> ## Content Index
> Fetch the complete content index at: https://romanempiretimes.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Arctic Ice Grew in Roman Times. Modern Warming Reversed It.
- URL: https://romanempiretimes.com/roman-era-ice-age-arctic-warming/
- Published: 2026-11-01T11:59:00.000Z
- Updated: 2026-11-01T11:58:59.000Z
- Description: New Arctic evidence places renewed ice growth in Rome’s imperial age. The findings connect ancient snowlines, the Little Ice Age and modern warming; but scientists differ on when a glaciation begins.
- Author: George Liapis
- Tags: Roman Empire News, Alps, Ice Age, Climate Change, Barnes Ice Cap, Baffin Island, Laurentide Ice Sheet, Little Ice Age, Late Antique Little Ice Age

Arctic ice was beginning to spread again around the time of the Roman Empire, according to new research into the last surviving remnant of North America’s great ice sheet. The researchers interpret that renewed growth as the earliest stage of another glaciation, interrupted by modern greenhouse warming.

The evidence comes from Canada’s Barnes Ice Cap, on Baffin Island, rather than from Roman territory. Published in *Science* on 22 October 2026, the study reconstructs a reversal from shrinking ice to renewed expansion roughly 2,000 years ago, followed by another reversal around 1960.

That places the proposed beginning in the broad historical setting of [the age of Rome’s first emperor](https://romanempiretimes.com/from-octavian-to-augustus-the-making-of-romes-first-emperor/). Whether this deserves the name “next ice age” depends partly on where scientists draw the starting line: the first returning ice, or the eventual growth of continental ice sheets.

![Landsat view of the pale Barnes Ice Cap surrounded by brown Baffin Island terrain, with branching meltwater channels etched across its surface.](https://storage.ghost.io/c/aa/80/aa80281a-abce-4f35-865f-05153c8cc335/content/images/2026/10/barnes-ice-cap-baffin-island-july-2026.jpg)

Barnes Ice Cap on Baffin Island, photographed by Landsat 9 on 12 July 2026\. Meltwater channels run across the exposed ice. Image: [NASA Earth Observatory / Lauren Dauphin](https://science.nasa.gov/earth/earth-observatory/a-bare-summer-for-barnes-ice-cap/?ref=romanempiretimes.com), using U.S. Geological Survey Landsat data.

## The plants that recorded the returning ice

The Barnes Ice Cap preserves ice inherited from the Laurentide Ice Sheet, which once covered much of Canada and the northern United States. At the frozen-bed locations studied, advancing ice buried mosses and small plants without grinding them away. Retreat has exposed those plants again, making their deaths a record of the ice margin’s advance.

The team combined 88 new radiocarbon dates from vegetation with beryllium-10 measurements from exposed rocks. The rock evidence places retreat behind the modern margin about 3,560 years ago. A sequence of plant dates records renewed advance from around AD 1000, with most sampled plants overrun during the last few centuries.

The earlier, Roman-era starting point comes from combining that Barnes record with 186 previously published plant dates from nearby ice caps and glaciers. Together, they indicate that the regional snowline descended onto Baffin’s plateau around the beginning of the Common Era, allowing ice to expand across the landscape again.

![A large boulder balanced on smaller stones above bare pink bedrock, with a low ice cap extending across the background on Baffin Island.](https://storage.ghost.io/c/aa/80/aa80281a-abce-4f35-865f-05153c8cc335/content/images/2026/10/baffin-island-cold-based-ice-boulder-miller-2023.jpg)

A boulder left by the Laurentide Ice Sheet rests on striated bedrock beside an ice cap in northern Baffin Island. Its survival illustrates how ice frozen to its bed can preserve a landscape. Photograph: Gifford H. Miller, [Miller et al. (2023), Figure 3](https://cp.copernicus.org/articles/19/2341/2023/?ref=romanempiretimes.com), [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/?ref=romanempiretimes.com).

## Summer decided whether the snow would stay

The decisive balance was between winter accumulation and summer melting. Changes in Earth’s orbit gradually reduced summer sunlight at high northern latitudes. As summers cooled and the snowline fell, more snow could survive from one year to the next, supporting renewed ice growth.

Snow and ice also reflect sunlight, reinforcing cooling as their coverage expands. The new study identifies Baffin’s broad plateau and surviving Laurentide ice as a favourable setting for this process. It proposes that the resulting growth represents glacial inception: the beginning of a much longer ice-building sequence.

By around 1960, rising summer temperatures had shifted the Barnes Ice Cap into retreat. The researchers associate that reversal with human-driven greenhouse warming overwhelming the orbital cooling influence. A separate Mediterranean study reconstructs a regionally warm interval during AD 1–500, so Arctic ice growth should not be read as a description of the weather throughout Rome’s empire.

