Antarctic sea ice has recorded its third-smallest winter peak extent since satellite records began 47 years ago, new data reveals.
Provisional data from the US National Snow and Ice Data Center (NSIDC) shows that Antarctic sea ice reached a winter maximum of 17.81m square kilometres (km2) on 17 September.
This is 900,000km2 below the 1981-2010 average maximum extent – the historical baseline against which more recent sea ice extent is typically compared.
According to one expert, the “lengthening trend of lower Antarctic sea ice poses real concerns regarding stability and melting of the ice sheet”.
Meanwhile, at the Earth’s other pole, Arctic sea ice reached its annual minimum on 10 September, ranking as the joint-10th lowest in the satellite record.
At 1.6m km2, the 2025 minimum shares the spot with 2008 and 2010. The NSIDC notes that all 19 of the lowest sea ice extents in the record have occurred in the past 19 years.
Antarctic peak
For decades, scientists have been using satellite data to track the annual cycle of sea ice growth and melt at the world’s poles. This is a key way to monitor the “health” of sea ice in both the Arctic and Antarctic.
The map below shows Antarctic sea ice on the day of its maximum extent for the year on 17 September 2025, where the yellow line shows the 1981-2010 average.
The NSIDC says that sea ice extent was “markedly below average” in the Indian Ocean and the Bellingshausen Sea, but “slightly above average” over the Ross Sea.
In an NSIDC press release announcing the Antarctic maximum, Dr Ted Scambos, a senior research scientist at the Cooperative Institute for Research In Environmental Sciences, said:
“The lengthening trend of lower Antarctic sea ice poses real concerns regarding stability and melting of the ice sheet. However, it may also be leading to greater snowfall over the continent, which would slow the progression of sea level rise.”
Antarctic sea ice growth
In its typical annual cycle, Antarctic sea ice grows during winter towards its annual maximum extent in September or October. It then melts throughout the spring and summer towards its March minimum.
Earlier this year, Antarctic sea ice recorded its second-smallest summer minimum on record.
At 1.98m m2, this was the fourth consecutive year that Antarctic sea ice had fallen below 2m km2, the NSIDC noted.
In its monthly sea ice updates, the NSIDC reported that sea ice then grew at a “near-average pace”. During this period, sea ice “expanded rapidly” in the last areas to lose ice, including the Ross Sea and eastern Weddell Sea, it said.
The NSIDC explained that sea ice rebounded quickly in the Ross Sea area because ice extent had retreated “slowly” there the month before – meaning that the upper ocean layer did not have time to accumulate heat which would slow the winter freeze.
In April, “the situation in the Antarctic remained fickle”, the NSIDC said. At the beginning of the month, sea ice extent neared “record-low” daily extents, but as the month progressed ice cover expanded “fairly quickly”, it said.
May had “below average growth” in Antarctic sea ice and saw the fifth lowest record for Antarctic sea extent.
As June began, the the Bellingshausen Sea and eastern Queen Maud Lord regions were “far behind” in ice re-growth, it said, adding that the Bellingshausen Sea was almost entirely ice-free as temperatures were 6-8C above average.
In June, Antarctic sea ice was 1.28m km2 below the 1981-2010 baseline, with “particularly low” sea ice extent in the Bellingshausen Sea and the Indian Ocean sector, according to the NSIDC. This was the third-lowest sea ice extent ever recorded for the month of June, it said.
Throughout July, Antarctic sea ice extent grew at a “slower-than-average” rate, according to the NSIDC. By the end of the month, Antarctic sea ice extent was 1.3m km2 below the baseline, it noted.
Arctic melt season
In the Arctic, sea ice cover typically reaches its high point in March, before dropping to its September minimum at the end of the northern-hemisphere summer.
The 2025 Arctic sea ice winter peak was the smallest since satellite records began. The peak, recorded on 22 March, was 1.31m km2 below the average maximum for the 1981-2010 historical baseline.
In March, Arctic sea ice extent averaged 14.14m km2 – the lowest in the satellite record, according to the NSIDC. It noted that, at the time, average air temperature was above the historical baseline across much of the Arctic region.
Arctic sea ice extent then “changed very little” throughout April, remaining “nearly constant” until the final days of the month, the NSIDC reported.
It added that the final days of April saw Arctic sea ice extent drop due to ice retreat along the coast of the Barents Sea.
According to data, the main reason why the April total extent remained largely flat was due to an increase of sea ice in the northeastern Barents Seas that “offset” losses elsewhere.
Below-average air temperatures over the northern Norwegian and Barents Seas was the most “notable feature” of April 2025, the NSIDC said.
May was marked by a decline in Arctic sea ice extent at a faster-than-average pace, the NSIDC noted, resulting in the seventh-lowest May extent on record.
