rely heavily on frequently referenced sources like the IPCC as the global scientific benchmark. If the studies it uses are conservative and synthesise the same into reports, the true description of what the reality is has been severely limited as the panel operates on assessment cycles.
“For SIDS, that suggests we need to read the science carefully, especially like that of the Small Islands chapter, and ask whether we are planning for the median, or for the reality experienced on the ground.
“Can SIDS and coastal states only plan for the median of current models, or can they be aided to prepare for what is increasingly observable in front of them?”  

Professor Tim Naish, Antarctic Research Centre, Te Herenga Waka, Victoria University of Wellington, comments:
Contact: +64 27 235 8101, timothy.naish@vuw.ac.nz (Professor Naish is currently in Japan, 4 hours behind NZ, but is happy to be contacted)

"This article suggests many local coastal hazard impact assessments around the world have underestimated future local sea-level rise, and the related risks, by not taking into account local factors such as land subsidence (sinking) and relying on inaccurate land elevation data.

"Fortunately, Aotearoa New Zealand is not one of those countries. For the majority of our coastline where people live, Land Information New Zealand have generated highly accurate land elevation data by using lasers from aircraft to measure distances. This is used widely in coastal hazard risk assessment.

"The NZSeaRise and Our Changing Coast programmes funded by MBIE Endeavour Fund over the last ten years provide location-specific projections of sea-level change to the year 2150 for the climate scenarios outlined in the IPCC's 6th Assessment Report. These sea-level rise estimates are currently available at 2km spacing along the coastline, but a new version with 200m resolution is about to be released.

"The sea-level projections allow users to estimate relative sea-level rises which account for global (e.g. melting land ice) and regional (e.g. corresponding large scale adjustments in Earth’s surface) changes as well as local changes from local vertical land movements. These vertical land movements are measured from satellite-borne radar systems and permanent GPS receivers across the country, yielding movement rates with the accuracy of +/-1 mm/year.

"In between earthquakes, 70% of the coastline of Aotearoa New Zealand is subsiding due to ongoing tectonic processes, fluid extraction and ground compaction. This land subsidence can be as high as the rate of global sea-level rise itself (+4 mm/year), significantly increasing the risk of, and reducing the time before, critical thresholds are breached.

"An online tool for the public and decision makers is also referred to in the Ministry for the Environment's Coastal Hazard and Climate Change Guidance. Councils and infrastructure providers around the motu are now using these estimates that include VLM for their planning work and risk assessments, thereby ensuring adaptation solutions can be found for those at risk communities and assets."

Conflict of interest statement: "I am the Co-Principal Investigator of the NZSeaRise and Our Changing Coast programmes, as well as the Chair of the World Climate Research Programme (WCRP) which helps co-ordinate the science for IPCC sea-level projections."

Gabriel Mara, Pacific Adaptation and Loss and Damage Analyst, Climate Analytics, comments:
Contact: +67 9 867 0756, press@climateanalytics.org (Gabriel is currently in Tuvalu, 1 hour behind NZ)

Note: This comment has been abridged. The full comment is available on Scimex.

“This paper finds something quite concerning; many of the projections have underestimated both what we are currently measuring and what we are likely to experience in the future. For Small Island Developing States (SIDS), that is far from a discrepancy. Because if the projections are conservative, then the policies, infrastructure standards and adaptation plans built on those projections are conservative also. And in our (SIDS) context, conservative does not mean safe. It means SIDS are further behind then currently believed.

“This is serious for Pacific SIDS: as parts of the western tropical and southwest Pacific have historically experienced rates of sea-level rise above the global average due to ocean dynamics. Sea level is not uniform, it shifts with currents, heat and gravity. So while globally the numbers can look one way, locally the lived experience is very different.

“The paper notes (in fig. 4) that compared to other regions, the Pacific shows lower exposure in aggregate terms. But we have to interrogate what “lower exposure” means. If exposure is measured by total population or absolute land area, then yes, small islands will appear smaller in the dataset. But proportionally? Existentially? Losing 1% of land in a large continental country is not the same as losing 1% of a small island state. In our context, that 1% could be an entire village or half a town, a burial ground, a freshwater lens and livelihood for subsistence survival.

