Arctic Futures: Five Signals of Change

The Arctic stands as a crucial frontier for our planet's future. But what do we need to be aware of? Here are five important signals of change across each STEEP parameter about the region and its future.

Often perceived as a distant, frozen wasteland, the Arctic is in fact a vital component of our planet's health and a barometer for environmental change. This vast, icy region acts like a canary in a coal mine, offering critical warnings about the state of our environment. The melting ice in the Arctic is not just a concern for the iconic polar bears but has far-reaching implications affecting weather patterns, sea levels, global trade, and even geopolitics. However, it's not all doom and gloom; by understanding and protecting the Arctic, we have the opportunity to safeguard our future. The time to act is now, to preserve this unique ecosystem and the communities that depend on it.

Social

**Risk of New Pandemic with Siberian Zombie Viruses**

The signal in 2023. Viruses preserved in Siberian permafrost, revived by researchers, presented as a potential source of a new pandemic.

Three years on. The science is real and the pandemic framing was not.

Jean-Michel Claverie and Chantal Abergel's team at Aix-Marseille have revived thirteen giant DNA viruses from Siberian permafrost, wolf remains and Kamchatka soil, the oldest around 48.500 years. Every one of them infects amoebae, specifically laboratory cultures of Acanthamoeba. The researchers describe the biohazard of their own method as negligible, on the grounds of roughly a billion years of evolutionary distance between amoebae and humans. The pandemic inference is an extrapolation made in interviews, that the same permafrost might also hold human-infecting viruses, and not a property of anything that has been isolated.

The genomic traces of poxviruses and herpesviruses are also real, and also do not mean what the coverage implied. They are degraded DNA fragments detected by metagenomics, in lower proportions than the amoeba-infecting viruses, and fragments are not infectious virus.

In February 2025 a team from the US Geological Survey, the Army Corps engineering laboratory and California State University Northridge published a formal rebuttal in mSystems, concluding that the risk from permafrost viruses is no greater than from viruses in temperate soils or aquatic systems. Attempts to isolate live smallpox from permafrost-preserved human remains have failed, yielding only degraded DNA. Only fragments shorter than 120 nucleotides have been recovered from 1918 influenza tissue. And humans have lived in the Arctic for millennia without evident large-scale consequences from permafrost exposure.

No outbreak has been attributed to thawing permafrost anywhere in the Arctic since. The 2016 anthrax outbreak in Yamalo-Nenets, which killed a boy and around 2.350 reindeer, remains the only precedent, and its mechanism was bacterial spores in a shallow reindeer carcass rather than deep permafrost virology.

Verdict: Deflated. This was a single laboratory, a single spokesperson and one newspaper. It did not survive contact with the wider virology community.

A better-founded successor signal has taken its place. Research attention has moved from viruses to the ancient resistome: antimicrobial resistance genes preserved in permafrost and glacial ice. A 2026 review in Critical Reviews in Microbiology covers the mobilisation of these genes as thaw proceeds, and a bacterium recovered from a 5.000-year-old ice cave in early 2026 proved resistant to modern antibiotics. That is the version of this signal worth carrying forward.

Question for the future. What surveillance would actually detect an emerging Arctic health risk, given that the one multilateral system anybody has proposed would need to include Russia?

Technology

**Arctic satellite broadband**

The signal in 2023. The Arctic Satellite Broadband Mission completing tests ahead of launch, promising reliable high-latitude connectivity.

Three years on. ASBM delivered exactly as specified, and the interesting question moved.

Two satellites launched on a Falcon 9 in August 2024 into highly elliptical orbit, which is the point, since geostationary satellites cannot see the high Arctic. Northrop Grumman built them and handed over a fully activated constellation in December 2024. Space Norway operates them, with hosted payloads from the US Space Force, the Norwegian Ministry of Defence, Viasat and the European Commission. It was the first time a US military payload flew hosted on an international commercial mission. Viasat's payloads entered service in April 2025.

Meanwhile the coverage problem was solved from a different direction. Starlink added several hundred polar-inclination satellites and now serves Alaska, Canada, Iceland, Svalbard, Norway, Sweden and Finland.

And then the question stopped being about bandwidth. Greenland banned Starlink in October 2024, nominally to protect the Tusass telecom monopoly and explicitly also on national security grounds. In October 2025 Tusass signed with Eutelsat OneWeb instead. Arctic connectivity became a sovereignty decision rather than a technical one, and no scan written in 2024 had a category for it.

Two subsea cable projects are running on the same logic. Far North Fiber would link Europe and North America through the Northwest Passage, targeting 2027. Polar Connect would cross the North Pole itself at depths up to 4.000 metres, connecting Europe and Asia; mapping began with the icebreaker Oden in summer 2025, about 45% of the route was surveyed by November, and the EU added €5 million in December 2025 towards a 2030 go-live. The stated rationale is that around 90% of Europe-Asia data currently transits the Red Sea.

Verdict: Developed, with the premise superseded. The infrastructure arrived. The scarce thing turned out to be sovereign control of it.

