Making Things That Add

Volume IV — Production and Regeneration

Nine movements, one square metre.

THE PLATE

A woman in a dark suit standing at the edge of a still green lake, seen from behind, mist on the forest beyond. — Plate IV.01The Declaration.A product and its declaration are two different objects, and the whole of this volume is about the distance between them.

THE LETTER

You are holding, or you are about to specify, or you are about to make, something that somebody has described as better for the world than the thing it replaces. You would like to know whether that is true, and you have discovered that finding out is harder than it should be — not because the information is hidden, but because it is there, in full, in a document you are entitled to read, and it does not say what the front of the brochure says.

That gap is the subject of this chapter, and the first thing worth saying about it is that it is very rarely a lie. In almost every case we will examine, the manufacturer published the numbers that undo their own headline, in a third-party verified document, on their own website, to an international standard, and nobody read past the first line. The declaration is honest. The summary of the declaration is where the trouble is, and the summary is what travels.

So this chapter does one thing. It defines net-positive manufacture precisely enough that an auditor could test it, and then it tests one product all the way through — a square metre of commercial carpet tile, chosen because its maker is the most rigorous in its industry and publishes everything.

The definition has four parts: the boundary, the baseline, the counterfactual, and the unit. You will find that changing any one of the four changes the sign of the answer, and that the physics never moves at all. That is not a scandal. It is the arithmetic of a system boundary, and once you can see it you can specify against it, contract against it, and price it.

This volume asks how making things adds rather than subtracts. This chapter sets the register for the ten that follow by insisting that adds and subtracts are only meaningful once somebody says, on the record, adds to what, measured against what, instead of what, and per what.

— The Editors

DISCOVERY

What is already working

Three things are working well enough to build on, and all three are unusual in that they were built by people who made their own claims harder to make.

Interface, and twenty-five years of a public target. In 1994 Ray Anderson set a carpet manufacturer the goal of taking nothing from the earth that could not be replaced, and called it Mission Zero. The company reported its closing position in 2019 and the figures are specific: a 69 percent reduction in the carbon footprint of its carpet tile products, bringing the average tile to 5.2 kg CO₂e/m²; a 96 percent reduction in greenhouse gas emissions globally; 89 percent renewable energy across its factories; 46 percent less energy per unit of production; 89 percent less water per unit; and 92 percent less waste to landfill. The waste programme alone is credited with avoiding on the order of $400 million of cost over its life — money that had previously left the building as offcuts.

Run the first of those backwards and you get the scale of the movement. A 69 percent cut ending at 5.2 kg CO₂e/m² implies a starting point of 16.77 kg CO₂e/m². The company did not merely improve a product. It removed roughly four-fifths of it.

Then it did something rarer. In 2020 Interface declared a carpet tile with a negative cradle-to-gate carbon figure, achieved by loading the backing with bio-based and recycled content that stores more carbon than the manufacture releases — a real, measured removal sitting inside a real, purchasable object. And in 2024 it ended its Carbon Neutral Floors programme, retiring the offset-based neutrality claim it had run since January 2019 and redirecting that budget into measured reduction. A company withdrew its own strongest marketing claim because the claim rested on a counterfactual and the reductions did not. Hold that; we return to it in the Arithmetic.

Ørsted, and a conversion with a date on it. The Danish utility formerly called DONG Energy reports that the emissions intensity of its generation and operations fell from a 2006 baseline of about 500 g CO₂e/kWh to a stated target of 10 g CO₂e/kWh — a 98 percent reduction — with 99 percent of its generated energy renewable, its last coal-fired station closed in 2024, and a further target of 1 g CO₂e/kWh by 2040. Whatever one concludes about the accounting, the coal stations are physically shut. A large thermal generator turned into a large wind developer inside two decades, and it is the most completely documented such conversion in the world.

The standard itself. ISO 14040 and ISO 14044, first published in 2006 and amended since, are the reason any of the above can be argued about at all. They set out life cycle assessment as four phases — goal and scope definition, inventory analysis, impact assessment, and interpretation — and they impose a discipline that is easy to underrate. The functional unit must be declared: the quantified performance a product delivers, not the product. The system boundary must be declared, unit process by unit process. Allocation between co-products follows a hierarchy that begins by trying to avoid allocation altogether, through subdivision or system expansion, before partitioning by physical relationships and only then by economic value. And where a study supports a comparative assertion disclosed to the public — a claim that this thing is better than that thing, made to a public audience, advertising and packaging included — the standard requires a critical review by a panel of interested parties, and forbids presenting a weighted single score as an objective comparison.

