Ocean Engineering

Scope, stated first, because this slot sits between four neighbours. The legal and financial machinery of inhabited floating settlement belongs to Floating Cities. Barriers, surge gates and nature-based shoreline defence belong to Coastal Defense Systems. The plant-level engineering of heat engines run on ocean temperature difference belongs to Ocean Thermal Energy Conversion. Treaty machinery for deliberate climate intervention belongs to Geoengineering Governance. What is left here is the engineering of the ocean as an industrial object: deep seabed mining, offshore and very large floating structures, ocean thermal energy conversion as an accounting problem, artificial upwelling as a thermodynamics problem, and ocean alkalinity enhancement as a measurement problem.

Established The organising claim: in every one of these programmes the binding constraint is an instrument, not a machine. Established Seabed mining reuses dynamic positioning, riser systems, subsea power and remote intervention that the offshore hydrocarbon industry has run at scale for decades. Frontier The one project that ever reached a granted environmental permit failed on capital and contracting. Established Ocean alkalinity enhancement has chemistry nobody disputes and a verification problem nobody has solved. Speculative Artificial upwelling has a stoichiometric objection that decides the question before any pump is built.

Established A second organising fact, about this brief rather than about the ocean. The research pack behind it ran under a provenance gate that refused the International Seabed Authority's own domain and every other primary agency source, and refused the open-access publisher where the ocean-alkalinity best-practice guide lives. Established Four peer-reviewed papers were read in full. Most of the rest is a verified bibliographic record with the text unread. Established Where that is so, this brief names the source and does not state its numbers. The gaps are marked in the prose rather than filled, and the marking is the point: a number without a fetched source is not a fact about the ocean, it is a fact about circulation.

Established The International Seabed Authority licenses exploration and has never licensed exploitation. It is the body established under Part XI of the UN Convention on the Law of the Sea to administer mineral resources of the seabed beyond national jurisdiction — the Area — as the common heritage of mankind. Established It has adopted prospecting-and-exploration regulations for all three resource classes: polymetallic nodules, seafloor massive sulphides, and cobalt-rich ferromanganese crusts. Established It has not adopted exploitation regulations. That distinction carries the whole topic, because an exploration contract confers a right to look, and there is no agreed text under which anyone may sell what they find.

Established The contract count drifts across sources, and the drift is documentable rather than mysterious. Miller, Thompson, Johnston and Santillo, writing in Frontiers in Marine Science in 2018, give 27 exploration contracts as of June 2017 — 17 for nodules, 6 for seafloor massive sulphides, 4 for cobalt-rich crusts. Established Komaki and Fluharty, in the same journal in 2020, give 30 contractors, of which 16 hold contract areas in the Clarion–Clipperton Fracture Zone. Frontier Secondary sources across 2024 to 2026 commonly quote 31; the primary register at the Authority's own site was unreachable for this brief and that third figure is therefore reported as circulating, not as verified. Established All three may be correct as of their respective dates. The failure mode is not the number; it is quoting a count without the date it belongs to, and it is endemic in coverage of this subject.

Established Contract areas are large and the ceilings differ by an order of magnitude between resource classes. Miller et al. record a ceiling of up to 75,000 km² per nodule contract and up to 10,000 km² for seafloor massive sulphides, with combined exploration area under contract exceeding 1.4 million km². Frontier The same source gives a cobalt-crust ceiling that this brief does not reproduce. The figure as printed is almost certainly a source-side compression of an underlying regulation that allocates crust contracts as a number of discrete blocks rather than as a single area, and the compressed value is implausible next to the nodule ceiling by three orders of magnitude. Established The regulation itself was unreachable here. The honest statement is that the crust ceiling is block-structured, that the figure in wide circulation is a collapsed form of it, and that this brief prints neither the collapsed figure nor a reconstruction.

Established In the absence of exploitation regulations, contractors set their own environmental thresholds. Christiansen and Bräger's 2023 review records that the Authority permits contractors to identify their own project-specific and area-specific impact thresholds, guided only by non-binding guidance, and that the Authority had recently decided to develop environmental thresholds for permissible levels of toxicity, noise, light and turbidity. Established As of that review, those thresholds did not exist. Frontier Whether numeric thresholds can be set at all is genuinely open, because a threshold is a dose–response statement and abyssal dose–response data are close to absent.

Established Contractor environmental disclosure is close to nil, and this is the sharpest verified fact in the topic. Komaki and Fluharty found that, as of 2020, only two contractors' environmental impact statements had ever been disclosed; no annual reports were public; release of confidential information rested with the Secretary-General without clearly stated criteria; and the Authority's database contained neither the geographic extent of coverage, nor content lists, nor data standards, nor sharing rules. Established It is worth sitting with the shape of that: the environmental case for and against seabed mining is being argued, in public, over a body of contractor data that the public has almost never seen.

Established The resource itself is a surface-lying, slow-growing, polymetallic concretion. Miller et al. give North Pacific nodule composition as manganese 22–27%, nickel 1.2–1.4%, copper 0.9–1.1%, cobalt 0.15–0.25%, iron 5–9%, with typical diameters of 4–10 cm at depths of 3,000–6,000 m. Established Growth proceeds at several millimetres to centimetres per million years. Established That last number is the most important in the subject and it is routinely mentioned and then not used. It means the resource is non-renewable on any timescale a human institution can contemplate, and it means that “recovery” of a nodule field is a category error rather than a slow process. The sediment can recover. The nodules cannot.

