Environment · 6 min read
Marine Carbon Dioxide Removal: Methods, Evidence, and Governance Questions
A bounded comparison of biological, chemical, and electrochemical methods, with measurement, ecological, permanence, and consent questions.

Marine Carbon Dioxide Removal: Methods, Evidence, and Governance Questions
The ocean already absorbs a substantial share of human carbon dioxide emissions. Proposals for marine carbon dioxide removal, or mCDR, would deliberately enhance biological, chemical, or physical processes so that more atmospheric CO₂ enters the ocean and remains stored for a meaningful period. Suggested methods include growing and sinking seaweed, adding nutrients to stimulate phytoplankton, increasing seawater alkalinity, and extracting dissolved CO₂ with electrochemical systems.
These proposals are not one technology at one readiness level. Some components are familiar from aquaculture, mining, wastewater treatment, or desalination, while the claimed climate service—additional, durable atmospheric removal at scale—can be difficult to measure. Local experiments may establish chemical response or biological growth without showing net removal after energy use, supply-chain emissions, ecological changes, and carbon leakage are included.
National Academies and U.S. government assessments support expanded research while emphasizing unresolved efficacy, durability, ecosystem, justice, and governance questions. That position is neither endorsement of commercial deployment nor a conclusion that every approach will fail.
Why the ocean is part of the removal debate
Climate assessments distinguish carbon dioxide removal from emissions reduction. Avoiding a tonne of new CO₂ and removing a tonne already in the atmosphere can both affect net emissions, but they involve different baselines, risks, and accounting. The Intergovernmental Panel on Climate Change finds that pathways limiting warming generally include some removal to counterbalance residual emissions, while stressing that removal cannot substitute for rapid emissions cuts.
The ocean covers about 70 percent of Earth and exchanges carbon continuously with the atmosphere. It stores carbon in dissolved inorganic forms, organisms, particles, and sediments over timescales from days to millennia. mCDR proposals try to redirect or accelerate parts of these cycles.
Removing dissolved CO₂ from surface water can allow additional atmospheric CO₂ to enter until the air–sea system moves toward a new balance. Increasing alkalinity can shift dissolved carbon into bicarbonate and carbonate forms, lowering surface-water CO₂ and potentially countering local acidification. Biological production can incorporate carbon into tissue, but climate benefit depends on what happens after organisms are eaten, respire, decompose, or sink.
Biological proposals move carbon through food webs
Ocean fertilization adds a limiting nutrient, often iron, to stimulate phytoplankton growth. Field experiments have shown that iron can trigger blooms in some high-nutrient, low-chlorophyll regions. The difficult question is how much additional carbon reaches depths where it remains isolated from the atmosphere, for how long, and with what ecological effects. A bloom visible from a ship or satellite is not itself a removal measurement.
Potential side effects include altered food webs, oxygen loss as organic matter decomposes, production of other greenhouse gases, and nutrient redistribution away from downstream ecosystems. Outcomes depend on location, season, species, and circulation. Past experiments are too small and short to establish commercial-scale net removal or ecological safety.
Macroalgae cultivation proposes to grow kelp or other seaweed and store its carbon through sinking, burial, products, or avoided emissions. Farms can provide habitat or local economic benefits, but carbon accounting is demanding. Existing ecosystems may already use the nutrients consumed by a farm. Harvested material may decompose quickly. Deliberately sinking biomass moves nutrients and organic matter into deep habitats whose responses are poorly observed.
Coastal restoration of mangroves, salt marshes, and seagrass is often discussed alongside mCDR. These habitats have well-established ecological benefits and can store carbon, but additionality, permanence, available area, and vulnerability to warming or development bound their climate contribution.
Chemical and electrochemical proposals alter seawater
Ocean alkalinity enhancement would add alkaline material or use electrochemistry to increase seawater’s capacity to hold inorganic carbon. Candidate minerals include lime-derived products and finely processed silicates. In principle, higher alkalinity lowers the partial pressure of CO₂ at the surface, drawing carbon from air and storing much of it as bicarbonate.
