§ Emissions still flat

There has now been a plateau in China’s CO2 emissions from fossil fuels and cement for more than two years, following a peak in March 2024.

Previous analysis for Carbon Brief described this as a “flat or falling” trend, which extended until the end of 2025. There was then a 2% increase in emissions year-on-year in the first quarter of 2026, resulting from a rise in the amount of “wasted” wind and solar power.

The latest analysis shows that this was followed by another decline in the second quarter of this year, when China’s emissions fell by 1%, as shown in the figure below.

For further details see: About the data.

Notably, China’s emissions fell in the second quarter despite an increase in coal use. For the first time ever, a drop in oil use was sufficient to drive a decline in emissions overall.

§ Oil use plummeted while coal grew

Within the overall 1% decline in China’s emissions in the second quarter of 2026, there were divergent trends when looking sector by sector and fuel by fuel.

The largest fall in CO2 emissions came from the consumption of petrol, diesel and jet fuel, with oil consumption in industry also falling, as shown in the figure below.

For further details see: About the data. 

Crude oil processing volumes fell 11% in the second quarter, but some of the fall was absorbed by drawing down oil product inventories, with Sinopec sales down 9%.

In total, China cut back oil imports by 32% in the second quarter. The million–barrel question has been how much of this was enabled by genuine reductions in oil consumption and how much by the drawdown of the country’s vast oil stockpile.

Energy mix numbers reported by the National Bureau of Statistics indicate that oil consumption fell by 3% in the first half of the year and around 9% in the second quarter. This shows that reduced consumption played a substantial role, while still leaving 60% of the fall in imports to be covered by the swing from building stockpiles to using them.

The sector with the largest increase in emissions during the second quarter of the year was power, where coal use grew 2.4% while gas-fired generation fell 1.2%. This was despite strong growth in wind and solar capacity over the preceding year, a significant rebound in hydropower generation, a small increase in nuclear power output and a slowdown in electricity consumption growth.

The explanation for the rise in emissions was – similar to the first quarter of 2026 – an increased amount of solar and wind generation being “wasted” due to the power market and grid not being adapted to increasing shares of variable renewable generation.

In other sectors, there was a fall in cement production, driven by falling construction volumes, which accelerated to 9% in the second quarter, from 8% in the first quarter. Crude steel output fell by 1% and pig-iron production by 3% in the second quarter.

Growth of coal use for chemical production slowed down in the second quarter, both compared with the previous quarter and the last year.

The rate of utilisation of installed coal processing capacity was already high before the current oil shock, so there was no headroom for production to increase even though rising oil prices made coal-chemicals more profitable. Oil-based chemical production also kept growing, with ethylene output up 17% and primary plastics production flat.

Coal use for heating continued to increase, with the sector’s coal consumption in the second quarter dominated by industrial heat, as there is little need for space heating at this time of year. Growth has continued despite the prominent drive for “zero-carbon industrial parks”, demonstrating the importance of the initiative for tackling industrial coal use.

§ What drove the fall in oil consumption?

The dramatic fall in China’s demand for oil imports during the Hormuz crisis has been widely hailed as the most important price stabilising factor for the global oil market. 

To understand the implications for China’s oil consumption and CO2 emissions going forward, it is important to unpack what enabled this reduction in imports.

A significant contribution comes from ongoing, structural reductions in transport oil demand driven by electrification. Sinopec had forecast 6% and 5% drops in diesel and petrol consumption this year, respectively, already before the start of the war on Iran. Actual sales fell 9% in the first half of the year.

Transportation levels show a slowdown in growth, but no outright decline. Cross-regional passenger trips were 0.1% higher year-on-year in the second quarter, while urban passenger trips were 2.9% higher. Commercial freight tonnage increased 2.4%.

The exception is air travel, where passenger numbers fell 7% in May-June, after 7% growth in the first quarter. However, this sector plays a minor role in overall transport oil consumption in China.

The stable or growing transportation levels show that the shift to electric vehicles, rail, public transport and other clean transportation, rather than a fall in mobility, played the key role in reducing oil consumption.

The rise in fuel prices that accompanied the Hormuz crisis only accelerated the structural shifts in transportation that were already underway.

Electric heavy-truck sales rose about 77% in the second quarter, year-on-year, with June sales more than doubling and the market share of electric trucks exceeding 45% of all new sales.

The total number of EVs on the road at the end of the quarter grew 33% year-on-year. Some 12.1m EVs were added, of which 8.1m were electric-only battery EVs.

EV usage saw even more of a shift. Charging volumes increased 60% in the second quarter, indicating that EVs already on the road were utilised much more than before, at the expense of petrol and diesel vehicles, with plug-in hybrid drivers likely favouring electricity over fuel. 

One factor enabling EV utilisation to grow was the increased use of electric taxis. Intense competition in the sector has pushed prices down at the same time as the use of private petrol vehicles has become more expensive.

Stronger subway and rail use also made a contribution. Rail-passenger traffic increased 5% in the first half of the year.

The fall in diesel demand has been particularly pronounced in the construction and mining sectors. The heavy machinery in the sectors is well-suited for electrification, in addition to which construction levels are also falling.

Based on reported growth in charging volumes, EVs helped avoid an estimated 19m tonnes of oil consumption (Mtoe) in the second quarter, up 50% year-on-year.

This took the total amount of oil displaced by EVs to 36 Mtoe in the first half of the year, as shown in the figure below, well exceeding, say, the total oil consumption of the UK over six months. Notably, trucks are the fastest-growing source of oil displacement, with avoided fuel use up 90% year-on-year in the first half of 2026. 

For further details see: About the data. 

The increase in avoided oil consumption due to EVs is equal to 4.5% of China’s oil imports in the same period in 2025. If EV sales and charging volumes continue their growth at the same rates in the second half of the year, avoided oil consumption will reach 80 mn tonnes, equal to the consumption of Mexico.

Estimated emissions avoided are 35 MtCO2, or 1.3% of China’s total CO2 emissions in the second quarter, after taking into account emissions from power generation for vehicle charging.

While the amount of oil displaced by the shift to EVs is significant – and is rising fast – the year-on-year increase in displaced oil still only accounts for a third of the drop in China’s oil consumption in the first half of the year, with the fall in consumption only accounting for half of the drop in imports. The remaining reduction is due to the shift from building to drawing down stockpiles, slower growth in chemical industry output, as well as behavioral adaptations by consumers and operational adaptations by businesses.

§ Coal power continued to rise despite clean-capacity growth

China saw record increases in solar and wind capacity over the past year. In addition, hydropower generation increased 9% in the second quarter of the year, compared with the same period in 2025, and there was a small 2% increase in nuclear-power output.

At the same time, the rate of power demand growth slowed down from 5.9% in the second quarter of 2025 to 5.2% in the same period in 2026.

Yet, power-sector emissions increased 3.0% in the first half of 2026, after falling 3.2% in the first half of 2025. Power generation from fossil fuels rose because of an increase in the amount of potential solar and wind generation that was wasted, as well as exceptionally poor wind conditions. Without those factors, coal-fired power generation and power-sector emissions would also have fallen in 2026.

Wind-power capacity has continued strong growth in 2026, with capacity additions in both the first and the second quarter of the year comfortably exceeding those in any year other than the record-setting 2025.

Solar power additions have slowed sharply from the rates seen in 2025, even falling behind 2024. Yet, they are in line with 2023, when more than 200 gigawatts (GW) was added by year-end.

Nuclear power development continues at pace, with eight new reactors approved in July and five reactors with 4.5GW total capacity expected to enter commercial operation this year. This includes China’s second commercial small modular reactor, Linglong One, with new policies paving the way for further development.

Reactor commissioning will pick up further next year: the government has approved 10 new reactor projects every year since 2022 and those projects will begin to come online. Meanwhile, 3GW of conventional hydropower was added, with a total of 6GW of projects targeting operation in 2026.

Taken together, this clean-energy growth puts China on track to add enough non-fossil generating capacity in 2026 to cover electricity demand growth of up to 5%, despite the slowdown in solar.

Power demand grew 5.3% in the first six months of 2026 and the energy regulator projects 5-6% for the whole year. This means that the increase in power-sector emissions seen in the first half would be reversed, once the obstacles to solar and wind sending their output to the grid are addressed – and once wind conditions revert to average levels.

Moreover, total energy demand growth has slowed down much more sharply than electricity demand, making it more feasible for clean-power generation growth to significantly exceed the increase in total energy consumption and to drive down fossil-fuel consumption.

For further details see: About the data.

The key reason for solar and wind curtailment in China is that neither the power-grid operating model nor the electricity market model require – or encourage – the flexible operation of coal-power plants, hydropower plants and inter-provincial transmission lines. 

This situation has been exacerbated by a wave of new coal-power plants entering operation, with newly added capacity reaching 30GW in the first half of 2026, the highest level since 2016. Another 25GW started construction, while less than 3GW was retired.

The electricity prices paid to coal-fired generators are fixed months in advance, as are the volumes of electricity that will be transmitted through long-distance power lines.

This removes the incentive for plants to adjust their output in response to conditions. This could include variations in solar and wind supply, or changes in power demand.

As a result, there is limited ability for the grid to absorb variable renewable power. Furthermore, coal plants are entitled to “capacity payments”, which require them to be available to generate, but do not reward them for operating flexibly.

One solution to integrate more solar and wind into the grid is increasing energy storage capacity. Battery storage capacity continued to grow, with 17GW added in the first half of 2026, bringing total installed capacity to 153GW. This represents a slowdown in storage additions, however, down from 23GW in the first half of 2025.

§ Outlook for China’s CO2 emissions

The key developments affecting the outlook for China’s emissions in the second quarter include the effects of the Hormuz oil-and-gas crisis, the release of a long list of sectoral five-year plans and a slowdown in energy consumption growth.

The rise in oil prices has caused a stronger shift in China’s transportation sector than anyone anticipated, with EV deployment and use accelerating from an already high base. This trend is unlikely to be reversed. It has also proven the value of electrification to China’s energy security strategy. 

The government is targeting a slight acceleration in the pace of electrification, aiming for electricity to make up 35% of energy end-use by 2030, up from 30% in 2025. This is a larger increase than achieved over the past five years, when the share of electricity rose from 26.5% in 2020 to 30% by 2025. The transportation sector plays a significant role in this, with a target for EVs to make up 30% of the vehicle fleet, up from 12% in 2025, and 25% of commercial vehicles.

Electrification both reduces emissions immediately and sets different sectors up for deep decarbonisation as electricity is much easier to produce without CO2 emissions than fuels. Faster transport sector electrification lowers the outlook for oil demand, increases the role of the sector in peaking and reducing emissions, plus means that more of China’s clean energy growth ends up displacing oil.

While transport emissions fell, power-sector emissions continued to rebound for the second quarter in a row. The increased coal-fired power generation and emissions can be attributed to increased solar and wind curtailment. Curtailment has emerged as the key obstacle to both continued rapid solar and wind capacity growth and full utilisation of existing capacity.

Several sectoral five-year plans published in recent months have laid out measures to improve solar and wind utilisation.

Long-distance transmission will continue to expand, helping to move wind and solar generation from remote “energy bases” to centres of demand. There is also a growing emphasis on local consumption of clean power. The power sector five-year plan, published in August, promotes direct purchases of clean electricity, smart microgrids, zero-carbon industrial parks and closer coordination between renewable resources and AI computing infrastructure

Yet the same plan further loosened the limits on the amount of wind and solar that can be curtailed.

The limit for curtailment was 5%, until it was relaxed to 10% in 2024 in provinces with good wind and solar resources. The new plan allows the limit to be increased further to 15% for some provinces, while keeping it at 5% and 10% for others. 

Looking at the 2025 data on reported curtailment, very few provinces had higher rates than 15% – only Tibet for wind and Qinghai and Tibet for solar.

Unless the most lenient limit is only applied to those two provinces, it means the plan would allow for higher levels of curtailment.

This is also true of the national average target of “around” 10% curtailment, given reported rates in 2025 were 94% and 95% for wind and solar, respectively.

Notably, monthly data on curtailment has not been published in recent months, raising the possibility that the indicator is being revised. Reported data has understated actual curtailment by a wide margin, compared to implied curtailment.

If the curtailment indicator is revised, such that it captures more of the actual curtailment, then this could make the headline targets stronger than they appear, in comparison to previously reported numbers.

The new five-year plans also lowered the overall level of ambition on coal use. Chinese president Xi Jinping announced in 2021 that China would “gradually reduce coal consumption during the 15th five-year period”, covering 2026-30. However, the target now is for coal consumption to “enter a plateau” during those five years.

The five-year plans call for “reasonably controlling coal-power capacity and generation”, signaling a higher bar for the approval for new coal-power projects, after the government’s active promotion of new coal power in recent years. This could also imply more retirements of older coal plants. However, there is 204GW of coal-power capacity under construction, even after the wave of new coal-power plants starting operation in 2025 and in the first half of 2026, making the implementation of the “reasonable control” more challenging.

It is the first time that the government has vowed to control “coal-power generation” and not just “generation growth”, as the energy regulator did in 2021, but the significance of that distinction is unclear.

The renewable energy five-year plan also broadens the concept of system reliability, which was a key justification for new coal power during the previous five years. Rather than relying primarily on coal-fired power for system stability, it increasingly looks to other options.

Alternatives include storage, flexible demand, EVs, “virtual power plants” and smarter system operation to provide balancing services. The plan also puts an emphasis on increasing the contribution of renewable energy to meeting demand peaks.

Therefore, while coal remains an important backup resource in the plan, reliability is no longer framed as something that can only be provided by coal.

The Chinese government has published numerous other sectoral five-year plans since its overarching plan came out in March. These include plans for the energy sector (“new-type energy system”), power system, renewable energy, carbon peaking, coal, climate-change mitigation, and the environment (“Beautiful China”). Some clear priorities emerge from these plans: electrification, electric vehicles, energy storage, offshore wind and “green”” fuels.

The energy plan also substantially increased ambition on the development of conventional hydropower, despite ecological and social risks and potential for tensions with neighbouring countries. The capacity additions will largely only materialise after 2030, however.

