Near final draft still going through red team accuracy verification process 10.15.26.

 

Short answer. Compared with the equator, climate models project warming about 2 times faster from roughly 60°N (Oslo, Helsinki, Anchorage), about 3 times faster from roughly 75°N (Svalbard), and about 3.5 times faster at 80–90°N (the high Arctic). Measured warming since 1979 is running even further ahead: about 2 times the equator’s rate already at 40–55°N, about 3 times at 60°N, close to 4 times at 65°N (the Arctic Circle), and 4.5 to nearly 6 times from 70°N northward. Measured against the global average instead of the equator, the “4 times faster” Arctic finding applies to the region north of the Arctic Circle (66.5°N).

Why it matters to everyone. The fastest-warming latitudes hold the parts of the climate system that can speed up warming for the whole planet: Arctic sea ice, snow cover, permafrost carbon, the Greenland ice sheet, mountain glaciers, boreal forests and the North Atlantic ocean currents. By about 2050 on the middle emissions path, the band from 70°N to the pole is expected to be roughly 5 to 8°C warmer than before industrialization, and several of these systems have estimated trigger points between 1.5°C and 2°C of global warming, which the world reaches between now and the 2050s. What happens there raises sea levels, adds carbon to the air, and shifts weather and water supplies far to the south.

 

Why the north warms faster. Scientists call it Arctic amplification. Melting snow and sea ice expose darker land and ocean that absorb more sunlight, the shallow cold air near the surface traps extra heat, and the atmosphere and ocean carry heat toward the pole. Near the equator, more of the extra energy goes into evaporating water instead of raising the temperature, so it warms more slowly than the global average.

 

Chart 1: how many times faster than the equator each band warms

Projected by climate models (median of 40 CMIP6 models)   Measured, 1979–2024 (Berkeley Earth). The equator (10°S–10°N) = 1×.

85–90°N
3.7×
5.7×
80–85°N
3.5×
5.4×
75–80°N
3.1×
5.1×
70–75°N
2.7×
4.5×
65–70°N
2.3×
3.7×
60–65°N
2.0×
2.9×
55–60°N
1.7×
2.3×
50–55°N
1.6×
2.0×
45–50°N
1.5×
2.1×
40–45°N
1.4×
2.3×
 
0× 1× 2× 3× 4× 5×

Chart 2: the latitude ladder — warming by 2050 and what speeds up at each parallel

Read from the top (the North Pole) down. The colored cell shows each band’s expected warming above pre-industrial levels in 2050 on the middle emissions path (low end = climate-model pattern, high end = measured pattern). Darker = hotter. “Trigger” means the level of global warming at which scientists estimate a system could tip into self-sustaining change (Armstrong McKay et al. 2022, best estimates).