Professor Gifford Miller explains how plants emerging from Baffin Island ice preserve evidence of past climates. This earlier field film was published by [INSTAAR on 16 January 2014](https://www.youtube.com/watch?v=B2hbesz-Npk&ref=romanempiretimes.com). Film: Matthew Kennedy / Earth Vision Trust.

## Can a single winter begin an ice age?

The first layer may be a single winter’s snow. What makes it the start of lasting glaciation is its survival through summer and the addition of snow in subsequent years. Baffin Island modelling shows how cooler summers and snow’s reflectivity can encourage that persistence, while suggesting that inception may require a long sequence of moderately cold, wet years.

Research does show that parts of the climate system can change within a year. Annual sediment layers in Germany’s Lake Meerfelder Maar record a sudden shift to stronger autumn-to-spring storminess at the onset of the Younger Dryas cold period, roughly 12,700 years ago. The researchers interpret it as a rapid change in North Atlantic winds. [Roman methods of coping with winter cold](https://romanempiretimes.com/unlike-other-civilizations-the-romans-used-these-methods-to-cope-with-winter-cold/) belong to the seasonal experience of an empire whose regions could follow different climatic patterns.

## What the Würm glaciers covered

The last glacial period, known in the Alps as the Würm, produced a very different landscape from the Roman-era Arctic advance. Modern reconstructions use dated glacier limits and the term Last Glacial Maximum.

A global synthesis places most ice sheets near their maximum extent between approximately 26,500 and 19,000–20,000 years ago. The Laurentide spread across northern North America. In Europe, the British–Irish, Scandinavian and Barents–Kara ice sheets formed a vast northern complex, with their margins reaching maximum positions at different times. In the Alps, glaciers filled valleys, crossed some passes and extended onto the surrounding forelands.

![World map with red lines outlining the ice-sheet margins about 20,000 years ago across much of Canada, northern Europe, Greenland, Patagonia and Antarctica; black lines show modern coastlines.](https://storage.ghost.io/c/aa/80/aa80281a-abce-4f35-865f-05153c8cc335/content/images/2026/10/global-last-glacial-maximum-ice-margins-gowan-2021.jpg)

Red lines show reconstructed ice-sheet margins approximately 20,000 years ago, against modern coastlines. Numbered locations follow the original study. Adapted from [Gowan et al. (2021), Supplementary Figure 1](https://www.nature.com/articles/s41467-021-21469-w?ref=romanempiretimes.com), [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/?ref=romanempiretimes.com); shaded relief and its elevation key are omitted.

Alpine geological evidence places well-documented maximum foreland advances around 25,500–24,000 years ago. A published ice-flow model reconstructs how accumulated ice moved through that mountain system. Reduced northern summer sunlight, lower greenhouse-gas concentrations and climate feedbacks helped the wider ice expansion. So much water was stored on land that global sea level fell approximately 130 metres (427 feet). Increasing northern summer sunlight later helped initiate retreat, beginning around 19,000–20,000 years ago.

![Map of the Alps showing modeled ice cover and flow in blue at the Last Glacial Maximum, with red geological ice limits and an inset chart of simulated ice volume across 120,000 years.](https://storage.ghost.io/c/aa/80/aa80281a-abce-4f35-865f-05153c8cc335/content/images/2026/10/alpine-last-glacial-maximum-seguinot-2018.jpg)

A model of Alpine ice at its maximum extent approximately 24,570 years ago, compared with mapped geological ice limits in red. The inset tracks simulated ice volume through the last glacial cycle. [Seguinot et al. (2018), Figure 4](https://tc.copernicus.org/articles/12/3265/2018/?ref=romanempiretimes.com), [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/?ref=romanempiretimes.com).

## The Little Ice Age had a Roman predecessor

The Little Ice Age usually associated with late medieval and early modern history appears directly in this Arctic research. A 2012 study of preserved plants and Icelandic sediments dated abrupt summer cooling and ice growth to AD 1275–1300, with a further intensification in 1430–1455\. These were later pulses within the longer history of returning northern ice.

That study proposed repeated volcanic eruptions as a trigger, with expanded sea ice and altered ocean heat transport sustaining the cooling. The 2026 Barnes record also contains a concentration of plant deaths around AD 1450\. The connection involves shared evidence and episodes of glacier growth, with different glaciers responding at different speeds.

Roman history contains a separate cold interval. Tree-ring research identifies a Late Antique Little Ice Age from AD 536 to about 660, following eruptions in 536, 540 and 547\. It overlapped the eastern Roman world of [the plague of Justinian](https://romanempiretimes.com/the-plague-of-justinian-when-death-came-to-constantinople/), although the researchers treat climate as one factor in complex historical changes. This sixth-century episode is similar, but distinct, from the later medieval Little Ice Age.