At the same time, energy consumption growth has slowed down markedly after the surge during and immediately after the “zero-Covid” period, making it more feasible for clean energy to meet all incremental demand.

If this trend continues, then total CO2 emissions will begin to fall even as power-sector emissions continue to plateau.

§ About the data

Data for the analysis was compiled from the National Bureau of Statistics of China, National Energy Administration of China, China Electricity Council and China Customs official data releases, as well as from industry data provider WIND Information and from Sinopec, China’s largest oil refiner.

Electricity generation from wind and solar, along with thermal power breakdown by fuel, was calculated by multiplying power generating capacity at the end of each month by monthly utilisation, using data reported by China Electricity Council through Wind Financial Terminal.

Total generation from thermal power and generation from hydropower and nuclear power were taken from National Bureau of Statistics monthly releases.

Total primary energy consumption is converted to the electricity equivalent using the substitution method.

Monthly utilisation data was not available for biomass, so the annual average of 52% for 2023 was applied. Power-sector coal consumption was estimated based on power generation from coal and the average heat rate of coal-fired power plants during each month, to avoid the issue with official coal consumption numbers affecting recent data.

CO2 emissions estimates are based on National Bureau of Statistics default calorific values of fuels and emissions factors from China’s latest national greenhouse gas emissions inventory, for the year 2021. The CO2 emissions factor for cement is based on annual estimates up to 2024.

For oil, total oil consumption is calculated based on energy mix data for the first quarter and first half of the year released by the National Bureau of Statistics. Consumption of transport fuels – diesel, petrol and jet fuel – is estimated based on the sales growth reported by Sinopec for the first quarter and the first half of the year, with monthly disaggregation based on production minus net exports. The consumption of these three fuels is labeled as oil product consumption in transportation, as it is the dominant sector for their use. Apparent consumption of other oil products is calculated as the residual.

Estimated non-energy use of fossil fuels is subtracted from total chemical industry fossil fuel consumption, and process emissions are calculated based on fossil fuel consumption with carbon retained in products subtracted. Emissions from the incineration of plastics are based on a peer-reviewed estimate of plastics incineration in 2022, combined with growth rates in the overall power generation from waste-to-energy plants. Metals industry process emissions are calculated using industrial output data and IPCC default emission factors.

Oil consumption displaced by EVs is estimated using China Association of Automobile Manufacturers’ sales data, via Wind Financial Terminal. The data breaks down vehicle sales by type and powertrain: passenger cars, buses, vans, semis and trucks of different sizes, each split into battery-electric and plug-in hybrid, with assumptions about how far each vehicle type is driven per year and the fuel economy of the conventional vehicle it replaces. 

Annual mileage and fuel-consumption assumptions are compiled from different sources, including the International Council on Clean Transportation. Each electric vehicle sold is credited with avoiding the fuel a comparable internal-combustion vehicle would have burned; plug-in hybrids are credited only with the portion of driving done on electricity (a utility factor of 64%).

The electricity and oil figures are calibrated to figures from China’s National Energy Administration, which put new-energy-vehicle charging at 142.3 TWh in 2025 and reported 56.9% year-on-year growth in the first half of 2026. The second half of 2026 is a projection: each vehicle segment’s actual second-half-2025 displacement is grown by its first-half-2026 year-on-year rate.

CREA data scientist Hubert Thieriot contributed to implementing and reviewing the CO2 emission methodology.

§ CO2 equivalent 

How much methane corresponds to one tonne of CO2?

The question is as difficult to answer as: ‘how much spaghetti equals a chicken?’ You could compare the two meals according to their calories, protein content or cost. Each metric can be convenient, but is only valid for that specific comparison – no amount of spaghetti is the same as a chicken. 

The same is true for the conversion of emissions of methane and other gases to CO2-equivalent emissions. It can be convenient, as it allows different gases to be compared or combined into a single number. This is why the metric is used in climate targets or evaluating the effectiveness of different mitigation options. 

But, because methane and CO2 have different atmospheric lifetimes and warming properties, any conversion is only valid for a chosen time horizon and a chosen baseline.

Depending on the assumptions baked into calculations, methane mitigation can either appear as an immediate priority or framed as almost unnecessary.

There are a number of metrics that scientists use to convert greenhouse gases – whether methane, hydrofluorocarbons or nitrous oxide – into CO2-equivalent emissions:

• “GWP20” measures how much heat a greenhouse gas traps in the atmosphere over a 20-year period, relative to CO2. It emphasises urgent methane mitigation but has been criticised for its implicit discounting of future damages.
• “GWP100” looks at a 100-year timeline. It gives more weight to long-term warming and is used in “integrated assessment models” (IAMs) used by scientists, national emission reporting to the UN and by the GHG Protocol used by companies.
• “GWP*” considers the rate of emissions, rather than warming over a fixed time horizon. Under GWP*, very limited methane reductions bring CO2-equivalent emissions to zero, meaning remaining methane emissions can be designated as causing “no additional warming”. (This interpretation remains controversial as it assumes the continuation of historical levels of warming.)

IAMs are the tools used to generate future emissions scenarios. Because they combine CO2 and methane emissions, the impact of methane emission cuts alone is difficult to isolate in existing emission scenarios.

IAM-generated scenarios also assume mitigation decisions driven by costs. Combinations of CO2 and methane emission pathways that are not purely cost-effective are, therefore, not represented, even though climate policy is messy and emission pathways are rarely cost-effective in the real world.

Only a few countries – including Japan, Mexico and South Korea – specify methane mitigation targets.

§ A different approach

In our study, we separate CO2 and methane emissions and treat them as independent. 

Instead of choosing a conversion method, we suggest that states and organisations set a limit on peak global warming first, then, based on their existing net-zero targets, determine the minimum compatible methane reduction target. 

Companies and countries around the world have set net-zero targets focused on CO2, as well as those that include all greenhouse gases. As a result, our research looks at the necessary methane reductions for both types of goal. We consider scenarios where companies or countries deliver linear – in other words, steady – emissions reductions to reach net-zero.

Using a simple climate model, we systematically combined methane and CO2 (or greenhouse gas) mitigation pathways starting in 2025 and calculated peak warming. 

The figure below shows how peak warming depends on both the year of reaching net-zero CO2 and the level of methane cuts.

The blue arrows in the figure show that to limit warming to 1.7C under a 2050 net-zero CO2 scenario, methane emissions would need to fall by at least 69% by 2050, relative to 2020. 

Our research also finds that, if an organisation or country’s 2050 net zero-target covers all greenhouse gases, its methane emissions would need to fall by 63% instead.

However, under current policies, methane emissions are expected to increase by around 20% by 2050, relative to 2020. We find that this pathway would result in peak warming above 2C by 2050 – even if global CO2 emissions were to reach net-zero by that date (see purple bar on the right-hand side of the figure above). 

The figure also shows how, if methane emissions remained at 2020 levels and net-zero CO2 was delivered by 2040 or later, warming would exceed 1.85C. This level of warming is above what has been argued as consistent with the Paris Agreement’s “well-below” 2C limit.

Conversely, cutting methane emissions by around one-third – in line with the Global Methane Pledge target for 2030 – could reduce peak warming by 0.15C, of which 0.05C could be delivered by interventions that come at no net cost. These are shown by the orange and red bars, respectively, on the figure above.

The table below highlights the minimum compatible methane cuts for three different peak warming levels and net-zero CO2 or greenhouse-gas emission targets.

Minimum methane emission reductions between 2020 and the year of net-zero emissions, consistent with peak warming of 1.7C, 1.8C, and 2.0C at 50% likelihood, assuming linear emission trajectories. For some net-zero targets and peak warming levels, there are no compatible methane mitigation targets (indicated by “–”).

§ Remaining carbon budget

The global carbon budget refers to the amount of cumulative CO2 emissions allowable while still meeting a particular global warming threshold.

The 2021 climate science report from the Intergovernmental Panel on Climate Change (IPCC) and a 2023 Nature study estimated that, by 2025, the remaining carbon budget for holding warming to 2C would be around 1,000-1,150bn tonnes of CO2 (GtCO2). 

We find that these estimates are founded on the assumption of methane reductions of 27-35% by 2050, relative to a 2020 baseline. (A 2024 Communications Earth & Environment study reached similar conclusions.) 

Under the GWP* metric, where methane emissions are only cut to maintain “no additional warming”, the remaining carbon budget would be constrained. The best estimate of a 2C budget shrinks by around 30% to approximately 750GtCO2.

Finally, if methane emissions are not cut at all in the future, our findings suggest that the remaining carbon budget for 1.7C of global warming has, in effect, already been exhausted.

Our analysis shows how peak warming depends on both CO2 and methane reduction – and how methane-specific targets can help refine existing net-zero targets.

Crucially, we show that complementing net-zero CO2 targets with stringent methane cuts is necessary to limit peak warming to well-below 2C.

§ A new generation of scenarios

To simulate how human activity could shape the climate of the future, climate modellers must estimate future levels of “radiative forcings” – the external drivers that cause global warming. These include atmospheric concentrations of greenhouse gases, air pollutants and land-use changes.

Given that no one knows how the future will unfold, modellers use a handful of scenarios that span a wide range of plausible outcomes.

The Scenario Model Intercomparison Project (ScenarioMIP) coordinates the development and running of emissions scenarios for climate models used in IPCC reports. 

In April 2026, high-level details about the new set of scenarios for CMIP7 were published in the journal Geoscientific Model Development (GMD).  

On 1 September, the underlying emissions data was released into the public domain by the ScenarioMIP team. 

There are seven new CMIP7 scenarios designed to drive model simulations for AR7. The first model runs took place in spring 2026 and initial results are expected later this year.

The previous SSP scenarios were starting to show their age. Finalised in 2015-17 using historical data ending in 2015, several years projected by the SSP scenarios were already in the past by the time AR6 concluded in 2021. Meanwhile, the world had changed considerably. 

(For a full guide to the SSPs, see Carbon Brief‘s 2018 explainer.)

§ Storylines and emissions levels

The most visible change in the new generation of scenarios is their names. Where the SSPs combined five socioeconomic “storylines” with radiative forcing targets (SSP1-2.6, SSP5-8.5, etc), the CMIP7 scenarios are named simply for the emissions trajectory that they follow.

(It is worth noting that, although the CMIP6-era scenarios are broadly referred to as the SSPs, shared socioeconomic pathways underlie both the CMIP6 and the new CMIP7 scenarios.)

The table below summarises the seven scenarios and the integrated assessment model (IAM) that produced each “marker” run – in other words, the specific IAM run used to generate the scenario that, in turn, will be used by CMIP7 climate models. IAMs run simulations of how the future energy system and emissions may evolve under different assumptions around socioeconomics, future technology costs and climate policy.

The table below also details how the scenario fares against a number of key metrics assessed by Carbon Brief, including CO2 emissions and warming outcomes. 

(For more on Carbon Brief’s approach, see: Methodology.).

Warming values are medians (with the 5-95% range) from the 841-member FaIR ensemble used in this article (see: Methodology); the marker model assignments come from the ScenarioMIP database. Note that scenario names in the database differ from the official CMIP7 names (for example, the high-to-low scenario appears as “SSP5 – Medium-Low Emissions_a”).

Each of the new scenarios is built on a set of updated SSP storylines similar to those used in the original SSP scenarios. These include assumptions about future population, technological and economic growth, as well as potential for international cooperation that shape the resulting emissions pathways. The socioeconomic assumptions underlying these revised SSPs were updated in 2024 with new population and economic projections. 

Most of the new emissions scenarios are now based on the “middle-of-the-road” SSP2 that assumes current socioeconomic trends broadly continue, with only one scenario using each of SSP1 (“sustainability”), SSP3 (“regional rivalry”) and SSP5 (“fossil-fuelled development”). None of the new scenarios uses SSP4 (“inequality”). 

The solid lines in the figure below show updated global human population, GDP and GDP per capita values in CMIP7 (solid lines), compared to the original SSPs from CMIP’s sixth phase (CMIP6), shown by the dashed lines.

The change in socioeconomic assumptions is substantial. Global population was revised upward in nearly every scenario, with the updated SSP2 projecting there will be 9.9 billion people in 2100 – an increase of 1 billion people compared to the 2013-era SSP. 

GDP was revised downward in the high-end growth scenarios (SSP1 and SSP5), slightly upward in SSP3 and SSP4 and was largely unchanged in SS2. 

The combination of these changes means that income per person in 2100 is around 10-25% lower in most scenarios, with only SSP3 and SSP4 seeing mostly unchanged income per capita. 

In short, the socioeconomic world underlying the new scenarios is somewhat more crowded and less wealthy per person than the one the SSPs originally imagined.

Another notable change is the shift in the SSP that underlies the highest future emissions scenario. 

In the original SSPs, the “very high” SSP5-8.5 scenario was based on SSP5, while the new “high” scenario in CMIP7 is based on SSP3. 

The GMD study explains that this is because IAM teams that developed the scenarios found that SSP3 and SSP5 variants produced similar emissions. They judged that the “fragmented” SSP3 world – which is characterised by large challenges to adaptation – to be more relevant for exploring high-end risks.

§ No more ‘baseline’ scenarios and other changes

In another important change, the authors of the CMIP7 scenarios decided to eliminate “baseline” scenarios that assumed a world without any climate policy. These scenarios were previously used as a counterfactual against which to compare climate-changed worlds.

Instead, the range of future emissions scenarios starts with current policies and explores ways that they could be strengthened, weakened, or kept the same. The high scenario explores a plausible “rollback of current mitigation policies“. 

The medium scenario, by contrast, extends climate policies officially implemented as of 2025, without assuming countries achieve their Paris Agreement pledges – known as nationally determined contributions (NDCs) – or net-zero targets that are not yet backed by legislation. 

In their GMD paper, the authors of the CMIP7 scenarios emphasise that the medium scenario “should not be considered as a ‘most likely’ scenario”, but that it can provide a benchmark against which the effect of future policy strengthening or weakening can be measured. It is roughly analogous in its emissions levels to the old SSP2-4.5 scenario.

