Treatment-caused uptake
Treatment-caused air–sea CO₂ uptake is estimated relative to an evolving untreated reference.
Carbon accounting
Ocean iron fertilisation: Carbon pathways from input to atmospheric outcome.
Five breakpoints separate material delivery, biological response, carbon partition, export and durable attribution. Evidence for a prior step never establishes the next.
Interactive evidence chain
Carbon-fate breakpoints.
Evidence limits and required measurements change at every transition from surface response to atmospheric accounting.
Breakpoint 01
Input
Material identity, mass, form, bioavailability, delivery and tracer.
Evidence boundary: A known input does not establish a biological response.
Measurement family
Intervention mass balance; dissolved/particulate iron; tracer recovery; spatial distribution.
Open measurement explorerBreakpoint 02
Biological response
Phytoplankton community, primary production, grazing, aggregation and ecological change.
Evidence boundary: A bloom does not establish additional atmospheric CO₂ uptake.
Measurement family
Chlorophyll; taxonomy; productivity; nutrients; grazing; toxins; oxygen; pH; N₂O, CH₄ and DMS.
Open measurement explorerBreakpoint 03
Uptake & partition
Treatment-caused air–sea CO₂ exchange and partition among dissolved, particulate and organic pools.
Evidence boundary: Surface drawdown does not establish export below the mixed layer.
Measurement family
pCO₂; DIC; alkalinity; air–sea flux; DOC/POC; matched counterfactual; gas transfer uncertainty.
Open measurement explorerBreakpoint 04
Export & transformation
Particle sinking, dissolved transport, aggregation, grazing and remineralisation through depth.
Evidence boundary: Export across one horizon does not establish durable atmospheric isolation.
Measurement family
²³⁴Th; traps; optical particles; sinking velocity; respiration; deep carbon; water-mass tracking.
Open measurement explorerBreakpoint 05
Attribution & durability
Counterfactual, re-equilibration, circulation, re-emission, nutrient displacement, non-CO₂ effects and life-cycle emissions.
Evidence boundary: A modelled isolation time does not establish verified net removal.
Measurement family
Reference evolution; inverse/forward models; winter mixing; downstream corridor; uncertainty, reproducibility and external validation.
Open measurement explorerSource: National Academies ocean-based CDR assessment (2022) ↗ and Buesseler et al. (2024) ↗. Conceptual accounting and measurement architecture. These sources do not contain a project-level calculation.
Depth is not a duration
Sinking crosses physical layers, not an automatic permanence threshold.
Remineralisation depth, water-mass circulation, ventilation and air–sea contact together affect how long carbon remains outside the atmosphere.
0 m · surface exchangemixed layer / seasonal boundary1,000 m · example observation horizondeep ocean · circulation continues
Surface: atmospheric uptake depends on treatment-caused disequilibrium and gas exchange.
Upper ocean: grazing and respiration can recycle carbon before it crosses a depth horizon.
At depth: material may remineralise and follow water-mass pathways.
Over time: circulation and ventilation govern return; depth alone is not a verified duration.
Source: conceptual depth section informed by ocean carbon-fate literature. Depth, circulation and ventilation relationships vary with location, season, particles and circulation.
Full-system accounting
Each gain has a counterfactual and subtraction.
Carbon fate & return
Shallow recycling, export depth, remineralisation, circulation, leakage and future re-emission determine carbon fate.
System effects
Nutrient displacement, non-CO₂ gases, material and vessel life-cycle emissions, uncertainty and reversals affect system accounting.
Carbon-accounting distinction: Chlorophyll, primary production, surface drawdown and particle export are distinct observations; none alone attributes tonnes of net atmospheric removal.
Regional controls
The same chain fails differently in each ocean.
Open questions
Equatorial Pacific
The IronEx II patch travelled roughly 1,500 km in 19 days. A coordinate is not a stable experimental boundary.
Open regional pathway →Open questionsSubarctic North Pacific
Western and northeast subarctic experiments cannot be collapsed into one regional response; their communities, seasons and physical settings differ.
Open regional pathway →Open questionsSouthern Ocean
Deep particle observations do not alone quantify additional atmospheric CO₂ uptake, later re-emission, nutrient displacement or life-cycle emissions.
Open regional pathway →