Color key (2050 band warming): under 2.5°C2.5–3.53.5–4.54.5–5.5over 5.5°C

Latitude band Warming in 2050
middle path 		What speeds up here How it affects the whole world
85–90°N
Central Arctic Ocean, North Pole 		6.4–8.1°C Summer sea ice disappears. The Arctic Ocean is likely to be practically ice-free in September at least once before 2050 (IPCC); observation-based projections say as early as the 2030s, and by the 2050s even with lower emissions (Kim et al.). Winter ice collapse needs far more warming (trigger about 6.3°C). Dark open water absorbs sunlight that ice used to reflect. An ice-free Arctic during sunlit months adds global heating equal to about 1 trillion tonnes of CO2, roughly 25 years of current emissions (Pistone et al.).
80–85°N
Franz Josef Land, north Greenland 		6.0–7.7°C Thinning sea ice and north Greenland melt. The last thick multi-year ice and north Greenland’s floating ice tongues are retreating. Greenland is already committed to at least 27 cm of sea-level rise from today’s climate (Box et al.); IPCC projects 4–13 cm from Greenland by 2100 on the middle path.
75–80°N
Svalbard, Barents Sea 		5.4–7.3°C Barents Sea ice loss and “Atlantification”. Warm Atlantic water now reaches further north and melts ice from below (Polyakov et al.). Parts of the Barents region have warmed up to about 7 times the global rate. Abrupt Barents winter-ice loss has an estimated trigger of about 1.6°C. Svalbard glaciers are shrinking. More open ocean in winter releases heat to the air and pushes the sea-ice edge north, strengthening the sea-ice feedback. Fish stocks and Arctic food webs shift.
70–75°N
Utqiagvik (Alaska), Siberian Arctic coast 		4.7–6.4°C Coastal permafrost erosion and the Greenland melt zone. Ice-rich coasts in Alaska and Siberia are eroding; subsea permafrost on the Siberian shelf is warming; Greenland’s melt area expands. Greenland ice sheet trigger: about 1.5°C (range 0.8–3°C). Once past its trigger, Greenland would lose most of its ice over many centuries to millennia: about 7 m of sea-level rise. A sudden methane-hydrate burst this century is judged very unlikely (IPCC).
65–70°N
Arctic Circle, Fairbanks area, Murmansk, Tromsø 		3.7–5.3°C Continuous permafrost thaw. About 25% of the frozen ground in the top 3 m is lost for every 1°C of global warming (IPCC). Abrupt thaw (ground collapse, thaw lakes) has a trigger of about 1.5°C and adds about 50% to emissions from gradual thaw. Tundra turns to shrubs and forest. Thawing permafrost releases CO2 and methane: about 3–41 billion tonnes of carbon per 1°C of warming by 2100 (IPCC), irreversible for centuries. Collapse of carbon-rich Yedoma permafrost (trigger about 4°C) would add 0.2–0.4°C more global warming.
60–65°N
Oslo, Helsinki, Anchorage, Reykjavik 		3.3–4.1°C Permafrost under towns and roads, glaciers, and the subpolar North Atlantic. About 70% of Arctic infrastructure and 3.6 million people are in areas where permafrost thaw threatens the ground by 2050 (Hjort et al.). Alaska, Iceland and Scandinavian glaciers shrink. Deep-water formation in the Labrador–Irminger Seas could collapse (trigger about 1.8°C). Alaska is the largest single source of glacier sea-level rise this century (Rounce et al.). A Labrador Sea convection collapse would cool the region by about 3°C and disrupt weather in Europe and North America within about 10 years.
55–60°N
Moscow, Edinburgh, Copenhagen 		2.9–3.2°C Boreal forest fire and stress; less snow. Fire seasons lengthen; Canada’s record 2023 season forced more than 200 communities to evacuate (Jain et al.). Southern boreal forest dieback has a trigger of about 4°C (range 1.4–5°C). Spring snow cover shrinks about 8% per 1°C of global warming (IPCC). Fires release stored carbon and smoke; less snow means less sunlight reflected. The Atlantic overturning circulation (AMOC) is very likely to weaken this century (IPCC).
50–55°N
London, Berlin, Calgary 		2.7–2.8°C Western Canada and US glaciers; earlier snowmelt. Glaciers in western North America are nearly gone at 2.7°C of global warming (Rounce et al.). Snowmelt arrives earlier, leaving less water in late summer. Rivers fed by snow and ice lose summer flow, stressing farms, cities and hydropower. North Atlantic circulation changes affect northwest European weather.
45–50°N
Milan, Montreal, Seattle 		2.6–2.9°C Alpine glaciers and mid-latitude heat. Central Europe’s glaciers are nearly gone at 2.7°C (Rounce et al.). In mid-latitude regions, the hottest days warm about 1.5 to 2 times faster than the global average (IPCC). Snow- and glacier-fed rivers such as the Rhine, Rhône, Po and Columbia lose summer flow; crop and power losses in major food and energy regions.
40–45°N
Madrid, New York, Beijing 		2.4–3.2°C Heatwaves, drought, wildfire, shrinking snowpack. Hottest days warm 1.5 to 2 times faster than the global average in many mid-latitude and dry regions (IPCC); mountain snowpack and water storage decline. Heat and drought hit major breadbaskets and big cities; water supply depends on shrinking snowpack.

Table 1: expected warming in each band at global milestones

Each band’s warming is shown above pre-industrial (1850–1900) levels at three global warming milestones. The first number uses the climate-model pattern; the second uses the faster pattern measured since 1979. Treat the range between them as the realistic span. The “Extra vs. equator” column shows how much more the band warms than the equator when the world is at 2°C.