The new low scenario explores a world where climate policy is rapidly strengthened and warming by 2100 is limited to below 2C. This makes it analogous to the old SSP1-2.6 scenario. 

The very-low scenario limits global warming to around 1.5C by 2100, similar to the old SSP1-1.9 scenario. However, it involves a greater degree of overshoot mid-century, reflecting the fact that global emissions did not begin to rapidly decline in 2020 as envisioned by SSP1-1.9. As the authors of the GMD ScenarioMIP paper point out: “At this point of time, some overshoot of the 1.5C seems unavoidable.”

In addition, there are a number of scenarios that start on one path before undertaking rapid mitigation. These high-to-low, medium-to-low and low-to-negative scenarios are intended to explore futures where mitigation is further delayed, followed by a rapid turn-around later in the century.

The scenario developers noted that there is no specific likelihood or probabilities assigned to any scenario, but rather only a judgement that all are within the realm of plausibility given where the world is today. They also said that “there might be potential futures outside the ScenarioMIP scenario range”.

§ Timescales and other changes

In addition to the shift away from baseline scenarios, there are three other notable design changes made in CMIP7. 

First, CMIP7 models will be driven by emissions of CO2 and other greenhouse gases, rather than set atmospheric concentrations. 

In every previous generation of models, the ScenarioMIP experiments required that modelling groups simulate future climate using the same set of CO2 concentrations. For CMIP7, models with an interactive carbon cycle are asked to run in “emissions-driven” mode for CO2, calculating atmospheric concentrations themselves based on emissions. 

This is a significant improvement. It means that the substantial uncertainty in carbon-cycle feedbacks will now show up directly in the range of projected warming, rather than being overlooked. (The change applies to CO2 only; methane, nitrous oxide and halocarbons remain prescribed as concentrations.) 

Second, emissions match observations up to 2023. IAM modellers were asked to stay close to observed trends up to 2025 to avoid emissions diverging from reality before models were run. Scenario differences only open up after 2026, avoiding an earlier problem of scenarios diverging from reality years before the models were even run.

Finally, the period over which models are being run has been extended from 2100 to 2150. This is important as the world is already more than a quarter of the way through the 21st century. 

The extended model runs out to 2150 will provide a more thorough exploration of the warming that people born in the coming decades may experience within their lifetimes. 

In addition, all scenarios have extensions to 2500 where temperatures are eventually stabilised. These allow scientists to explore changes to long-term Earth-system processes, such as ice sheets and sea level, as well as whether warming is reversible. 

§ A narrower range of future CO2 emissions

Overall, the new scenarios provide a notably more narrow range of future CO2 emissions than the SSP scenarios used in CMIP6. 

The figure below shows net global CO2 emissions (combining fossil-fuel and land-use emissions) for the seven new scenarios, alongside the five SSP scenarios used for climate model runs in CMIP6 (e.g. SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5).

At the bottom of the range, the new scenarios closely track their predecessors: the very-low scenario reaches net-zero CO2 around mid-century much like SSP1-1.9, while the low scenario lands close to SSP1-2.6. 

The chart below shows total emissions for the same scenarios for the period 2024-2100.

The lowest emissions scenarios now have somewhat higher total emissions, reflecting the failure of the world to rapidly reduce emissions after 2020 that occurred in the lower SSP emissions scenarios, such as SSP1-1.9 and SSP1-2.6. The very-low scenario results in 310bn tonnes of CO2 (GtCO2) cumulative emissions between 2024 and 2100, compared to around 110GtCO2 in SSP1-1.9.

At the top end, the change is particularly dramatic. The high scenario in CMIP7 reaches 55GtCO2 per year in 2100. The previous high scenario, SSP5-8.5, by contrast, reached around 126GtCO2 per year in 2100. 

In cumulative terms – which is what matters most for global warming – high reaches around 3,820GtCO2 over 2024-2100, half the roughly 7,600GtCO2 of SSP5-8.5 and about three-quarters of the 5,140GtCO2 of SSP3-7.0.

To put it another way: the top of the new scenario range sits between SSP2-4.5 and SSP3-7.0 in cumulative emissions terms, which is territory that CMIP6 treated as its middle ground.

To make the scale of this shift clear, Carbon Brief analysed the CO2 emissions trajectories in each of the prior generations of high-end emissions scenarios, using the same simple climate model – FaIR – to calculate future warming.

Below, four different generations of emissions scenarios are examined. The SRES scenarios were originally published in 2000 and used in the IPCC’s third (2001) and fourth (2007) assessment reports (and the corresponding CMIP3 model runs). The RCPs were developed in the early 2010s and used in the IPCC fifth assessment report (AR5; 2013) and CMIP5, while the SSPs were developed in the late 2010s and used in the IPCC AR6 report and CMIP6.

Over the past two decades, the highest emissions scenarios all resulted in comparable amounts of warming in 2100: SRES A1FI (the highest SRES scenario) reached 4.6C in 2100 (5-95% range; 3.5-6.1C), RCP8.5 reached 4.9C (3.7-6.5C) and SSP5-8.5 reached 4.6C (3.5-6.2C). 

(RCP8.5 edges out its successor despite lower CO2 emissions because it assumed considerably more methane and nitrous oxide.) 

CMIP7’s high scenario comes in remarkably lower, at 3.3C (2.6-4.4C).

The downward revision of future emissions in CMIP7 reflects two key changes since RCP8.5 was published back in 2011. First, the plausible baseline of a repeal of current policy has fallen. Cheap solar, wind and batteries, a global plateau in coal use and more than $2tr per year in clean-energy investment mean that a rollback in climate policy would not result in coal deployment levels assumed in the RCP8.5 scenario.

The GMD study states that CMIP6’s high-emission levels “have become implausible, based on trends in the costs of renewables, the emergence of climate policy and recent emission trends”. 

(For more, see Carbon Brief’s recent factcheck of false claims around the retirement of the SSP5-8.5 emissions scenario. Also see Carbon Brief’s recent interview with Prof Detlef van Vuuren, a key architect of both the old SSPs and new scenarios.)

Second, part of the apparent decline reflects a correction of how scenarios are communicated – rather than real-world progress. The old high-end scenarios always represented an estimate of worst-case scenarios at the time, rather than likely outcomes. 

Genuine progress in reducing emissions probably accounts for around 0.7C of the roughly 1.7C gap between SSP5-8.5 and today’s current-policy trajectory, with the remainder reflecting that the baseline was never particularly likely.

§ What the new scenarios mean for future warming

To compare warming outcomes across scenario generations on a like-for-like basis, Carbon Brief ran both the seven CMIP7 scenarios and the CMIP6 SSP emission scenarios through the same simple climate model. 

(This is FaIR v2.2, using the 841-member ensemble calibrated and constrained to match the assessment of climate sensitivity in IPCC AR6, historical warming and ocean heat uptake). 

These values may differ from the ultimate results that are found by CMIP7 climate models, but give a sneak peak of what those results may look like when they become available.

The seven scenarios produce warming in 2100, relative to pre-industrial (1850-1900), that ranges from 1.6C (with a 5-95% range of 1.1-2.5C) in the very-low scenario to 3.3C (2.6-4.4C) in high, with the current-policy medium scenario reaching 2.9C (2.2-3.9C). Warming also continues after 2100 in both the medium and high scenarios.

The figure below shows the range of 2100 warming (5th to 95th percentile) relative to the preindustrial period expected in each of the old SSP scenarios and the new CMIP7 ones, along with a central estimate (white dots).

The largest changes are, unsurprisingly, at the top. CMIP7’s high scenario (3.3C in 2100) produces less warming than SSP3-7.0 (3.7C in the same ensemble) and far less than SSP5-8.5 (4.7C). 

The entire CMIP6 “high” tier (e.g. SSP5-8.5 and SSP3-7.0) now sits above anything in the new scenario set, at least up to 2100. Extended beyond 2100, however, high keeps climbing towards levels the previous extreme scenarios reached earlier.

At the low end, the picture is more similar. The very-low scenario (1.6C in 2100) lands close to SSP1-1.9 (1.5C) and low (1.8C) is essentially indistinguishable from SSP1-2.6 (1.8C) in 2100. 

However, the new low scenario involves more rapid late-century emissions reductions and greater amounts of carbon removal than its SSP analogue, while the very-low scenario involves greater overshoot of 1.5C mid-century.

§ Crossing warming thresholds

In addition to calculating 2100 and 2150 warming, Carbon Brief has calculated the likelihood of passing different global warming levels (2C, 2.5C, 3C, 4C and 5C) over time in the new CMIP7 scenarios. 

The chart below uses the IPCC approach of calculating the crossing year based on a 20-year average, rather than when a single year exceeds the warming level.

Under the medium scenario, which reflects a world where current policies are maintained, passing 1.5C is essentially locked in. 

Most models cross the threshold by the late 2020s or early 2030s. The 2C limit is crossed around 2050 on average and 3C by around 2110. The chance of exceeding 4C is around one-in-four by 2150, but, ultimately, rises to roughly 50% if emissions continue after that point.

Under the high scenario, 2C arrives in the 2040s, 3C in the 2080s and the chance of exceeding 4C by 2150 is around 60% (and around 95% by 2300). Even 5C is reached by 2150 in roughly 20% of climate model simulations.

The lower scenarios tell a different story. In the very-low scenario, the chance that peak warming (which the IPCC determines using a 20-year average of warming) ever exceeds 1.5C is around 90%. This reflects the fact that passing 1.5C is almost unavoidable at this point. 

However, the chance of surpassing 2C sits at around 30% and the scenario has warming falling after mid-century as more CO2 is removed from the atmosphere than is added. 

§ Carbon dioxide removal

Every scenario that has global warming peaking and declining requires pulling CO2 back out of the atmosphere. Otherwise, warming from CO2 emissions will persist for millennia. 

CO2 removal (CDR) remains one of the few levers available to reduce future temperatures – particularly given additional warming caused by cuts to aerosol pollution. 

The chart below shows the total CDR deployment in each of the different scenarios by year, reflecting the sum of both land-based and engineered approaches (top), as well as the total CDR deployment between 2024 and 2150 (bottom).

Every scenario that deeply cuts global emissions in CMIP7 also involves a large amount of CDR. 

The low-to-negative scenario pulls a cumulative 2,360GtCO2 out of the atmosphere by 2150, roughly 60 years of today’s emissions run in reverse. 

The high-to-low scenario has around 1,480GtCO2 cumulative CDR, medium-low has 1,450GtCO2 and low has 1,360GtCO2. 

Even the very-low scenario, which seeks to minimise CDR use, requires 655GtCO2 of removals between 2024 and 2150.

The degree to which scenarios rely on “engineered” removals – such as the use of biochar or direct air capture – or land-based removals – including afforestation and reforestation – ranges  across models. 

In the low scenario, roughly one-third of the removals is from the land “sink”, while low-to-negative relies almost entirely on engineered methods, with direct air capture alone reaching around 16GtCO2 per year by 2100.

The chart below shows the deployment of engineered removals by year (top), as well as the total engineered CDR used between 2024 and 2150 (bottom). The lower plot also includes a breakdown between the portion of CDR that requires geologic storage (e.g. DAC and BECCS) and the portion that does not (e.g. enhanced weathering and biochar) and compares the total to a recent “prudent” total CO2 storage limit published in the scientific literature.

(For more on limits to carbon storage capacity, see Carbon Brief’s 2025 guest post.)

The amount of CDR going toward geological storage is most highest in the low-to-negative scenario, which injects around 1,750GtCO2 of BECCS and direct-air-capture CO2 underground by 2150. 

The high-to-low and low scenarios each commit around 800GtCO2 to storage by 2150. This is within the range of available geologic storage, but would require that the storage industry handles more CO2 than the mass of oil currently moved by the fossil-fuel industry. 

That said, there are other potential CDR approaches – such as enhanced rock weathering, surficial mineralisation and ocean alkalinity enhancement – that do not require injection of CO2 into geologic formations. In-situ mineralisation approaches that inject CO2 into alkaline rock formations such as basalt or peridotite could also open up more potential CO2 storage.

It is worth noting that the amount of CDR deployed in these scenarios would require planetary-scale engineering at the cost of trillions of dollars, while many of the engineered CDR approaches are still relatively early-stage technologies. 

§ No single climate future

The goal of scenarios is to span a range of possible futures. While it may be tempting to treat current climate and energy policies – and the medium scenario – as a forecast, there is no reason to expect that they will not change in the future. 

It is likely that policies will continue to be strengthened, as has been the case over the past two decades. However, they may also be weakened if national priorities or politics change, as has happened in the US during the two terms of the Trump administration.

In the new CMIP7 scenarios there is no “business-as-usual” scenario, but rather a narrower range of futures than was available in CMIP6, reflecting greater clarity among scientists on where the world is heading in terms of future energy use and emissions. 

The fact that the worst-case scenarios of the past have become increasingly implausible is good news. However, this is tempered by the fact that the very-low emission scenarios have, in turn, become harder to achieve given that global emissions have yet to decline.

There is also real uncertainty in the climate-system response to emissions. This is due to uncertainty around how sensitive the climate is to a build-up of CO2 in the atmosphere, as well as how the carbon cycle will respond to emissions. 

The CMIP7 medium scenario – which has a central estimate of 2.9C of warming by 2100 – still has around a 3% chance of reaching 4C by that date. If emissions continue, those odds increase to 25% by 2150. This remains far outside anything resembling a safe outcome for the climate system.

The scenarios are now being run using the new CMIP7 models, whose emissions-driven runs will fold carbon-cycle uncertainty directly into projections. These projections will subsequently be analysed in the reports of AR7. 

Ultimately, it will be decisions made by governments, businesses and individuals that decide which of these seven futures become closest to reality.

§ Methodology

Emissions scenarios shown in this article are the seven CMIP7 ScenarioMIP scenarios set out in van Vuuren et al. (2026), harmonised to observed 2023 emissions, with rule-based extensions to 2500 generated using the FLEX methodology. Emissions through 2100 match the ScenarioMIP database; extension trajectories are indicative and may differ from the final published extensions.