It added that ice loss in May was “primarily” in the Barents Sea, Bering Sea and the Sea of Okhotsk.
In June, Arctic sea ice extent averaged 10.48m km2 – the second-lowest average on record for the month, the NSIDC said. It noted that sea ice hit record-low levels over 20 June and 26 June and tracked at “near-record” low levels through the month. The Barents and Kara Seas were both “nearly ice-free” by the end June.
Hudson Bay ice extent was also “considerably below average” throughout June and northern parts of Baffin Bay were nearly ice-free, it said.
By the end of July, daily sea ice extent in the Arctic had fallen to 7.66m km2 – the third lowest in the satellite record, the NSIDC reported. It noted that, for most of the month, Arctic sea ice extent tracked close to levels recorded for 2012 – the year in which Arctic sea ice extent reached its lowest-ever September minimum.
Throughout August, the NSIDC reported that sea ice “rapidly melted and compacted” north of Alaska in the Beaufort Sea, with sea ice extent averaging at 5.41m km2 – the seventh lowest on record.
Dr Zack Labe – a climate scientist at Climate Central – tells Carbon Brief that northern Siberia saw August air temperatures more than 5C above the 1981-2010 average, resulting in “a striking amount of open water along the Atlantic side of the Arctic that would normally be ice-covered”.
At an annual minimum of 1.6m km2, this year’s Arctic minimum is “pretty unremarkable”, Labe tells Carbon Brief, and “adds to the evidence of a clear slowdown in the rate of summer Arctic sea ice loss”.
However, Labe stresses that this is “not surprising” – referencing a recent study which “clearly shows how internal variability can temporarily drive periods of slower melt in a warming climate, as well as periods of rapid melt, such as in the early 2000s”. (For more on this research, read Carbon Brief’s guest post).
He adds:
“It is only a matter of time before summertime melt accelerates again. This is not a good news story, especially since in many other months we still see a clear downward trend…
“While the past decade of summers may give the appearance of a slowdown, regional extremes such as in the Kara Sea this year underscore that the Arctic is already radically different from past decades. The driver is clear – human-caused climate change.”
Satellite switch
For decades, NSIDC has tracked sea ice using data from weather satellites run by the US Navy. However, earlier this year, Mongabay reported that NSIDC scientists “noticed holes in the data they were receiving”.
The article explains:
“When scientists inquired with the Department of Defense (DoD), they were told not all data were being downloaded and access to the data had been deprioritised. Soon after, the DoD said it would stop sharing…data altogether, citing military cybersecurity risks in the old systems.”
NSIDC scientist Walt Meier told Science that while the US satellites “are up there and functioning…we’re not getting all the data anymore, at least regularly”.
The DoD then set a cut-off date to “cease distribution data from the Defense Meteorological Satellite Programme” on 31 July.
In June, the NSIDC announced that it would “explore switching to a different sensor” aboard a Japanese satellite that was launched in 2012.
The only other option available to NSIDC was a “series of Chinese weather satellites, which the country is already using to produce its own record of sea ice”, Science noted. It added that a new US DoD weather satellite, launched last year, is “also capable of collecting similar data, but its data have not yet been made public”.
The switch was completed by the July cut-off date and NSIDC reprocessed all data for 2025 to use the new data source to ensure “consistency through the year”.
The post Antarctic sea ice winter peak in 2025 is third smallest on record appeared first on Carbon Brief.
Antarctic sea ice winter peak in 2025 is third smallest on record
The launch of the Africa Energy Bank (AEB) has been put back yet again, raising doubts about the institution’s future ability to finance fossil fuel projects – its main objective – as global lenders retreat from such investments over climate concerns, experts told Climate Home News.
The bank, which had been billed to launch in September after a series of delays, is now scheduled to begin operations in November, according to the head of the African Energy Chamber, an advocacy body for the continent’s oil and gas sector.
Even as the world aims to transition away from fossil fuels, many African leaders have made clear they want to continue exploring and extracting the continent’s large oil and gas deposits – estimated at around 125 billion barrels of crude and over 600 trillion cubic feet of gas – to boost economic development.
As a group, Africa sided with a number of powerful oil-and-gas producing nations in blocking progress on negotiations to craft a global roadmap to transition away from fossil fuels at last year’s UN COP30 climate talks, although some countries did individually support the proposal.
-
Campaigners oppose Dangote’s planned Kenya refinery over climate and ecological risks
Plans to build Africa’s second largest oil refinery in ecologically sensitive region criticised as “environmental recklessness” -
Tanzania pushed African nations to oppose fossil fuel transition at COP30
While chairing the African group at the summit, the gas-dependent nation advised the continent’s ministers to block language on transitioning away from fossil fuels
Meanwhile, major projects under development across the continent – including the 1,443-km East African Crude Oil Pipeline (EACOP) and Dangote’s 700,000-barrel-per-day Kenyan refinery – show that African governments see oil and gas as playing a significant role in meeting their energy and economic needs for many years to come.