Question for the future. If connectivity is now a sovereignty question, who else in the Arctic will choose a provider on political grounds rather than performance?

Economy

**US Subsidies Threaten Arctic Circle Town's Clean Energy Effort**

The signal in 2023. American subsidies under the Inflation Reduction Act pulling clean-tech investment away from Europe, with Freyr's move from Mo i Rana to the United States as the case.

Three years on. This is the signal that reversed, and the outcome is worse than a straight reading of it would have suggested.

The pull disappeared. The One Big Beautiful Bill Act, enacted on 4 July 2025, dismantled most of what created it. The consumer and commercial electric vehicle credits terminated on 30 September 2025. The technology-neutral clean electricity credits now require construction to begin before 5 July 2026 or the project to be in service by the end of 2027. Foreign entity of concern rules expanded substantially, applying to nearly all remaining credits. The advanced manufacturing credit that supports battery cells survives longest, phasing down after 2030.

The company left anyway, and stopped making batteries. Freyr bought a solar cell facility in Texas from Trina Solar, rebranded as T1 Energy, moved its headquarters to Austin and began trading on the NYSE in March 2025. It abandoned its planned Georgia battery gigafactory and put the site up for sale. It is now an American solar manufacturer.

Europe's champion went bankrupt. Northvolt filed in Stockholm on 12 March 2025 with around 5.000 employees. The American lithium-sulfur company Lyten agreed to buy the assets in August 2025 and completed in February 2026, committing to more than 600 additional hires. So Europe's flagship battery company is American-owned, and Norway's became a Texan solar company.

And Mo i Rana found a different use for its power. In August 2026 the municipality approved rezoning part of the Giga Arctic site for data centre development: 15.000 square metres of an 86.000 square metre campus, 50 MW initially, potentially operating from 2027, with a queue position for 396 MW of grid capacity and a 50-year lease. A second site in the town, VivoPower's 41,5 MW development, opened lease bidding to AI operators in April 2026.

European battery manufacturing has held rather than recovered: around 251 GWh of cell capacity as of September 2025 on roughly €72 billion invested since 2017, with South Korean firms controlling about four fifths of operational European capacity and 2025's new investment going to Hungary and Spain. The European Commission's answer, the Battery Booster announced in June 2026, is €1,5 billion in interest-free loans rather than grants.

Verdict: Reversed. The mechanism the 2024 signal identified was correct. The direction flipped, the sector contracted on both sides of the Atlantic, and the Arctic industrial site the signal was about is now being marketed to AI companies.

Question for the future. If cheap Arctic hydropower is worth more to inference compute than to battery cells, what does that do to the northern Scandinavian industrial thesis, and to the grid?

Environment

**Accelerated Arctic Warming**

The signal in 2023. A sharp decline in Arctic reflectivity accelerating warming, alongside the loss of Russian research station data.

Three years on. The trend is confirmed and worse.

The albedo study is Dreike and colleagues in Scientific Reports, published in August 2023 and publicised by Sandia National Laboratories in January 2024. What it found is narrower and more interesting than the headline: reflected sunlight north of 80°N fell 20 to 35% over an 80-day window around the summer solstice, across the years 2014 to 2019 only. Sea ice coverage over the same window and period fell only 7 to 9%. The implication is that the reflectivity of the surviving ice itself dropped by 20 to 25%, through melt ponds, thinning and weathering. Surface darkening matters roughly three times more than ice retreat for total reflected sunlight. The data came from broadband radiometers on GPS satellites, accessed through the US Space Force, which is an unusual repurposing of a military dataset. We could find no Aarhus University involvement in that paper.

The data loss finding is separate and is Aarhus-led. López-Blanco and colleagues, in Nature Climate Change in January 2024, found that 17 of the 60 INTERACT research stations above 59°N, or 28%, are Russian. Excluding them biases the observed network towards warmer, wetter, higher-biomass sites, and for precipitation and vegetation biomass the resulting distortion is of similar magnitude to about 80 years of projected climate change. The network's blind spot is as large as the signal it exists to detect. Nothing has improved: political-level Arctic Council meetings have still not resumed.

What has happened since is more consequential than either. In December 2024 the NOAA Arctic Report Card declared the Arctic tundra a net source of carbon dioxide for the first time, counting wildfire emissions. That is the single most important Arctic finding of the whole period.

The 2025 Report Card records the warmest Arctic year since 1900 for the period October 2024 to September 2025, with the oldest and thickest ice down more than 95% since the 1980s and Greenland losing around 129 Gt in 2025. New topics this year include the Atlantification of the central Arctic Ocean and rivers running orange with iron released by thawing permafrost across more than 200 Alaskan watersheds.

Winter is now where the records fall. March 2025 produced the lowest annual sea ice maximum in the 47-year record, and March 2026 statistically tied it. The September minimum, counter-intuitively, has plateaued: the 2025 minimum was only tied tenth lowest, and NSIDC notes that while the 19 lowest minima are the last 19 years, there is no significant trend within that period. A scan watching only the summer minimum would have missed the shift.

