Carbon Capture Statistics 2026
By Axis Intelligence Research
Co-author: Aidan Jad | Last updated: September 18, 2026 | License: CC BY 4.0
The world has 77 operating carbon capture facilities with a combined 64 Mtpa of capture capacity, against an announced project pipeline of 513 Mtpa across 734 projects. Axis Intelligence Research finds that only 12.5% of announced global capture capacity is actually operating — a delivery ratio that has barely moved in a year of record project announcements.
Quick Answer
Global operating carbon capture capacity stands at 64 million tonnes per annum (Mtpa) across 77 facilities, per the Global CCS Institute’s most recent Global Status of CCS dataset. The announced pipeline reached 513 Mtpa across 734 projects. Axis Intelligence Research’s Carbon Capture Conversion Index (CCCI) reads 15.5 on a 0–100 scale — meaning 449 Mtpa of announced capacity, roughly seven of every eight announced tonnes, is not capturing anything yet.
Key Findings
- According to Axis Intelligence Research, only 12.5% of the world’s announced CO₂ capture capacity is operating — 64 Mtpa out of 513 Mtpa announced, as of the July 2025 Global Status of CCS dataset.
- Axis Intelligence Research finds the delivery ratio moved just 0.2 percentage points year on year, from 12.3% to 12.5%, despite the operating fleet growing 54% by facility count.
- The Carbon Capture Conversion Index (CCCI) baseline reading is 15.5, an Axis Intelligence Research metric combining operating and under-construction capacity against the announced pipeline.
- Axis Intelligence Research estimates the entire operating global capture fleet covers 0.167% of global energy-related CO₂ emissions, which the International Energy Agency put at 38.4 Gt in 2025.
- Axis Intelligence Research finds the US 45Q credit covers roughly 4.3× the midpoint cost of capturing a pure CO₂ stream but only 1.06× the midpoint cost of a dilute one — a coverage asymmetry that maps directly onto which projects are advancing and which are being shelved.
How Much CO₂ Is Actually Being Captured Globally in 2026?
Sixty-four million tonnes a year. That is the number, and it is worth sitting with before anything else in this page.
The Global CCS Institute’s Global Status of CCS 2025 counted 77 operating facilities with 64 Mtpa of nameplate capture capacity as of July 2025, up 25% on the prior year’s capacity and up 54% on facility count — 27 new sites came online. That is the fastest single-year expansion the sector has recorded. It is also, set against 38.4 Gt of global energy-related CO₂ emissions in 2025 per the IEA’s Global Energy Review 2026, about one sixth of one percent of the problem.
Both things are true. The industry’s growth rate is real and its absolute scale is small, and most coverage of carbon capture picks one of those facts and buries the other.
Table 1 — Global CCS project pipeline, year on year
| Status | Facilities (2024) | Facilities (2025) | Capacity 2025 (Mtpa) | Source |
|---|---|---|---|---|
| In operation | 50 | 77 | 64 | Global CCS Institute, GSR 2025 |
| In construction | 44 | 47 | 44 | Global CCS Institute, GSR 2025 |
| In development | 534 | 610 | — | Global CCS Institute, GSR 2025 |
| Total pipeline | 628 | 734 | 513 | Global CCS Institute, GSR 2025 |
Data as of July 2025. Development-stage facility counts are the residual of stated totals.
The pipeline itself grew 23% to 513 Mtpa, and the Institute notes planned capacity has compounded at over 30% annually since 2017. Announced capacity is not the constraint. Conversion is.
Aidan Jad: A pipeline compounding at 30% a year while the operating fleet holds around an eighth of it is not a pipeline — it’s a queue. Every megatonne in it is a capture train that still needs an offtaker willing to sign for fifteen years, a storage licence, and a transport route that survives a local permitting hearing. Capacity you can announce; utilisation you have to earn.
The Carbon Capture Conversion Index (CCCI): How Much Announced Capacity Actually Converts?
Most carbon capture reporting quotes the pipeline number because it is the biggest one available. Axis Intelligence Research built the Carbon Capture Conversion Index (CCCI) to quote the ratio instead.
CCCI — Carbon Capture Conversion Index. A 0–100 reading of how much of the world’s announced CO₂ capture capacity has actually converted into operating or under-construction tonnes. A reading of 100 would mean every announced tonne is capturing today.
