The engine reads the world as a typed graph; this is the public, sourced data that sits underneath it. Energy intensity per production chain, the value ladder from raw to processed, and per-chain techno-economic benchmarks — each figure carries a fidelity tier and a named public source. These are order-of-magnitude reference inputs, refined per engagement. Auditable, not bankable — reference figures, not claims.
A library of ~20 production chains and a comparator set of 11+ nations, drawn entirely from public sources — IEA, IRENA, World Bank, FAO, BNEF, published techno-economic studies. Every figure is a range, tier-tagged and sourced. Nothing on this page is a client deliverable: no project IRR, no bankable return (precise enough for a lender to act on), no value-side pools. It is the falsifiability surface (every claim here can be checked, and proven wrong if it is wrong) — the "show your work" companion to the engine.
These numbers are order-of-magnitude screening inputs, drawn from public data and refined per engagement. They are deliberately stated as ranges — the world varies by technology vintage, scale, and location, and a single point estimate would be dishonest. Two markers travel with every figure: a fidelity tier stating how trustworthy the provenance is, and the public source it came from. A composed result is only ever as trustworthy as its weakest input.
Order-of-magnitude. Indicative figures for triage and first contact — enough to rank options and classify a chain, not to bank a project. Most figures on this page.
Sourced and project-grade — backed by published techno-economic studies and operator benchmarks. The feasibility-grade readout that survives a board, IC, cabinet, or coalition partner.
Every figure is a band. Energy intensity, price, and capex all vary with technology, scale, and vintage. The range is the honest unit — a point estimate would overstate confidence.
Reference inputs are stated as intensities and multiples (per tonne, per unit, raw→processed ratio), not jurisdiction-specific absolute returns. Those are engagement work, gated above this page.
Bankable (T3) figures exist — the engine runs through to them — but they are shared only under NDA, once calibrated for a specific deal. The gating is on publication, not on the engine's reach.
Each production chain ties a piece of engineering physics — the electricity needed per unit of output — to its dispatch flexibility (how easily it can shift power use to match available surplus). High intensity plus high flexibility is the strongest abundance-coordination signal: a load that can soak curtailed (surplus power that would otherwise be thrown away) midday surplus rather than draw firm power at the evening peak. The figures below are the process-energy basis. They are screening-grade and span the technology range; calibrate against a specific site before sizing.
| Chain | Process energy (per unit) | Holding / other | Dispatch flex | Public source |
|---|---|---|---|---|
| SWRO desalination | 3.0–3.5 kWh/m³ T2 | — | Very high — solar-direct, tank-buffered | IRENA 2024; Elemental Water Makers; MDPI Water 2024 (modern SWRO + energy recovery) |
| Cold chain — freeze + hold | 70–130 kWh/t freeze T1 | 4–8 kWh/t/day cold store | High — blast-freeze cycle time-shifts to solar peak | ScienceDirect fish-refrigeration analysis; FAO fisheries energy use; NRCan benchmarking |
| Ice plant (fishing fleets) | 90–130 kWh/t (≈0.25–0.40 kWh/kg) T1 | — | Very high — ice is a thermal battery | NIST automatic ice-maker study; FAO fisheries energy use; E3S Conferences 2021 |
| Fisheries processing | 380 kWh/t (150–200 loin-only) T1 | 8 m³/t water | Partial — refrigeration flexible; lines less so | ScienceDirect fish-refrigeration (factory avg.); ACIAR 2023 Pacific tuna landscape |