Temperature projections use FaIR v2.2 with the fair-calibrate v1.4.5 constrained ensemble (841 members set out in Smith et al. (2024), which matches the AR6 assessed climate sensitivity (ensemble ECS median 3C, 5-95% 2.0-5.1C), historical warming and ocean heat content. 

Historical emissions (1750-2022) use the FaIR historical emissions dataset, with scenario emissions spliced in after 2023. 

Solar and volcanic forcing are updated through 2025 from the Climate Indicator forcing timeseries; future volcanic forcing ramps to the 1850-2021 climatological background by 2035 (following the CMIP7 protocol) and solar forcing follows a SOLARIS-HEPPA-derived cycle projection to 2300. 

All warming is expressed relative to 1850-1900.

SSP comparisons run the RCMIP-harmonised CMIP6 scenario emissions through the FaIR ensemble, which yields 2081-2100 warming 0.1-0.3C below the AR6-assessed values at the high end (e.g. SSP5-8.5: 4.2C vs 4.4C assessed), reflecting differences between the AR6 assessment and the FaIR configurations used here. Updating the volcanic dataset to use CMIP7 values (which revises the eruption-rich 1850-1900 baseline period) raises all reported anomalies by 0.03-0.05C.

For CDR, the scenario database reports the technology split (for example, BECCS, direct air capture, enhanced weathering, biochar, ocean-based, soil carbon management). Agriculture, forestry and other land-use  (AFOLU) removals are available only as a net flux, so are shown as the net sink where negative. Soil-carbon management is grouped with land-based rather than engineered removal, and the geological storage comparison uses BECCS plus direct air capture only. 

The figure showing high-end scenarios for the past four CMIP generations runs SRES A1FI through the same ensemble using the A1G MiniCAM model from the SRES database v1.1, spliced onto historical emissions at 2000, and covering CO2 (fossil and land use), methane, nitrous oxide and sulphur; SRES-era ozone-precursor projections (nitrous oxide, carbon monoxide and volatile organic compounds) lie outside the calibrated range of FaIR, so RCP8.5 values are used instead. RCP8.5 uses RCMIP v5.1 emissions, with 13 minor halogenated gases absent from the RCP database following SSP5-8.5.

§ Humid heat

High temperatures can be deadly. For example, the heatwaves that swept across Europe in the summer of 2026 have been linked to tens of thousands of “excess deaths”.

A prominent 2021 study found that children born in the 21st century will be exposed to more extreme weather events in their lifetimes than their parents and grandparents. 

Four years later, a study conducted by scientists from the same team found that more than half of children born in 2020 – around 62 million people – will experience “unprecedented lifetime exposure” to heatwaves, even if warming is limited to 1.5C.

Now, the latest research from the same team finds that children already face greater  exposure to dangerous levels of humid heat than adults as a result of human-caused climate change.

Extreme heat is particularly dangerous when combined with high humidity. In hot weather, the human body produces sweat to cool itself down. However, as humidity increases, sweating becomes less effective.

The study uses indoor wet-bulb globe temperature – a measure of temperature that takes humidity into account – to calculate heat stress. It defines a “heat-stress day” as any day with a wet-bulb globe temperature above 28C, as this is considered the threshold for “moderate heat stress”

The authors then use climate models to simulate global temperature patterns in the present-day climate. (The authors use the climate of 2023, in which human activity has caused 1.3C of warming, to represent the “present-day”.)

They then count the number of heat-stress days that each country records on average, per year. The authors then repeat this exercise, simulating a pre-industrial climate without human-caused warming.

By comparing the number of heat-stress days in the present-day climate with the number in a pre-industrial climate, the authors can determine how many extra heat-stress days were driven by climate change. They refer to these as “extra” or “attributable” heat-stress days.

The authors find that “low-latitude” countries, located in the tropics, record the most extra heat-stress days.

For example, the paper finds that people living in Côte d’Ivoire currently face 112 heat-stress days every year. It adds that around half of these are due to human-caused climate change. 

In contrast, Germany sees only 0.1 heat-stress days per year in today’s climate on average, which is largely attributable to human-caused climate change.

Rosa Pietroiusti, a PhD student at Vrije Universiteit Brussel and lead author on the study, explains why this number may seem lower than expected.

She tells Carbon Brief that the paper “really focuses on humid heat, at levels that are relatively rarely felt in Europe”. She adds:

“Our data also doesn’t capture the urban heat island effect, due to the resolution of the data we use, which also would lead to underestimations of heat stress locally, and lead to a mismatch with what people are experiencing at local scales, particularly in cities.”

§ Inequality

Extreme heat affects some people more severely than others. Children, people over 65 and those with pre-existing medical conditions or certain disabilities are among the most vulnerable. This is because their bodies are less able to regulate their temperature. 

The authors use gridded demographic data to determine the age structure of each country. From this, they calculate how many people from each age cohort are exposed to extra heat days as a result of climate change. 

The research finds that globally, 583 million children under the age of 10 already live through at least 20 attributable heat days every year. This accounts for 44% of all children in this age bracket. 

In comparison, 190 million people aged 60-69 face at least 20 attributable heat days per year, accounting for 30% of this age cohort.

The authors find that children face the greatest exposure to humid heat for two main reasons. 

First, there are more young people alive today than older people, with 1.3 billion children aged under 10 in the world, compared to 0.6 billion people aged 60-69.

Second, they find that countries in Africa and Asia typically have rapidly growing populations with more young children. In contrast, many countries in the northern hemisphere – which are typically cooler – have older populations.

The map below shows how many extra stress heat days each country currently faces as a result of human-caused climate change. Darker reds indicate a higher number of attributable heat days. The blue circles show the percentage of the population under the age of 10, with larger circles indicating a higher percentage.

§ Warming world

The authors also repeat their analysis for a 1.5C and 2C warmer world. They use population estimates from the SSP2 scenario, which projects that the world’s population will peak at more than nine billion in the second half of the 21st century, with most growth occurring in low-latitude regions – especially in sub-Saharan Africa.

The research finds that, in today’s climate, 11% of all under 10s currently experience 100 or more extra heat-stress days per year due to climate change. In worlds warmed by 1.5C and 2C, the percentage rises to 13% and 23%, respectively.

In contrast, only 6% of all people aged 60-69 currently face 100 or more extra heat-stress days each year due to climate change. This number rises to 9% and 17% for 1.5C and 2C worlds, respectively. 

These results are shown in the plot below. The three rows represent the climates of 2023 (top), a 1.5C world (middle) and a 2C world (bottom). The columns show different age cohorts, from the oldest on the left to the youngest on the right.

Each circle contains 100 coloured dots, with each dot representing 1% of the age cohort. 

The colour of the dot represents exposure to annual heat-stress day, with darker dots indicating more heat-stress days. Grey dots mean that people experience fewer than one extra heat-stress day per year due to human-caused climate change, while black dots mean more than 150 extra heat-stress days due to climate change.

The figure shows that higher warming levels expose more people to heat stress and that younger cohorts tend to be worst affected.

For example, the top-right circle represents heat stress for under 10s in the present-day climate. Three of these dots are coloured black, indicating that 3% faced at least 150 attributable heat-stress days in 2023.

Pietroiusti tells Carbon Brief the study uses wet-bulb globe temperature because it is a “well-established heat stress metric”. However, she notes that it was not “explicitly defined to focus on children”. She continues:

“A really important step forward in the research community would be to link up climate science and health science experts to do research on what metrics are really most representative of, for example, health impacts and educational impacts that children will be suffering.”

§ Vulnerability

Dr Qinqin Kong, a postdoctoral researcher at the departments of medicine and health policy at Stanford University, who was not involved in the study, praises its “robust” methodology.

He tells Carbon Brief that the research provides “a timely quantitative evidence for discussions of climate justice, children’s rights and intergenerational equity”. 

However, Kong suggests that the paper “may overstate the contrast between children and the elderly and underestimate the relative burden of older adults”.

He says:

“The elderly may also be more vulnerable due to their social circumstances. Children often benefit from parental supervision and caregiving, whereas many older adults live alone, have limited mobility and face barriers to accessing cooling or emergency assistance during heat events.”

Kong also notes that “people and societies in the mid-latitudes [for example, across much of Europe and North America] are less adapted to heat”, which may make them vulnerable to its impacts.

For example, he says that Europe “shows substantially stronger relative risk of heat mortality likely due to less heat-acclimatised populations, lower air conditioning prevalence and urban designs that don’t favour heat dissipation”.

Similarly, Dr Daniel Vecellio – a researcher at the University of Nebraska, who was not involved in the study – tells Carbon Brief that children are an “understudied cohort”. 

However, he says there is “reason for hope” because “children are typically pretty good behavioural adapters to extreme heat” and because people who are “chronically exposed to extreme heat” will “have a better chance at better acclimatisation”.

Pietroiusti tells Carbon Brief that global reporting on heatwaves is often skewed towards wealthier nations. 

For example, she notes that large-scale databases of disasters, such as EM-DAT, often underrepresent heatwaves in Africa, due in part to a lack of news coverage and formal reporting. She adds:

“Studies like this, which start from the climate data, can start to fill some of these gaps.”

She adds that the paper should inform discussions about climate justice, noting that children in developing countries, who are most severely affected by the increase in heat-stress days, “have contributed the least to historical greenhouse gas emissions”. 

§ The plan would increase UK emissions

The report by Onward is based on modelling by advisory firm Transira Energy, which compares two pathways out to 2050.

One is a “business-as-usual” scenario based on current “net-zero” policies. (Nevertheless, this only achieves a clean power system by 2045 – far short of the 2030 Labour target.)

The other is an “alternative policy pathway” (APP), developed by Onward, which assumes the UK’s 2050 economy-wide net-zero target is abandoned after the next election in 2029.

The latter says it places “greater emphasis on reducing the cost of electricity”, which includes fewer renewables, no electrification goals and more gas and nuclear power capacity.  

This mirrors the policy platform set out by the Conservatives, who argue that “net-zero” drives up energy costs and that climate change can be tackled without such targets.

In fact, the Conservatives say their “common sense” approach would make it easier to cut emissions, as shadow energy secretary Claire Coutinho states in the report foreword:

“If we want those emissions to fall, then we need people to want to use electric cars and electric heating – then our priority should be to make electricity cheap.”

Yet, this argument is firmly contradicted by the report itself. 

The APP results in an extra 524MtCO2 being emitted between 2030 and 2050 – equivalent to the annual emissions of South Africa.

The Transira Energy analysts say this is “explained by an increased share of unabated gas-fired capacity”.

Finally, it is worth noting that the UK’s net-zero target is based on the fact that the planet will continue warming until global emissions reach net-zero. Without such targets, climate change – and its impacts – will get worse.

§ The plan would slow electrification

Contrary to Conservative claims, uptake of heat pumps and electric vehicles is actually expected to be slower in the alternative scenario, “despite lower electricity costs”.

This is due to the removal of supportive government subsidies and mandates, such as the boiler upgrade scheme and the 2030 ban on the sale of new petrol and diesel cars.

Overall electricity consumption is 7% lower in the APP, compared to the current pathway. 

Daniela Quiroga, a senior associate at Copenhagen Infrastructure Partners, questions this reliance on lower electricity demand in the APP, telling Carbon Brief:

“While this is an interesting scenario to explore, it overlooks potentially important feedback effects – mainly, as electricity prices and the capital costs of electrification technologies fall, uptake would be expected to increase.”

A related point was made in a LinkedIn post by Tara Singh, chief executive of trade body RenewableUK, who noted:

“APP makes the electricity system cheaper partly by electrifying Britain less – while leaving the fuel costs that replace electricity outside the model.”

For example, Singh estimates that the extra petrol and diesel fuel expenditure to replace the missing electric vehicles (EVs) on the road could be around £65-95bn over two decades. These costs are not included in the APP scenario.

The only sector that sees increased power demand is data centres, due to policy support to “prioritise” new grid connections for these facilities.

Quiroga notes that the costs of accelerating data centre connections “are not mentioned at all” in the report. 

In short, the proposed pathway involves removing grants that help households buy EVs and heat pumps, while providing more policy support for the AI industry.

Finally, Onward stresses the UK’s “high spark gap” – referring to the electricity-to-gas price ratio. This makes switching from gas boilers to heat pumps less appealing for consumers, given the relatively high price of electricity, compared to gas.

However, Matt Elliott, lead economic analyst at the Energy and Climate Intelligence Unit (ECIU), says the analysis does not indicate this gap would substantially change in the proposed APP. He tells Carbon Brief:

“The report claims that electrification would happen even without specific policies, simply due to lower retail electricity prices driving consumer choice. However, its own modelling indicates that the gas-electricity price ratio would actually rise in the early years and end up only marginally lower than today by 2050.”

In other words, in the APP the price of electricity compared to gas would not fall sufficiently to drive consumers towards heat pumps without subsidies or other incentives.

Rather than scrapping net-zero policies, analysts have suggested shifting tax and policy levies from electricity to gas, or breaking the link between wholesale gas prices and electricity, as more effective ways to reduce the spark gap. 

§ Gas prices are unlikely to remain low and stable

The “alternative” scenario pushed by the Conservatives continues to rely heavily on gas for both electricity generation and heating.

This includes constructing new gas power plants in a bid to lower electricity prices, despite the fact that gas is the main driver of high electricity prices in the UK.

In recent years, the largest spikes in energy prices have been triggered by wars in Ukraine and the Middle East, which have disrupted fossil-fuel supplies and sent gas prices spiralling.

(Indeed, the report was published on the same day the Office for National Statistics announced that inflation had jumped to its highest rate in four months, due to energy costs surging because of the impact of the Iran war on global oil and gas supply chains.) 

Despite this, the scenario set out by Onward assumes that gas prices drop to pre-conflict levels and remain that way for the next two decades.

Ashutosh Padelkar, research lead at Aurora Energy Research, tells Carbon Brief that the gas price assumptions are “hard to fathom” and significantly at odds with future expectations, from both Aurora and other market analysts. 

Analysis by E3G and ECIU in 2025 concluded that four years of energy spikes caused by the post-pandemic demand surge and Russia-Ukraine war had cost the UK £183bn.