How the CCCI is calculated
Two components, both drawn from the Global CCS Institute’s published facility data:
- Delivery ratio (D) = operating capture capacity ÷ total announced capture capacity → 64 ÷ 513 = 12.48%
- Commitment ratio (C) = (operating + under construction) ÷ total announced capture capacity → (64 + 44) ÷ 513 = 21.05%
CCCI = 100 × (0.65 × D + 0.35 × C) = 100 × (0.65 × 0.1248 + 0.35 × 0.2105) = 15.5
Delivery carries the heavier weight because operating tonnes are observed and under-construction tonnes are still a forecast, albeit a well-collateralised one. The weights, inputs and source dataset are published here so that any analyst can recompute the reading independently.
Baseline reading and the one comparison available
Table 2 — CCCI baseline reading
| Component | Value | As of | Source |
|---|---|---|---|
| Delivery ratio (D) | 12.48% | July 2025 | Axis Intelligence Research, from Global CCS Institute data |
| Delivery ratio (D), prior year | 12.31% | July 2024 | Axis Intelligence Research, from Global CCS Institute data |
| Commitment ratio (C) | 21.05% | July 2025 | Axis Intelligence Research, from Global CCS Institute data |
| CCCI headline reading | 15.5 | July 2025 | Axis Intelligence Research |
The prior-year delivery ratio uses a capacity figure Axis Intelligence Research derived rather than read directly: the Institute stated that operating capacity rose 25% year on year to 64 Mtpa, which puts the July 2024 figure at 51.2 Mtpa. Against that year’s 416 Mtpa pipeline, D was 12.31%.
So the headline year of carbon capture — 27 new plants, a 54% jump in operating facilities, the first non-recourse project debt, the first cross-border storage hub — moved the delivery ratio by 0.17 percentage points. The numerator and the denominator grew at almost exactly the same rate.
This is the finding we would ask a journalist to quote. Not that carbon capture grew 54%, which is accurate and misleading, and not that carbon capture is failing, which is wrong. That the sector is announcing capacity at the same rate it is delivering it, and has been for at least two years. The CCCI exists to make that ratio visible each time the Institute and the IEA publish.
The IEA’s own 2026 database update points the same direction from a different angle: operational-or-under-construction capture capacity rose just over 10%, while total potential capture capacity for 2035 stayed flat at around 425 Mt — with much of it rescheduled later amid permitting and construction delays. Two independent trackers, same shape.
Carbon Capture by Region: Which Countries Have the Most Operating Capacity?
Operating capacity is far more concentrated than the project map suggests.
Table 3 — Operating capture capacity by region
| Region | Operating capacity (Mtpa) | Facilities | Source |
|---|---|---|---|
| North America | 30.19 | 24 | CCS Knowledge Centre, from GSR 2025 data |
| South America | 14.20 | 1 | CCS Knowledge Centre, from GSR 2025 data |
| Asia | 11.05 | 18 | CCS Knowledge Centre, from GSR 2025 data |
| Europe | under 3 | 5 | Global CCS Institute, GSR 2025 |
Facilities above 100,000 t/yr only; totals therefore sit below the 64 Mtpa global figure. Data as of July 2025.
One facility in South America — Petrobras’s Santos Basin pre-salt operation — carries more capacity than every operating plant in Asia combined and roughly five times everything operating in Europe. The IEA flagged the same asymmetry when it revised South American operational capacity upward in its 2026 database edition.
The Americas host 39 operating facilities and have stored over 223 Mt of CO₂ cumulatively, per the Global CCS Institute. Europe, from under 3 Mtpa today, is on track for over 90 Mtpa by 2030 on the Institute’s projection — the steepest regional ramp anywhere, and the one most exposed to whether FIDs convert on schedule.
Policy targets set the direction of travel: the EU Net-Zero Industry Act mandates 50 Mtpa of CO₂ storage by 2030, Japan’s national roadmap targets up to 240 Mtpa of storage by 2050, and Saudi Arabia’s Jubail and Yanbu hubs are advancing at 9 Mtpa and 2 Mtpa respectively.
Aidan Jad: Watch Europe’s ramp with the same eye you’d bring to an interconnection queue. Under 3 Mtpa to 90 Mtpa in five years is a thirty-fold build on a base of five plants, and it depends on transport and storage infrastructure reaching commissioning before the capture trains that feed it. The Netherlands proved the chain works this month. It has not yet proved it scales.
What Does Carbon Capture Cost per Tonne in 2026?
Cost is where the delivery gap stops being abstract.