In 2022, at a gathering of the African Petroleum Producers’ Organization (APPO) in oil-rich Angola, ministers from its member states adopted a resolution to create the Africa Energy Bank to finance projects for the production, use and trade of oil, gas and broader energy sources.
African control over energy resources
An article on the APPO website explains that the bank was conceived as a way to overcome “disenchantment” with fossil fuels among “the international community” which it said had crystallised around the “energy transition” concept.
“If Western countries, after having long taken advantage of the energy sources they now revile to develop, can afford the luxury of abandoning them, this is not the case in Africa,” it adds, noting that many of the continent’s economies are still largely dependent on oil and gas revenues.
A separate web page about the bank, also hosted on APPO’s website, says its objectives include financing the exploration, production and refining of oil and gas, as well as supporting member states in transitioning from fossil fuels to cleaner energy sources “while ensuring energy security”.
Said Addi, a former executive with Shell and energy commodities trading house Gunvor, said the new bank was judged necessary because financing for hydrocarbons from many traditional international lenders has become constrained.
In trying to fill this financing gap, Africa is not simply setting up another fund to support oil and gas, he added. “It is also an attempt to give African countries greater control over how their energy resources and infrastructure are financed,” he explained.
Nigeria to host the AEB
The energy bank – a joint initiative of APPO and the African Export–Import Bank (Afreximbank) – has so far suffered several delays and is almost two years behind schedule. The initial plan was to start operations in January 2025, with Nigeria as the host country, but the bank’s opening was delayed to June of that year to allow Nigeria time to finalise the construction of the bank’s headquarters in Abuja.
After the government announced the completion of the offices in late November 2025, a new launch date was set for January 2026, which was moved back to April, June and then September. Now it has shifted again to November, raising concerns that the institution may be losing momentum.
Former Shell executive Addi said that if the capital is eventually paid in, the bank becomes operational and its first projects are commercially credible, then the delays will be regarded as normal teething troubles in setting up a multilateral institution. But, he added, scepticism will be justified if it continues to stall.
Uganda may see lower oil revenues than expected as costs rise and demand falls
Baron Lamarré, an oil and gas expert and former Petronas oil trader, said that missing “three deadlines in a row is not normal”, and warned that if the timeline slips again, “the story flips from ‘ambitious institution finding its footing’ to ‘good idea that lost momentum before it found any’.”
The Nigerian government, APPO and Afreximbank did not respond to requests for comment by the time of publication.
The funding challenge
The Africa Energy Bank is targeting base capital of $5 billion, with plans to scale up to $120 billion within five years by mobilising private-sector funds. However, it is expected to start operations with initial seed capital of $500 million.
The funding plan is to have the 18 member countries of the APPO contribute $83 million each to the bank as equity for a combined $1.5 billion. Afreximbank, other non-APPO African countries and investors outside the continent are expected to provide the remaining $3.5 billion.
But even the initial $500 million has not been easy to mobilise. In May, APPO Secretary-General Farid Ghezali called on members to deliver on their pledges towards the startup goal before the end of June. But the delays suggest this may not have been met, with experts saying Africa may be finding it difficult to self-fund its oil and gas projects in the absence of international capital.
Lamarré said every extension of the deadline points to the fact that “raising fossil fuel capital in Africa without the majors and their financing networks is brutally hard”.
Why the global electrification agenda misses the point on Africa’s energy crisis
Since 2020, Western lenders, export credit agencies and insurers have been in steady retreat from African hydrocarbons, he said, while oil majors are divesting their African assets, handing over fields to smaller local operators whose credit ratings are not high enough to borrow cheaply.
Even capital from China and the Gulf, which has partially filled the gap, cannot match the volume, tenor or pricing that Western investors once offered, Lamarré argued.
“If mobilising the first $500 million of seed capital [for the AEB] has taken this long, that’s the clearest signal yet of how steep the climb to $120 billion looks,” he said, noting that the continent’s energy financing gap is as large as $30 billion-$45 billion per year.
Africa’s investment landscape, meanwhile, has been shifting. While foreign direct investment dropped from a 2024 peak, inflows remained roughly one-third above the continent’s long-term average in 2025, according to the 2026 World Investment Report from UN Trade and Development (UNCTAD). They are concentrated in a few sectors including critical minerals needed for renewable energy technologies, battery manufacturing and advanced industrial production.
At the same time, data on global energy investment from the International Energy Agency (IEA) shows that fossil fuel investment in Africa has declined over the last decade.