The Onward report acknowledges that the new scenario is “more exposed to a future gas price shock” than the current net-zero scenario. It suggests that a new spike could increase fuel costs in the gas-reliant scenario by another £6bn in 2040.

However, Onward argues that the impact of gas price spikes on consumers would be “significantly smaller” than the shock following Russia’s invasion of Ukraine. This is owing to existing renewable energy contracts and future nuclear power construction in the APP.

In the press release accompanying the new report, Conservative leader Kemi Badenoch is clear that “our plan means using our own oil and gas in the North Sea”.

This mirrors rhetoric that has been widespread on the right of UK politics, stressing the importance of expanding North Sea drilling as a way to cut energy bills.

However, given the relatively small volumes remaining in the North Sea, the UK will likely remain reliant on gas imported from the US and the Middle East. 

Gas prices will still be set globally and remain subject to geopolitical turmoil, no matter where the UK sources its supplies.

Given this, Johnny Gowdy, director of the thinktank Regen, tells Carbon Brief that the scenario presented by the Conservatives is “a call to rely on imported gas, with global gas prices”.

§ The plan assumes gas plants are cheap to build

The Conservative plan involves building new gas power plants, in order to meet part of the nation’s growing electricity demand without relying on renewables.

Onward states that the UK “has lost firm generation capacity” – such as gas and nuclear plants – and replaced it with “intermittent”, or variable, power in the form of wind and solar. 

To remedy this, its alternative pathway involves building an extra 21 gigawatts (GW) of gas power plants by 2050 – equivalent to around 20 new facilities. This is roughly a 70% increase from the UK’s current capacity.

However, the small print in the accompanying Transira Energy report explains that it assumes capital expenditure – the cost of building the power plants – is £650 per kilowatt (kW).

This is considerably lower than other recent analyses, which tend to cite capital expenditure figures that are more than double this estimate.

For example, a 2025 GridLab report notes that new US gas power plants set for completion in 2026 and 2027 had a cost range of $1,116/kW (£819/kW) to $1,427/kW (£1046kW).

However, it adds that more recent projects are “routinely reporting” costs of $2,000/kW (£1467/kW) or more. Other sources have reported up to $2,800/kW (£2054/kW).

Gas power plant costs have increased significantly in recent years – a trend that has been attributed to a tight supply of gas turbines worldwide.

This, in turn, is the result of increased demand for gas turbines to power data centres and countries transitioning from coal to gas. 

The International Energy Agency (IEA) says data-centre demand in the US is “limiting the availability of turbines for near-term deployment elsewhere in the world”.

§ Nuclear faces high costs and delivery challenges

The Onward report champions a substantial increase in nuclear power capacity. 

However, it fails to explain how this could be facilitated or why its cost assumptions are lower than the most recent nuclear projects in the UK. 

Within the report’s net-zero scenario, there is 13.3GW of nuclear power by 2050, roughly double the current capacity. It notes that this will be financed under the regulated asset base (RAB) model – a government-backed funding approach announced in 2022.   

Under the APP scenario, nuclear power capacity more than triples from current levels to 20GW by the middle of the century, all backed by the RAB model. 

The report adds: 

“Reducing nuclear construction costs and timelines becomes the core energy priority of the UK government, with measures to improve the availability of sites and grid connections.” 

The report acknowledges that the APP scenario “faces significant cost headwinds from expensive nuclear capacity”. 

However, it suggests that large-scale nuclear power stations built in the 2040s could cost £122-£138 per megawatt hour (MWh) in 2025 terms.

Hinkley Point C – which in 2018 became the first new nuclear power plant to begin construction in the UK since the 1980s – has a “strike price” of £138/MWh for 2030. (This is the fixed price for the electricity it will generate, guaranteed by the power plant’s contracts for difference agreement.)

This price is at the top end of Onward’s forecast range for “levelised cost of electricity” (LCOE) – the average total cost of building and operating an asset over its lifetime. 

As such, the report suggests, on average, costs will fall over the course of the decade from 2030, but provides little detail as to how this would happen. 

As Richard Howard, global research director at Aurora, wrote on LinkedIn, the cost assumptions for nuclear are “optimistic”. He adds: 

“It assumes that the LCOE of nuclear will fall 10-20% below the *original* cost of Hinkley Point C, when we know that nuclear costs escalated massively since the HPC deal was struck. The UK does not have a great track record of managing down the costs of nuclear.”

In fact, Sizewell C – a replica of Hinkley Point C in the early stages of construction in Suffolk, which received a final investment decision in 2025 – has a considerably higher strike price of £150/MWh in 2039. 

Hinkley Point C is the first new nuclear power plant to be built in 30 years in the UK. It has been beset by delays and nearly doubled in cost since it was originally approved.

A footnote in the Transira Energy report adds that its calculations for the cost of nuclear include expected capital expenditure for new large-scale plants ranging from £10,000/kW to £12,500/kW.

While the 3.26GW Hinkley Point C was originally supposed to have a price tag of £18bn, which would equate to £5,521/kWh, costs have repeatedly increased. More recent estimates from developer EDF suggest a figure of £10,736/kW, closer to Onward’s figure.

However, if this is adjusted for inflation for 2026, this jumps closer to £14,724/kW. 

As such, the upfront cost of new nuclear is already around £2,500 more per kilowatt than the assumptions in the report for 10 years from now. 

The report provides limited information about how these costs would fall so substantially. 

It suggests that the recommendations from the 2025 Fingleton review should be implemented in full to cut the cost of the technology.

The Fingleton report – a full review of the UK’s nuclear sector by the Nuclear Regulatory Taskforce, led by John Fingleton – found an “overly complex” and “bureaucratic” system was holding back the nuclear industry. It advocated for “smarter regulation”, as an overhaul of the planning regime. 

In March 2026, the Labour government committed to full implementation of the Fingleton review by the end of 2027. Despite this, the Onward report includes the implementation of the Fingleton review in the APP scenario, but not the net-zero scenario. 

§ The report’s high network cost estimates do not ‘add up’

The biggest drop in costs outlined in the Onward APP scenario comes from a reduction in network costs, but experts have said that this “just doesn’t add up”.

Network costs are broadly made up of the price of building, maintaining and operating the transmission and distribution systems. 

A reduction in network spending accounts for £137bn of the £320bn in “savings”, compared to the net-zero scenario that sees significant network expansion to help facilitate more renewables on the grid. 

This drop is “thanks to a higher utilisation of firm power system with supply located closer to demand”, the report says.  

In particular, the report points to discrepancy between the “best wind resources” being located in the north of Scotland, while the major centres of demand are in the southeast of England. As such, currently grid expansion is needed to avoid constraints or the requirement to curtail generation in windy periods with low demand. 

By avoiding the connection of geographically dispersed generation assets, such as 78GW of generation, storage and interconnectors, the APP scenario can reduce total network costs by 43%, according to the report. 

Staffell tells Carbon Brief that the £137bn saving has “a convincing story to it – if we build more fossil and nuclear capacity we can utilise the system better”.

However, he adds that Onward gives “so little detail about how this works that it’s hard to comment”.

The Transira Energy report notes that the APP still includes £19bn in investment for the electricity network, covering the cost to maintain the existing system and connect new gas and nuclear generation.

However, this 86% drop in new transmission investment compared to the BAU scenario leans on “flawed logic”, according to Tara Singh from RenewableUK.

On LinkedIn, she explained that it “rests on an extraordinarily aggressive assumption about how little grid Britain will need”, adding: 

“Onward assumes £137bn of new transmission assets under BAU between 2030 and 2050, but only £19bn under their plan, even though by 2050 it still has 32m EVs/hybrids, more than 6m additional heat pumps, 45GW gas, 20GW nuclear and – particularly strikingly – 62 terawatt hour (TWh) a year of datacentre demand. Is this grid figure credible…?”

Beyond this, the report also attributes a significant portion of the proposed savings to cuts in “balancing costs”. These are the costs to the system operator of balancing electricity supply and demand. 

It claims that having more firm generation located closer to demand and existing transmission infrastructure will “save billions of expenditure on network expansion and balancing costs”. 

Onward suggests that under the APP scenario, the cost of keeping generation and demand balanced would fall by £67bn. 

However, claiming savings by both cutting network expansion and balancing costs amounts to “double counting” and “just doesn’t add up”, according to Aurora’s Padelkar.

He tells Carbon Brief that including both high capital expenditure for the electricity network and high balancing costs in the BAU scenario is “difficult to reconcile”. 

Expanding the electricity network would reduce constraints, reducing the need for constraint management. Such a move would lower balancing costs. 

As noted by the National Energy System Operator (Neso), retaining the current transmission network into 2030, with no expansion, would mean constraint costs could reach around £12.7bn a year. But building new network capacity could cut costs by as much as 75%.  

Padelkar says: 

“They’re saying ‘we continue to invest in the network’…But somehow the network [balancing] costs just don’t come down…This is basically saying ‘we’re paying both to fix the problem and to have the problem’. You can have one of the two, but you can’t have both.”

Despite the claim that the APP approach will lead to the cheapest electricity, Padelkar says that the report does not present a “consistent picture” as to how the system would operate, pointing to the approach to network and balancing costs. He adds: 

“Overall, we would expect that once these figures are correctly accounted for, that renewable energy would remain the cheapest form of a form of decarbonisation. I would even further flip the argument around, to say that decarbonisation is not a prerogative [on] its own, but because it also achieves lowest total costs.”

§ The system integration costs are ‘far out of line with mainstream thinking’

A central argument in the Onward report is that the costs of renewables are higher than often claimed by proponents, due to the wider system costs of having a large amount of “intermittent” generation. 

As such, it proposes pulling back support for wind and solar, and instead putting focus on “firm generation” sources, particularly gas and nuclear power. 

This relies heavily on the claim that “system integration costs” for wind and solar are much higher than is being “properly revealed” in either contracts for difference (CfD) auctions or levelised costs estimates. 

(CfD’s are power contracts between generators and the government, which work as the UK’s main method for supporting the development of renewables by providing long-term price certainty to developers.)

Therefore, when assessing the overall cost of renewable energy, the cumulative network investment, balancing and ancillary services system costs necessary to manage such variable generation must be considered, it suggests.

The existence of integration costs is not widely understood, but the scale of their impact is disputed.

The report continues that if these costs are taken into account, the “marginal system integration costs” of renewables are “much higher than their individual levelised costs”.

Onward suggests that the cost to integrate additional offshore wind, onshore wind and solar onto the electricity system is £125/MWh. This is far higher than the cost of generating electricity from these sources in the first place.

The figure has been challenged by a number of commentators, with Staffell telling Carbon Brief that this is “very far out of line with mainstream thinking”. 

Analysis published in Nature suggests that if 80% of the electricity mix comes from renewables, the system integration cost is around €30/MWh (£26/MWh). 

Elsewhere, engineering firm Afry put the total cost of electricity at around £55-75/MWh in a high-renewable system. This is “less than [Onward’s] integration cost alone”, Staffell adds.

The high price tag of the £128/MWh marginal integration “is derived by apportioning additional balancing and transmission costs solely to 60GW of new wind and solar deployed from 2030 onwards”, explains Callum MacIver, research fellow at the University of Strathclyde and the UK Energy Research Centre. 

He adds:

“[This figure] only looks at the cost side and there is not enough published detail on where the renewables are deployed and the transmission upgrades it triggers to critique the scale of the numbers presented. 

“It also excludes potential wider system benefits of further renewables deployment, including reduced wholesale prices, avoided fuel and carbon costs and reduced exposure to future external gas price shocks, which are properly examined by looking at overall system costs and testing various sensitivities including different gas price futures.”

Writing on LinkedIn, Adam Bell – a partner at consultancy Stonehaven – suggests that the £125/MWh system costs are “really egregious”. He explains: 

“The ‘system costs’ of renewables…rests on assuming that all additional network upgrades and balancing costs for a net-zero system after 2029 are attributable to additional renewables deployed in that net-zero system. 

“Many of those costs relate to existing renewables as well as nuclear, so this likely overstates system costs by an order of magnitude [roughly 10-fold].”

Furthermore, the system costs for the APP scenario are not fully accounted for in the report. Regardless of the technology mix, old network and generation assets will need replacing, adding additional costs to the system. 

§ The proposed changes could undermine investor confidence

The APP scenario involves stripping back all support for renewables going forward.

It calls for the CfD scheme to end in 2030. Pre-existing CfD contracts would continue under APP, but after this decade, all further support would “exclusively” be for nuclear power.

Additionally, the renewable obligation (RO) payments for existing wind and solar would end from 2033. These are legacy contracts signed ahead of the scheme closing to new applicants in 2017. Payments are expected to continue until 2037.

(Onward makes an exception for the large-scale biomass power plant owned by Drax, which already has a contract with the UK government to switch from an RO to a low-carbon “dispatchable CfD”. This switch is included under both the net-zero and APP scenarios, in recognition of the “importance of its contribution to generation and to system stability”.)

Both the CfD and RO schemes have contributed significantly to the expansion of the renewable energy sector in the UK. For example, despite coming to an end in 2017, nearly 30% of current electricity supplies are still covered by RO contracts.

It is unclear from the report what the 10GW of capacity currently expected to receive the RO would do beyond 2033. 

Writing on Bluesky, Tom Haddon, senior economist at Arup, says that if, as the APP scenario proposes, the UK “bin[s the] RO”, this could force 10GW of renewable capacity still on the system to simply shut down after 2033.  

Such a dramatic change to a longstanding support system could have an impact on investor confidence. 

Padelkar tells Carbon Brief that energy investors are often involved in numerous technologies. He adds: 

“You wouldn’t be able to say ‘yeah, not going to continue honouring this contract [for renewables], but I expect you to sign this new one for me [to build new nuclear]’. That just wouldn’t work.”

As such, there is no guarantee that investors would agree to enter into government-backed RAB contracts to develop nuclear power plants, having just seen government-backed RO contracts being reneged on four years early.

§ Carbon market ‘savings’ are ‘just rearranging things on a spreadsheet’

One of the large chunks of “savings” identified to bring down electricity prices in the Onward report is £94bn from “lower wholesale prices, thanks to the removal of carbon taxes”.

