Carbon sequestration tap points

Public science page — research and education synthesis. Not a deployment recommendation. Not a carbon-credit product.

Abstract

The ocean’s dissolved inorganic carbon reservoir is vastly larger than the atmosphere, and the contemporary ocean sink already removes a substantial share of anthropogenic CO₂ (Wanninkhof et al., 2013; Gruber et al., 2023; Friedlingstein et al., 2023). Sequestration, however, is not identical to air–sea uptake. It is the time carbon spends out of atmospheric contact — set by chemical form, export pathway, remineralization depth, water-mass isolation, and (rarely) sedimentary burial (Siegel et al., 2021, https://doi.org/10.1088/1748-9326/ac0be0 Kwon et al., 2009, https://doi.org/10.1038/ngeo612 Boyd et al., 2019, https://doi.org/10.1038/s41586-019-1098-2).

This synthesis organises that chain as twelve tap points: air–sea gas exchange; Revelle / alkalinity state; nutrient- and iron-limited fixation; gravitational export (aggregation and ballast); particle injection pumps; mesopelagic remineralization depth; microbial routing to refractory dissolved organic carbon; the carbonate counter-pump; mode-water and deep subduction; basin residence below roughly 1,000 m; sedimentary burial; and coastal blue-carbon accretion. Proposed marine carbon-dioxide-removal (mCDR) approaches are then located on the same map. The central literature message is blunt: raising surface biomass does not automatically produce durable sequestration, and chemically durable alkalinity pathways still face equilibration, detection, and supply constraints (National Academies of Sciences, Engineering, and Medicine [NASEM], 2022; Buesseler et al., 2024; Renforth & Henderson, 2017).

Keywords: biological carbon pump; particle injection pumps; solubility pump; microbial carbon pump; ocean iron fertilisation; ocean alkalinity enhancement; marine CDR; sequestration timescale; blue carbon.

1. Purpose and definition

Ocean Flux publishes this page so readers can see where carbon actually becomes hard to return to the air, which knobs control that transition, and which research pathways touch which knobs — without treating any method as a ready climate product.

A tap point is a process node where carbon (i) changes chemical form, (ii) changes vertical or lateral reservoir, or (iii) changes return time to the atmosphere by about an order of magnitude. The metaphor is hydraulic: reservoirs linked by valves. Opening a valve increases flux; changing the downstream pipe changes how long carbon stays isolated. Export can rise without durable storage; alkalinity can rise without detectable air–sea uptake if equilibration is incomplete (NASEM, 2022; Siegel et al., 2021).

The framing builds on the classical pump vocabulary of Volk and Hoffert (1985), the microbial carbon pump (Jiao et al., 2010, 2024), the multi-pathway biological pump including particle injection pumps (Boyd et al., 2019), transit-time assessments of storage duration (Siegel et al., 2021), and research strategies for ocean-based CDR (NASEM, 2022).

2. Reservoirs, fluxes, and what “sequestered” means

2.1 Relevant pools (order-of-magnitude context)

Approximate contemporary inventories compiled in ocean biogeochemistry syntheses place atmospheric carbon near ~880 Pg C (and rising), ocean dissolved inorganic carbon (DIC) near ~38,000 Pg C, and ocean dissolved organic carbon near ~660 Pg C, with marine biomass only of order a few Pg C (Sarmiento & Gruber, 2006; Hansell, 2013; Gruber et al., 2023). These stock figures are Assumption / order-of-magnitude on this page — useful scale, not Ocean Flux measurement products.

The leverage of biology is not the standing stock of plankton. It is the rate at which a small organic inventory is produced, exported, and respired, thereby helping maintain a surface-to-deep DIC gradient (Boyd et al., 2019; Nowicki et al., 2022).