“While the study does not directly assess whether the underestimated projections feed into IPCC assessments, it does highlight something important. Many national policies rely heavily on frequently referenced sources like the IPCC as the global scientific benchmark. If the studies it uses are conservative and synthesise the same into reports, the true description of what the reality is has been severely limited as the panel operates on assessment cycles.

“For SIDS, that suggests we need to read the science carefully, especially like that of the Small Islands chapter, and ask whether we are planning for the median of current models, or for the reality experienced on the ground."

Conflict of interest statement: "Gabriel has previously served as Climate Change Officer for the Fiji Government, supporting national adaptation policy, the NAP process, the development of Fiji’s SOP for Planned Relocation and the operationalisation of the Climate Relocation of Communities Trust Fund. Internationally, he has represented Fiji as an Adaptation Negotiator under the UNFCCC and later served as G77 and China Coordinator for NAPs."

Dr Emma Ryan, Senior Lecturer in Environmental Science at the University of Auckland, comments:
Contact: +64 21 0876 8939, e.ryan@auckland.ac.nz

"This study is a welcome addition to global sea‑level rise research. Many worldwide assessments rely on coarse resolution global satellite data. By using a more accurate and globally-consistent way of measuring coastal land elevation, the authors show that earlier global studies may have underestimated how many places are exposed to future sea‑level rise. This is an important improvement in our global understanding.

"However, global studies can only take us so far. They are not designed for local planning or decision‑making. While they can highlight global and regional differences, the impacts of sea level rise on coasts vary a lot from place to place, depending on factors like land sinking or rising (vertical land motion, often caused by tectonic factors), sediment movement, storm patterns, waves, and long‑term shoreline change patterns.

"For Aotearoa New Zealand and Pacific Island nations, high‑quality local data remains essential. New Zealand and Pacific Island nations need more investment in obtaining and maintaining accurate elevation datasets for both the coastal zone and the nearshore. New Zealand is already investing in high-quality local data. The New Zealand Coastal Change Database (coastalchange.nz) provides long‑term, nationally consistent coastal erosion rates that can be used to underpin local or regional coastal hazard and risk assessments.

"Other national research programmes are updating localised sea‑level rise projections with vertical land movement data, which will give communities more accurate information for regional and local planning. New Zealand and Pacific Island nations should prioritise local data for decision making about sea-level impacts."

Conflict of interest statement: "No conflicts of interest. I currently have research projects funded by the New Zealand government (e.g. Natural Hazards Resilience Platform, MBIE) and Department of Conservation, and the Coastal People: Southern Skies centre for research excellence. I am a researcher on the project linked to coastalchange.nz."

Dr Dalila Gharbaoui, Political and Social scientist, Adjunct Research Fellow, University of Canterbury, comments:
Contact: +64 22 679 5320, dalila.gharbaoui@canterbury.ac.nz

“This study offers an important correction to how we understand coastal risk. By showing that more than 99% of global coastal hazard assessments have misinterpreted the relationship between sea‑level height and land elevation, the authors highlight a major methodological issue in how exposure is calculated. It indicates that a significant number of global and regional coastal hazard assessments may be underestimating current exposure, particularly in the Indo‑Pacific.

“For Pacific Island nations and for vulnerable coastal regions in Aotearoa New Zealand, these findings suggest that some current hazard maps and planning tools may be underestimating present‑day risk. Many communities already experiencing more frequent inundation, shoreline loss, or saltwater intrusion may be living with conditions more severe than what official assessments show. For governments across the region, this means some planning frameworks, coastal hazard maps, and investment decisions may be based on outdated baselines. This has direct implications for land‑use planning, infrastructure design, and long‑term adaptation pathways.

“This is also a reminder that climate impact assessments are continually improving as data and modeling advance, but this study reinforces a broader issue raised in climate mobility research: many widely used assessments rely on global averages or projections that do not fully capture local realities, creating a tension between modeled definitions of "habitability" and the conditions communities are already experiencing on the ground.