This refers to removing power plants from the UK emissions trading scheme (UK ETS) from 2031. 

Onward argues that this reduces the cost of gas power plants, which frequently set wholesale power prices under the marginal pricing system.

Staffell tells Carbon Brief that this is a “concern” when considering the report’s findings:

“That is £94bn no longer going into the government coffers, so it’s not saving the country any money; it’s just rearranging things on a spreadsheet. This lowers electricity bills, but does that get compensated for by higher taxes elsewhere, or do we have to take on a larger national deficit, or does it go hand-in-hand with cutting public services?”

Tom Edwards, a consultant at Cornwall Insight, wrote on Bluesky that it would be “madness” to simply remove the UK ETS and “expect things to remain stable”.

The UK currently sources around a tenth of its electricity via interconnectors that link its grid up with Ireland and parts of mainland Europe. It also exports electricity to other European countries when it has surplus supply.

These relationships would be complicated if the UK abandoned its carbon price on electricity altogether. 

The UK and EU have been negotiating over linking their carbon pricing systems, which would involve the UK navigating the EU’s carbon border adjustment mechanism (CBAM).

Alongside ending support for renewables, the new Onward scenario also removes subsidies for new interconnectors, although it says “existing interconnectors will continue”. 

The Transira Energy analysis says there would be “new cross-border trading arrangements” from 2031. Such “arrangements” would, presumably, need to be negotiated from scratch with the EU.

Specifically, the report proposes a “carbon reference price” for electricity sold to the EU to “prevent carbon leakage and the distortion of cross-border electricity flows”.

Adam Berman, policy director at Energy UK, pointed out that the post-Brexit trade and cooperation agreement between the UK and the EU includes a legal commitment by the UK to maintain a carbon price on electricity. He wrote that the Onward proposal “would run contrary to that agreement”.

§ The report ‘grossly simplifies’ long-duration energy storage

The Onward report states that it would cancel support for long-duration energy storage (LDES), such as large batteries and pumped hydropower.  

This follows the government recently launching a “cap-and-floor scheme” to support the technology. In June 2026, the nation’s energy regulator Ofgem identified 16 LDES that it is “minded to” support under the new scheme. 

LDES can store power across days, weeks or even seasons, helping to boost electricity system security. Analysis by analytics company LCP Delta suggests that rolling out LDES technologies could cut energy system costs in the UK by more than £24bn between 2030 and 2050.

Onward lists support for storage systems – including LDES, as well as smaller batteries, which are only briefly mentioned in the report – as one of the “costs of an intermittent-first, low-carbon electricity system”. 

As such, alongside cuts to support for renewable energy technology, the APP scenario includes ending the cap-and-floor scheme for LDES. (See: The proposed changes could undermine investor confidence)

The report suggests that even if all 16 of the projects shortlisted by Ofgem were built, the total would only provide around five and a half hours of generation. 

It adds: “This is not enough to make it through a winter spell of low wind and sun”. 

This assertion is based on the total storage capacity of all the projects being 136GWh. 

However, the report “grossly simplifies the operation of LDES”, explains Padelkar. He adds: 

“This assumes a rate of discharge that the fleet doesn’t have. Further, this LDES capacity would play a key role in reducing the balancing and ancillary costs, even in the early 2030s, by helping absorb cheap wind generation in Scotland in constrained periods and then discharging it when the transmission from Scotland to the south of Great Britain is not constrained.”

The role of LDES is more complex than simply all projects providing the entire electricity demand for the nation in one go. The projects are designed to act together with other assets to absorb excess supply, smooth out peaks in demand and step in to provide cheaper power when prices spike. 

§ What is LDES?

LDES is a broad category of technologies, with some variation in definition. 

The UK government defines it as technologies that can store energy for anywhere from four hours up to years. Ofgem uses a slightly different threshold of eight hours and upwards. 

Sir Chris Llewellyn Smith, emeritus professor of physics at the University of Oxford and lead author of a Royal Society report on large-scale electricity storage, tells Carbon Brief: 

“[The Department of Energy Storage and Net-Zero] (DESNZ) seems to describe it as including things which we would regard as some short duration or medium duration [storage]. It’s a big confusion…For us, long duration is stuff that can last not just into seasons, but into years and into decades.”

LDES can be used to support several different aspects of the electricity system, including the integration of variable renewable energy. 

Currently in the UK, there is 2.8 gigawatts (GW) of LDES, made up of four pumped-hydro energy storage assets in Scotland and Wales.

(This article refers to the UK throughout, but strictly relates to the island of Great Britain made up of England, Scotland and Wales. Northern Ireland is part of the separate all-Ireland electricity system.)

The largest of these existing sites is the Dinorwig power station in North Wales, sometimes referred to as the “electric mountain”. This is a 1,728 megawatt (MW) station opened in the 1980s, which is used to manage short-term surges in electricity demand. 

For example, during England’s football World Cup match against the Democratic Republic of Congo on 1 July 2026, electricity demand rose by around 1.2GW at half-time and 1.7GW at full-time. This is equivalent to the total electricity demand for the cities of Glasgow and Leeds, combined.

Pumped storage, alongside batteries, has been used to keep the electricity system balanced during such moments by providing enough electricity to keep the system secure very quickly. 

As the UK’s electricity system becomes increasingly dominated by variable renewables, however, the need for LDES to manage peaks and troughs of generation is growing.

George Martin, principal for power system modelling at analytics company LCP Delta, tells Carbon Brief that wind power creates a particular need for LDES. He says:

“[LDES is] really important for the system, particularly in a wind-driven system. You get more peaks and troughs in your renewable output and, [while] short duration [storage] can obviously help with that, with things like ‘dunkelflaute’, long-duration storage is what is needed.”

As such, the UK is working to expand the capacity and duration of storage available through LDES, as well as the range of technologies this system is based on. 

For example, in May 2026 the UK’s largest vanadium “flow battery” site opened, co-located with a 3MW solar farm in Uckfield, East Sussex. (A flow battery stores energy in liquid chemical mixtures that are pumped between tanks, via an electrochemical cell.)

The Uckfield site consists of 90 vanadium flow batteries, which can be used to store 21 megawatt-hours (MWh) of electricity. This is equivalent to seven hours of peak output from the attached solar farm and is roughly enough electricity to power 3,000 homes for a day. 

The batteries can be used to store surplus daytime solar generation, which can then be used in the evening and overnight.

Other LDES technologies with a longer storage capacity could be used to similarly help manage power supply and demand, but over weeks, months or seasons. This could include compressed-air energy storage, hydrogen storage and others.

The diversity of LDES technologies reflects the range of roles it is expected to play in the electricity system in the UK. This could be meeting short-term surges, helping to utilise surplus renewable energy generation or providing longer-term flexibility. 

§ What types of LDES are available?

There are numerous types of energy storage technology, although most fall into four main categories: mechanical; thermal; chemical; and electrochemical. 

For example, a pumped-hydro project uses surplus energy to pump water uphill to a reservoir. The mechanical energy is released when the water flows down through a turbine.

Thermal storage could be a tank of gravel that is heated up, then later used to warm up water. Electrochemical storage is familiar in the form of batteries.

Finally, chemical storage relates to energy stored in molecular bonds, for example, making hydrogen from water. (Similarly, the energy in fossil fuels, which is ultimately derived from the sun, is a form of chemical storage.)

A key consideration for each LDES technology is the amount of energy it can store, measured in watt-hours (Wh). For example, a 1MW battery with four hours of storage contains 4MWh of electricity. It can therefore be used to deliver 1MW continuously for up to four hours.

Another consideration is whether the energy can be stored for long periods before use – and whether it is economic to do so.

In recent years in the UK, battery energy storage – predominantly lithium-ion batteries with a duration of one to four hours – has dominated the storage sector. The lithium battery sector in the UK has grown from almost nothing in 2015 to more than 6GW today.  

However, as lithium-ion batteries have only tended to hold a few hours of storage, they cannot help support the grid during longer periods of low renewable energy generation. 

Technologies such as vanadium-redox flow batteries, compressed-air energy storage or hydrogen salt-cavern storage could potentially help manage supply and demand over days, weeks or even years. 

A range of LDES technology options are shown in the table below.

Each option has specific advantages and disadvantages; for example, while pumped hydro storage has a high upfront cost, it has a long lifespan of over 50 years. As such, its capital cost per kilowatt hour (kWh) is lower than many other storage options over time. 

(Pumped hydro is the most established LDES technology in the world, but no new projects have been built in the UK since the 1980s.)

While it has historically been a short-duration form of storage, some lithium-ion batteries can now store power for much longer chunks of time. 

Lithium-based grid batteries now often offer 8-12 hours of storage and – as shown in the table above – even longer durations are possible

As Ed Porter, director for Europe at data company Modo Energy, quipped on LinkedIn following the cap-and-floor scheme results: 

“Lithium [is] going far beyond 8 hours; that debate must surely be dead now.”

While even 12 hours is of limited use for gaps in generation of days, weeks or seasons, there are numerous benefits to lithium-ion batteries in comparison to other LDES technologies. For example, the cost of these batteries has fallen by an average of 20% per year over the last decade. 

Given the variation in technologies – including scale, lifespan, commercial readiness and aspects such as necessary geography – comparing the costs of each technology is challenging. 

However, utilising a diverse set of storage technologies is expected to be particularly beneficial for electricity systems, according to experts. 

Julia Souder, CEO of industry group the LDES Council, tells Carbon Brief: 

“The UK is leading the charge on technology diversity. We’re witnessing matching different LDES solutions to the real differences in market structure and country needs.

“But make no mistake: a handful of LDES technologies will do the heavy lifting over the next decade. We’re seeing that play out in which technologies are winning through the UK government’s new cap-and-floor mechanism for long duration storage.”

§ How much LDES will the UK need?

LDES is expected to be a key component of the UK’s electricity system in the future, particularly as it moves away from easily stored and dispatched fossil fuels such as gas. 

The government has set a target of “clean power by 2030”, in the lead-up to the wider net-zero by 2050 goal. 

In 2024, the Labour administration set out an “action plan” for reaching the 2030 target, which included substantial increases to electricity generation technologies. 

This included setting widely discussed targets to double offshore wind, triple onshore wind and quadruple solar capacity by 2030, alongside rebuilding the UK’s nuclear fleet. 

But the action plan also set a less well-known target for 4-6GW of LDES, to help balance this renewables-dominated electricity mix. This is in addition to 23-27GW of short-duration battery energy storage, new interconnectors and a big push to develop consumer-led flexibility. 

There is also a major expansion of LDES to 3.8-5.3GW by 2030 in the most recent “future energy scenarios” report from the National Electricity System Operator (Neso), as shown in the chart below. 

Neso’s pathways show LDES rising to between 16.6GW and 13.2GW by 2050, mainly dependent on how hydrogen is used in the electricity system. 

The Neso report notes that few LDES schemes are likely to come online before 2030, due to the long project development and planning times, as well as high capital expenditures.

§ Which types of LDES is the UK planning to use?

While the UK is pursuing a diverse range of LDES, certain technologies are likely to make up the bulk of LDES in the next decade or so. 

This is evident in the technologies that have bid successfully into the UK government’s new “cap-and-floor” mechanism for LDES. 

The scheme was first announced in 2024 and is designed to guarantee a minimum level of revenue for energy storage operators – the “floor” – as well as to put a limit on profits via the “cap”. 

(The mechanism will be funded through electricity bills. However, Ofgem expects it to be broadly cost-neutral over time.)

Similar mechanisms have been used to support the development of other technologies in the UK, in particular those with high upfront costs, such as interconnectors. Ultimately, it minimises the risk for developers by guaranteeing a certain level of future revenue. 

In 2025, 171 LDES projects with a total capacity of 52.6GW applied to enter the cap and floor scheme, which is administered by Ofgem. Of these, 77 projects (28.7GW) were deemed eligible to enter a second “assessment” phase. 

These were made up of nine different technologies, as shown in the figure below. However, lithium-ion batteries dominated the process, making up more than 20GW of the 29GW total. 

No pure vanadium-flow batteries, liquid-air energy storage, iron-air batteries, sodium-sulphur batteries or hydrogen batteries were deemed eligible for the second phase. 

(Conventional hydrogen storage was not eligible to bid into the process either, but could be supported through other means. The government is expected to release an updated hydrogen strategy later in 2026.)

Ultimately, Ofgem announced in June 2026 that it was “minded to” support 7.6GW of LDES capacity, spread across 16 projects. Of this total, 4GW is expected to be online by the end of the decade, at the bottom end of the range said to be required for the clean power 2030 target. 

The 16 projects are listed in the table below. They comprise four technologies: pumped storage hydro (3.9GW); lithium batteries (3.6GW); one vanadium-zinc flow battery (65MW); and one compressed- air energy storage site (50MW).

Welcoming Ofgem’s initial decision on the cap-and-floor mechanism, energy minister Michael Shanks said in a statement: 

“Forty years after the country’s last pumped storage facility, this government is getting Britain building again… 

“We are [going] further and faster in delivering the clean-power mission by rolling out a new generation of pumped-hydro storage and state-of-the-art batteries – making more of the clean, homegrown power we already produce, cutting waste, lowering bills and strengthening our energy security.”

Collectively, the provisionally successful projects can provide between eight and 32 hours’ worth of electricity storage. The top ten projects in terms of duration that applied for the mechanism – those with at least 12 hours’ worth of storage – all moved forward.

Following Ofgem’s “minded-to” decision, the regulator launched a consultation that ended on  7 August 2026. It will now make a final decision on the projects that will be supported through the “cap and floor” mechanism. 

Martin tells Carbon Brief that “it’s not over” yet, with Ofgem likely to face scrutiny over the methodology it used to determine these final results. He adds: 

“There’s going to be a lot of activity and a lot of responses to that consultation. I don’t expect the overall amount of capacity that’s been awarded to change, although they could increase it – it could only go up, probably.

“But there might be some change in what projects end up getting approved as a result, or maybe they end up making some changes for the next window [of applications for LDES support].”