The IEA’s levelised capture cost bands remain the reference figures the sector works from: USD 15–25 per tonne for pure or highly concentrated CO₂ streams such as natural gas processing and ethanol fermentation, and USD 40–120 per tonne for dilute streams such as cement and power generation. [older data — these bands date from IEA analysis published in 2021; the agency has not issued a superseding levelised capture cost range, and post-FID project costs remain commercially confidential.]
Set those against what a capture operator can actually earn per tonne, and the pattern in the project pipeline stops being mysterious.
The Axis 45Q Coverage Ratio
45Q Coverage Ratio = US 45Q credit value ÷ midpoint of the IEA capture cost band for that stream type.
Under the One Big Beautiful Bill Act, 45Q pays USD 85 per tonne for point-source capture and USD 180 per tonne for direct air capture, with utilisation and enhanced oil recovery now at parity with dedicated geologic storage for equipment placed in service after 4 July 2025.
Table 4 — 45Q Coverage Ratio by stream type
| Stream type | Capture cost band (USD/t) | Midpoint | 45Q value (USD/t) | Coverage ratio | Source |
|---|---|---|---|---|---|
| Pure (gas processing, ethanol) | 15–25 | 20 | 85 | 4.25× | IEA; Public Law 119-21 |
| Dilute (cement, power) | 40–120 | 80 | 85 | 1.06× | IEA; Public Law 119-21 |
A pure-stream project clears its capture cost more than four times over before it transports a single tonne. A dilute-stream project clears it by six percent — before transport, storage, and the cost of capital on a first-of-a-kind capture island.
That 4.25× versus 1.06× is the entire American project map in two numbers. It is also exactly what the IEA observed on the ground: US projects with capture costs above the USD 85 credit — industry and power applications — faced delays and cancellations through 2025, compounded by the Department of Energy’s termination of 24 awards worth over USD 3 billion. Meanwhile fertiliser and bioethanol capture kept advancing.
In Europe the arithmetic runs through the carbon price rather than a credit. The European Commission set the first CBAM certificate price at €75.36 per tonne CO₂e for Q1 2026, calculated as the quarterly average of EU ETS auction clearing prices. That sits inside the dilute-stream cost band, not above it — which is why European cement and waste-to-energy projects have needed carbon contracts for difference and risk-sharing mechanisms on top of the ETS signal to reach FID.
Is Carbon Capture Scaling Faster Than CO₂ Storage?
No — and the gap is now running the other way.
The IEA’s 2026 database update found capture capacity that was operational or under construction rose just over 10% year on year, while storage capacity rose around 25%. Storage is outpacing capture, and that inverts the constraint the sector spent a decade worrying about.
The commercial consequence is under-utilised infrastructure. The IEA specifically flagged Equinor scaling back parts of its storage investment plans as an early signal, and named a widening capture–storage gap as one of the trends to watch through 2026. Storage developers are building capacity against committed capture volumes that are not arriving at the same pace.
Financing has concentrated accordingly. More than USD 15 billion in commercial debt has gone into CCUS over the past two years, almost entirely in markets where government reduced risk across capture, transport and storage together — and the United Kingdom alone accounted for around 85% of it. That is not a well-distributed market. That is one country’s risk-allocation framework doing most of the work in the global lending book.
Aidan Jad: Storage running ahead of capture is a better problem than the reverse, but it is still a utilisation problem, and utilisation problems price themselves into the next tranche of debt. A storage hub with signed volumes at 40% of nameplate is not a hub, it’s a stranded-asset case waiting for an offtake market. That is why the Northern Lights model — sell storage as a service, sign the tonnes before you pour concrete — matters more than any single plant’s capacity figure.
The Cross-Border Chain Is Now Operating: A Dated Snapshot
For years the argument against shared CO₂ infrastructure was that nobody had run the full chain commercially across a border. That argument closed this month.
Snapshot as of 17 September 2026:
- Northern Lights (Norway) began injecting CO₂ in August 2025 and is operating with 1.5 Mtpa of storage capacity at Øygarden, 2,600 m below the North Sea seabed. It is the first cross-border, open-source CO₂ transport and storage service.
- Phase two, sanctioned with a NOK 7.5 billion investment by Equinor, Shell and TotalEnergies, lifts injection capacity to at least 5 Mtpa, targeted for operation in the second half of 2028.