| Copra / coconut & agri processing | 40–55 kWh/t hot-press oil; 600 kWh/t cold-press VCO T1 | 150–200 kWh/t drying | Partial — drying step an excellent solar sink | Anderson International Copra Processing 2022; SPC Pacific Copra; Commonwealth coconut value chain |
| Agri value-add (drying / flour) | 150–300 kWh/t processing T1 | 80–200 kWh/t solar-assisted drying | Partial — drying ideal solar-thermal load | SPC Pafpnet Breadfruit Market Study 2018; FAO Family Farming; SARE EW17-004 |
| Aquaculture (RAS / pond / cage) | 2–8 kWh/kg pond/cage; 7–29 kWh/kg RAS T1 | ~500 m³/t water (recirc.) | Partial — aeration pumps load-follow solar | Landbasedaq 2024 RAS review; mispeces.com; MDPI Energies 2022; Aquahoy (7 kWh/kg salmon) |
| Bottled water export | 3.5–5.0 kWh/m³ total T1 | 8–12 kWh/1,000 bottles line | Partial — RO step flexible; bottling less so | iBottling 2026 setup cost; MDPI Sustainability 2021 (bottling machines); Elemental Water Makers |
| Electric terraforming (bore / soil) | 500–2,000 kWh/m bore; 50–200 kWh/m³ earthworks T1 | — | Very high — pure deferrable surplus absorber | Engineering bore-drilling standards; BJG Inari entropy-arbitrage thesis |
| Mineral / metal refining | 400–500 kWh/t steel (EAF); 2,000–3,000 kWh/t Cu (EW); 6,000–8,000 kWh/t Al T1 | — | Partial–high — EAF batch; electrowinning steppable | Wikipedia / Stanford / Sanrui EAF (350–520 kWh/t); IEA Critical Minerals 2024 |
| Green hydrogen / P2X | 52–55 kWh/kg H₂ (PEM / alkaline, system-level) T2 | ~22 L/kg water | Very high — designed for intermittency | DOE Hydrogen Program Record 24005 (52.5 kWh/kg); IEA Global Hydrogen Review 2025; BNEF 2024 |
| Ammonia / fertiliser (green H₂) | 10,000–12,000 kWh/t NH₃ (electrolysis + Haber-Bosch) T1 | ~2 m³/t water | Partial — electrolysis flexible; synthesis steady | Oxford Energy ET40 2024 (10–12 MWh/t); IRENA green-ammonia pathway; ScienceDirect 2024 |
| Green steel (DRI-EAF via H₂) | 2,700–3,500 kWh/t steel (vs 5,000+ BF-BOF) T1 | 55 kg H₂/t · 1.4 t ore/t | Partial — EAF batch-flexible; DRI steadier | SSAB / HYBRIT DRI-EAF benchmark 2024; IEA Iron & Steel Roadmap 2024 |
| Edge data centre | 8,760 MWh/yr per MW IT load (PUE 1.5–2.0) T1 | 1–3 L/kWh cooling | Load-following — batch/AI 20–30% shiftable | McKinsey 2024 cost of compute; LBNL 2024 US Data Center Energy Usage; Thunder Said Energy |
| Freeze-drying (horticulture) | 30,000–50,000 kWh/t FD product T1 | — (input moisture removed) | High — sublimation fully schedulable | IEA / IRENA project cost databases; ISHS horticultural energy studies; USDA grain benchmarks |
| Ecotourism / resort load | 40–80 kWh/guest-night (20–35 eco-lodge) T1 | ~300 L/guest-night water | Partial — desal, laundry, pool flexible | IRENA 2014 RE Opportunities for Islands Tourism; Huawei Solar Palawan case study |
| 100% RE + battery (foundation) | N/A — foundation generator T2 | 2.0 ha/MW land | Dispatchable — battery soaks midday surplus | IRENA 2024 Renewable Power Generation Costs (solar 1,200 USD/kW Pacific); BNEF 2025 BESS 210 USD/kWh |
Reading rule: the chains with the strongest abundance-coordination signal sit at high intensity and high dispatch flexibility — desalination, ice, freeze-drying, electrolysis, EAF refining. They turn curtailed midday solar into stored value rather than spilling it.
The value ladder shows the price per unit at each stage of a chain. The gap between raw input and the downstream stage is the value-capture opportunity available to a jurisdiction that moves down the chain — and where electricity sensitivity peaks, cheap surplus is the lever that makes the move viable. Prices are public, ranged, and vintage-flagged; several chains (lithium, rare earths) swing more than 50% across a cycle, so the multiple matters more than any single price.
The fish chain is the clearest island case: the gap from whole fresh fish to sashimi-grade is 15–30×, and it is almost entirely a cold-chain and energy-reliability problem — not a resource one. The full chain table follows.