"These findings reinforce the urgent need to reassess existing coastal hazard maps, infrastructure planning, and adaptation strategies. They also carry major implications for climate finance. Funding decisions must reflect the accurate level of risk communities face today, as these gaps influence how we quantify loss and damage, including cultural and non‑economic losses central to communities' identity and wellbeing. Accurate baselines are essential for ensuring climate‑finance mechanisms like the Loss and Damage Fund direct support in proportion to the real risks faced by communities.

"This study highlights the need for locally grounded, regularly updated risk assessments to inform durable, equitable, well‑targeted adaptation policy and effective climate‑finance decisions."

Conflict of interest statement: "N/A."

Professor Silvia Serrao-Neumann, Environmental Planning, The University of Waikato, comments:
Contact: +64 7 837 9171, silvia.serrao-neumann@waikato.ac.nz

"My quick read of the paper indicates that it won't affect Aotearoa as much because it focuses on the implications for the Global South, especially low-lying coastal areas in Southeast-Asia.

"Aotearoa has the Searise portal which accounts for the vertical land movement cited in the paper as one of the critical errors. In summary, we don't lack the information on the impacts of sea level rise on our coasts. Our impediment is more of a political nature, as in being more proactive in decisions to avoid further developing coastal areas at-risk of sea level rise and related coastal hazards such as erosion and inundation."

Conflict of interest statement: "I don't have any conflict of interest to declare."

  

Prof Andrew Shepherd, Director of the Centre for Polar Observation and Modelling, Northumbria University, said:

“Sea levels are much higher than we had thought – about 30 cm higher – due to the swell of ocean currents. This means that 80 million people are living below sea level today, 50 million more than we had realised. Although this is bad news, it raises an important question: how do these communities cope – especially where sea defences have not been built? With around a metre of further sea level rise already locked in from existing global warming, what they are doing today may be exactly what the rest of the world needs to learn.”

 

 

Dr Natasha Barlow, Technical Director, Costal Resilience, Haskoning engineering consultancy, said:

“This study highlights the continuing need for consistent treatment of datasets, so they are comparable and use the same reference points. Transitions from national reference points (datums) to global datums, along with varied use of satellite data, GPS and models, mean this has not always been applied consistently or accurately, as demonstrated in the authors’ detailed assessment.

“The study focuses on global datasets, which are valuable for providing a broad picture of climate‑related hazards and risks. However, these datasets carry considerable uncertainty because of the limited precision achievable when estimating land‑level elevation at a global scale. They often assume coastal flooding behaves like a simple ‘bathtub’ model. Such global elevation models cannot account for the detailed factors essential in city‑, community‑ or region‑scale flood assessments; for example, interconnected water systems and groundwater storage, coastal erosion, or the socio‑economic circumstances that shape people’s ability to respond to flooding and adapt to climate change.

“The authors also emphasise an important point: these issues disproportionately affect the Global South. This is due not only to inconsistent application of reference points in coastal datasets, but also to limited availability of high‑quality elevation data (such as LIDAR) needed for accurate regional and local flood‑risk estimates. Many densely populated delta and estuary regions in the Global South face significant risk from sea‑level rise while lacking the resources required to respond effectively to increasing flood hazards.”

 

 

Prof Jonathan Bamber, Director of the Bristol Glaciology Centre, University of Bristol, said:

“Sea level rise (SLR) is one of most serious and certain consequences of global heating. It is a field I have worked in for some two decades and I was genuinely surprised by the results of this study, which is not about reassessing how much SLR there might be in the future but about using the correct estimate of present-day sea level. This might seem obvious, but unless practitioners and others working in the field understand the nuances of the different types of reference surfaces used and how they relate to coastal sea level it is easy to miscalculate the offset between coastal elevation and the sea surface height in that location.

“What the authors demonstrate is that in the majority of impact studies, that is exactly what has happened. The wrong assumptions are made about what present-day sea level is and it turns out that it has generally been underestimated in key sensitive coastal areas. This has important implications for impacts of future SLR in terms of the area and number of people potentially affected in low lying areas such as south east Asia and the Nile delta. It does not affect how much SLR might occur in the future.