Read that list again as what it is: a set of rules written specifically to stop the person doing the arithmetic from choosing the answer. They are not always obeyed. But they exist, they are public, and they are the reason the rest of this chapter is possible. A manufacturer who publishes a module-level declaration has handed a stranger the means to disagree with them, on the record, with their own numbers. Very few industries have done that. This one has.

THE ARITHMETIC

The four declarations, and the square metre that moves under them

Definition. A manufacture is net positive with respect to an impact category when, over a declared boundary, against a declared baseline, compared with a declared counterfactual, and expressed per declared functional unit, the total carries a negative sign — the system removes more of that impact than it adds.

Four declarations. Not one of them is optional, and a claim missing any of them is not a claim, it is a sentence.

Now the worked product. Every figure below is read from a third-party verified environmental product declaration for a modular carpet tile, published to ISO 14025, ISO 21930 and EN 15804+A2. Nothing here is modelled; the arithmetic is reproducible in lib/verify/IV_01.py.

The unit. One square metre of floorcovering, nominal mass 3.34 kg, reference service life 15 years.

The boundary as declared. Product stage A1–A3; transport to site A4; installation A5; maintenance B2; transport of waste C2; disposal C4. Modules B1, B3, B4 and B5 are not declared, and — the sentence that matters — B2 is declared for one year, against a service life of fifteen.

The modules, in kg CO₂e/m².

  A1-A3  product stage          2.760     of which  fossil    4.730
                                                     biogenic -1.980
                                                     land use  0.0124
  A4     transport to site       0.323
  A5     installation            0.238
  B2     maintenance, ONE year   0.403
  C2     transport of waste      0.00710
  C4     disposal                0.104

Two things are visible immediately. The first is that the biogenic credit is 41.9 percent of the fossil burden — which is why a backing with more bio-based content crosses zero, and why a cradle-to-gate figure of about −0.30 kg CO₂e/m² is achievable and was achieved.

The second is the cut this chapter is built on. Sum the declared modules and the tile is 3.835 kg CO₂e/m². Now hold the product for its own declared service life — change nothing but the arithmetic of the number the document already gives you — and B2 becomes 0.403 × 15 = 6.045 kg CO₂e/m², and the total becomes 9.477 kg CO₂e/m², which is 3.43× the cradle-to-gate figure.

  module shares          as declared (B2, 1 yr)     over the 15-year life
  A1-A3                        72.0 %                      29.1 %
  A4 + A5                      14.6 %                       5.9 %
  B2  maintenance              10.5 %                      63.8 %
  C2 + C4                       2.9 %                       1.2 %

The largest single climate impact of a carpet tile is not the carpet. It is the vacuum cleaner. Sixty-four percent of the declared life cycle is the cleaning regime the manufacturer itself specifies, and the entire industry conversation — recycled nylon, bio-based backing, take-back — is conducted about the twenty-nine percent. Nothing about this is concealed; the declaration states the reference service life and states that B2 covers one year. The sentence is there. It is simply not the sentence anyone reads.

Where the sign flips. One square metre, six defensible boundaries.

  A1-A3, biogenic credit taken, bio-loaded backing      -0.300   NEGATIVE
  A1-A3, this product as declared                        2.760
  A1-A3, biogenic credit refused (fossil only)           4.742
  A1-A5 + B2(1 yr) + C2 + C4, as declared                3.835
  A1-A5 + B2(15 yr) + C2 + C4, over the service life     9.477
  ... and end of life by incineration                   11.325

A spread of 11.625 kg CO₂e/m² and a change of sign, on one object, with no physical difference between the cases. The last line is the one worth sitting with: the declaration's own biogenic bookkeeping records 6.22 kg of CO₂ removed into the product and 4.55 kg emitted from it during manufacture, with 0.218 and 0.0396 kg for packaging — leaving 1.848 kg of stored biogenic carbon still in the tile at the end of its life. The declared disposal scenario is 95.7 percent landfill, 4.3 percent incineration, no recycling, and treats that stored carbon as staying stored. Send the same tile to an energy-from-waste plant, which is the normal European fate, and it comes back.

A carbon-negative product whose declared grave is a landfill is storing carbon in a hole. That is a real removal for exactly as long as the hole holds, and the boundary should say so.

The honest negative, in two parts

First: the counterfactual that no auditor can attest. Most published net-positive claims are not inventory claims at all. They are avoided emissions — the difference between what happened and what would otherwise have happened — and the second term is a description of a world that did not occur.

