Direct Air Capture Scale Claims Need Net-Removal Accountability
Direct air capture is often discussed as if every captured tonne of CO2 is automatically a durable climate removal. That shorthand is operationally unsafe. This paper synthesizes deployment trackers, public direct-air-capture hub policy, life-cycle assessment guidance, carbon-removal certification law, geologic-storage regulation and peer-reviewed process, scenario and community evidence. The synthesis shows that DAC scale claims need a denominator that follows the tonne from atmospheric capture through energy supply, process materials, transport, injection, storage monitoring, liability and public reporting. Gross capture capacity, project announcements and credit purchases are useful signals, but none of them alone establishes net removal. The contribution is a seven-gate accountability stack that separates hardware scale from certified climate value. DAC can be a meaningful part of a residual-emissions strategy only when claims are expressed as net, durable, auditable removals rather than as nameplate capture or procurement volume.
Introduction
Direct air capture (DAC) has moved from laboratory and pilot discussion into public procurement, hub funding, infrastructure planning and corporate carbon-removal portfolios. The climate rationale is clear: some net-zero pathways use carbon dioxide removal (CDR) to counterbalance residual emissions that are hard to eliminate, and official climate-policy language increasingly treats durable removals as part of the long-run mitigation portfolio [[cite:ipcc2022,eu2024]]. The risk is equally clear: once DAC becomes a procurement or industrial-policy symbol, the phrase "one tonne captured" can be mistaken for "one tonne durably removed."
That substitution is the paper's target. A DAC contactor may separate CO 2 from ambient air, but the climate claim depends on the full system: energy source, heat integration, sorbent or solvent supply chain, compression, transport, injection, storage permanence, monitoring, liability, additionality and double-counting controls. IEA tracking and the State of CDR show an expanding but still early-stage field, not a mature commodity market in verified net removals [[cite:iea2025,statecdr2026]].
This paper asks: what evidence must accompany a DAC scale claim before it can be treated as a net climate-removal claim? The answer is a source-grounded accountability stack. The stack does not reject DAC; it rejects claim shortcuts. It treats capacity, capture, storage and certified removal as related but distinct states of evidence.
Methods
I used a conceptual-synthesis method. On 2026-06-26, I ran six AlexandrAI graph searches and twelve external searches across official policy pages, assessment reports, peer-reviewed papers, research-agenda reports, certification law and lifecycle guidance. Screening prioritized sources that could answer one of four questions: what DAC is expected to do in climate pathways; what makes gross capture differ from net removal; what storage and monitoring evidence is required; and what policy or community constraints limit credible scale claims.
The full-read corpus contains thirteen sources. They include deployment and CDR-status trackers [[cite:iea2025,statecdr2026]], mitigation assessment context [[cite:ipcc2022]], public DAC-hub policy [[cite:doe2025]], DAC-specific LCA guidance [[cite:doelca2024]], a carbon-removal certification framework [[cite:eu2024]], geologic-storage regulation [[cite:epa2025]], a negative-emissions research agenda [[cite:nasem2019]], peer-reviewed process and scenario work [[cite:keith2018,realmonte2019,deutz2021]], community-evidence research [[cite:scott2024]] and durable-offsetting principles [[cite:oxford2020]].
Each source was coded against seven accountability gates: gross capture quantification, lifecycle emissions, energy and resource boundary, transport and storage, durability and liability, additionality and double counting, and independent public MRV. A source counted for a gate only if it directly addressed the evidence requirement or a close operational proxy. Table 1 summarizes how the evidence families were used. Figure 1 reports the gate counts; Table 2 turns the same synthesis into a claim-screening ledger.
Results
First, the evidence base separates deployment scale from climate value. IEA and State of CDR tracking make DAC visible as a growing CDR pathway, but neither a facility announcement nor nameplate capture capacity proves net removal [[cite:iea2025,statecdr2026]]. DOE hub materials similarly demonstrate public investment in DAC infrastructure, yet they are program evidence rather than final verification that a given tonne was removed and stored [[cite:doe2025]]. The correct denominator is therefore not "announced tonnes per year"; it is "net tonnes durably stored and independently verified."
Second, lifecycle subtraction is not optional. DOE LCA guidance and the EU certification framework both require attention to energy, materials, transport and associated direct or indirect emissions when quantifying net benefit [[cite:doelca2024,eu2024]]. Deutz and Bardow show why this matters for industrial DAC: the climate value of a DAC process can change with energy supply and upstream assumptions [[cite:deutz2021]]. A gross-capture claim that omits the emissions needed to run, build and supply the system is therefore not yet a removal claim.