“The authors estimate that the offset between the true sea surface height and that used in impact studies is around 24-27 cm. To put that in context, that is more than the total SLR that has occurred since the beginning of the 20th century. They highlight an important limitation in most coastal impact studies but their conclusions about how this will affect estimates of future coastal inundation are less certain. First, they assume a higher accuracy for estimates in mean sea surface height at the coast than is actually the case. The best way to assess this is against tide gauge data, which are actually located at the land/ocean boundary. Second, the elevation of the land surface in the coastal zone has relatively large errors, which compounds the difficulty in determining the relative difference between the height of the land and sea at the coast.”

 

 

Prof Daniela Schmidt, Professor of Earth Sciences, University of Bristol, said:

“Methodological improvements of sea level rise are important due to the concentration of people, settlements, infrastructure and biodiversity hotspot in these places where land meets the sea. The study claims to quantify  SLR impacts and coastal hazard assessment. Like so many others, it solely sees the impact of sea level rise and the resulting losses and damages on people via land loss and impacts on settlements. These are important and need long term consideration such as the impact of sea level rise on the Thames Estuary and potential need for further infrastructure. The timing for adaptation options will ultimately depend on the rate of sea level change, something this paper is not addressing. 

“But coastal sea level rise will also alter our coastlines, result in erosion, and destroy important ecosystems which in turn also provide protection for our coastlines. It is fundamental to sea level rise projections consider the losses holistically and consider the impact drowning our wetlands and mudflats, or losing mangroves and coral as they cannot keep up with sea level rise, will have on natures ability to draw down CO2, to protect the costs and to nourish humans and biodiversity. No risk assessment is complete without a more holistic view.”

 

 

Declared interests

Andrew Shepherd: “No competing interests.”

Jonathan Bamber “is a professor of Earth Observation and glaciology at the University of Bristol and a visiting fellow at the Technical University Munich (TUM) and receives funding from the European Space Agency on a study aimed at closing the sea level budget globally and regionally. He also receives funding from TUM on coastal sea level and the EU on glaciers and ice sheets.”

Daniela Schmidt: “No competing interests”

Natasha Barlow “is Technical Director at Haskoning, a global engineering and environmental consultancy whose clients include coastal communities, policy makers, and authorities facing risks from sea‑level rise. Natasha previously spent 20 years as an academic. She has no conflict of interest in this study and had no involvement in the work. While Haskoning uses datasets of this kind in its projects, Natasha’s perspective on the research reflects her own experience of addressing complex local and regional challenges and the importance of using the highest‑quality datasets in hazard assessment.”

For all other experts, no reply to our request for DOIs was received.

  

• 한국과학기술미디어센터 초대 이사장에 노정혜 서울대 교수(조선일보)
• 한국과학기술미디어센터장에 이근영 사회평론아카데미 기획위원(연합뉴스)
• 언론-과학자 잇는 ‘가교’ 되겠다…한국과학기술미디어센터 개소(매일경제)
• 대형 산불 때 정치인 찾아가던 기자들이 달라졌다… 과학기술미디어센터, 과학 보도를 바꿔놓다(한국일보)

• 과학자, 연구자에게는 의견과 해설이 온전한 맥락과 함께 제공되는 안전한 발언 공간이 돼줍니다. 선의를 위해 한 논평이 기사화 과정에서 왜곡되거나 부정확하게 변질될 우려를 줄이는 완충 작용을 합니다. 
• 기관 홍보 담당자에게는 기관의 성과를 기자들에게 보다 객관적이고 정교하게 알리고, SMC 글로벌 네트워크를 통해 영향력을 높일 기회를 제공합니다. 
• 기자에게는 사안을 해석하는 데 도움이 될 치우침 없는 종합적인 정보를 빠르고 풍성하게 제공하고, 이를 통해 기사에서 과학과 기술을 보다 자유롭고 편리하게 활용하도록 돕습니다.

지속적으로 SMCK의 전문가 해설 및 의견을 받아보길 원하신다면 아래 링크에 등록해주세요.
📌 기자 등록하고 전문가 의견 받기 >