The feasibility of reaching gigatonne scale CO₂ storage by mid-century

Integrated assessment models (IAMs) used in IPCC mitigation scenarios anticipate widespread global deployment of geological CO₂ storage, with mid-century injection rates typically exceeding 6 GtCO₂ yr⁻¹ (range 1–30 GtCO₂ yr⁻¹). These projections often assume ubiquitous, easily accessible storage resources (global totals of >1,000–10,000 Gt) and broad applicability of CCS across sectors. However, real-world CCS deployment has lagged far behind such expectations due to project cost, technology readiness, and limited revenue streams, with most current injection associated with EOR rather than dedicated storage. Many IAMs lack explicit, geologically grounded constraints on storage resource use, injectivity, and regional scale-up rates, and the presence or absence of simple constraints (e.g., supply cost curves or global capacity limits) does not systematically correlate with the storage actually deployed in model outcomes. The research question addressed here is whether gigatonne-scale CO₂ storage by 2050 is feasible when explicitly accounting for geological resource availability, geographies of deployment, and realistic growth-rate limits. The study aims to provide geographically resolved, physically and empirically anchored constraints on feasible scale-up trajectories and to assess the plausibility of AR6 pathway projections.

The paper reviews how IAMs represent CO₂ storage, noting widely varying global storage potential assumptions (∼1,000 to >10,000 Gt) and heterogeneous or absent subsurface constraints across models. Prior evaluations using more restrictive storage capacities, physics-based models of injectivity and plume migration, or analogues from hydrocarbon drilling and production indicate significant regional limitations that are not captured by IAMs and suggest slower scale-up in regions like China compared with the USA. Logistic growth models, commonly used for extractive industries and technology diffusion, have been proposed to bound plausible growth trajectories for CO₂ storage at global and regional scales, embedding effects of depletable resources and typical S-shaped adoption dynamics. Strengths of logistic models include simplicity and the ability to leverage historical analogues; trade-offs include limited granularity, lack of explicit economic/policy drivers, and validity primarily over multi-decadal horizons. These insights motivate a geographically resolved logistic framework constrained by current project pipelines and assessed storage resources.

The authors develop geographically resolved projections of CO₂ storage scale-up across ten regions with active or planned CCS: Australia, Canada, China, Indonesia, South Korea, Thailand, the UK, the USA, the EU (with offshore focus via Norway’s North Sea assets), and the Middle East (aggregating Saudi Arabia, Qatar, UAE). Regional readiness levels follow the GCCSI Readiness Index (2018). The core framework is a symmetric logistic model describing storage rate Q(t) and cumulative storage P(t) parameterized by resource base C (Gt), growth rate r (yr⁻¹), and peak year t_p, with t_p constrained to occur after 2050 to focus on the formative, near-exponential growth phase. The model is anchored to external constraints on cumulative storage by 2030 derived from announced capture capacities of existing and planned CCS projects (GCCSI Global Status 2022). Storage resource bases use central estimates from the OGCI Storage Resource Catalogue and national assessments, with uncertainties spanning one to two orders of magnitude; analyses bracket these via a hypothetical minimum (0.1× central) and hypothetical maximum (10× central). Growth rate caps are set at up to 20% yr⁻¹ for all regions except China (up to 25%) reflecting historical precedents and observed acceleration in large-scale infrastructure development. For each region, 1,000 random iterations explore combinations within predefined parameter spaces for C, r, and t_p that match the 2030 cumulative constraint (least-squares fit). Global aggregate storage distributions are generated by summing regional rates across random combinations (1,000 iterations). Six scenarios are analysed: Reference (central resource, r up to 20%/25% in China), Minimum (resource 0.1×, r up to 10%), Maximum (resource 10×, r up to 20%/25% China), Growth10% (central resource, r up to 10%), US1Gt (Reference bounds with USA capped at 1.04 Gt yr⁻¹ by 2050), and EUUSChina (Reference bounds with USA 1.04 Gt yr⁻¹, EU 0.33 Gt yr⁻¹, UK 0.175 Gt yr⁻¹, China 0.216 Gt yr⁻¹). The model output is compared with AR6 Scenario Explorer projections for 1.5 °C (limited and high overshoot) and 2 °C (>50% and >67% likelihood) climate categories.