Latitude band
example places 		Times the equator’s warming
models / measured 		World at 1.5°C
about 2030–2035 		World at 2°C
about 2041–2060 		World at 3°C
2070s on high emissions; after 2100 on middle path 		Extra vs. equator
at 2°C world
Equator (10°S–10°N)
reference 		1× 1.3°C 1.7°C 2.6°C —
40–45°N
Madrid, New York, Beijing 		1.4× / 2.3× 1.8–2.5°C 2.5–3.3°C 3.7–5.0°C +0.7 to +1.6°C
45–50°N
Milan, Montreal, Seattle 		1.5× / 2.1× 2.0–2.3°C 2.6–3.0°C 4.0–4.5°C +0.9 to +1.3°C
50–55°N
London, Berlin, Calgary 		1.6× / 2.0× 2.1–2.2°C 2.8–2.9°C 4.1–4.4°C +1.0 to +1.2°C
55–60°N
Moscow, Edinburgh, Copenhagen 		1.7× / 2.3× 2.2–2.5°C 3.0–3.3°C 4.5–5.0°C +1.2 to +1.6°C
60–65°N
Oslo, Helsinki, Anchorage, Reykjavik 		2.0× / 2.9× 2.6–3.2°C 3.4–4.2°C 5.1–6.4°C +1.7 to +2.5°C
65–70°N
Arctic Circle, Fairbanks area, Murmansk, Tromsø 		2.3× / 3.7× 2.9–4.1°C 3.8–5.5°C 5.8–8.2°C +2.1 to +3.7°C
70–75°N
Utqiagvik (Alaska), Siberian Arctic coast 		2.7× / 4.5× 3.6–5.0°C 4.8–6.6°C 7.2–9.9°C +3.0 to +4.9°C
75–80°N
Svalbard, Barents Sea 		3.1× / 5.1× 4.2–5.6°C 5.6–7.5°C 8.3–11.2°C +3.8 to +5.7°C
80–85°N
Franz Josef Land, north Greenland 		3.5× / 5.4× 4.6–5.9°C 6.1–7.9°C 9.2–11.9°C +4.4 to +6.2°C
85–90°N
Central Arctic Ocean, North Pole 		3.7× / 5.7× 4.9–6.2°C 6.5–8.3°C 9.8–12.4°C +4.8 to +6.5°C

Table 2: temperature rise in each band by calendar year

This answers “how hot, where, and when.” Each cell is the band’s yearly-average warming above pre-industrial levels in that year (low end = climate-model pattern, high end = measured pattern). The top row shows the global average for that year. All values are estimates calculated with the formula at the end of this page.

 

Middle path (IPCC SSP2-4.5 central estimate). The IPCC’s middle-of-the-road emissions path.

Latitude band 2030 2040 2050 2070 2100
World average 1.5°C 1.7°C 1.9°C 2.3°C 2.8°C
Equator (10°S–10°N) 1.3°C 1.5°C 1.7°C 2.0°C 2.4°C
85–90°N 4.9–6.2°C 5.6–7.0°C 6.4–8.1°C 7.5–9.5°C 9.2–11.6°C
80–85°N 4.6–5.9°C 5.2–6.7°C 6.0–7.7°C 7.1–9.1°C 8.6–11.1°C
75–80°N 4.2–5.6°C 4.7–6.3°C 5.4–7.3°C 6.4–8.6°C 7.8–10.4°C
70–75°N 3.6–5.0°C 4.1–5.6°C 4.7–6.4°C 5.5–7.6°C 6.7–9.3°C
65–70°N 2.9–4.1°C 3.3–4.6°C 3.7–5.3°C 4.4–6.3°C 5.4–7.6°C
60–65°N 2.6–3.2°C 2.9–3.6°C 3.3–4.1°C 3.9–4.9°C 4.8–5.9°C
55–60°N 2.2–2.5°C 2.5–2.8°C 2.9–3.2°C 3.4–3.8°C 4.2–4.7°C
50–55°N 2.1–2.2°C 2.3–2.5°C 2.7–2.8°C 3.2–3.4°C 3.9–4.1°C
45–50°N 2.0–2.3°C 2.2–2.6°C 2.6–2.9°C 3.0–3.5°C 3.7–4.2°C
40–45°N 1.8–2.5°C 2.1–2.8°C 2.4–3.2°C 2.8–3.8°C 3.4–4.6°C

 

+20% case (Universe Institute Chart 2, current emergency-planning case). Global warming 20% above the middle path each year.