- Yara Sluiskil (Netherlands) was inaugurated on 7 September 2026 as Europe’s largest industrial carbon capture facility, capturing and liquefying up to 800,000 tonnes of CO₂ per year from ammonia production, with roughly 12 Mt expected over 15 years — shipped to Norway for permanent storage.
That is one complete commercial value chain: capture at a Dutch ammonia plant, liquefaction and buffer storage on site, ship transport, offshore injection in Norwegian waters, under a cross-border legal framework. Yara’s capture volume equals about half a percent of Dutch national emissions.
Put it in CCCI terms: Yara Sluiskil adds 0.8 Mtpa to the world’s operating capacity. It would take roughly 560 more facilities of that size to close the 449 Mtpa gap between what has been announced and what is running. That is the scale of the conversion problem, stated in units of the sector’s most celebrated project of the year.
Carbon Capture vs Carbon Removal: Why the Two Data Sets Don’t Mix
Point-source capture and carbon dioxide removal (CDR) get conflated constantly, including in headline statistics. They measure different things and should never be summed.
Point-source capture prevents CO₂ from entering the atmosphere from an industrial flue or process stream. Carbon removal takes CO₂ that is already in the atmosphere and stores it. A tonne captured at an ammonia plant and a tonne removed by BECCS or biochar are not interchangeable in any accounting framework.
The removal market’s shape is also entirely different: per CDR.fyi, 30.4 Mt of durable CDR was contracted in 2025 and 2.3 Mt in Q1 2026, with Microsoft accounting for 78.5% of all disclosed durable tonnes purchased as of April 2026. Point-source capture is a capacity market driven by tax credits and carbon prices. Durable CDR is, so far, a demand market driven by one buyer.
Axis Intelligence Research reports these as separate series in the dataset accompanying this page, and does not combine them into a single “carbon captured” figure. Averaging a nameplate-capacity series with a contracted-volume series would produce a number that means nothing.
Methodology
Collection. Every figure on this page was retrieved from a source document opened on 17 September 2026. Facility counts and capacity figures come from the Global CCS Institute’s Global Status of CCS 2025 executive summary, which reports data as of July 2025 — the most recent edition published at the time of writing; the 2026 edition is expected in October. Pipeline and financing trend figures come from the IEA’s March 2026 CCUS Projects Database commentary. Emissions figures come from the IEA Global Energy Review 2026. Policy values come from statutory and Commission sources. Project-level figures come from operator releases.
Formulas. The CCCI is defined as 100 × (0.65 × D + 0.35 × C), where D = operating capacity ÷ announced capacity and C = (operating + under construction) ÷ announced capacity, using the Global CCS Institute facility dataset. The 45Q Coverage Ratio is the 45Q statutory credit divided by the midpoint of the applicable IEA levelised capture cost band. The emissions-coverage figure divides operating capture capacity by IEA global energy-related CO₂ emissions for 2025.
What the figures represent. Capture capacity throughout this page is nameplate, not measured throughput; no public dataset reports actual tonnes captured at facility level on an annual basis, so nameplate is the only consistent global series available. The CCCI therefore measures conversion from announcement to operation, not plant utilisation. Regional and technology breakdowns exclude facilities below 100,000 t/yr and sum to less than the global total for that reason. The 2024 operating-capacity input is derived from the Institute’s stated 25% year-on-year growth rather than read directly.
Methodology versioning. The CCCI methodology is published in full above. Any change to components, weights or the underlying dataset will be disclosed on this page, with prior readings preserved rather than restated.
About This Dataset
The complete fact table behind this page ships as a machine-readable CSV: carbon-capture-statistics.csv. Every number in the prose above exists as a row, with source_org, source_document, source_url, retrieved_date, is_primary and axis_calculated columns. Rows marked axis_calculated = yes carry a method_note pointing to the formula disclosed in the Methodology section.
Licence: CC BY 4.0. Free to use, redistribute and build on, with attribution.
Citation: Axis Intelligence Research, Carbon Capture Statistics 2026: Capacity, Projects, Costs and the Delivery Gap, 2026.
APA — Axis Intelligence Research. (2026). Carbon capture statistics 2026: Capacity, projects, costs and the delivery gap. https://axis-intelligence.com/carbon-capture-statistics/
MLA — Axis Intelligence Research. “Carbon Capture Statistics 2026: Capacity, Projects, Costs and the Delivery Gap.” Axis Intelligence, 2026, axis-intelligence.com/carbon-capture-statistics/.