| Chain | Raw ($/unit) | Processed ($/unit) | Multiple | Public source |
|---|---|---|---|---|
| Fish → sashimi-grade | $0.8–2.0k/t whole T1 | $15–40k/t sashimi · $3–6k/t fillet | 15–30× | FAO; ITC Trade Map 2024; Tsukiji/Toyosu auction data |
| Tuna (onshore processing) | NZD 4–8/kg raw landed T1 | NZD 15–25/kg processed | 3–4× | ACIAR 2023 Pacific tuna landscape; IRENA Pacific Lighthouses; MSC WCPO socio-economics |
| Vanilla (green → cured) | green bean T1 | cured / graded | 3–5× | SPC Pacific agri value-chain studies; FAO crop value-add benchmarks |
| Copra → coconut oil / VCO | copra (dried meat) T1 | $1,100/t CNO · $2,200/t VCO | ~2× VCO premium | IndexBox Asia-Pacific Coconut Oil 2024; Commonwealth coconut value chain; SPC |
| Raw milk → WPC80 | $300–500/t milk solids T1 | $4,000–8,000/t WPC80 · $8–15k/t WPI | 12–25× | GlobalDairyTrade; NZ Dairy Companies; FAO |
| Field veg → freeze-dried | $200–350/t fresh T1 | $15,000–80,000/t FD | 50–120× | ITC Trade Map 2024; ISHS; published FD facility benchmarks |
| Raw seaweed → carrageenan | $50–200/t wet T1 | $4,000–12,000/t extract | 25–60× | FAO Aquaculture 2024; published hydrocolloid market data |
| Cu ore → cathode | $5–15/t ore T1 | $9,000–10,000/t cathode | 600–2,000× | LME spot; ICIS / Fastmarkets; ITC Trade Map 2024 |
| Bauxite → Al ingot (green) | $30–55/t bauxite T1 | $2,200–3,100/t ingot (+$200–400/t green premium) | 50–80× | LME spot; Norsk Hydro green-premium data; IEA |
| NdPr oxide → NdFeB preform | $70–120/kg oxide T1 | $160–280/kg preform | 2.5–4× | Penney et al. 2026 (NdFeB techno-economic); ICIS / Fastmarkets |
| Spodumene → LiOH | $800–1,500/t concentrate T1 | $12,000–22,000/t LiOH | 10–18× | BNEF Battery Metals Outlook 2024; ICIS / Fastmarkets (2024 trough) |
| Curtailed MWh → AI compute | $0–30/MWh spot T1 | $200–800/MWh AI training | 10–30× | BNEF; published hyperscaler compute economics · (fibre prerequisite) |
| Green H₂ → green ammonia | $3–8/kg H₂ T1 | $500–1,200/t NH₃ | ~2× (energy basis) | BNEF H2 Scorecard 2024; IRENA Green Hydrogen Cost Review 2024 |
Where electricity is a small share of opex (food, fish), the multiple is large and the binding constraint is cold chain and market access, not energy cost. Where electricity dominates opex (hydrogen ~80%, polysilicon ~60%), the price of surplus is the lever. Public price sources: LME spot; ICIS / Fastmarkets; BNEF; FAO; GlobalDairyTrade; ITC Trade Map 2024.
For each chain the framework carries an order-of-magnitude capital cost, a capex-per-unit-of-capacity figure, and a viability threshold — the electricity price at or below which the chain becomes broadly cost-competitive against the fossil-powered global incumbent. These are greenfield screening estimates; actual costs vary by location, scale, and technology vintage, so apply a ±25% location band before any sizing. Capex order-of-magnitude is shown for relative comparison, not as a project budget.
| Chain | Capex OOM | Capex per unit capacity | Viability ≤ $/MWh | Public source |
|---|---|---|---|---|
| SWRO desalination | ~$2M T2 | $1,500–5,000/ML·day | <$80 | MDPI Water 2024 RO cost studies; IRENA 2024; assumptions.json |
| Cold chain / logistics hub | ~$2.5M T1 | $800/t capacity | Any (timing) | ScienceDirect fish-refrigeration; FAO; published cold-store benchmarks |
| Ice plant (fishing fleets) | ~$0.5M T1 | $150k/(t·day) ice | Any (timing) | NIST ice-maker study; FAO fisheries energy use |
| Fisheries processing | ~$1.2M T1 | $0.08M/(t·day) · ~$10M for 10kt/yr | <$80 | IFC Solomon Islands tuna loan; World Bank Sustainable Fisheries; ACIAR 2023 |
| Copra / agri processing | ~$0.8M T1 | $400/(t·yr) | <$100 | Anderson International 2022; SPC; IndexBox 2024 |
| Agri value-add (drying / flour) | ~$0.3M T1 | $200/(t·yr) | <$100 | SPC Pafpnet 2018; FAO; SARE EW17-004 |
| Aquaculture (RAS / pond) | ~$1.5M T1 | $80k/(t·yr) | <$60 | Landbasedaq 2024; MDPI Energies 2022; FAO Aquaculture 2024 |