## Why another model puts glaciation far ahead

Other scientists have produced a substantially different timetable. A 2025 study led by Christine Kaufhold used a coupled climate, carbon-cycle and ice-sheet model to place the next natural glacial inception around 50,000 years in the future. Its results also show that additional human carbon emissions can delay that transition further.

The model defines inception through rapid expansion and roughly a doubling of present Northern Hemisphere land-ice area. The Barnes researchers instead identify the first renewed growth of an existing ice-sheet remnant as the beginning. 

The new study’s central finding is the reversal preserved in the landscape: northern ice returned during the Common Era, expanded through later cold episodes, then retreated under modern warming. Its authors propose that Romans lived during the earliest stage of a glaciation whose later development was interrupted. How that first local growth relates to the threshold for continental glaciation remains the scientific question.

#### ****🏛️ Sources used:**

  
[Young, N. E., et al. (2026). Late Holocene reexpansion of the Laurentide Ice Sheet. Science 394, 455–459.](https://academiccommons.columbia.edu/doi/10.7916/rqk6-rn37?ref=romanempiretimes.com)

  
[Miller, G. H., et al. (2023). Moss kill dates and modeled summer temperature track episodic snowline lowering and ice cap expansion in Arctic Canada through the Common Era. Climate of the Past 19, 2341–2360.](https://cp.copernicus.org/articles/19/2341/2023/?ref=romanempiretimes.com)

  
[Margaritelli, G., et al. (2020). Persistent warm Mediterranean surface waters during the Roman period. Scientific Reports 10, 10431.](https://www.nature.com/articles/s41598-020-67281-2?ref=romanempiretimes.com)

  
[Birch, L., Cronin, T., and Tziperman, E. (2017). Glacial Inception on Baffin Island: The Role of Insolation, Meteorology, and Topography. Journal of Climate 30, 4047–4064.](https://groups.seas.harvard.edu/climate/eli/reprints/Birch-Cronin-Tziperman-2017.pdf?ref=romanempiretimes.com)

  
[Brauer, A., et al. (2008). An abrupt wind shift in western Europe at the onset of the Younger Dryas cold period. Nature Geoscience 1, 520–523.](https://www.tgacademyscience.com/assets/news/Abrupt-93.pdf?ref=romanempiretimes.com)

  
[Clark, P. U., et al. (2009). The Last Glacial Maximum. Science 325, 710–714.](https://stuff.mit.edu/~heimbach/papers%5Fpaleo/science%5Fclark%5Fetal%5F2009%5Flgm.pdf?ref=romanempiretimes.com)

  
[Hughes, A. L. C., et al. (2016). The last Eurasian ice sheets—a chronological database and time-slice reconstruction, DATED-1\. Boreas 45, 1–45.](https://nva.sikt.no/registration/0198cc830e8a-a6d1c924-949d-42c8-9db1-cf6a0b95b18c?ref=romanempiretimes.com)

  
[Seguinot, J., et al. (2018). Modelling last glacial cycle ice dynamics in the Alps. The Cryosphere 12, 3265–3285.](https://tc.copernicus.org/articles/12/3265/2018/?ref=romanempiretimes.com)

  
[Miller, G. H., et al. (2012). Abrupt onset of the Little Ice Age triggered by volcanism and sustained by sea-ice/ocean feedbacks. Geophysical Research Letters 39, L02708.](https://glaciers.pdx.edu/fountain/readings/HoloceneClimate/MillerEtAl2012%5FLittleIceAge.pdf?ref=romanempiretimes.com)

  
[Büntgen, U., et al. (2016). Cooling and societal change during the Late Antique Little Ice Age from 536 to around 660 AD. Nature Geoscience 9, 231–236.](https://www.repository.cam.ac.uk/items/072ba4aa-014e-4f87-8fc7-334ebf876f1f?ref=romanempiretimes.com)

  
[Kaufhold, C., et al. (2025). Timing of a future glaciation in view of anthropogenic climate change. Communications Earth & Environment 6, 1016.](https://www.nature.com/articles/s43247-025-02867-0?ref=romanempiretimes.com)

  
[Eichenberger, U., et al. (2020). Feuille 1143 Le Locle, avec partie de la feuille 1123 Le Russey. Atlas géologique de la Suisse 1:25,000, Notice explicative 172\. swisstopo.](https://prod-swishop-s3.s3.eu-central-1.amazonaws.com/product-documents/GA25-ERL-172.pdf?ref=romanempiretimes.com)