2.2 The contemporary ocean sink (sourced context)

Observation-based syntheses show that the ocean has absorbed on the order of one quarter of anthropogenic CO₂ over recent multi-decadal windows, with a mean sink near −2.7 Pg C yr⁻¹ for 1990–2019 in Gruber et al. (2023) and consistent Global Carbon Budget accounting in Friedlingstein et al. (2023). Wanninkhof et al. (2013, https://doi.org/10.5194/bg-10-1983-2013) remain a canonical observation-based statement of global ocean carbon uptake magnitude, variability, and trends — cited here as sink context, not as an Ocean Flux credit or explorer output.

In short: air–sea uptake is transfer into the ocean carbon system. It is not, by itself, sedimentary burial or a credit tonne.

2.3 Operational definition of sequestration

Sequestration is time out of atmospheric contact. Siegel et al. (2021) show that interior discharge of CO₂ (or of organic carbon that will become CO₂) has a highly skewed transit-time distribution: medians often decades to centuries, with a heavy tail. Retention over a roughly 50-year horizon is generally higher for injection deeper than ~1,000 m in many basins, with important exceptions (for example, parts of the western North Atlantic). Biological-pump carbon that remineralizes in the upper mesopelagic returns toward the surface on that horizon at high fractional rates in their assessment — a reminder that depth is a first-order tap, not a detail (Siegel et al., 2021; Kwon et al., 2009).

Practical duration classes used here (teaching labels, not credit grades):

• Short: years to ~two decades (mixed layer, seasonal thermocline, shallow mesopelagic).
• Medium: decades to ~150 years (mode waters, much of the gravitational pump).
• Long: centuries to ~1,000 years (deep Pacific/Indian isolation pathways).
• Geologic: ≫10⁴ years (burial that survives early diagenesis; durable alkalinity storage on weathering timescales — research framing).

3. Natural pumps (compact)

Table 1. Natural ocean carbon pumps. Process identities are DOI-backed in the text. Global review Pg C / GtCO₂ bands from the working extract remain Assumption on this site until each figure is tied to that paper’s wording.

Solubility pump. Cold, dense water carries DIC into the interior. Rising atmospheric pCO₂ drives invasion; circulation carries that DIC down. Falling buffer capacity (rising Revelle factor) is a tap that climate change itself is tightening (Wanninkhof et al., 2013; Gruber et al., 2019, 2023).

Soft-tissue biological pump. Photosynthesis converts DIC to organic carbon; a fraction leaves the euphotic zone. Almost all is respired again in the mesopelagic and abyss; only a small residual reaches long-term burial (Dunne et al., 2007; Nowicki et al., 2022). Boyd et al. (2019, https://doi.org/10.1038/s41586-019-1098-2) argue that particle injection pumps (mixed-layer subduction, eddy-driven subduction, migrant pumps) likely sequester on a scale comparable to gravitational settling and help close mesopelagic carbon-budget gaps. On this site, PIPs are treated as teaching context that sits beside gravitational export — not a licence to invent durable-burial product numbers from eddy anomalies alone.

Carbonate counter-pump. Calcification raises surface pCO₂ while exporting alkalinity; dissolution at depth does the reverse. Ballast can speed organic sinking, so the rain ratio is a signed tap (Hain et al., 2014; Boyd et al., 2019).

Microbial carbon pump. Microbes can route a slice of production into refractory DOC that persists for centuries to millennia without requiring deep sinking (Jiao et al., 2010, 2024; Legendre et al., 2015). MRV of small refractory increments against a large DOC inventory remains hard.

4. Evaluation axes

Following mCDR research assessments (NASEM, 2022), each tap can be scored qualitatively on:

1. Additionality — extra atmospheric CO₂ removed, or only rearranged ocean carbon?
2. Efficiency — carbon moved per energy, nutrient, or material.
3. Durability — return time after the intervention stops.
4. MRV — detectable against the natural background?
5. Ecological risk — food-web shifts, oxygen loss, N₂O, toxins, biodiversity.
6. Scale headroom — literature’s climate-scale potential if used responsibly — not a deployment recommendation.

High efficiency with weak durability is the classic OIF tension (Boyd et al., 2007; Buesseler et al., 2024; Jiang et al., 2024). High durability with hard MRV is the present alkalinity story (Renforth & Henderson, 2017; NASEM, 2022).