Established The footprint arithmetic of a commercial operation is published and is larger than the mined area. Christiansen and Bräger, assuming 10 kg of dry material per m², put commercial extraction at 100–200 km² per year, giving 6,500–14,000 km² directly mined over a 20-year operation and 28,000–56,000 km² affected once sediment plume dispersal is included. Established The benthic impact zone is at least two to four times larger than the mined area itself. Established Any per-tonne impact comparison that uses the mined area alone understates the footprint by a factor of several, and most of them do.

Established Plume behaviour is measured and modelled, and the distances are long. The same review reports deposits thicker than 1 cm more than 5 km from the operation within 10 days; deposits 1 mm thick more than 50 km away after one year; and fine particles at or below 10 µm potentially remaining in the water column for a year and travelling up to 1,000 km. Established The protective threshold proposed in that literature is no blanketing above 1 mm, justified on the ground that 1 mm is roughly a thousand years of natural sedimentation at these sites. Frontier Hold those two facts together: the candidate regulatory limit is exceeded at fifty kilometres within a year of operation, on the review's own figures.

Established The one long-interval measured recovery result is DISCOL, and it says something more specific than either side quotes. Simon-Lledó, Bett, Huvenne, Köser, Schoening, Greinert and Jones resurveyed in 2015 a plough-harrow disturbance made in 1989 in the Peru Basin at roughly 7°06′S, 88°27′W and 4,140–4,160 m depth — an interval of 26 years. Established Plough tracks covered about 20% of the roughly 1,100 ha experimental area; the resurvey imaged 11.1 ha (5.86 ha disturbed, 5.25 ha reference), counting 7,284 individuals across 97 morphospecies. Established Suspension feeders inside plough tracks remained at about 40% of the density found elsewhere. Deposit feeders showed no diminished presence — they had recovered. Established Faunal composition remained markedly distinct across disturbance levels and heterogeneity diversity remained significantly lower in disturbed zones; megafauna densities ran at about 600 individuals per hectare inside tracks against about 800 per hectare in the intermediate zone one to fifty metres away.

Established DISCOL is a lower bound on commercial impact, not a model of it. The plough disturbed the sediment; it did not remove the nodules. Commercial collection removes the substrate across a contiguous hundred to two hundred square kilometres a year. Established Because a substantial fraction of abyssal epifauna is attached to nodule hard substrate, removing the nodules removes the habitat irreversibly — a different and worse impact pathway than sediment blanketing, and the one DISCOL cannot speak to at all. Established Christiansen and Bräger state the consequence directly: mined nodules and their associated fauna regrow only over millions of years, producing permanent benthic community shifts at mined sites.

Established Solwara 1 is the only seafloor mine that ever reached a granted permit, and it is the most instructive datum in the topic. Miller et al. record Nautilus Minerals' project in the Bismarck Sea off Papua New Guinea at 30 km offshore, 1,600 m water depth within a deposit range of 1,500–2,000 m, over an area of 0.112 km², at grades of about 7% copper and 6 g/t gold, with planned throughput of 1.3 Mt/yr and a mine life of 2.5 years. Established An environmental permit was granted in 2009 for a 25-year term and commercial operation was scheduled for early 2019. Frontier It never operated; the company entered insolvency and the host state's equity position was written off. Frontier The specific sums lost were not verifiable for this brief and are not stated. What is verifiable is the shape of the failure: permit granted, technology within reach, and the money did not close.

Frontier The two-year rule is real, and the account of it in general circulation is not verified here. Two peer-reviewed treatments exist, both by Pradeep A. Singh in The International Journal of Marine and Coastal Law, one from 2021 and one from 2022; their bibliographic records were confirmed and their texts were not obtained. Speculative The circulating account — that Nauru, sponsoring a subsidiary of a listed mining company, invoked a provision of the 1994 Part XI Implementation Agreement in June 2021, starting a two-year clock obliging the Council to complete exploitation regulations, failing which it must consider and provisionally approve a plan of work — is reported here as circulating, not as verified. Established The paragraph number in that account is exactly the class of citation that should not be reproduced without reading the instrument, and this brief does not reproduce it. The direction of the correction is safe: the provision creates an obligation to consider, not an authorisation to mine.

Established On ocean thermal energy conversion, artificial upwelling and ocean alkalinity enhancement, this brief states the structure of each problem and none of the numbers. No source on ocean thermal conversion was reachable at all; the artificial-upwelling material failed the same way, and its single load-bearing citation is flagged as a hazard rather than a source. Established The canonical ocean-alkalinity best-practice volume — including the verification chapter by Ho and colleagues and the field-experiments chapter by Cyronak and colleagues — exists, is fully open access, and could not be retrieved through the gate this brief ran behind. Established Their records are verified, their contents unread, and their numbers therefore absent from this page by decision rather than by oversight.