Third, storage converts capture evidence into removal evidence only when monitoring and liability are specified. Geologic storage rules such as EPA Class VI requirements define a regulatory route for CO 2 injection and post-injection obligations, while the EU framework links permanent removals to storage duration, conservative quantification, monitoring and liability [[cite:epa2025,eu2024]]. This turns permanence from a label into an operating duty. A tonne that is captured but not tracked into a durable storage claim remains an intermediate process output.
Fourth, large-scale DAC is a system claim, not only a plant claim. Scenario work finds that DAC can appear at large scale in deep mitigation pathways, but those pathways also reveal energy-system implications that must be disclosed [[cite:realmonte2019]]. Process work shows that technical costs and performance estimates depend on process configuration and assumptions [[cite:keith2018,nasem2019]]. Community evidence adds a further boundary: support for DAC deployment is conditional on governance, local effects and benefit structures, not automatic acceptance of the technology [[cite:scott2024]].
Accountability Model
The synthesis supports a simple equation for DAC claims. Let R net be certified net removal, C gross be measured atmospheric CO 2 capture, E life be lifecycle greenhouse gas emissions associated with the activity, and L risk be conservative deductions or liabilities for leakage, reversal, double counting and unresolved monitoring risk. A credible claim should be stated as:
R net = C gross - E life - L risk
The equation is intentionally conservative. It follows the EU framework's requirement to quantify net benefits after associated emissions and conservative uncertainty treatment, while aligning with DAC LCA guidance that treats energy and supply-chain inputs as part of the accounting boundary [[cite:eu2024,doelca2024]]. It also makes a governance point: if a project cannot disclose enough evidence to estimate the terms, the claim should remain at the lower evidence state, such as "capture capacity" or "DAC project under development."
This stack sharpens the novelty boundary. Prior sources establish DAC process feasibility, scenario relevance, LCA sensitivity, regulatory storage requirements and certification principles [[cite:keith2018,realmonte2019,deutz2021,epa2025,eu2024]]. The contribution here is to join them into a claim-state model: announced capacity , gross captured CO 2 , stored CO 2 , and certified net removal are progressively stronger evidentiary states. Policy and procurement documents should name the state they are claiming.
Discussion
The accountability stack has three practical implications. First, procurement should buy evidence states, not slogans. A purchase agreement for future DAC tonnes can fund scale-up, but it should not be reported as delivered climate benefit until gross capture, lifecycle subtraction, storage and independent verification are complete. Second, public funding should report learning milestones separately from removal outcomes. DOE DAC hubs can be valuable even before they deliver verified removals, but the performance claim should match the evidence state [[cite:doe2025]].
Third, certification should protect emissions-reduction priority. IPCC mitigation pathways and the Oxford Principles both warn against treating removals as a reason to delay emissions cuts [[cite:ipcc2022,oxford2020]]. DAC has its strongest role in residual-emissions strategies and possibly net-negative phases, not as a blanket substitute for avoidable emissions. A net-removal ledger should therefore travel with a residual-emissions ledger: what emission source is being counterbalanced, why it is residual, and how the removal will remain durable.
The strongest counterargument is that strict accounting could slow early DAC deployment by raising transaction costs before the industry matures. That is a real risk. The answer is not to weaken accounting, but to separate labels. Early projects can be described as demonstrations, capacity additions, learning investments or future-removal offtakes. They should become net-removal claims only when the evidence passes the seven gates. This distinction lets industrial policy support learning while preventing premature climate accounting.
The second counterargument is that voluntary-market protocols can handle these distinctions privately. Some can, but private certificates do not replace public comparability. EU certification language points toward standardised baselines, conservative quantification, independent auditing and registry controls precisely because removal claims become public climate claims [[cite:eu2024]]. The accountability stack therefore favors auditable disclosure over trust in project branding.
Limitations
This paper is a conceptual synthesis, not an empirical measurement of DAC plants. It does not estimate current global verified DAC removals, compare plant-level cost curves, or rank DAC technologies. The source coding in Figure 1 is transparent but author-coded; another reviewer could reasonably merge or split gates differently. Dynamic market trackers and company project pages were screened but not used as primary evidence because the paper emphasizes durable public claims rather than the latest procurement volume.
The paper also focuses on DAC with storage. DAC-derived CO 2 used in short-lived products, fuels or other utilization routes would require a different claim ledger because the storage term and displacement claims change. Finally, regulations and certification methodologies are evolving. Any operational project should apply the current jurisdiction-specific rules, not treat this synthesis as a compliance checklist.
Conclusion
Direct air capture can become climate-relevant only when its claims are disciplined by evidence. The central result of this synthesis is that DAC scale should be reported as a sequence of states: announced capacity, measured gross capture, stored CO 2 , and certified net removal. A tonne moves to the next state only when the evidence follows it. That discipline protects the credibility of DAC, because it lets early deployment proceed without allowing future capacity, procurement volume or gross capture to masquerade as delivered durable climate benefit.