The framing has also changed. A December 2024 paper in Nature Communications moved the question from an ice-free summer to a first ice-free day, plausibly as early as 2027 in extreme cases, with ice-free summers remaining a 2030s to 2050s proposition.

One alarm softened. A July 2026 study found Greenland meltwater drives strong AMOC weakening without a shutdown, and no tipping point in view in the updated model.

Verdict: Developed. Two record winter maxima, the warmest year on record, and a tundra that has crossed from sink to source.

Question for the future. With a quarter of the Arctic observation network unavailable and no prospect of it returning, how much of what we think we know about Arctic change is an artefact of where we can still measure?

The signal in 2023. Nordic Response, over 20.000 troops from 13 nations exercising in northern Norway, Finland and Sweden, as evidence of heightened Arctic security concern.

Two years on. The exercises grew and the threat model changed.

Nordic Response ran from 3 to 14 March 2024 as part of Steadfast Defender, with 20.000 soldiers from 13 nations, around 110 aircraft and 50 vessels. Joint Viking followed in March 2025 with around 10.000 troops from nine nations. Cold Response 2026, from 9 to 19 March, involved around 32.000 personnel from 14 nations, the largest Arctic exercise since the Cold War.

The reason it grew is not the one a 2024 scan would have assumed.

Greenland. From December 2024 the incoming US administration described control of Greenland as a necessity. Donald Trump Jr landed in Nuuk on 7 January 2025. Denmark announced its first Arctic package, DKK 14,6 billion for naval vessels, long-range surveillance drones and satellite capacity, at the end of that month. Greenland's general election on 11 March 2025 produced a surprise win for Demokraatit, gradualist on independence, and Jens-Frederik Nielsen became Premier at the head of a broad four-party coalition, a result read as a rejection of both rapid independence and American annexation. JD Vance visited Pituffik Space Base on 28 March 2025, an itinerary cut back from a planned civilian visit after Greenlandic and Danish objections; the base commander was relieved of command in April after writing to staff that the administration's positions were not reflective of Pituffik.

In May 2025 the Danish Defence Intelligence Service named the United States as a national security concern for the first time in its history.

In October 2025 Denmark announced a second Arctic and North Atlantic agreement worth DKK 27,4 billion, alongside a separate DKK 27,4 billion for 16 additional F-35s, announced together in Belfast. The Arctic tranche funds a new Arctic Command headquarters in Nuuk, maritime patrol aircraft, two Arctic vessels, icebreaker access, drones, radar in East Greenland and a Denmark-Greenland submarine cable.

January 2026 was the crisis. Greenland's government rejected US demands outright and Copenhagen and Nuuk issued a joint statement describing unacceptable pressure. Denmark reinforced Greenland, and Sweden, Norway and France followed, in what became Operation Arctic Endurance, eventually involving eleven European countries. NATO officials and analysts described it openly as a tripwire force, European allies deterring the United States. Trump announced tariffs on eight European countries unless they backed annexation. On 21 January at Davos, after meeting the NATO Secretary General, he withdrew the tariff threat, said he would not use force, and announced a framework on Arctic security and minerals, to which neither Denmark nor Greenland was a party.

NATO launched Operation Arctic Sentry on 11 February 2026, a twelve-month mission under a JFC Norfolk whose command had recently transferred from the United States to the United Kingdom and whose area of responsibility now includes Denmark, Finland and Sweden. Cold Response 2026 was its first major exercise. In June the Danish foreign minister said a trilateral agreement could conclude by the end of 2026. At the Ankara summit in July, Trump restated that the United States should control Greenland.

Around this, the Nordic build-up continued. Norway added NOK 115 billion to its long-term defence plan in March 2026, accelerating the Finnmark Brigade by two years. Finland published a new Arctic defence strategy in November 2025 which notes that the United States increasingly views the Arctic through the lens of global competition, shaped by an America First policy, which is a striking sentence for a NATO member to publish.

Two footnotes that carry more than they look. Russia's Arctic expansion appears to have slowed since 2024 as resources go to Ukraine, and its Arctic military activity fell that year. And Northern Sea Route traffic declined for the first time in years, from 37,9 million tonnes in 2024 to 37,02 million in 2025, with petroleum products down 43%; the binding constraints on Arctic shipping turned out to be sanctions and hull availability, not ice.

Verdict: Developed, in a direction the scan could not have framed. A signal about allies deterring Russia became a story about allies deterring an ally. The Kingdom of Denmark chaired the Arctic Council through the period in which the United States threatened to annex part of it.

Question for the future. If a small state's security guarantor becomes a source of risk, what does hedging look like, and how many other states are working through the same arithmetic without saying so?

The signal in 2024. Nordic Response, over 20.000 troops from 13 nations exercising in northern Norway, Finland and Sweden, as evidence of heightened Arctic security concern.

Two years on. The exercises grew. The threat model changed.