Established The offshore-structure side of ocean engineering is mature and its central problem is hydroelasticity. A pontoon-type very large floating structure is long enough relative to wavelength that it deforms elastically in waves rather than moving as a rigid body, which makes its response a coupled fluid–structure problem; effectively the whole peer-reviewed literature on these structures, three decades of it, is about predicting and reducing that response. Established This matters to seabed mining for one reason: it demonstrates that the technology-readiness argument for ocean industry is stronger than critics usually allow, and therefore that the constraints which actually bind are elsewhere.

Frontier Can an international regulator set a numeric environmental threshold for a habitat whose dose–response curves do not exist? The Authority has decided to try, for toxicity, noise, light and turbidity. There were none as of the 2023 review. A threshold requires an exposure–effect relationship; abyssal fauna have essentially none published; and the resulting question — whether to set a threshold on precautionary grounds and discover its adequacy in operation, or to wait for data that a moratorium would prevent anyone collecting — is a genuine circularity rather than a delay.

Frontier The most consequential recent move in the regulatory debate is a change of object: constrain the machine, not the outcome. Christiansen and Bräger propose best environmental practice as a design constraint set — collection depth limited to under ten centimetres so as not to breach the oxidised–anoxic sediment boundary, separation of minerals from sediment as close to the seabed as possible with induced rapid resedimentation behind the collector, a prohibition on discharge into epipelagic and mesopelagic waters with return flow only below the oxygen minimum zone, avoidance of noise input at SOFAR channel depths of 700 to 1,300 metres, and mining blocks shaped round or approximately square to minimise edge-to-area ratio. Frontier These are specifications, not impact limits, and they are enforceable by inspection of a machine rather than by measurement of an ocean. Speculative That is a materially different regulatory technology and it is under-covered on both sides of the argument.

Frontier In-situ plume measurement has moved from model to field within the last five years. Christiansen and Bräger cite field plume studies by Muñoz-Royo and colleagues in 2021 and 2022 and by Haalboom and colleagues in 2022 as the empirical basis for the dispersal figures. Frontier Their venues and identifiers could not be confirmed here and they appear in the reading list without links for that reason. The plume is now something measured behind a real collector rather than only simulated.

Frontier High-frequency in-situ alkalinity sensing is the instrument the alkalinity-enhancement field was waiting for. Two records were verified: a lab-on-chip total alkalinity sensor presented at OCEANS 2024 in Halifax by Motahari and colleagues, aimed explicitly at long-term monitoring, reporting and verification; and a 2026 Communications Engineering paper by Zabihihesari and colleagues on high-frequency in-situ total alkalinity measurement for field trials. Frontier Neither text was read and no performance figure from either appears here. Speculative The existence of a 2026 engineering paper on this exact measurement is itself the tell — in-situ alkalinity sensing, not chemistry and not deployment, was the rate-limiting instrument. Frontier A 2024 preprint by Milde and colleagues extends the same effort to coastal enhanced weathering, which says that verification is now being designed per delivery route rather than in general.

Frontier Who may sue for damage to the common heritage is now a formally posed question with no answer. Davenport's chapter in the twenty-five-year review of the Authority asks whether the Authority itself can be the claimant for damage to the marine environment of the Area. Speculative If the Area belongs to everyone and harm occurs, the identity of the claimant is undefined, the measure of damages is undefined, and the defendant may be a single-purpose subsidiary of a sponsoring state's registry. Frontier This is a structural hole and it is upstream of the money: a lender cannot price a liability whose claimant is unnamed.

Frontier The payment regime is not agreed, which means project economics cannot be closed. Wilde's evaluation of the payment regime for nodule mining in the Area is a verified record; its content is unread here. The structural consequence stands regardless of what it concludes: a royalty rate or profit-share design that does not yet exist cannot be entered into a project model, and a project model that cannot be built cannot be financed.

Speculative A 2026 preprint frames the Authority's exploitation-phase failure as ongoing and structural rather than as delay. Schwabach's SSRN paper carries that argument in its title. It is a preprint, it is not peer reviewed, and it appears here as framing rather than as evidence. Frontier A related fragility deserves stating: the analytic frame in wide circulation for the post-two-year-rule situation traces substantially to two papers by one scholar. That is not a criticism of the scholarship. It is an observation that a field with one interpreter has no independent check.

Frontier On the structures side, the live ideas are response mitigation and applicable law. Nguyen's doctoral work treats incident wave energy at a floating structure's perimeter as a resource to be harvested rather than a load to be resisted, solving two problems with one device. Frontier Singla and Pu report passive response reduction using permeable and porous perimeter barriers. Frontier Tay applies neural-network surrogates to hydroelastic response, which matters because real-time prediction is a precondition for an operable structure rather than merely a designable one. Speculative And Fialkoff's 2026 treatment of floating nuclear power under the law of the sea is the most useful adjacent item in the field: floating reactors force the legal system to answer what law applies to a large, permanently moored, non-ship structure with people on it — and they have sponsors with the resources to force the answer.

Established The workback chain for commercial seabed mining under an agreed regime has ten links, and the binding one is money. This ordering is deliberate and it inverts the usual presentation, in which technology comes first and finance is a footnote.