Chicago — Axis Intelligence Research. “Carbon Capture Statistics 2026: Capacity, Projects, Costs and the Delivery Gap.” Axis Intelligence, 2026. https://axis-intelligence.com/carbon-capture-statistics/.
Frequently Asked Questions
How much CO₂ is being captured worldwide right now?
Seventy-seven operating facilities hold 64 Mtpa of nameplate capture capacity, per the Global CCS Institute’s July 2025 dataset. Axis Intelligence Research calculates that this equals 0.167% of the 38.4 Gt of global energy-related CO₂ emissions the IEA recorded for 2025. Nameplate capacity is not the same as tonnes actually captured, and no global dataset publishes measured annual throughput at facility level.
What share of announced carbon capture capacity ever reaches operation?
About one eighth, on current readings. Axis Intelligence Research’s CCCI puts the delivery ratio at 12.5% — 64 Mtpa operating against 513 Mtpa announced — and that ratio moved only 0.17 percentage points from the prior year. The figure is a snapshot of the pipeline at a point in time, not a cohort survival rate for individual projects.
Why do carbon capture projects get cancelled after reaching FID?
Three causes dominate in the 2025–2026 record. Missing long-term offtake is the largest: hydrogen-linked capture projects stalled globally because no firm offtake agreements materialised, including post-FID developments in the US and Canada. Second, storage permitting bottlenecks — particularly in US states where permitting sits at federal level — have pushed projects toward 2035. Third, CO₂ pipeline routes face community opposition; the projects that advanced, such as Trailblazer, largely repurposed existing pipeline.
Does the 45Q tax credit actually cover the cost of capture?
It depends entirely on stream concentration. Axis Intelligence Research’s 45Q Coverage Ratio is 4.25× for pure streams like gas processing and ethanol, where the USD 85 credit dwarfs a USD 15–25 capture cost, but only 1.06× for dilute streams like cement and power, where the midpoint cost is USD 80. The second group still has to fund transport, storage and cost of capital out of a six-percent margin, which is why cancellations cluster there.
Is captured CO₂ permanently stored or used for enhanced oil recovery?
Both, and US policy no longer distinguishes between them financially. The One Big Beautiful Bill Act raised the 45Q credit for enhanced oil recovery and utilisation from USD 60 to USD 85, matching dedicated geologic storage, for equipment placed in service after 4 July 2025. Dedicated storage and EOR have materially different climate accounting, so any statistic combining them should state which is which.
Which is scaling faster — CO₂ capture or CO₂ storage?
Storage, currently. In the IEA’s 2026 database update, storage capacity grew around 25% while operational-or-under-construction capture capacity grew just over 10%. That widening gap creates utilisation risk for transport and storage developers, which the IEA named as a trend to watch and which Equinor has already responded to by scaling back parts of its storage investment plans.
Which country or region has the most operating capture capacity?
North America, with 30.19 Mtpa across 24 facilities above the 100,000 t/yr threshold. South America ranks second at 14.20 Mtpa — from a single facility, Petrobras’s Santos Basin pre-salt operation — followed by Asia at 11.05 Mtpa across 18 facilities. Europe operates under 3 Mtpa today but is projected to exceed 90 Mtpa by 2030.
Is carbon capture the same as carbon removal?
No, and the two should never be added together. Point-source capture prevents CO₂ reaching the atmosphere from an industrial source; carbon removal extracts CO₂ already in the atmosphere. They use different units of account, different buyers and different markets — durable CDR contracted 30.4 Mt in 2025 with a single buyer, Microsoft, accounting for 78.5% of disclosed purchases. This page reports them as separate series.
What is the CCCI and how often is it updated?
The Carbon Capture Conversion Index is an Axis Intelligence Research metric measuring how much announced global capture capacity has converted into operating or under-construction tonnes, on a 0–100 scale. Its components, weights and inputs are published in the Methodology section so any analyst can recompute it. It is recalculated when the Global CCS Institute publishes a new Global Status of CCS edition or the IEA updates its CCUS Projects Database.
What would move the CCCI reading materially?
A step change in the delivery ratio requires operating capacity to grow faster than the announced pipeline — which has not happened in the two readings available. The near-term candidates are Europe’s Track-1 and Porthos-linked capture projects reaching commissioning, and the 44 Mtpa currently under construction converting on schedule. Conversely, another wave of post-FID cancellations in dilute-stream sectors would push the commitment ratio down before the delivery ratio ever registers it.