“Trojan horse” for fossil fuels
While the Africa Energy Bank struggles to get off the ground, climate campaigners have criticised its primary aim of financing oil and gas on the continent at a time when the world is starting to move away from high-carbon fuels to cleaner alternatives.
Bhekumuzi Dean Bhebhe, founder of Africa Change Lab, described the bank as a “Trojan horse”, arguing that its focus on fossil fuel financing runs counter to the global energy transition and the African Union’s Agenda 2063 goals of sustainable development and inclusive growth.
The energy bank, he warned, “risks locking Africa into a new cycle of debt, dependency and fossil fuel entrenchment”, adding that its financing blueprint does not pave the way for a climate-resilient future. “In truth, it is to deepen the same extractive, carbon-heavy pathways that the continent should be moving away from,” he added.
Ugandan farmers use British court to try to stop East Africa oil pipeline
Kenya-based climate and energy expert Joab Okanda said the AEB’s plan to finance oil and gas is “a misplaced priority” and it should instead back clean energy in line with the policies of some of Africa’s major export markets like Europe.
In addition, the new bank could struggle to mobilise enough resources to advance large-scale oil and gas projects, he added, noting that its proposed $5-billion initial capital is equivalent to the cost of the East African Crude Oil Pipeline alone.
The AEB’s aim of backing more fossil fuels should be flipped “to support countries that are oil-dependent to start working on their transition plans”, Okanda said.
The post Launch of Africa Energy Bank delayed again in blow to oil and gas hopes appeared first on Climate Home News.
Launch of Africa Energy Bank delayed again in blow to oil and gas hopes
A booklet published by the UK’s opposition Conservative party includes a “cost of net-zero” that appears to have been doubled by a spreadsheet error.
The “common sense” policy document argues that “what people ultimately want is a government competent enough to solve the problems for which it takes responsibility”.
In a section that says “sophisticat[ed]…modelling” should not be a substitute for “political judgement”, the “Right Way” document disparages various estimates of the cost of net-zero.
The Conservative document then claims – incorrectly – that the government’s official adviser, the Climate Change Committee (CCC), had put the cost of net-zero at close to £1tn. It says:
“In 2020, the CCC estimated that its route to net-zero would cost £957bn.”
In fact, the CCC’s 2020 estimate was exactly half this amount – £478bn – and last year it published a revised figure of £108bn, largely as a result of the falling cost of electric vehicles (EVs).
Spreadsheet error
The Conservative party’s erroneous claim appears to stem from another report that had accidentally added up numbers twice, using a spreadsheet published by the CCC in 2020.
The 2020 spreadsheet contains a table listing the additional investments that would be needed to build a net-zero economy, from low-carbon electricity generation through to heat pumps and EVs.
These extra capital expenditures, listed as “CAPEX”, add up to a total of £1.38tn over the 30 years of 2020-50. They are set against operational savings, listed as “OPEX”, of £0.90tn.
Added up over 2020-50, the combined CAPEX and OPEX figures come to a total of £478bn.
In addition to the annual sectoral CAPEX and OPEX figures, the CCC’s 2020 spreadsheet also has a line giving combined totals for each year. It appears that someone has added all of these numbers together, resulting in the savings and costs being counted twice.
This double-counted total for the cost of net-zero amounts to £957bn – as shown in the image below – and it appears to be the source of the claim in the Conservative booklet.
At the time of publication in 2020, the CCC said that the £478bn net cost of net-zero amounted to less than 1% of GDP over 30 years – and that the large investment needed would not only result in savings due to lower fossil-fuel imports, but that it would boost GDP overall, by around 2%.
In 2025, the CCC revised its estimates for investment costs and operating savings to £670bn and £562bn respectively, giving a net total of £108bn over 2025-50, or less than 0.2% of GDP.
Earlier this year, the committee said that cutting emissions to net-zero would cost less than a single fossil-fuel price shock and that doing so would have benefits worth £110bn per year.
Paper trail
The erroneous claim in the Conservative document is referenced to the CCC’s 2020 advice on the UK’s sixth “carbon budget”, which, as explained, does not contain the £957bn figure.
The earliest online use of the £957bn figure found by Carbon Brief is a 12 January 2026 article in the Spectator, by retired engineer and self-described “accidental energy analyst” David Turver.
A day later, Turver repeated the mistaken number in a report for the free-market Institute of Economic Affairs. His report cites figure 5.3 of the CCC’s 2020 advice.
However, as set out above, the CCC spreadsheet containing the data for figure 5.3 only adds up to £478bn, half the figure claimed by Turver.
It appears that Turver accidentally added up all of the numbers in the CCC spreadsheet, without noting that it already included a line