Alongside the cap-and-floor process being run by Ofgem, the government introduced legislation via the Planning and Infrastructure Act to support the introduction of the scheme. 

Additionally, in August 2026, Innovate UK – the UK’s national innovation agency – announced new funding for “ultra-long” duration battery energy storage. 

Up to £3m will be invested in demonstration projects as part of the first phase of the funding, with £10m available in the sector to support the development of technologies capable of storing and discharging at least 100 continuous hours of electricity. 

In a statement responding to the new funding, Dr Jamie Speirs of the University of Strathclyde and co-director of the UK Energy Research Centre, said achieving the UK’s low-carbon ambitions will rely on “unlocking” LDES to support a highly renewable system. He added:

“By providing flexibility across hours, days and even seasons, LDES could enable a resilient, low-carbon electricity system – reducing curtailment, strengthening security of supply and ensuring that intermittent renewables can maximise their contribution to the grid in all conditions. 

“Investing in innovation opportunities such as this call to support market deployment of LDES technologies is a key way to support these technologies to market, giving us the best chance to meet our net zero targets.”

Phase one of the funding is open for applications until 30 September, with grants of between £350,000 and £700,000 available for the successful projects.

Seamus Garvey, professor of dynamics at the University of Nottingham, welcomes the new funding. However, he cautions that more needs to be done to ensure the future markets for medium- and long- duration storage are not compromised by early commitments to storage at shorter timescales. He tells Carbon Brief:

“Energy storage will be required over many timescales and as we decarbonise further and further, the requirements for longer durations grow and grow. 

“One key problem in my opinion is that because we are tending to buy into lots of short-duration stores now, we are actually removing pieces of market that could be accessible by longer duration stores and that is making the (already-difficult) problem of financing these stores ever more difficult.” 

§ How could LDES impact energy bills?

The rollout of LDES technologies is widely expected to help reduce energy bills as the UK transitions to a clean-energy system.

There is still a significant amount of uncertainty over the development of LDES, due to the wide range of options, nascent stages of development and lack of market maturity. Nevertheless, most research agrees that it will cut electricity system costs by the middle of the century, relative to a world where LDES is not used. 

For example, adding 20GW of LDES could reduce electricity system costs by £16-51bn between 2030 and 2050, compared with a scenario that has limited flexible capacity, according to analysis for the Department for Energy Security and Net Zero (DESNZ), by thinktank Regen and LCP Delta. The analysis, published in 2023, found that 20GW of LDES could reduce costs by around £26bn. 

Analysis by LCP Delta in 2025 found that building 20GW of established medium-sized LDES technologies – pumped hydro with a capacity of 8-12 hours – by 2050 would have a system benefit of more than £10bn. 

LDES could reduce total UK electricity system costs by £7-13bn annually by 2040-2050, according to a report from the Transition Finance Council – a public-private body launched by the City of London Corporation and the UK government – citing a range of other studies. 

The council says this would predominantly be by avoiding “curtailment”, where some generators are paid to switch off because the electricity grid cannot accommodate their output. It says that LDES would defer the need for additional grid investment and would reduce balancing costs, including curtailment.

(In the financial year 2024-25, balancing costs reached £2.7bn, adding around £40 to the average household electricity bill. Some £1.9bn of this – £28 per household – related to constraints, where wind is “curtailed” and gas plants are switched on elsewhere.)

Curtailment is a particular issue in Scotland, where much of the UK’s wind capacity sits behind congested sections of the national electricity network. Porter notes on LinkedIn that this helps explain why 79% of the LDES projects by storage capacity are located in northern Scotland. 

Martin says LDES will allow the UK to “use our renewable fleet more efficiently”. He adds: 

“[LDES] is able to increase renewable energy and then decrease gas generation during high-demand periods, and that brings all sorts of benefits to the system.

“It reduces emissions, it reduces the overall cost of the system, it can help reduce bills for consumers. So those are the types of benefits that we’ll see as a result of [more] LDES being [on the system].”

The Transition Finance Council report adds that despite the upfront cost, LDES quickly pays for itself. It estimates that each gigawatt of long-duration flexibility on the system requires around £2-2.5bn in investment, but yields annual system savings of £0.5-1bn once operational. 

As such, even accounting for the upfront cost of developing LDES, the technologies would provide £30-60bn of electricity system savings over 25 years, the council says. It adds that this means LDES “will repay itself several times over”. 

§ Record lows

At full capacity, Lakes Mead and Powell can hold a combined 68 cubic kilometres of water – enough to supply all household consumption in the contiguous US for nearly 1.5 years. However, the water level in both reservoirs has been declining for decades.

The chart below shows the water level of Lake Mead, in metres above mean sea level. The reservoir, which began to fill in 1935 following the construction of the Hoover Dam, has a “full pool” maximum capacity of 347.60 metres. The water level in Lake Mead reached a record low of 317.11 metres on 7 August. 

The following chart shows the water level of Lake Powell, in metres above mean sea level. Lake Powell’s full-pool level is 1,127.76 metres. 

While the reservoir reached its maximum capacity several times in the 1980s, it has not done so since. On 15 August, the water level in Lake Powell was recorded at a new record-low of 1,072.87 metres.

Both reservoirs have continued to decline in the days since breaking their respective records. The downward trend will largely continue in both lakes until next spring, when the snowpack in the mountains of the Upper Colorado River Basin begins to melt, says Dr Jack Schmidt, a senior research scientist at Utah State University’s Center for Colorado River Studies. He tells Carbon Brief:

“The big dilemma of the moment is that we’re only in the middle of August, and we have no assurance of what the coming winter will be. The only thing we can be sure of is that we will be depleting overall total basin reservoir storage from now until, roughly, early April.” 

§ Compounding factors

The record lows across the two reservoirs are the result of several compounding factors, experts tell Carbon Brief. 

Since the turn of the 20th century, the amount of water flowing along the Upper Colorado River has declined by about 20%. Research suggests that half of this decline can be attributed to human-induced climate change. 

Most of the river’s streamflow comes from the snowpack of the Upper Colorado River Basin, which stretches across five western US states but is primarily located in Colorado and Utah. 

This region has been gripped by a historic “megadrought” for more than a quarter of a century. Nearly half of the megadrought’s intensity over 2000-18 is attributable to climate change, according to a 2020 study.

At the same time, the increasing population in the US south-west has put added pressure on the Colorado River’s water supply. The number of people obtaining some or all of their water from the Colorado system has grown by 15 million (around 60%) since 1992. 

Schmidt tells Carbon Brief:

“There’s an ultimate cause of the present water crisis, and there’s a proximate cause. The ultimate cause is a warming climate, a warming planet and a pretty clear correlation between warming conditions and decreased runoff in the Colorado River Basin.

“The proximate cause is that in this messy democratic republic of ours, big policy decisions that match the variability of the climate occur painfully slowly – with intense political negotiations – and only incrementally.”

On 31 July, the US Bureau of Reclamation, which manages water resources in the western US, released an environmental impact statement on its proposed post-2026 strategy for managing Lakes Powell and Mead. The strategy itself has not been released yet.

Schmidt notes that the statement does appear to give the Bureau flexibility to “respond to crisis” by reducing the delivery of water to several states. However, he adds:

“They acknowledge it won’t work if we just stay critically dry, and of course every climate model for the 21st century, especially with a continually warming planet, says that that’s exactly what’s going to happen.”

§ How have land-use emissions changed?

Deforestation, forest degradation, loss of peatlands and harvesting trees for wood all release carbon into the atmosphere.

Collectively, these emissions are known as land-use, land-use change and forestry (LULUCF) emissions, referred to here as land-use emissions. 

Each year, global land-use emission trends are analysed in the Global Carbon Budget report. The report, produced by dozens of scientists, documents how human-caused greenhouse gas emissions are changing over time.

Key findings from the annual report are released each year in the autumn, before being published formally in an academic journal the following year following a peer-review process.

(For more on the findings of the 2025 report, read Carbon Brief’s summary.)

The latest edition of the Global Carbon Budget report notes that, in the four decades to 1999, net CO2 emissions from land-use change remained “relatively constant”, sitting at around 6.6bn tonnes of carbon dioxide (GtCO2) per year.

However, since the late 1990s, global land-use emissions have been falling. 

The 2025 report estimates that land-use emissions over 2015-24 averaged at 5GtCO2 a year. This is around 23% lower than the average over 1995-2004 and 19% lower than 2005-14, it says.

In contrast, global emissions from fossil fuels and cement have increased every decade since 1959, rising from an average of 11GtCO2 in the 1960s to 35.9GtCO2 over 2015-24, it says. 

“Preliminary data” included in the report suggests that land-use emissions in 2025 clocked in lower than their 2014-25 average, at 4.1GtCO2, as fossil-fuel and cement emissions reached a new high of 38.1GtCO2. 

(For more on how land-use emissions are calculated, see: Why are estimates of land-use emissions uncertain?)

The chart below shows how land-use emissions have been falling in the 21st century and have helped to temper the overall rise of human-caused emissions.

§ Why have land-use emissions fallen?

The Global Carbon Budget attributes falling land-use emissions since the late 1990s to decreasing emissions from deforestation, in particular “permanent deforestation”. 

Permanent deforestation refers to the complete removal of trees for the conversion of forest to another land use, such as agriculture, mining or the construction of towns and cities. This sets it apart from other forms of deforestation, such as logging and rotational farming, where the canopy is removed on a more temporary basis.

The Global Carbon Budget also points to “increasing [CO2] removals” from forest regrowth as a reason for falling land-use emissions since the turn of the century.

(For more on the countries and policies that have driven these changes, see: Which countries are behind falling land-use emissions? and: Which countries are leading on forest regrowth?) 

Looking at more recent trends, the report attributes a “steep drop” in land-use emissions in the decade since 2015 to the “combined effect” of a “peak” in peat fire emissions in 2015, as well as a “long-term decline” in deforestation emissions in many countries over 2010-20.

The chart below shows how deforestation and forest growth have been responsible for the bulk of change to land-use emissions over the 21st century.

Over 2015-24, the sequestration of CO2 through reforestation and afforestation efforts offset two-thirds of deforestation emissions, according to the Global Carbon Budget report.

Specifically, it notes that deforestation was responsible for an average of 6.96GtCO2 of emissions each year over 2015-24. Forest growth, on the other hand, removed 4.76GtCO2 a year. 

Just under half – 2.2GtCO2 – of carbon removals over 2015-24 was from afforestation and reforestation efforts and the remaining 2.56GtCO2 were driven by forest regrowth from shifting cultivation cycles, it says. 

Forest regrowth from shifting cultivation refers to the recovery of a forest after a plot has been farmed for a short period and then abandoned.

This is shown in the chart below below, which shows how carbon removals from forest regrowth have offset emissions from deforestation. 

In the near-term, the Global Carbon Budget attributes its projection of a drop in land-use emissions between 2024 and 2025 to the “end of El Niño conditions”. 

(The naturally occurring weather phenomenon typically leads to the drying out of peatlands in the tropics and causes more planned deforestation fires to burn out of control.)

Prof Pierre Friedlingstein, director of the Global Carbon Budget office and a professor at the University of Exeter, tells Carbon Brief there is “no indication” of what might happen in the future, but adds that land-use emissions trends over the 21st century are “going in the right direction”. He says:

“If you are optimistic, you hope the trend will not reverse and start increasing again. But we don’t know for sure. The assumption, given current land policies across the world, is that deforestation should continue to decline.”

§ Which countries are behind falling land-use emissions?

The countries that contributed the most to land-use emissions over 2015-24 were Brazil, the Democratic Republic of the Congo (DRC) and Indonesia, according to the Global Carbon Budget.

It notes that these three countries together contributed more than half – 57% – of global land-use emissions. 

Over the first quarter of the 21st century, falling land-use emissions in Brazil and Indonesia have combined with increased afforestation and reforestation in China to drive down overall land-use emissions, according to the Global Carbon Budget.

This is illustrated in the chart below, which shows how China’s land-use emissions have dropped below zero, as Brazil and Indonesia’s emissions have declined.

Friedlingstein says that the decline in land-use emissions since the 2000s has been “primarily driven by a decline in deforestation in Brazil”.

He tells Carbon Brief that tree clearance in the South American country rose in the 1990s then started to fall after a peak in the 2000s:

“There was a bit of up and down – mainly due to politics and who was in charge in Brazil – [whether the president] was [Luiz Inácio] Lula [da Silva] or [Jair] Bolsonaro. But the long-term trend in Brazil is a decline in deforestation due to forest protection policies.”

These policies included a 2004 “action plan” for the prevention and control of deforestation in the Amazon, a 2006 soy moratorium, which banned the purchasing and financing of soya produced in deforested areas of the Amazon, as well as the expansion of protected areas across Brazil during the second half of the 2000s.

Prof Julia Pongratz, a professor of physical geography and land-use systems at the University of Munich and contributor to the Global Carbon Budget, says Brazil is the “single most important contributor to the early-2000s global land-use change emissions peak and subsequent decline”.

She says that the largest contributor to an “acceleration” in the decline of global land-use emissions in the past decade has been Indonesia, which she notes has “rewetted more peatland area since 2017 alone than Europe in its entire history”. 

Around the world, peatlands are exploited and damaged by humans for a range of purposes, including converting the land for agriculture and peat extraction for horticulture and fuel. Peatland wetting refers to the process of restoring water levels in drained peatlands in order to return them to their natural, waterlogged conditions, which allows for peat formation and carbon storage. 

Another reason for Indonesia’s downward trend in land-use emissions is that there have been fewer spikes in emissions caused by fires related to human land-use activities over the last decade, says Pongratz.

Emissions from ecosystem fires are not always counted towards national and regional land-use emissions budgets, which estimate the sum of human-caused emissions. Deforestation fires and those related to peatland drainage are included, whereas fires caused by droughts and heatwaves are not. 

Pongratz says it is “hard to separate natural and land-use drivers completely”, given that deforestation and peatland fires often “get out of control and cause spikes in emissions” during dry El Niño conditions. 

(For more on uncertainties in land-use emissions data, see: Why are estimates of land-use emissions uncertain?)