| Bottled water export | ~$2M (1.8–2.4) T1 | $350k/(ML·yr) | Brand-led | iBottling 2026; MDPI Sustainability 2021; Fiji Water precedent |
| Freeze-drying (horticulture) | $0.3M+ T1 | $8,000–20,000/(t·yr) FD | <$50 | IEA / IRENA cost databases; ISHS; published FD facility data |
| Copper electrorefining | ~$20M T1 | $1,000–2,000/(t Cu·yr) | <$60 | IEA Critical Minerals 2024; Nyrstar / Rio Tinto energy benchmarks |
| Mineral / metal refining (EAF) | ~$20M T1 | $1.5M/MW | <$25–60 | Stanford US steel energy use; IEA Iron & Steel 2024; S2S library CM-3 |
| NdFeB alloy / preforms | — T1 | $5,000–10,000/(t·yr) | <$45 | Penney et al. 2026; IEA Critical Minerals 2024 |
| Green hydrogen / P2X | ~$15M T1 | $2.0M/MW electrolyser (2,000–2,600 $/kW) | <$35 | IEA Global Hydrogen Review 2025; BNEF 2024; DOE Hydrogen Shot |
| Ammonia / fertiliser (green H₂) | ~$20M T1 | $1,500/(t·yr) | <$30 | Oxford Energy ET40 2024; IRENA green-ammonia pathway |
| Green steel (DRI-EAF) | ~$500M T1 | $800/(t·yr) | <$25 | SSAB / HYBRIT 2024; IEA Iron & Steel Roadmap 2024 |
| Edge data centre (AI / HPC) | ~$8M T1 | $9–15M/MW IT | <$70 | McKinsey 2024 cost of compute; LBNL 2024; Thunder Said Energy |
| 100% RE + battery (foundation) | ~$9M T2 | $1.2M/MW · battery 210 $/kWh | Foundation | IRENA 2024 Renewable Power Generation Costs; BNEF 2025 Storage Outlook |
Viability threshold reads as: at or below this electricity price, the chain is broadly cost-competitive without subsidy. Foundation chains (RE + battery) and timing-driven chains (cold storage, ice) are not threshold-gated — their value is in dispatch, not in beating a fossil price. Capex sources: IEA / IRENA project cost databases; BNEF LCOE 2024; published project announcements.
A jurisdiction is read against a public archetype set — eleven reference nations plus sub-national profiles — scored on the same dimensions: renewable penetration, industrial electricity tariff, curtailment, manufacturing share, and consenting velocity (how fast permits and approvals move). One typed structure is what makes an island, an emerging leapfrogger (a fast-developing nation skipping older energy stages), and a continental grid comparable. The figures are public-agency benchmarks; the table is reference context, not a ranking or a recommendation.
| Nation | Renewable % | Industrial tariff $/MWh | Curtailment | Manuf. % GDP | Consent (yr) |
|---|---|---|---|---|---|
| New Zealand | ~85% T2 | 70–110 | 0.5–1.5 TWh | ~11% | 7–12 |
| Iceland | ~99% T2 | 25–45 | ~0 | ~18% | 2–4 |
| Chile | ~57% T2 | 55–80 | 8–12% solar | ~15% | 4–7 |
| UAE | ~17% T2 | 20–40 | n/a | ~9% | 1–3 |
| Morocco | ~42% T2 | 70–100 | n/a | ~17% | 3–6 |
| Norway | ~96% T2 | 35–60 | <1% | ~9% | 3–6 |
| Germany | ~60% T2 | 90–140 | 10–15 TWh | ~21% | 4–8 |
| South Australia | ~73% T2 | 80–120 | 15–25% intervals neg. | ~7% | 2–4 |
| South Korea | ~10% T2 | 60–90 | growing | ~27% | 3–6 |
| Namibia | ~38% T2 | 75–110 | n/a | ~13% | 3–7 |
| Kenya | ~91% T2 | 90–130 | n/a | ~9% | 4–8 |
Sources: IEA World Energy Outlook 2024; IRENA Renewable Capacity Statistics 2024; World Bank WDI 2024; national energy-agency publications (MBIE, Orkustofnun, CNE, NVE, Bundesnetzagentur, AEMO, KEPCO, NamPower, KenGen). Curtailment for several jurisdictions is not formally published and is shown as the agency estimate or "n/a". Tariff bands are spot-weighted; large industrial users negotiate below.
This is the public reference layer. A scoping conversation is the right first step toward a calibrated read — an hour to understand the question, then a screening classification against this data, or an honest signal that the framework isn't a fit yet. We will not ship a generic capability deck.
A scoping call is free. Published figures are public-sourced, screening-to-feasibility grade — never investment, financial, or legal advice.