5. Twelve tap points (catalogue)

T1 — Air–sea gas exchange. Transfer is set by the air–sea pCO₂ disequilibrium and gas-transfer physics (Wanninkhof et al., 2013). Interventions usually change surface pCO₂ (biology, alkalinity, direct ocean capture), not the transfer velocity itself. Without sustained undersaturation, deeper plumbing does not pull atmospheric carbon. This tap is transfer, not burial.

T2 — Revelle factor and alkalinity. Alkalinity sets how much DIC a given pCO₂ can store. Adding alkalinity lowers pCO₂ and, after re-equilibration, stores carbon mainly as bicarbonate on long timescales (Renforth & Henderson, 2017). Constraints include precipitation if saturation is driven too high, incomplete coastal equilibration, and life-cycle costs of minerals and energy (NASEM, 2022).

T3 — Nutrient- and iron-limited fixation. Modern surface production is rarely CO₂-limited; it is limited by N, P, Si, Fe, and light (Moore et al., 2013; Tagliabue et al., 2017). Roughly one-third of the surface ocean is HNLC. Mesoscale iron-enrichment experiments showed Fe can raise production and restructure communities (Boyd et al., 2007; Smetacek et al., 2012). Fixation is not sequestration unless coupled to T4–T6 / T10–T11.

T4 — Aggregation, pellets, and ballast (gravitational export). Packaging and mineral ballast set sinking speed and remineralization depth. Community composition (for example, heavily silicified diatoms) matters more than chlorophyll alone (Boyd et al., 2019; Buesseler & Boyd, 2009).

T5 — Particle injection pumps. Submesoscale fronts, mixed-layer pump dynamics, and vertical migrants inject carbon below the seasonal mixed layer without fast gravitational sinking (Boyd et al., 2019; Omand et al., 2015). Teaching only on this site: PIPs inform siting and interpretation; they do not mint new explorer burial keys.

T6 — Mesopelagic attenuation / remineralization depth. Small deepenings of remineralization can strongly affect atmospheric CO₂ because more carbon is handed to longer-isolation water masses (Kwon et al., 2009, https://doi.org/10.1038/ngeo612). Surface greening that leaves T6 unchanged buys little durability (Siegel et al., 2021).

T7 — Microbial routing to refractory DOC. Long potential duration, difficult additionality and MRV (Jiao et al., 2010, 2024). Research tap — not a credit tap.

T8 — Carbonate production and dissolution. Signed: more surface calcification is not automatically more atmospheric benefit.

T9 — Mode-water and deep convection subduction. Physical subduction stores anthropogenic DIC already taken up at the surface. Climate weakening of overturning pathways is a paper-result / teaching story for water-mass integrity — not an Ocean Flux model default (Gruber et al., 2023).

T10 — Basin residence below ~1,000 m. Same carbon, different basin, different clock (Siegel et al., 2021). Siting charts for durability claims should use transit-time information, not chlorophyll maps alone. The 1,000 m figure is a heuristic, not a law.

T11 — Sedimentary burial. The main biotic path that routinely reaches geologic duration — and a small flux relative to water-column respiration (Dunne et al., 2007). Paleoceanographic dust–Fe coupling (Martínez-García et al., 2014) and glacial deep-sea organic-carbon burial pulses (Cartapanis et al., 2016) are existence proofs under past climates, not a modern OIF operations manual. Do not invent burial keys from air–sea transfer or eddy anomalies, and do not treat contested seafloor-disturbance remobilization petagrams as established global magnitudes.

T12 — Coastal blue-carbon accretion. Mangroves, marshes, and seagrasses bury carbon at high rates per area; global potentials are often described as far below open-ocean gigatonne ambitions unless cultivation is scaled and coupled to durable export — treat any Tg-class total as Assumption pending claim-specific DOI wording (Duarte et al., 2013; Krause-Jensen & Duarte, 2016; Macreadie et al., 2019). Habitats score high on co-benefits and lower on climate-scale headroom than some open-ocean proposals claim.