Frontier Link 1, and binding: project finance against a single deposit. The asset in seafloor mining is the vessel, not the orebody. A purpose-built production vessel with a riser and a collector is a capital item with a decade-plus economic life; Solwara 1 proposed to amortise one against a 2.5-year mine life at a single site. Established That is not bankable, and it did not get banked. Speculative The fix is structural rather than technical: a multi-deposit pipeline that converts the vessel from a project asset into a fleet asset. No contractor holds one. A regime could be adopted tomorrow and no vessel would be built.

Frontier Links 2 and 3 are the inputs a lender would demand and both are open: an agreed payment regime, and a resolved liability-and-standing rule. Their unresolved state is not incidental to link 1; it is why link 1 cannot clear. A financier prices royalty and prices liability. Neither has a number.

Frontier Link 4, the secondary binding link: measurability. A candidate blanketing threshold of 1 mm is worthless without a monitoring system that can detect 1 mm of accumulation at the ranges the plume reaches — 5 km, 50 km, and out toward the thousand-kilometre transport distance for the finest fraction. Established No such system exists. A regime adopted with an unmeasurable threshold reverts to contractor self-certification, which is precisely the status quo it was meant to replace.

Established Link 5: baselines. Ninety-seven morphospecies in eleven hectares at one Peru Basin site implies survey effort at the hundred-square-kilometre-per-year scale that nobody has funded, and most Clarion–Clipperton species remain undescribed. Established Link 6: disclosure. Two impact statements and no annual reports is not a data foundation for anything. Frontier Link 7: heavy marine vessel availability. Production vessels, cable and riser lay capacity, and heavy-lift construction tonnage are shared with offshore wind, oil and gas decommissioning and subsea cable programmes, all of which are bidding for the same yards and the same crews. Speculative That is a real queue and it is the supply-side constraint recorded in this brief's tech tree, but it binds after finance, not before: nobody is waiting for a berth in a shipyard they have not ordered from.

Frontier Link 8: the demand case, and it is deteriorating. Nodule value rests on nickel and cobalt. Battery chemistry has been migrating toward lithium iron phosphate and sodium-ion, which use neither. Speculative A twenty-to-thirty-year asset whose value rests on a chemistry that may be substituted away inside the permitting timeline has weak option value, and that is a financing fact before it is a geological one.

Established Dose–response data for abyssal fauna. Every numeric environmental threshold the Authority proposes to set is downstream of exposure–effect relationships that do not exist for these communities. This is the single research dependency whose absence blocks the largest number of other links.

Established Taxonomy and baseline density at contract-area scale. The DISCOL resurvey found 97 morphospecies in 11.1 hectares. Extrapolate that sampling intensity to a hundred square kilometres a year of production and the survey burden is the dominant scientific cost of the industry, before anyone argues about whether the surveys are adequate.

Frontier Validated plume models at three scales. The near field behind the collector, the mid field at tens of kilometres, and the far field for the sub-ten-micron fraction are three different physical problems, and the field measurements that would close them — the Muñoz-Royo and Haalboom work — are recent, few, and not read for this brief.

Speculative A stoichiometric measurement that decides artificial upwelling. The question is whether the carbon-to-nitrogen ratio of exported organic matter exceeds the dissolved-inorganic-carbon-to-nitrogen ratio of the water being pumped up by a margin large enough to yield net drawdown. That is one measurable quantity. Until it is measured under field conditions, artificial upwelling is a proposal rather than a technology.

Frontier A counterfactual baseline method for marine carbon removal. Alkalinity added at a point is easy to quantify; carbon removed is inferred from a model of what would have happened otherwise. That baseline is the unsolved link, and it is the same unsolved link across most marine carbon dioxide removal, which makes it a shared dependency with Carbon Capture at Scale rather than a peculiarity of alkalinity.

Frontier A net-power reporting convention for ocean thermal conversion. Not a technology — a convention. The literature's chronic defect is gross output reported without the parasitic pumping load, and the dependency is a published protocol that makes net-to-gross a required field rather than an optional one.

Established 1. A full-scale collector trial with plume instrumentation at 5, 50 and 500 kilometres. The measurement: sediment accumulation depth and suspended load as functions of range and time, over at least one seasonal cycle, with the collector operating at commercial throughput rather than at pre-prototype scale. The decision it closes: whether a 1 mm blanketing limit is a specification or an aspiration.

Frontier 2. A demonstrated monitoring system with 1 mm detection at abyssal depth. The measurement: minimum detectable accumulation, with a stated false-positive rate, from a platform that can be deployed at the ranges in experiment 1. Until this exists, every threshold in the draft regulations is unenforceable by construction.

Established 3. A DISCOL analogue that removes the nodules. The single largest gap in the recovery literature is that the best long-interval experiment disturbed sediment and left the hard substrate in place. The measurement: epifaunal density and composition on cleared versus nodule-bearing seabed at ten, twenty-five and fifty years. Speculative This experiment takes half a century to return its main result, which is an argument for starting it now and against waiting for it before deciding.