Latitude band 2030 2040 2050 2070 2100
World average 1.8°C 2.1°C 2.3°C 2.8°C 3.3°C
Equator (10°S–10°N) 1.6°C 1.8°C 2.0°C 2.4°C 2.9°C
85–90°N 5.9–7.5°C 6.9–8.7°C 7.6–9.7°C 9.2–11.6°C 10.8–13.7°C
80–85°N 5.5–7.1°C 6.4–8.3°C 7.2–9.2°C 8.6–11.1°C 10.1–13.1°C
75–80°N 5.0–6.7°C 5.8–7.8°C 6.5–8.7°C 7.8–10.4°C 9.2–12.3°C
70–75°N 4.3–6.0°C 5.0–6.9°C 5.6–7.7°C 6.7–9.3°C 7.9–10.9°C
65–70°N 3.5–4.9°C 4.0–5.7°C 4.5–6.4°C 5.4–7.6°C 6.3–9.0°C
60–65°N 3.1–3.8°C 3.6–4.5°C 4.0–5.0°C 4.8–5.9°C 5.7–7.0°C
55–60°N 2.7–3.0°C 3.1–3.5°C 3.5–3.9°C 4.2–4.7°C 4.9–5.5°C
50–55°N 2.5–2.6°C 2.9–3.1°C 3.2–3.4°C 3.9–4.1°C 4.5–4.8°C
45–50°N 2.4–2.7°C 2.8–3.2°C 3.1–3.5°C 3.7–4.2°C 4.3–5.0°C
40–45°N 2.2–3.0°C 2.6–3.5°C 2.9–3.9°C 3.4–4.6°C 4.1–5.5°C

 

+30% case (Universe Institute Chart 3). Global warming 30% above the middle path each year.

Latitude band 2030 2040 2050 2070 2100
World average 1.9°C 2.2°C 2.5°C 3.1°C 3.6°C
Equator (10°S–10°N) 1.7°C 1.9°C 2.2°C 2.7°C 3.1°C
85–90°N 6.2–7.9°C 7.2–9.1°C 8.3–10.5°C 10.2–12.8°C 11.8–14.9°C
80–85°N 5.8–7.5°C 6.8–8.7°C 7.8–10.0°C 9.5–12.3°C 11.1–14.2°C
75–80°N 5.3–7.1°C 6.1–8.2°C 7.0–9.4°C 8.6–11.6°C 10.0–13.4°C
70–75°N 4.6–6.3°C 5.3–7.3°C 6.1–8.4°C 7.4–10.3°C 8.6–11.9°C
65–70°N 3.7–5.2°C 4.2–6.0°C 4.9–6.9°C 6.0–8.5°C 6.9–9.8°C
60–65°N 3.3–4.0°C 3.8–4.7°C 4.3–5.4°C 5.3–6.6°C 6.2–7.6°C
55–60°N 2.8–3.2°C 3.3–3.7°C 3.8–4.2°C 4.6–5.2°C 5.4–6.0°C
50–55°N 2.6–2.8°C 3.0–3.2°C 3.5–3.7°C 4.3–4.5°C 5.0–5.3°C
45–50°N 2.5–2.9°C 2.9–3.3°C 3.3–3.8°C 4.1–4.7°C 4.7–5.5°C
40–45°N 2.3–3.2°C 2.7–3.7°C 3.1–4.2°C 3.8–5.1°C 4.4–6.0°C

Table 3: tipping points by latitude — global trigger, when it is reached, and local temperature at that point

Triggers are the global warming levels estimated by Armstrong McKay et al. (2022): best estimate, with the range in brackets. “When reached” is the year the global curve first hits the best-estimate trigger on each path. “Local warming at trigger” = trigger × that band’s amplification (an estimate). “Now” means the path already reaches it by 2026. Crossing a trigger starts a change; the change itself can take years to millennia.

