How fast did sea levels rise after the last ice age?
After the last ice age, some 11,700 years ago, sea levels rose rapidly due to global warming and the melting of the huge ice sheets that covered large parts of North America and Europe. During the early Holocene, the period after the last ice age, sea levels rose by about one metre a century at some stages. This rapid rise is seen as an important reference point for predicting future sea level rise, particularly now when we are seeing a similar situation with rapidly melting ice sheets due to global warming.
Until now, the rates and extent of sea level rise during the early Holocene were poorly understood because of a lack of sound geological data from this period. Researchers from Deltares and, among others, Utrecht University, the Netherlands Organisation for Applied Scientific Research (Geological Survey of the Netherlands), Delft University of Technology, the Netherlands Institute for Marine Research (NIOZ), Wageningen University and Research and the University of Amsterdam have now been able to draw on a unique dataset from the North Sea region to make highly accurate estimates of sea level rise in the early Holocene for the first time. The results of their study were published today in the leading journal Nature. Marc Hijma, a geologist at Deltares and the lead author of the study: “With this groundbreaking research, we have taken an important step towards a better understanding of sea level rise after the last ice age. By drawing on unique data from the North Sea region, we can now better understand the interaction between ice sheets, climate and sea level. This provides insights not only for scientists but also for policymakers so that we can prepare better for the impacts of current climate change.”
I think we can say that we now have a good understanding of North Sea drowning over the entire offshore area.
Global sea level rise: 38 metres
The researchers analysed data from the area that was once Doggerland, a land bridge between Great Britain and mainland Europe. This area flooded as sea levels rose. By analysing the peat layers from this area, dating them and applying modelling techniques, the researchers showed that, during two phases in the early Holocene, global sea level rise briefly peaked at more than a metre per century. By comparison, the current rate of sea level rise in the Netherlands is about 3 mm annually, the equivalent of 30 centimetres per century, and is expected to increase. The researchers also concluded that global sea levels rose by about 38 metres between 11,000 and 3,000 years ago, clarifying previous discrepancies between ice sheet reconstructions and sea level data for this important period.
Text continues below image
Better understanding of sea level rise
The study helps to close the gap between previous reconstructions of ice sheets and sea level data, which were often contradictory. The findings are of major importance for our current understanding of sea level rise and they provide valuable insights for the future. Given the current rise in greenhouse gas concentrations, climate models indicate that sea levels will rise several metres by 2300, with some scenarios anticipating a rise of more than one metre per century. The rate of sea level rise in the early Holocene now provides an important reference point for scientists and policymakers, allowing them to better understand, and prepare for, expected rises in the future.
Differences now and then
An important difference with the early Holocene is that the consequences of sea level rise are far greater today and in the future. This is due to a growth in population and the current presence of infrastructure, cities and economic activity.
Text continues below image
Legacy data and new data
From Utrecht University, physical geographer Kim Cohen contributed to the research. His expertise is landscape change in north-western Europe from the ice ages to the present. Enthusiastically, he points to the care taken when compiling the dataset of cores with peat dates from the seabed. ‘I’m fond of the older North Sea data, collected from 1960 to 2000 and happy that we did not simply replace that with ‘better, newer’ data collected by ourselves. From the outset, we have strived to use the new data to determine what old data stand up. That worked out well.’
Careful assembly of the data set
The legacy data had been collected from many scattered locations: random hits in individual sea-bed cores that happened to bring up a layer or lump of peat. ‘The new data, in contrast, was gathered with a keen strategy: using seismics, we tracked down locations with peat layers that ran up against buried topography of former small river valley sides. Then the ship turned around and immediately took series of seabed cores to harvest a cluster of data points over a depth range of a few meters. With that new data, for many areas we could then filter the loose old points on quality: where they plot in-between series of new points: that's good, they stay in the list. Where they don't: we no longer trust the old ones, and they drop out.’ The overall dataset for the area this way became a carefully worked-up construct, along the yardstick of international protocols. ‘We also treated the depth uncertainties of points collected longer ago very systematically, in the same way as for our new measurements. And because we looked at the data points as series, we were also able to constrain the radiocarbon-dating outcomes more narrowly than we could when the legacy data was originally collected. That all helped improve our goal of determining the global sea level history for the period a lot more accurate than was possible before.’