The chemistry is well understood at basic equilibrium level; implementation is not. Mining, grinding, heating, transporting, and dispersing material require energy and can generate emissions. Minerals may contain trace metals. Rapid local changes in pH or particle concentration can affect organisms even if broader chemical change would reduce acidification. Dissolution rates and ocean mixing determine when and where atmospheric uptake occurs.
Electrochemical approaches pass seawater through systems that separate acid and base streams or remove dissolved CO₂. They can produce measurable outputs in controlled equipment, but energy demand, membrane durability, treatment of byproducts, and the carbon intensity of electricity affect net removal. Coupling to desalination or existing coastal infrastructure may share costs, though it also constrains suitable sites.
Artificial upwelling and downwelling attempt to move water vertically. Bringing nutrient-rich deep water upward can stimulate production but may also release naturally carbon-rich water to the atmosphere. Sending surface water downward does not guarantee long storage if circulation returns it quickly. Physical transport is therefore not equivalent to verified atmospheric removal.
Measurement must follow carbon across boundaries
A credible claim begins with a counterfactual: how much atmospheric CO₂ would have been removed without the intervention? Natural ocean variability is large, and currents move signals beyond a project boundary. Researchers need observations and models to separate intervention effects from changing temperature, biology, and circulation.
Measurement, reporting, and verification must follow more than carbon concentration at the treatment site. It should include upstream emissions, air–sea exchange, changes in other greenhouse gases, ecological responses, lateral transport, and storage duration. For some approaches, the atmospheric response is delayed and spread over a large area, making direct project-level measurement impractical.
Permanence is not binary. Carbon stored in deep water for decades differs from geological storage for millennia, yet both can affect peak warming depending on timing. Accounting systems must specify a time horizon and address reversal. Those choices are partly scientific and partly policy judgments about equivalence.
The National Academies’ 2022 research strategy evaluated six broad ocean-based approaches and recommended substantial, staged research rather than premature scaling. NOAA’s 2023 and 2024 strategies similarly emphasize observations, modeling, ecosystem effects, standards, and decision support. The repeated demand for research reflects the gap between plausible mechanisms and verified climate performance.
Governance is developing before evidence is complete
Ocean interventions cross jurisdictions and affect shared resources. National permits, environmental review, the London Convention and Protocol concerning marine pollution, maritime law, fisheries rules, Indigenous rights, and local consent may all become relevant depending on method and location. Legal classification remains unsettled for some activities.
Private companies are already conducting pilots and selling or proposing removal credits. Early finance can support learning, but credit revenue creates incentives to claim removal before measurement standards mature. Transparent protocols, independent data access, preregistered monitoring, and liability rules can reduce conflicts of interest. They cannot resolve ethical questions about who may alter a marine system and who bears uncertain risk.
Governance also has a research dilemma. Demanding commercial-scale evidence before permitting bounded experiments can prevent learning; allowing scaling on the basis of laboratory evidence can expose ecosystems and communities. Staged trials with stopping rules and increasing monitoring are one proposed response. Whether a given trial is acceptably bounded depends on reversibility, exposure, and local authority.
What is established is that several mCDR methods have physically plausible mechanisms and that small experiments can test parts of them. It is also established that emissions reduction remains the most direct way to limit further ocean acidification. What is contested is which approaches merit larger trials and how removal should be credited across time and space. What remains unknown includes net cost, durable capacity, and ecological effect at climate-relevant scale.
The field is therefore at a comparison stage, not a verdict. The relevant question is not whether “the ocean” can remove carbon—it already does—but whether a specific intervention causes additional, measurable, durable removal whose benefits exceed its full environmental and social costs.
Sources
4 sources, every one with a link you can follow.
- National Academies of Sciences, Engineering, and Medicine, A Research Strategy for Ocean-Based Carbon Dioxide Removal and Sequestration (2022).
- NOAA, Carbon Dioxide Removal and marine methods.
- NOAA, National Marine Carbon Dioxide Removal Research Strategy (2024).
- IPCC, Sixth Assessment Report, Working Group III: carbon dioxide removal.