System and where Global trigger
best (range) 		When reached
middle / +20% / +30% 		≈ Local warming at trigger How long the change takes Effect on the whole world
Arctic summer sea ice: practically ice-free September
70–90°N, Arctic Ocean 		— first ice-free September likely before 2050 (IPCC); as early as the 2030s (Kim et al.) — seasonal; returns each summer once reached Adds heating equal to about 25 years of CO2 emissions once ice-free in sunlit months (Pistone et al.)
Greenland ice sheet collapse
60–83°N 		1.5°C
(0.8–3.0) 		2031 / now / now 3.6–5.0°C about 10,000 years (1,000–15,000) About 7 m of sea-level rise in the long run; already committed to 27 cm or more
Abrupt permafrost thaw
55–75°N 		1.5°C
(1.0–2.3) 		2031 / now / now 2.9–4.1°C about 200 years About 50% more permafrost emissions than gradual thaw alone
Barents Sea winter ice: abrupt loss
70–80°N 		1.6°C
(1.5–1.7) 		2033 / now / now 4.4–6.0°C about 25 years Regional warming; strengthens the sea-ice feedback
Labrador–Irminger Sea convection collapse
55–65°N, subpolar North Atlantic 		1.8°C
(1.1–3.8) 		2042 / 2030 / now 3.1–3.8°C about 10 years (5–50) About 3°C regional cooling; disrupted weather in Europe and North America; shifted monsoons
Mountain glacier loss (outside the ice sheets)
Alps, Rockies, Alaska, Scandinavia 		2.0°C
(1.5–3.0) 		2053 / 2039 / 2033 2.6–3.0°C about 200 years (50–1,000) Lost dry-season water supply; sea-level rise
Southern boreal forest dieback
50–60°N 		4.0°C
(1.4–5.0) 		after 2100 / after 2100 / after 2100 6.0–6.7°C about 100 years Carbon release and ecosystem loss (net global effect small because bare land reflects more sunlight)
Boreal forest spreading into tundra
60–70°N 		4.0°C
(1.5–7.2) 		after 2100 / after 2100 / after 2100 7.7–10.9°C about 100 years Darker forest absorbs more sunlight: about 0.14°C extra global warming
Permafrost collapse (carbon-rich Yedoma)
60–75°N, Siberia and Alaska 		4.0°C
(3.0–6.0) 		after 2100 / after 2100 / after 2100 7.7–10.9°C about 50 years (10–300) Adds 0.2–0.4°C of global warming
Atlantic overturning circulation (AMOC) collapse
North Atlantic, all latitudes 		4.0°C
(1.4–8.0) 		after 2100 / after 2100 / after 2100 — about 50 years (15–300) Strong regional cooling of northwest Europe; global weather and monsoon shifts. IPCC: very likely to weaken this century; medium confidence of no abrupt collapse before 2100
Arctic winter sea ice collapse
70–90°N 		6.3°C
(4.5–8.7) 		after 2100 / after 2100 / after 2100 — about 20 years Adds about 0.6°C of global warming

For reference, the lower ends of the trigger ranges for Greenland (0.8°C), abrupt permafrost thaw (1.0°C), Labrador Sea convection (1.1°C), southern boreal dieback (1.4°C) and the AMOC (1.4°C) are at or below today’s global warming of about 1.4°C. Separate statistical early-warning studies put a possible AMOC tipping window at 2037–2109 (central estimate 2065 after correction), but a 34-model study finds collapse this century unlikely; the evidence is genuinely split.

How warming at high latitudes speeds up warming and damage everywhere

  1. Less ice and snow, more absorbed sunlight. An ice-free Arctic in sunlit months would add about 0.71 W/m² of global heating, equal to about 1 trillion tonnes of CO2 or about 25 years of current emissions (Pistone et al. 2019). Northern Hemisphere spring snow cover shrinks about 8% for every 1°C of global warming (IPCC), removing more reflective surface.
  2. Permafrost carbon. The frozen north stores roughly twice as much carbon as the atmosphere. Thaw releases about 3–41 billion tonnes of carbon per 1°C of global warming by 2100 (IPCC), and the IPCC states this loss is irreversible on century time scales. Abrupt thaw adds about half again; Yedoma collapse adds 0.2–0.4°C (Armstrong McKay et al.). Every extra tonne shrinks the carbon budget left for the rest of the world.
  3. Sea-level rise for every coast. Greenland is already committed to at least 27 cm (Box et al.) and holds about 7 m in total. Glaciers worldwide lose about a quarter of their mass at 1.5°C (about 9 cm of sea-level rise) and about 41% at 4°C (about 15 cm) by 2100 (Rounce et al.), with Alaska the largest single source.
  4. Ocean circulation. Meltwater and warming weaken deep-water formation in the Labrador and Irminger Seas and the wider AMOC. Effects reach Europe’s weather, North American storms and tropical monsoons that feed billions of people.
  5. Water and food in the mid-latitudes. Shrinking snowpack and glaciers at 40–55°N cut summer river flow for farms, cities and hydropower; mid-latitude hottest days rise 1.5–2 times faster than the global average (IPCC).
  6. Fire and forests. Longer boreal fire seasons release stored carbon and smoke that crosses continents, as in Canada’s record 2023 season (Jain et al.).
  7. Weather patterns (still debated). Some research links Arctic warming to a wavier jet stream and stuck weather patterns in the mid-latitudes; the IPCC rates this link as low confidence and scientists disagree (Cohen et al. 2020).