Take a developer generating 10.0 TWh a year from offshore wind, whose lifecycle intensity the IPCC's Fifth Assessment puts at a median of 12 g CO₂e/kWh (range 8.0–35). What did it displace?

  coal, pulverised        820 g/kWh  (740-910)  ->  avoided   8.080 Mt CO2e/yr
  gas, combined cycle     490 g/kWh  (410-650)  ->  avoided   4.780 Mt CO2e/yr
  EU-27 grid average 2023 242 g/kWh             ->  avoided   2.300 Mt CO2e/yr
  onshore wind that an
  auction would have
  built anyway             11 g/kWh  (7.0-56)   ->  avoided  -0.010 Mt CO2e/yr

A spread of 8.090 Mt on one wind farm, from a large positive to a small negative, decided entirely by one sentence in a methodology note written by the party making the claim. Nothing physical distinguishes the four numbers.

And the claim usually fails a simpler test before it reaches the counterfactual at all. The World Business Council for Sustainable Development's 2023 guidance on avoided emissions is explicit that avoided emissions are a separate ledger and are not netted against the claimant's own scope 1, 2 and 3 inventory. A net positive that subtracts the one from the other is not permitted by the standard it cites. This is precisely what Interface did when it retired Carbon Neutral Floors in 2024: the neutrality had rested on purchased offsets, the offsets rested on counterfactual baselines, and the company chose measured reduction over an unverifiable subtraction. The most credible thing in this chapter's Discovery is a company withdrawing a claim.

Second: the case where regeneration loses, and it loses on price. Specify the lower-carbon tile over the 2019 average and the gate-stage cut is 5.2 − 2.76 = 2.44 kg CO₂e/m². At a mid-market installed price of $35.00/m², a 5.0 percent specification premium is $1.75/m² — which implies an abatement cost of $717/t CO₂e.

  at $ 50/t  the cut is worth $0.122/m2   does not clear
  at $100/t  the cut is worth $0.244/m2   does not clear
  at $190/t  the cut is worth $0.464/m2   does not clear
  at $250/t  the cut is worth $0.610/m2   does not clear

On carbon alone, at that premium, the specification loses against every internal carbon price in general corporate use. That is the honest answer and it should be said first, not buried. The condition the argument needs — and usually lacks — is a buyer whose decision is not made on carbon alone.

But hold the two negatives side by side, because together they resolve. A 30 percent reduction in maintenance intensity over the service life removes 1.813 kg CO₂e/m² — 0.74× the entire gate-stage cut — at no capital cost at all. (The declaration does not break B2 into vacuuming and extraction, so no split is asserted here; the thirty percent is a stated scenario and is labelled as one.) The specification that loses at $717 a tonne sits next to an operational change that costs nothing and delivers three-quarters as much. The boundary told you where to spend the money. That is what a boundary is for.

DREAM

What becomes ordinary

In the version of this that has already happened, a product declaration is read the way a set of accounts is read: by someone who knows which line is load bearing, and who turns to it first.

A specifier opening an environmental product declaration looks at the reference service life before the headline, because the headline is a rate and the service life is the multiplier. Declarations that cover one year of maintenance against fifteen years of life are not treated as dishonest; they are simply annualised before comparison, by a spreadsheet everyone has, in the same motion as converting a currency. Nobody finds this remarkable. It is what it means to read a document.

Manufacturers compete on the boundary rather than inside it. A firm that declares A1 to C4 with a full-life maintenance module and a realistic end-of-life scenario is understood to be making a stronger claim than a firm declaring cradle-to-gate, in the way a company reporting audited figures is understood to be making a stronger claim than one reporting management estimates. The wider boundary becomes the premium position, because it is the harder one, and the market prices difficulty.

Avoided emissions are reported, and reported separately, in their own statement, with the counterfactual named on the same page and a sensitivity range beside the headline. A developer reporting 2.300 Mt against a grid-average baseline shows the 8.080 and the −0.010 alongside it, and is respected for it, because a number without its range is understood to be a number that has not been looked at. No one nets that statement against an inventory. The two documents do not have a common denominator and everybody knows it.

And the cleaning contract is in the sustainability conversation. The facilities manager who moves an extraction-cleaning regime and documents the change is reported in the same paper as the procurement officer who moved the yarn, because the arithmetic put them in the same order of magnitude and the arithmetic was done before the meeting rather than after it. The building's declared whole-life carbon is a number with four contributors, and the people responsible for all four are in the room.

None of this requires a new standard. ISO 14040 and 14044 already require the functional unit, the boundary and the critical review. EN 15804 already defines the modules. What is missing is not a rule. It is the habit of turning to the service life first — and a contract that makes the habit pay.