Open access and many applications
The dataset accompanying the article, both published open access, thus unlocks new and legacy data neatly lined up, bundling all information relevant to North Sea drowning. ‘I think we can say that we now have a good understanding of North Sea drowning over the entire offshore area. Before, we understood it well for the last 8,000 years or so, the time frame that the coastline had reached the modern shores, where data is more easily in reach. But now, we have extended that to 11,000 years ago when coastlines were still far out, beyond the Dogger Bank. From Doggerland to the situation of today as a continuum, so to say. This is, of course, useful for coastal and offshore research for in and along the North Sea. For instance, think about better mapping out the disappearance of Doggerland, as an environmentally and archaeologically lost world. One can also use it to renew research on when and how the Wadden Isles first came to be. And the work includes findings on how rapidly the Netherlands and the North Sea floor are naturally subsiding, in the past even faster than today. In turn, this will be vital input to addressing present-day human-caused contributions to land subsidence and sea-level rise.”
Teamwork
Cohen highlights the international cooperation and teamwork. ‘The paper married results from German and Dutch sea-going research. Likewise, the geophysical modelling of sea level change and vertical land motions due to the disappearing European ice sheets - super important in the analysis steps - was the result of international teamwork with contributions from English, German and Dutch researchers. We worked some eight years on it, on and off. All of it grew as a spontaneous initiative making use of personal and institutional contacts of the authors. But setting drill cores in the North Sea is bottom-up research anyway, of course.’
Publication
Marc P. Hijma, Sarah L. Bradley, Kim M. Cohen et al., ‘Global sea-level rise in the early Holocene revealed from North Sea peats’, Nature, https://doi.org/10.1038/s41586-025-08769-7
Some related North Sea Holocene sea-level rise publications:
de Wit, K., Cohen, K. M., & Van de Wal, R. S. W. (2025). HOLSEA-NL: a Holocene water level and sea level indicator dataset for the Netherlands. Earth System Science Data, 17, 545–577. https://doi.org/10.5194/essd-17-545-2025
Hoebe, P. W., Cohen, K. M., Busschers, F. S., van Heteren, S., & Peeters, J. H. M. (2024). Early Holocene inundation of Doggerland and its impact on hunter-gatherers: An inundation model and dates-as-data approach. Quaternary International, 694, 26-50. https://doi.org/10.1016/j.quaint.2024.05.006
[Dutch] Amkreutz, L. W. S. W., Cohen, K. M., Hijma, M. P., & Ode, O. (2021). Verdrinkend land in kaart. In L. Amkreutz, & S. van der Vaart-Verschoof (Eds.), Doggerland: Verdwenen wereld in de Noordzee (pp. 33-37). Sidestone Press. https://www.sidestone.com/books/doggerland-verdwenen-wereld-in-de-noordzee
Hijma, M. P., & Cohen, K. M. (2019). Holocene sea-level database for the Rhine-Meuse Delta, The Netherlands: implications for the pre-8.2 ka sea-level jump . Quaternary Science Reviews, 214, 68-86. https://doi.org/10.1016/j.quascirev.2019.05.001
Vermeersen, B. L. A., Slangen, A. B. A., Gerkema, T., Baart, F., Cohen, K. M., Dangendorf, S., Duran-Matute, M., Frederikse, T., Grinsted, A., Hijma, M. P., Jevrejeva, S., Kiden, P., Kleinherenbrink, M., Meijles, E. W., Palmer, M. D., Rietbroek, R., Riva, R. E. M., Schulz, E., Slobbe, D. C., ... Van Der Wegen, M. (2018). Sea-level change in the Dutch Wadden Sea. Geologie en Mijnbouw/Netherlands Journal of Geosciences, 97(3), 79-127. https://doi.org/10.1017/njg.2018.7
Hijma, M. P., & Cohen, K. M. (2010). Timing and magnitude of the sea-level jump preluding the 8,200 yr event. Geology, 38(3), 275-278. https://doi.org/10.1130/G30439.1