When will the world reach each milestone?

  • 1.5°C: likely in the early 2030s; the IPCC finds 1.5°C is likely to be exceeded in 2021–2040 under the middle emissions path (SSP2-4.5) and higher paths. The +20% and +30% cases are already above it.
  • 2°C: more likely than not in 2041–2060 under the middle path (the central curve reaches it in 2053). Under the Universe Institute’s precautionary +20% and +30% cases, 2°C arrives in about 2039 and 2033.
  • 3°C: not reached this century on the middle path in the IPCC best estimate (about 2.7°C for 2081–2100), but reached around the 2070s on the high path (SSP3-7.0, about 3.6°C for 2081–2100). The +20% and +30% cases reach it in about 2080 and 2068.

How to read these numbers

  • These are yearly averages across each whole band, land and ocean together. Inland areas warm more than coasts and oceans; winters warm much more than summers in the far north.
  • The measured ratios are larger than the model ratios. Research finds most climate models underestimate how fast the Arctic has warmed, and the tropical oceans have warmed slowly in recent decades, which raises the measured ratios. That is why both are shown as a range.
  • The milestone dates are for the world as a whole. Because the far north warms faster, it reaches any given local temperature rise decades before the equator does.
  • Tipping-point triggers are expert estimates with wide ranges. A trigger being reached means a change can become self-sustaining; it does not mean the full change happens that year.

Formulas and estimates used on this page

Where no published number exists for a specific 5° band or year, this page calculates one. The formulas are:

  1. Band amplification (measured): Rmeas = (band warming trend, °C per decade, 1979–2024) ÷ (global warming trend, same period). Data: Berkeley Earth gridded land-and-ocean temperatures, averaged over each 5° band with area weighting. Global trend = 0.20°C per decade; equator (10°S–10°N) = 0.15°C per decade.
  2. Band amplification (models): Rmodel = (band warming, 2081–2100 minus 1995–2014) ÷ (global warming, same periods), median of 40 CMIP6 models under SSP2-4.5. The models put equator warming at 0.87 × global warming.
  3. “Times the equator”: band warming ÷ equator warming, from the same data.
  4. Band temperature in a year: Tband(year) = Tglobal(year) × R, shown as a range from Rmodel (low) to Rmeas (high). Tglobal(year) comes from the IPCC AR6 SSP2-4.5 central curve as tabulated in the Universe Institute timeline, and from that curve × 1.20 and × 1.30 for the precautionary cases.
  5. Extra warming versus the equator: Tglobal × (R − 0.87).
  6. Year a tipping trigger is reached: the first year Tglobal(year) ≥ trigger on each path. For 1.5, 2.0, 2.5 and 3.0°C the crossing years stated in the Universe Institute timeline are used.
  7. Local warming at a trigger: trigger × R for the band where most of that system sits (for example, Greenland uses 70–75°N; permafrost uses 65–70°N; Alpine glaciers use 45–50°N).

Limits. The ratio method (often called “pattern scaling”) assumes each band’s share of warming stays roughly constant as the world warms. The measured ratios come from 46 years that included fast sea-ice loss and slow tropical-ocean warming, so the high end may overstate future ratios; the model ratios may understate them. Band values are annual averages over very large areas and are not local forecasts. All calculation scripts and results are kept with this project.