Qualitative score sheet (not a deployment ranking)

Table 2. Qualitative scores. “Headroom” restates literature order-of-magnitude discussion — not Ocean Flux targets and not a recommendation to deploy.

6. Interventions mapped onto taps

Table 3. Intervention-to-tap map for education. Durability-if-stopped is the fate of already-moved carbon after cessation — not the fate of the climate problem.

Ocean iron fertilisation. Twelve mesoscale experiments (1993–2005) established Fe limitation responses in HNLC waters (Boyd et al., 2007). EIFEX documented deep export in at least one Southern Ocean diatom bloom (Smetacek et al., 2012). Syntheses agree that production responses are clearer than durable sequestration (Buesseler et al., 2024; Jiang et al., 2024). Assessment tables sometimes quote gigatonne-class OIF potentials under sustained, favourable, low-side-effect assumptions; those magnitudes remain Assumption here (not demonstrated operations, not Ocean Flux targets). Paleoglacial dust–burial coupling is not a modern permit.

Ocean alkalinity enhancement. Acts on T2 then T1 with chemically attractive durability if alkalinity stays in solution (Renforth & Henderson, 2017). Open research problems include dissolution kinetics, air–sea equilibration, detection of a small signal on a large natural sink, and full life-cycle accounting (NASEM, 2022).

Blue-carbon restoration. Direct T12 with co-benefits unmatched by open-ocean methods. Not a gigatonne tap on current area and accretion rates in most syntheses (Duarte et al., 2013; Macreadie et al., 2019). Still among the more methodologically developed marine pathways at project scale — without converting Ocean Flux into a credit vendor.

Protecting existing pumps. Reducing unnecessary degradation of margin habitats and of overturning and buffer capacity is carbon management of the largest active reservoir — conservation first, not branded mCDR (Gruber et al., 2023). Contested petagram-scale remobilization figures from seafloor disturbance are not asserted as Facts here.

7. Measurement, reporting, verification, and governance

Natural fluxes dwarf near-term pilots. Detecting an intervention requires process tracers at the claimed tap, an air–sea flux increment (not chlorophyll alone), a sequestration-time estimate from depth and water-mass age, life-cycle accounting, and attribution against observing systems such as SOCAT / GLODAP / BGC-Argo backgrounds (Wanninkhof et al., 2013; Siegel et al., 2021; NASEM, 2022; Buesseler et al., 2024).

Governance can close a scientific tap. Ocean fertilisation has been constrained under London Protocol / CBD decision pathways; high-seas EIA frameworks continue to evolve; research permitting is not a commercial licence (NASEM, 2022). A responsible sequence restated from research academies is: deep emissions cuts first; transparent research with environmental assessment; no climate-scale deployment until durability, ecology, and MRV are demonstrated in the biome of use.

Ocean Flux’s public posture matches that sequence: ecology and restoration first; Flux Explorer headlines stay pathway tonnes of C and N; Fe/P and related quantities stay drivers; credit maths stay off.

8. Conclusions

Carbon sequestration in the ocean is a chain. Public discussion often names the first link — grow more phytoplankton, dissolve more rock — and treats the rest as automatic. The literature says otherwise.

• Air–sea exchange is transfer.
• Fixation (including iron-limited blooms) is a conditional tap.
• Particle injection and remineralization depth are first-order biotic durability controls (Boyd et al., 2019; Kwon et al., 2009).
• Alkalinity can be chemically durable and still hard to verify at climate scale (Renforth & Henderson, 2017; NASEM, 2022).
• Burial is geologic when it happens — and rare relative to respiration (Dunne et al., 2007).
• Blue carbon is real, valuable, and area-limited (Macreadie et al., 2019).

For readers using this page: score a method by which of T1–T12 it actually moves, how those taps fail when the method stops, and whether that failure is visible in an MRV plan. Ocean Flux offers this synthesis for education and research framing — not as commercial OIF and not as a credit prospectus.

References (APA 7th Edition)