Pongratz notes that international trade regulations that have helped to drive down land-use emissions in Brazil and Indonesia have had a lesser effect in the DRC, where the root drivers of deforestation are different:

“Emissions in the DRC have increased, then stayed high in the last two decades. This is partly related to population growth and expanding smallholder and subsistence farming. 

“The picture is different in Brazil and Indonesia, which are much more driven by export; international regulations aiming at curbing deforestation thus have larger effects in these countries.”

§ Which countries are leading on forest regrowth?

Reforestation and afforestation schemes that draw down carbon from the atmosphere have helped to reduce the overall emissions from land-use change over the course of the 21st century.

As noted above, the 2025 Global Carbon Budget report highlights how the removal of carbon from forests offset two-thirds of deforestation emissions over 2015-24. 

The report says that China, the EU and US account for the highest levels of carbon sequestration from reforestation and afforestation, collectively drawing 1.1GtCO2 per year over the 2015-24 period. 

This, it says, is “partly related to expanding forest area as a consequence of the forest transition in the 19th and 20th centuries and subsequent regrowth of forest”.

The chart below, which draws from the latest edition of the “state of carbon dioxide removal” report, shows how carbon uptake by forests has increased over the last 20 years in a number of countries, most notably in China.

In China, a raft of reforestation and improved land management policies were introduced in the 1990s which have led to the rehabilitation of tens of millions of hectares of forests. Research has shown the schemes have significantly increased the country’s uptake of carbon and switched its land from a carbon source to a carbon sink. 

The Global Carbon Budget highlights that substantial carbon removal from reforestation and afforestation occurred in other regions, such as Brazil, Russia and Indonesia. However, in these regions, emissions from deforestation and other land-use changes “dominate”, it says.

§ Why are estimates of land-use emissions uncertain?

Tallying the world’s emissions from land-use change is complex.

The Global Carbon Budget estimates an uncertainty range of 2.6GtCO2 per year for its average annual global land-use emissions figure for 2015-24 – more than half the overall figure of 5GtCO2.

To calculate overall land-use emissions for the annual Global Carbon Budget report, researchers create an average from three land-use models: BLUE, OSCAR and LUCE.

These models combine satellite and statistical information on land cover and land-use changes from global and regional datasets.

Pongratz, who is involved in the LUCE model, explains that scientists can measure the exchange of CO2 between land and atmosphere, but are not able to determine whether CO2 is being released or sequestered from a managed area as a result of human activities or other climate or environmental factors. She continues: 

“For this, you need to turn to modelling, where you can isolate drivers – and, again, models are uncertain and the land-use input imperfect. This is why we use all available model estimates – three at the moment.”

The Global Carbon Budget highlights that its three different models treat different components of the land-use emissions “budget” differently. 

While models agree “relatively well” about emissions from permanent deforestation, they take different approaches in their approach to shifting cultivation patterns, which increases both emissions and removals, as well as wood harvesting, it says.

Moreover, it notes that land-use emissions and removals occur on different timelines. While carbon removals generated by forest growth and soil recovery are “slow”, there is an “instantaneous component” to emissions from deforestation, it says.

(For more on the challenges in analysing changes to the global carbon cycle, see Carbon Brief’s recent in-depth interview with Prof Philippe Ciais, one of the world’s leading experts on land-use emissions.)

The Global Carbon Budget notes that its confidence in its 2025 projection for overall land-use emissions remains “low” given that the figure is based on deforestation, degradation and peat fire emissions, which are “only a proxy” for land-use change.

The report notes that 2023 is the final year in which it calculates land-use emissions directly from land-use statistics across all three bookkeeping models. For more recent years, full statistics are not yet available across the models and scientists instead turn to short-term proxies.

§ What does the plan say about peaking coal?

Five-year plans are a key tool in Chinese governance, used to guide economic and social development across the economy.

The plan for coal is the latest topic-specific document to address climate and energy matters within the 15th five-year plan period of 2026-30. It is subordinate to the overarching 15th five-year plan, which covers China’s broad socio-economic strategy. 

Other topic-specific plans for the period cover climate change, developing a “new-type energy system” and renewable energy, among other topics.

The coal plan opens by stating that coal is a “foundational [source of] energy” for China:

“[Coal is] vital to the national economy, people’s livelihoods and national energy security, and plays a crucial role in providing foundational support and systemic regulation within the energy supply system.”

However, the plan also covers the 15th five-year plan period (2026-2030), the final five-year period before China is expected to have peaked its carbon emissions.

The 15th five-year plan period marks a time of “significant transformation” for the coal industry, the plan says.

Policy documents issued in April 2026 called for the “strict control” of fossil fuels and created a framework for local governments to be graded on coal use in their region. 

Coal has traditionally been the largest source of energy in China and is responsible for around 80% of its emissions. 

But its role is gradually being superseded by non-fossil energy, which accounted for more than half of the country’s power mix in 2025. In the first half of 2026, coal supplied less than 50% of power generation, while its share of total energy consumption fell to 51.4%, as shown below.

The five-year plan for coal signals “continuity” of China’s aim of “safeguarding energy security while advancing the low-carbon transition”, says Kevin Tu, non-resident fellow at Columbia University’s Center on Global Energy Policy. 

Another key factor behind the plan is concerns from policymakers around energy security, exacerbated by the conflict in the Middle East.

In an article published in early August, the Communist party-affiliated People’s Daily noted the “severe volatility” the war has created in energy markets, adding that “China’s energy system has withstood these shocks”.

It quoted NEA head Wang Hongzhi stating in a press conference that “coal is [China’s] greatest source of confidence in ensuring a stable energy supply”. 

The conflict will “reinforce coal’s role in China’s energy system”, both as a source of energy and as a feedstock for commodities, Li Shuo, China climate hub director at the Asia Society Policy Institute, tells Carbon Brief.

The plan outlines a number of aims to be achieved by 2030, starting with a goal to “further strengthen” the coal industry’s “ability to be a ‘bottom-line guarantee’”.

The other targets in the plan, to be achieved by 2030, include:

• Peaking coal consumption;
• “Basically establishing” a modern coal-industrial system;
• Optimising the “layout” of coal production and development;
• Increasing the proportion of “high-quality, advanced” coal-production capacity;
• “Clearly improving” levels of “safe, green development” and “clean, efficient use” of coal;
• Increasing the share of coal produced by “large-scale, modernised coal mines” to 87%;
• Developing a diversified coal-based industrial structure;
• Improving mechanisms to ensure a “dynamic balance” between supply and demand.

The large share of China’s CO2 emissions that come from coal and China’s carbon-peaking and neutrality targets are not the main focus of the five-year plan.

“This is clearly neither a coal phase-out nor phase-down plan,” Tu tells Carbon Brief. He adds that it grants China “considerable flexibility…over the pace of the transition”. 

A pledge to peak coal consumption during the five-year plan period is reiterated several times in the document. Notably, the plan says that China will “promote coal consumption successfully reaching a peak”. 

This, it says, is “guided” by China’s “dual-carbon” goals for peaking and neutrality, but is also based on the premise of “guaranteeing the secure supply of energy” 

However, the plan does not provide a government-endorsed target year for peaking consumption. 

State-affiliated organisations, such as Xinhua, have suggested that coal consumption is “expected to peak around 2027”. Independent analysis has stated that emissions from coal consumption may have already peaked.

“The absence of a 2027 deadline is significant, but I would be careful not to over-interpret it,” Tu tells Carbon Brief.

While a 2027 peak for coal remains possible, in his view, it is dependent on factors such as “electricity-demand growth, renewable generation, industrial activity, weather conditions and coal demand from the chemical sector”.

Similarly, Li believes that it will be “market and technological progress”, rather than state directives, that determine exactly when coal consumption and emissions will peak.

“Beijing’s regulatory interventions, if any, will be limited to making sure the peaking timelines do not blow past 2030,” he says.

§ What does the plan say about China’s coal production?

The plan does not set a concrete target for coal production during the five-year plan period. In contrast, total coal production targets for 2015 and 2020 had been set in the 12th and 13th five-year plans.

The plan also reduces a target for “reserve production” capacity, which was first announced in 2024.

The plan reiterates that, by 2030, China should “establish a coal reserve-production capacity of 100m metric tonnes or more per year”. This was first mentioned in the 15th five-year plan for building a “new-type energy system”, published in June.

Despite China’s rapid buildout of renewable energy, reserve coal capacity is necessary, argues state news agency Xinhua. It says that, to balance the variability of renewable energy, coal will shift to “playing a supporting and regulating role to safeguard energy supply”. 

Nevertheless, the new reserve goal is lower than the target of 300m tonnes of coal set when China first announced the establishment of the system in 2024. 

“Overall, this five-year plan is targeted at the coal industry, not the energy transition”, says Yang Biqing, energy analyst at Ember, although the energy transition and the peaking of coal consumption form the overarching context for the plan.

Provinces in northern China will continue to provide the majority of China’s coal, according to the plan.

It reiterates a pledge from the new-type energy five-year plan that China will continue building “coal-supply security bases” in the provinces of Shanxi, Inner Mongolia, Shaanxi and Xinjiang. It says these bases will supply more than 80% of China’s coal by 2030.

This does not indicate a change in direction, as coal production is already increasingly concentrated in northern China. In 2025, 82% of China’s coal came from these four provinces.

New or expanded coal mines in these provinces – with the exception of southern Xinjiang – must have a minimum annual production capacity of 1.2m tonnes, says the plan.

This is an “important signal”, Tu tells Carbon Brief. He notes that the plans suggest that “China’s coal transition is not simply about reducing the quantity consumed”, but also about creating a “more concentrated, efficient, flexible and resilient” coal system.

The plan also calls for a more centralised approach to managing coal. It states that in 2026-2030, any new production capacity must be “included in the single ledger” – essentially meaning that it must be approved by the central government – before it can be implemented. 

Yang tells Carbon Brief that this could indicate that the government is trying to prevent a potential “rush” to get new capacity approved as coal consumption starts to plateau and fall.

§ What does the plan say about coal’s greenhouse gas emissions?

The plan includes sections on the need to “accelerate” the low-carbon transition of the industry, as well as the “clean and efficient use” of coal.

The former section largely focuses on the production and processing of coal, while the latter addresses emissions associated with its consumption. 

Suggested policies include promoting energy efficiency, water conservancy and electrification, coupled with greater use of renewable-energy sources at coal mines.

In addition to promoting a successful peaking of coal consumption, the plan also re-affirms existing policies around promoting energy efficiency and carbon-emission reduction.

It calls for “accelerate energy conservation and consumption reduction in key coal-consuming industries”, largely through methods already established by existing policies.

This includes phasing out inefficient coal-fired equipment, replacing coal-fired equipment with “clean energy” alternatives, reducing use of “dispersed coal” and promoting clean heating sources such as distributed solar heating and waste heat utilisation.

Tom Wang, executive director of People of Asia for Climate Solutions, describes the plan as “more of a coal exploration plan, rather than a coal transition plan”. He tells Carbon Brief that while several policies call for “green” or “smart” development, the plan does not address the greenhouse gas emissions underpinning each step of coal extraction, processing and combustion.

Another major focus is on utilisation of coalbed methane, a significant source of China’s methane emissions.

China will “implement work plans to increase coalbed-methane reserves and production”, the plan says, including a “rapid ramp-up” of production in deep coalbed-methane sites.

Affixed to the main five-year plan is an appendix further detailing plans for coalbed methane.

It notes that utilising coalbed methane has “multiple benefits”, such as improving safety, “increasing the supply of clean energy” and reducing emissions. [Methane is a fossil fuel.]

The government is targeting 26bn cubic metres of coalbed-methane production and 6.5bn cubic metres of mine-gas utilisation by 2030, it says.

At least 18bn cubic metres will be sourced from the Ordos Basin, a region spanning several northern provinces, according to an action plan published by the NEA.

In its coverage of the Ordos action plan, the state-run newspaper China Daily said that developing coalbed methane is a “vital strategic move to optimise [China’s] energy mix and ensure domestic gas supply”.

Reporting by Xinhua and economic news outlet Jiemian said that coalbed methane could help China become an “energy powerhouse” and “secure [its] energy self-sufficiency”, respectively. 

In addition, the coal industry will “steadily advance methane-emission control” and “actively participate in the reduction of non-carbon dioxide greenhouse gas emissions”, according to the appendix.

However, Sun Xiaopu, senior China counsel at the thinktank Institute For Governance and Sustainable Development, tells Carbon Brief, the plan “does not establish an absolute methane-emissions reduction target”. 

She notes that the implications for emissions may only become clear as implementation frameworks for meeting the utilisation targets are released.

§ How does the plan tell coal companies to evolve?

Despite reaffirming the importance of coal, the plan emphasises that the overall role of the fuel in China will change. It adds that the coal industry must adapt to this changing reality.

As the coal industry “modernises”, coal companies must “strengthen management” of mine closures and exit plans. They must also plan for a “smooth transition” and “prudently handle” workforce relocation, debt resolution and ecological restoration, it says.

Companies should also be supported in expanding into industries such as “power, new energy and chemicals”, according to the plan.

A number of major coal producers, as well as at least one oil giant, have already established wings focused on “new energy”.

But the focus on the use of coal to make chemicals is one of the “most consequential parts of the plan”, says Tu.

China must promote the shift to coal being used “equally” as a fuel and a feedstock, the plan says.

The plan urges policymakers to push through “construction of strategic coal-to-oil and gas bases”

The chemicals sector is China’s fastest source of emissions growth, although it remains well behind power and other industries in terms of total emissions. 

Tu notes that the plan calls on the coal-chemicals industry to decarbonise production, such as through low-carbon power, green hydrogen and carbon capture, utilisation and storage.

As such, he says, the policy signal is “not to exit coal chemicals, but to make them more efficient, higher-value and potentially less carbon-intensive”.  

Li echoes this, telling Carbon Brief that the sector is “likely to receive a major boost from the conflict in Iran”. He adds: 

“We will probably see further capacity expansion in the sector and I doubt environmental arguments will convince Chinese authorities to take a different approach.”