Frontier 4. Conservative tracer co-release in an alkalinity field trial. The measurement: dilution and transport of the added alkalinity signal against natural variability, tracked by a co-released conservative tracer, so that the model doing the attribution has an observational anchor. This is the second link in the alkalinity chain and it is the last one that is straightforwardly measurable.

Speculative 5. The upwelling stoichiometry measurement. Co-located measurement of upwelled water dissolved inorganic carbon and nutrient concentrations, and of the elemental ratio of the organic matter actually exported below the winter mixed layer, at one site, over one production season. One number decides the concept.

Established 6. A net-power ocean thermal test reported to a single protocol. Gross output, cold-water pumping load, warm-water pumping load, and auxiliaries, published together, over a duty cycle that includes storm conditions. Frontier The reason this belongs in a required-experiments list rather than an engineering list is that the historical record is contaminated by gross-only reporting, and a clean modern number would be a contribution in itself.

Established The collector is the constrained component, and the constraints are specified. Collection depth under ten centimetres, so that the oxidised sediment layer is not breached and the anoxic material beneath it is not mobilised. Mineral separation as close to the seabed as possible, with induced rapid resedimentation immediately behind the machine. Frontier Whether a collector can lift nodules at commercial throughput without entraining the top decimetre is a tractable mechanical problem in ground pressure and pickup geometry, and it is one of the few places in this subject where the frontier is genuinely engineering.

Established The return-water system is the second constrained component. Discharge into epipelagic and mesopelagic waters prohibited; return flow only into fully oxygenated water below the oxygen minimum zone. Frontier That places a depth requirement on the riser return that interacts directly with pumping power and with vessel station-keeping, and it is a requirement most published system concepts do not meet.

Established Noise is a design variable, not an externality. Avoiding acoustic input at SOFAR channel depths of 700 to 1,300 metres is a specification on the machinery, the riser and the vessel, because sound injected into that duct propagates ocean-basin distances.

Established Block geometry is a specification too. Round or approximately square mining areas minimise edge-to-area ratio and preserve dispersal corridors; elongated blocks maximise the disturbed perimeter for a given tonnage. Speculative A whole-of-a-deposit approach — mining a deposit out rather than high-grading it — minimises areal consumption and is directly contrary to the economics that make marginal projects viable. Making it happen requires a royalty structure that prices area rather than tonnage, which nobody has drafted.

Established Everything upstream of the collector is borrowed and mature. Dynamic positioning, riser systems, subsea power distribution and remote autonomous intervention were solved at scale by offshore hydrocarbons and are being re-solved at higher volume by offshore wind. Frontier The consequence is uncomfortable for both camps: the readiness argument for seabed mining is stronger than critics allow, which is exactly why the binding constraints are legal, financial and ecological rather than technical.

Established Offshore hydrocarbons and offshore wind are the parent industries and the competitors for the same steel. Nodule collection reuses their station-keeping, riser and subsea power technology almost wholesale. It also queues behind them for heavy marine vessels, cable-lay tonnage and yard slots, which is what makes vessel availability a supply-chain constraint rather than a procurement detail.

Established Very large floating structures. Three decades of hydroelasticity research, response-mitigation devices, and machine-learning surrogates give the ocean-structure field a depth that seabed mining does not have. Its own binding constraints are financial and jurisdictional, which is the subject of Floating Cities.

Speculative Floating nuclear power will probably answer the applicable-law question before any floating settlement does. A large permanently moored non-ship structure with people on it and a high-consequence hazard aboard forces regulators to characterise it, and the sponsors have the resources to make them. Whatever regime results will be inherited. See Small Modular Reactors.

Frontier Battery chemistry is the demand-side adjacency and it is moving the wrong way for nodules. Lithium iron phosphate and sodium-ion cathodes use no nickel and no cobalt; see Advanced Battery Technologies. A resource whose case is metals-for-electrification is exposed to a substitution it cannot influence.

Frontier Marine carbon dioxide removal shares its verification problem with the rest of the removal sector, and alkalinity enhancement's counterfactual-baseline link is the same link that binds elsewhere; see Carbon Capture at Scale and Climate Engineering.

Speculative Asteroid mining is the closest structural analogue in the whole corpus — a common-heritage-adjacent resource, an unsettled property regime, a purpose-built capital asset against a single body, and an economics that only closes across a pipeline. See Asteroid Mining and Space Law and Governance.

Frontier One adjacency runs the other way, outward from the ocean. This is the only place where an international body has ever been given the job of licensing extraction from a commons in advance of the extraction. Speculative Every institution that will eventually have to do the same for orbit, for the polar seas or for the deep subsurface has exactly one working precedent to study; see Resource Economies and Arctic Engineering.

Established The Authority is simultaneously the regulator of exploitation, the administrator of the common heritage, and the recipient of the payments exploitation generates. That is not a fringe critique; it is the standing structural observation in the peer-reviewed literature, and it sits alongside the verified disclosure record of two environmental impact statements and no public annual reports. Frontier The institutional-defence position exists too and belongs on the page: Jares's chapter in the twenty-five-year review argues the Authority's cooperation with the United Nations as a success, and Wood's historical chapter in the same volume traces how an institution designed in the 1970s survived the 1994 renegotiation at all.