DESIGN

The declaration card, and the governance around it

The mechanism is a single page that travels with every product claim, internal or external, and it has four fields. It is deliberately not a methodology; it is a form, and forms are what survive staff turnover.

Field one — the boundary. Which modules are counted, named as modules, plus what is excluded and why. A1–A3 only is a legitimate entry. A1–A3 with the word "lifecycle" next to it is not.

Field two — the baseline. What this is being compared with, in the same units, from the same or a declared source, with its date. "Industry average" is not a baseline unless the average is named, dated and sourced. The 2019 figure of 5.2 kg CO₂e/m² is a baseline; "conventional carpet" is a mood.

Field three — the counterfactual. What would have happened instead, written as a scenario in one sentence, with the sensitivity range beside it. If the claim does not depend on a counterfactual — if it is an inventory claim about a measured system — the field reads none, and that is the strongest entry on the card.

Field four — the unit. The functional unit and the service life it is held for. Not the product. A square metre for fifteen years, a kilowatt-hour delivered, a tonne-kilometre moved, a wash at 40°C.

The sequence. Build the card for what you already make before you build it for what you propose to make. The existing product is a measurement and the proposed one is a forecast, and a measurement is harder to argue with. This follows the discipline of I.01: start from the positive core, because you are looking for the place where the company is already ahead and has not costed it.

The governance. Three rules, and only three.

One: the card is signed by the person whose budget moves if it is wrong. Not the sustainability function. The buyer, the specifier, or the product manager. A declaration signed by the person who benefits from the claim is a marketing document; signed by the person exposed to it, it is a control.

Two: the boundary is frozen at the moment a comparison is made, and any change to it is a restatement. This is ordinary accounting practice — a change in accounting policy is disclosed, prior periods are restated, and the effect is quantified. Carbon claims have no equivalent convention and should adopt this one wholesale. Most of the movement in published product footprints between successive years is boundary movement, and nobody says which.

Three: where a claim is disclosed to the public and asserts superiority over an alternative, it goes to critical review by a panel that includes someone with an interest in disagreeing. ISO 14044 already requires this for comparative assertions disclosed to the public. Treat it as the floor rather than the ceiling, and treat "disclosed to the public" as covering the brochure, the website and the sales deck, because it does.

The cost of the whole apparatus is one page per product line and one meeting a quarter. The return is that your own claims stop being falsifiable by anyone who downloads your own declaration, which is the position almost every manufacturer is in today and almost none has noticed.

DESTINY

How it holds when nobody is pushing

It holds because it is cheaper than the alternative, and the alternative is arriving.

The regulatory direction is toward declared boundaries rather than headline claims — product-level declarations in construction procurement, substantiation requirements for environmental marketing, and disclosure regimes that ask for the methodology alongside the number. A company that already keeps declaration cards has its answer written. A company that does not will reconstruct it under time pressure, from records that were never kept for the purpose, which is the expensive way.

It holds because the card creates a second reader. The signature of the person exposed to the claim is the whole mechanism: it puts a pair of eyes on the boundary that has a reason to look.

Now the failure modes, named. It fails when the card is owned by the sustainability function, because then it is a report rather than a control, and reports are written to be filed. It fails when the boundary is frozen but the baseline is not, so the claim improves every year by re-choosing what it is compared against. It fails when the counterfactual field is filled in with a phrase rather than a scenario — "versus the market" is a phrase, and it is the single most common way an unauditable claim passes review. It fails when the functional unit is the product rather than the service, because then a tile that lasts eight years compares favourably with one that lasts twenty. And it fails quietly, and most often, when the whole-life arithmetic is never done at all: when B2 is read as 0.403 and not as 6.045, and the company spends four years and a great deal of money optimising twenty-nine percent of its own footprint.

DELIGHT

What it feels like

There is a specific pleasure in reading a declaration properly for the first time. It is the pleasure of a document that turns out to be generous — pages of modules, uncertainty, scenarios and stated assumptions, produced at real expense by people who were trying to tell the truth, sitting unread behind a marketing line that flattens all of it into one adjective.

You open it expecting to catch someone. Instead you find they already wrote down the thing that undoes them, on page seven, in a table, with a footnote explaining exactly why. And the feeling is not triumph. It is something closer to company: somebody else cared about this enough to put it in writing, and they did it years ago, and they left it where you would find it.

Then the second pleasure, which is the durable one. Once you have read one declaration all the way through, you cannot stop seeing the boundary in everything — the payback period that excludes the fit-out, the cost per unit that excludes the returns, the productivity figure that excludes the overtime. It is the same move every time, and learning it on a carpet tile teaches you to read an annual report. The square metre was never the point. It was the exercise.