• The maximum feasible global storage rate by 2050 is 16 GtCO₂ yr⁻¹ across all scenarios, but only if the USA provides at least 60% (≈10 GtCO₂ yr⁻¹) of the total; under the Reference scenario (central resources), the feasible upper bound is 13 GtCO₂ yr⁻¹.
• Approximately 8% (56/689) of AR6 model pathways exceed 16 GtCO₂ yr⁻¹ in 2050 and are deemed infeasible within the model’s geologic and growth constraints, implying sustained growth >20% and resource needs exceeding theoretical maxima for individual countries.
• If USA deployment is limited to 1.04 GtCO₂ yr⁻¹ by 2050 (US Long-Term Strategy), the global total cannot exceed 6 GtCO₂ yr⁻¹. When also constraining EU to 0.33 GtCO₂ yr⁻¹, UK to 0.175 GtCO₂ yr⁻¹, and China to 0.216 GtCO₂ yr⁻¹ (EUUSChina scenario), the global total is capped at ≈5 GtCO₂ yr⁻¹.
• Sustained growth limited to ≤10% yr⁻¹ (Growth10% and Minimum scenarios) restricts global storage to ≤1 GtCO₂ yr⁻¹ by 2050, below the requirements of AR6 1.5 °C and 2 °C pathways.
• AR6 projections substantially overestimate feasible deployment in China, Indonesia, and South Korea (e.g., up to 6.7 GtCO₂ yr⁻¹ for China would require >30% sustained annual growth for at least 20 years), while projections for Australia, Canada, and the USA are within the model’s feasible ranges.
• Achieving ≥3.6 GtCO₂ yr⁻¹ globally (lower quartile of AR6 2 °C pathways) requires Australia and Canada to each deliver at least 0.2 GtCO₂ yr⁻¹ by 2050 (about a 10× increase from their 2030 plans); Indonesia, South Korea, Thailand, and the Middle East have no minimum required contribution at that threshold.
• As global targets rise, the geography shifts: to reach 13 GtCO₂ yr⁻¹ in the Reference scenario, the USA would need to store nearly 70% (≈8.4 GtCO₂ yr⁻¹), implying r ≈18–20% sustained and reliance on a large resource base (≈central estimate of 506 Gt). Even modest Chinese contributions would require >20% sustained growth to 2050.
• A convergent benchmark from distinct modelling approaches suggests a feasible global CO₂ storage rate of about 5–6 GtCO₂ yr⁻¹ by 2050, with the USA around 1 GtCO₂ yr⁻¹ as the largest single contributor.

The study directly addresses the feasibility of gigatonne-scale geological CO₂ storage by 2050 under geologically and operationally grounded constraints. It shows that while gigaton-scale deployment is technically feasible, the upper bounds of AR6 projections require growth rates and regional allocations—particularly a dominant U.S. contribution—that are far from current policy and business frameworks. The results reconcile discrepancies between IAM projections and empirical constraints by highlighting that overestimation of feasible deployment in several Asian countries (notably China, Indonesia, and South Korea) inflates global totals in many pathways. Conversely, Western regions (USA, EU, UK, Canada, Australia) have projections and national targets largely consistent with feasible ranges. The geography of feasible deployment tightens as higher global totals are targeted, making outcomes increasingly dependent on high U.S. growth and resource utilisation. The findings imply IAMs should incorporate explicit subsurface resource and growth constraints to improve realism, and policymakers should calibrate expectations and infrastructure planning (transport networks, cross-border agreements) to the most plausible regional contributions. The work provides a tractable, data-driven framework to bound mid-century deployment that complements more granular cost-supply and physics-based injectivity assessments.

By combining a geographically resolved logistic growth framework with constraints from announced 2030 capacities and assessed storage resources, the study provides feasible bounds for global geological CO₂ storage by 2050. The global maximum across scenarios is 16 GtCO₂ yr⁻¹ (Reference scenario upper bound 13 GtCO₂ yr⁻¹), but achieving the high end requires the USA to supply the majority share of storage. Under constraints aligned with government roadmaps, plausible global totals are 5–6 GtCO₂ yr⁻¹ by mid-century, with the USA contributing ≈1 GtCO₂ yr⁻¹. Many AR6 pathways exceeding 16 GtCO₂ yr⁻¹ are infeasible given geologic and growth limits, and several regions—especially China, Indonesia, South Korea—are overrepresented in IAM deployment projections. Future research should integrate these multi-decadal growth constraints with granular cost-supply curves and physics-based injectivity and pressure-management models to capture year-to-year variability, improve regional siting, and refine resource estimates through learning-by-doing and field data.

• Logistic growth modelling provides multi-decadal averages and lacks high temporal fidelity; it does not explicitly represent annual volatility, market incentives, regulatory drivers, or public acceptance dynamics.
• Storage resource estimates carry 1–2 orders of magnitude uncertainty; while bracketing (0.1× to 10× central) is used, actual accessible capacity depends on future site characterisation, engineering design, and pressure management.
• Peak year is constrained to occur after 2050, focusing on the formative phase and possibly overstating late-stage availability if resources prove more limited or injectivity constraints bind earlier.
• EU resource considerations are simplified (offshore focus, limited onshore consideration due to social and regulatory uncertainties), and regional aggregations (e.g., Middle East) abstract intra-regional heterogeneity.
• The framework assumes all captured CO₂ is stored geologically and does not differentiate capture sources; transport, storage permitting, and cross-border logistics are addressed qualitatively, not modelled explicitly.
• High-end feasible outcomes depend heavily on unprecedented U.S. deployment and sustained high growth rates, which currently lack established business models and policy frameworks, increasing uncertainty in those trajectories.