Frontier Three institutional artefacts would have to be drafted, and none is a scientific result. A payment regime specific enough to underwrite a project model. A liability rule naming a claimant and a measure of damages for harm to the Area. A sponsoring-state due-diligence standard that means something when the sponsor is a small state with limited administrative capacity and a large commercial counterparty. Speculative Harrison's chapter on checks and balances on the Authority's regulatory powers is the natural entry point to the third.

Frontier Reform proposals exist and have not been adopted. Komaki and Fluharty propose independent review committees, defined disclosure criteria, and multinational monitoring collaboration. The specificity is the useful part: these are drafting instructions, not aspirations.

Established One institutional fact about this brief's own instruments belongs here, because it is a fact about how citations get corrupted in machine-assembled scholarship. During compilation, a publisher's article server returned, twice, under a correctly DOI-addressed URL, the full text of an entirely unrelated paper — a study of connective-tissue fibroblasts in mouse digit-tip regeneration — in place of the requested article. Established It rendered confidently, with no error and no low-confidence signal. Established An agent that had asked a narrower question would have attributed mouse-limb content to a marine DOI, or, worse, recorded “the source does not contain these data” as a finding about the source. Established The compiler withheld that paper's effect sizes entirely rather than fill them from memory, and this brief reports the defect rather than the numbers. Speculative The relevance is not incidental: this subject is governed by assessment reports and regulatory annexes assembled from exactly this kind of retrieval, and a silent substitution in that pipeline is indistinguishable from scholarship until someone checks the title.

Established The common heritage principle creates an unusual consent problem: the beneficiary is everyone, and nobody is asked. The Area is administered for humankind as a whole, which means that the constituency with standing to object is coextensive with the species and is represented by a Council of states, most of which have no maritime interest in the matter and some of which sponsor contractors.

Frontier The substitution argument is the strongest ethical case for mining and it is unverified as a matter of fact. Nodules require no blasting, no overburden stripping, no tailings dam, no acid rock drainage, and displace no human community; the terrestrial alternatives are Indonesian laterite with rainforest clearance and coal-fired processing energy, and Congolese cobalt with artisanal mining and child labour. Speculative On land use, freshwater toxicity and human-rights harm per tonne of contained metal, nodules may well win. Established But that is a substitution claim, and nothing in the regime makes production substitutive rather than additive. Historically, new supply has been additive. The ethical argument therefore rests on a market outcome nobody has committed to producing.

Established Irreversibility is the asymmetry that most impact accounting cannot handle. A terrestrial mine site can in principle be rehabilitated. A nodule field regrows at millimetres per million years. Undiscounted footprint metrics treat a permanent loss and a fifty-year loss as commensurable; discounted ones treat the permanent loss as nearly free. Neither is right, and no agreed abyssal ecosystem-service metric exists to fix it.

Frontier Sponsoring-state exposure is a distributive question dressed as an administrative one. Sponsors are frequently small states with limited supervisory capacity, and the potential liability for a failure of due diligence sits with them while the operating expertise sits with a foreign contractor. Papua New Guinea's written-off equity in the one project that reached permitting is the available data point.

Speculative And there is a claim on the other side that deserves to be stated rather than sneered at: that a regime which forbids seabed extraction while consuming the metals produced by the worst terrestrial mines has exported a harm rather than avoided one. The answer is not that the claim is wrong; it is that the substitution it assumes has not been arranged.

Established The Area is the largest thing on Earth that has never been allocated. Roughly half the planet's surface lies beyond national jurisdiction, and the seabed beneath it is the only major resource province governed on an explicitly non-appropriative principle. Whatever is decided there is a precedent for every other commons a technological civilisation reaches.

Speculative The strongest civilizational reading is that this is a rehearsal. The seabed regime is the first serious attempt to govern extraction from a commons before extraction begins, with a benefit-sharing mechanism written in advance. If it fails — if the exploitation regulations arrive unenforceable, or the payment regime arrives after the vessels — that failure is the operative precedent for lunar and asteroid resources, where the same arguments are already being made with the same structure.

Frontier The pattern this brief shares with its neighbours is that the missing artefact is legal or financial, not physical. Seabed mining wants a bankable multi-deposit pipeline under an agreed payment and liability regime. Floating settlement wants a mortgageable interest in a berth. Coastal retreat wants a statutory power to withdraw protection. Speculative These are drafting problems. That is a hopeful finding and an uncomfortable one: the things currently considered impossible in ocean engineering are impossible for reasons that are, in principle, writable, and the writing is nobody's research programme.

Speculative The pessimistic reading is that the ocean gets industrialised in the order of least resistance rather than least harm. Offshore wind, subsea cable, aquaculture and hydrocarbon decommissioning are all expanding, none of them under a common-heritage regime, and cumulative marine impact accrues without a body competent to sum it. Frontier On that reading the seabed argument is a proxy fight over the one activity that has a regulator, while the activities that do not have one expand unexamined.

These horizons track the seabed-mining chain primarily, because it is the only sub-topic with dated institutional milestones; the marine carbon-removal and thermal-conversion horizons are tracked against their binding measurement rather than against a calendar.