Bibliography

  1. IPCC (2021). Summary for Policymakers. In: Climate Change 2021: The Physical Science Basis. Working Group I, Sixth Assessment Report. Cambridge University Press. B.2.1: Arctic warming “above two times the rate of global warming”; B.2.3: coldest Arctic days warm about three times the global rate; B.2.5: Arctic “likely to be practically sea ice-free in September at least once before 2050”; B.5.2: permafrost carbon loss irreversible at centennial scales; C.3.4: AMOC very likely to weaken. www.ipcc.ch/report/ar6/wg1/chapter/summary-for-policymakers/
  2. Fox-Kemper, B., et al. (2021). Ocean, Cryosphere and Sea Level Change. IPCC AR6 WGI, Chapter 9. Spring snow cover down about 8% and top-3 m permafrost volume down about 25% per 1°C of global warming; Greenland 0.04–0.13 m of sea-level rise by 2100 under SSP2-4.5. www.ipcc.ch/report/ar6/wg1/chapter/chapter-9/
  3. Canadell, J. G., et al. (2021). Global Carbon and other Biogeochemical Cycles and Feedbacks. IPCC AR6 WGI, Chapter 5 (permafrost carbon feedback, methane hydrates). Summary in: Woodwell Climate Research Center (2022), Review of permafrost science in IPCC’s AR6 WG1. www.ipcc.ch/report/ar6/wg1/chapter/chapter-5/
  4. Rantanen, M., Karpechko, A. Yu., Lipponen, A., et al. (2022). The Arctic has warmed nearly four times faster than the globe since 1979. Communications Earth & Environment 3, 168. doi.org/10.1038/s43247-022-00498-3
  5. Carbon Brief (2022). The Arctic has warmed ‘nearly four times faster’ than the global average (explainer of Rantanen et al.; Barents Sea up to about 7 times; model underestimation). www.carbonbrief.org/the-arctic-has-warmed-nearly-four-times-faster-than-the-global-average
  6. Armstrong McKay, D. I., Staal, A., Abrams, J. F., et al. (2022). Exceeding 1.5°C global warming could trigger multiple climate tipping points. Science 377, eabn7950. (Trigger thresholds, timescales and impacts in Table 3.) doi.org/10.1126/science.abn7950
  7. Kim, Y.-H., Min, S.-K., Gillett, N. P., Notz, D. & Malinina, E. (2023). Observationally-constrained projections of an ice-free Arctic even under a low emission scenario. Nature Communications 14, 3139. doi.org/10.1038/s41467-023-38511-8
  8. Pistone, K., Eisenman, I. & Ramanathan, V. (2019). Radiative heating of an ice-free Arctic Ocean. Geophysical Research Letters 46, 7474–7480. doi.org/10.1029/2019GL082914
  9. Box, J. E., Hubbard, A., Bahr, D. B., et al. (2022). Greenland ice sheet climate disequilibrium and committed sea-level rise. Nature Climate Change 12, 808–813. doi.org/10.1038/s41558-022-01441-2
  10. Rounce, D. R., Hock, R., Maussion, F., et al. (2023). Global glacier change in the 21st century: Every increase in temperature matters. Science 379, 78–83. doi.org/10.1126/science.abo1324
  11. Hjort, J., Karjalainen, O., Aalto, J., et al. (2018). Degrading permafrost puts Arctic infrastructure at risk by mid-century. Nature Communications 9, 5147. doi.org/10.1038/s41467-018-07557-4
  12. Polyakov, I. V., Pnyushkov, A. V., Alkire, M. B., et al. (2017). Greater role for Atlantic inflows on sea-ice loss in the Eurasian Basin of the Arctic Ocean. Science 356, 285–291. doi.org/10.1126/science.aai8204
  13. Jain, P., Barber, Q. E., Taylor, S. W., et al. (2024). Drivers and impacts of the record-breaking 2023 wildfire season in Canada. Nature Communications 15, 6764. doi.org/10.1038/s41467-024-51154-7
  14. Cohen, J., Zhang, X., Francis, J., et al. (2020). Divergent consensuses on Arctic amplification influence on midlatitude severe winter weather. Nature Climate Change 10, 20–29. doi.org/10.1038/s41558-019-0662-y
  15. Previdi, M., Smith, K. L. & Polvani, L. M. (2021). Arctic amplification of climate change: a review of underlying mechanisms. Environmental Research Letters 16, 093003. doi.org/10.1088/1748-9326/ac1c29
  16. Rohde, R. A. & Hausfather, Z. (2020). The Berkeley Earth Land/Ocean Temperature Record. Earth System Science Data 12, 3469–3479. Data used: Land_and_Ocean_LatLong1.nc. doi.org/10.5194/essd-12-3469-2020
  17. Eyring, V., Bony, S., Meehl, G. A., et al. (2016). Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6). Geoscientific Model Development 9, 1937–1958. Model output used: monthly surface air temperature, historical and SSP2-4.5, Pangeo/Google Cloud CMIP6 archive. doi.org/10.5194/gmd-9-1937-2016
  18. Universe Institute (2026). The Polycrisis and Climate Change Risk & Survival Timeline, v2.1. Source of the annual global temperature curve, the +20% and +30% precautionary cases, and the AMOC timing summary. universeinstitute.org (full forecast page)