• 10 yr: Frontier Exploitation regulations adopted in some form is the modal expectation, and it does not by itself produce a mine. Frontier A full-scale collector trial with instrumented plume measurement at multiple ranges is achievable and likely. Speculative A first commercial nodule operation reaching financial close within the decade requires either a state balance sheet or a multi-deposit portfolio that does not currently exist. Frontier On alkalinity enhancement, expect field trials with tracer-anchored verification and continued absence of an agreed counterfactual baseline.
• 25 yr: Speculative If mining happens at all, this is the horizon on which the first twenty-year operation would be part-way through and the 28,000–56,000 km² plume-affected footprint would be a measured quantity rather than a projection. Speculative A registry-grade marine carbon removal methodology is plausible here and would be the first time a modelled inference of this kind carried commercial value at scale. Handwave A commercial ocean thermal plant competitive on cost per net kilowatt against offshore wind plus storage is possible and has no current pathway.
• 50 yr: Speculative A second-generation seabed regime, revised in the light of measured impact from first-generation operations, is the realistic institutional horizon; the first regime will not survive contact with data. Speculative The DISCOL-analogue substrate-removal experiment proposed in section 6, if started soon, returns its main result in this window. Handwave Artificial upwelling at climatically meaningful scale remains improbable and would require the stoichiometric objection to have been answered decisively in its favour.
• 100 / 250+ yr: Handwave On this horizon the interesting question is not whether the abyss was mined but whether the areas mined in the twenty-first century are still distinguishable. Established On the growth-rate evidence, they will be: nodule fields removed now will not have regrown by any margin at 250 years, and the mined blocks will be legible as a stratigraphic and ecological boundary for longer than any institution now existing has lasted. Handwave Whether that is recorded as a mistake or as a cost depends on a substitution that has not yet been arranged.

• Depends on Depends on the measurement and instrument work that would make an environmental threshold enforceable: validated near-, mid- and far-field plume models; a monitoring capability with millimetre-scale detection at abyssal depth; abyssal dose-response data that do not exist; and, for the marine carbon-removal branch, high-frequency in-situ alkalinity sensing and a counterfactual baseline method. None of these is a bottleneck this brief expects to bind first.
• Requires (not on this map) Three non-brief constraints are recorded here and the ordering between them matters more than the list. Settled jurisdiction and a liability-and-benefit-sharing regime are the two institutional artefacts: an exploitation text with enforceable environmental standards, and a rule naming who may claim for damage to the common heritage and on what royalty or profit-share basis the proceeds are divided. Heavy marine vessel availability is the supply-chain constraint, and it is real — production vessels, riser and cable-lay tonnage and heavy-lift capacity queue behind offshore wind and decommissioning for the same yards. But this brief's evidence puts a fourth constraint ahead of all three, and it is not in the vocabulary: bankable project finance against a single deposit. The only seafloor project ever to hold a granted environmental permit failed there, with a purpose-built vessel amortised against a 2.5-year mine life. The two institutional tokens are best read as the inputs a lender would price rather than as ends in themselves, and the vessel token binds only after somebody has ordered a vessel.
• Enables Enables, if it resolves, a template for extraction from a commons under a benefit-sharing regime that is directly transferable to lunar and asteroid resources, and a per-tonne impact accounting for battery metals that the terrestrial supply chain currently lacks. Enables nothing at all if the regime is adopted without a measurement capability, because an unenforceable threshold returns the field to contractor self-certification.
• Adjacent Adjacent to offshore wind and hydrocarbon decommissioning, which supply the station-keeping, riser and subsea-power technology and compete for the same vessels; to very large floating structure engineering and floating nuclear, which are working the applicable-law question this field will inherit; to battery chemistry, which sets the demand case; and to marine carbon dioxide removal, which shares the verification problem exactly.

Handwave “The Clarion–Clipperton Zone holds more nickel than all land reserves.” Resource tonnages of this kind circulate without a grade cutoff, without a recovery factor, and without a recovery cost. Established A resource statement is not a reserve statement; converting one into the other requires a demonstrated economic extraction method, and there is none at commercial scale. Established No nodule has ever been commercially recovered at a profit. Any tonnage quoted should be paired with that sentence.

Speculative “Mining the seabed avoids the environmental harm of terrestrial mining.” This is a substitution claim wearing the clothes of a comparison. Established Nothing in the regime obliges or incentivises the retirement of a terrestrial mine, and new supply has historically been additive. Frontier State the claim conditionally — if nodule production displaced laterite and artisanal cobalt, then the land-use and human-rights comparison would plausibly favour it — or do not state it.

Established “DISCOL showed no recovery after 26 years.” False as stated, and it is the critics' error. Deposit feeders showed no diminished presence. They recovered. Established The accurate statement is partial, taxon-specific recovery with persistent community-composition change.

Established “DISCOL showed the seabed recovers in a few decades.” Equally false, from the same paper, and it is the enthusiasts' error. Established Composition remained markedly distinct, heterogeneity diversity remained significantly lower, and suspension feeders sat at about 40% of reference density. Established More decisively: DISCOL ploughed the sediment and did not remove the nodules, so it carries no information whatever about hard-substrate-obligate fauna, which is the impact that defines commercial mining. Treat it as a lower bound and say so.

Frontier “The two-year rule made deep-sea mining legal in July 2023.” The provision, on every reading available to this brief, obliges the Council to consider an application, not to grant one, and the Council retained discretion to decline. Speculative “Became legal” is a category error about what the provision does. Established The paragraph number in circulation is not reproduced here; the two peer-reviewed treatments that would settle it are named in the reading list and their texts were not obtained.

Established “The mined area is the impacted area.” The benthic impact zone is two to four times the mined area, and for a twenty-year operation the plume-affected area runs 28,000–56,000 km² against 6,500–14,000 km² mined. Established Comparisons built on mined area alone are wrong by a factor of several, in a consistent direction.

Established “The Authority is a mining regulator like any other.” It is the regulator, the administrator of the common heritage, and the recipient of exploitation payments at once, and its verified contractor-disclosure record stands at two environmental impact statements, ever. Frontier The institutional critique is mainstream peer-reviewed literature, not activism — which does not make it correct, but does mean it cannot be dismissed as advocacy.

Established “There are 31 exploration contracts.” There may well be. Established What is verified here is 27 as of June 2017 from Miller and colleagues, and 30 in 2020 from Komaki and Fluharty. Frontier The figure of 31 circulates in secondary sources without a date attached and the primary register was unreachable for this brief. An undated contract count is not a fact; it is a citation with the timestamp removed.

Frontier “A cobalt-crust contract is capped at 20 square kilometres.” A figure of this shape appears in an otherwise reliable fetched source and this brief does not reproduce it. Established A ceiling three orders of magnitude below the nodule ceiling is implausible on its face, and the underlying regulation allocates crust contracts as discrete blocks rather than as a single area — so the circulating figure is very probably the block size with the block count dropped. Established The regulation was unreachable here, so this brief names the compression and prints neither the compressed figure nor a reconstruction. That is what a flagged number looks like when the flag is honoured.

Speculative “OTEC is proven; it just needs scaling.” Historical demonstrations have frequently been reported at gross output. The relevant figure is net, and the parasitic load of pumping large volumes of cold water up from around a kilometre is a thermodynamic consequence of pipe diameter and length rather than an engineering defect to be optimised away. Established This brief could not verify a single ocean thermal output figure and therefore prints none — not the historical demonstrations, not the modern grid-connected plants, not the efficiencies. Frontier What survives is the rule: never reproduce an ocean thermal output number without establishing whether it is gross or net. Speculative And the comparator has moved. The case was strongest when the island alternative was diesel; against contemporary solar-plus-storage it is a cost-per-net-kilowatt question, not a feasibility question.

Speculative “Artificial upwelling cools the surface and sequesters carbon.” The objection generally made against it is that deep water is not merely nutrient-rich but dissolved-inorganic-carbon-rich, so pumping it up brings carbon dioxide up with the nutrients, and the net effect on air–sea flux is the difference between two large numbers of opposite sign. Speculative Separately, the surface cooling is a redistribution of heat within the ocean rather than a removal of it, and stopping the pumps produces a rapid rebound. Frontier The single paper carrying most of this argument in secondary accounts could not be verified here — title, identifier or findings — and it is flagged as a citation hazard rather than cited as a source. Established That is not pedantry: this research cluster produced two confirmed cases of a well-known paper being cited for something it does not say, and at that base rate an unread source should not be leaned on.

Frontier “Ocean alkalinity enhancement is measurable because alkalinity is measurable.” Alkalinity addition is measurable, at the source, easily. Speculative Carbon removal is not directly measurable at field scale: it occurs as the surface ocean re-equilibrates with the atmosphere over a period governed by mixed-layer depth and gas transfer, by which time the added signal has diluted against a naturally variable background. Frontier The measured quantity and the credited quantity are different quantities, and the credited one is a modelled inference. Established The canonical treatment of this — the monitoring, reporting and verification chapter of the ocean alkalinity best-practices guide — is open access, was verified to exist, and could not be retrieved through this brief's provenance gate, so no precision, timescale or detectability figure from it appears here.

Speculative “Deep-sea mining will be decided by science.” The evidence assembled here says otherwise. Frontier It will be decided by the payment regime, by who may sue for damage, by battery chemistry, and by the size of the moratorium coalition. Established Science sets the constraint set; not one of the binding links in this brief's workback chain is a scientific measurement. Speculative That generalises: where a technology is mature and a market is absent, reading the disagreement as a scientific one predicts the wrong outcome.

Speculative “Solwara 1 failed because the technology was not ready.” The permit was granted in 2009 for a twenty-five year term and operation was scheduled for early 2019. Frontier What did not close was the capital: a purpose-built production vessel against a 2.5-year mine life at a single site, with an under-capitalised sponsor and an unfinanced host state. Speculative The generous reading is that Nautilus was idiosyncratic and a major with a balance sheet could have carried it. Frontier The structural reading, which this brief prefers, is that the asset in seafloor mining is the vessel and not the deposit, that vessels are financeable only against a portfolio, and that no contractor holds one. Avoiding a repeat means inventing the pipeline first.