OPERATIONALIZE THIS

At the level of finance

The instrument: the Verified Product Carbon Covenant. A supply agreement whose price per functional unit moves with an independently verified product declaration, over a boundary frozen at signature. It is a carbon contract for difference pointed at a product rather than a plant, and it can be written into an ordinary supply contract as a schedule.

The mechanics.

Quantity and unit. Contracted quantity in functional units, with the service life stated. Worked here at 40,000 m² of a large fit-out, held for 15 years.
Boundary, frozen. The modules counted, named at signature: A1–A3, A4, A5, B2 annualised across the reference service life, C2 and C4. Any change is a restatement, disclosed, with the prior position quantified.
Baseline. The supplier's own verified declaration at signature — 2.760 kg CO₂e/m² at the gate for the product worked above. Not an industry average, because an industry average is a number the supplier does not control and therefore cannot be held to.
Counterfactual: excluded by covenant. No avoided-emissions credit enters the price. This is the clause that makes the instrument auditable, and it aligns the contract with the WBCSD's separate-ledger requirement rather than against it.
Price adjustment. $0.55/m² per kg CO₂e/m² of verified movement against the baseline, symmetrical. A verified move to 2.260 pays the supplier $0.275/m², or $11,000 across the contract; a move to 2.510 pays $0.138/m², or $5,500; a move the wrong way to 3.010 rebates the same amount to the buyer. Symmetry is what stops this being a subsidy.
Verification. By the declaration's programme operator, annually, on the same schedule as the EPD's own validity. No new verifier is created; the existing one is contracted to.
Term. The shorter of the EPD validity period and the supply term, with a mandatory re-baseline on renewal and the movement disclosed.

The balance-sheet treatment. For the buyer, the premium or rebate is part of the cost of the fitted asset and is capitalised and depreciated with it — it is not an operating expense and it is not a donation. For the supplier, the adjustment is variable consideration under the revenue standard, recognised when the verification constraint is resolved, which is why the verification date belongs in the contract rather than in a policy.

The counterparty. The supplier you already buy from, on the contract you are already renewing. This instrument has no external counterparty, requires no lender, and creates no new legal entity. That is its whole commercial appeal: it is a schedule, not a transaction.

The number that decides it. One inequality, on the front page:

     internal carbon price  x  verified movement per unit
     ------------------------------------------------------  >  1
                  price adjustment per unit

For the worked case, that inequality does not hold on carbon alone — the $1.75/m² premium against a 2.44 kg/m² cut implies $717/t CO₂e, and no internal carbon price in general use clears it. Say so in the paper. Then put the second number beside it: across this contract the whole-life quantity governed is 379.1 t CO₂e, of which the gate stage is 110.4 t and maintenance is 241.8 t. The covenant governs the smaller share.

So the ninety days, and the recommendation that follows from the arithmetic.

Day	Action	Artifact
1–15	Pull the declarations for your three largest specified products	Three EPDs, downloaded
16–30	Build the declaration card for each; annualise every B module to the stated service life	Three cards, signed by the buyer
31–45	Rank your abatement options by implied $/t across the whole boundary	One ranked table
46–60	Draft the covenant schedule for the one supplier already ahead	Contract schedule
61–75	Negotiate the frozen boundary and the verification date	The signed schedule
76–90	Move the operating variable the ranking put first	The maintenance specification, changed

Note the last row. In this worked case the covenant is signed and the cleaning contract is rewritten, and the second one is where the tonnes are. The instrument buys you a supplier who can be held to a number. The boundary tells you where to spend the money, and it was never where the brochure said.

APPRECIATIVE QUESTIONS

Twelve, for a room

Discovery — what is already working

Which of our suppliers already publishes a module-level declaration we could recompute ourselves — and what did we learn the last time somebody read one all the way through?
Where have we already made a claim narrower and more defensible than it had to be, and what did that cost us commercially — if anything?
Think of a product or process here whose real impact sits somewhere nobody expected. Who found it, and what were they looking at when they did?

Dream — what becomes possible

If every claim we made carried its boundary, its baseline, its counterfactual and its unit on the same page, which of our current claims would get stronger?
Imagine our competitors' declarations were as readable as our own. What would we want ours to say on the line they turn to first?
If the widest boundary became the premium position in our market, what would we be selling three years from now that we do not sell today?

Design — what we build

Who in this business should sign the declaration card — meaning, whose budget moves if the number is wrong?
What would it take for us to freeze a boundary and restate prior periods the way we already do for a change in accounting policy?
Which single operating variable, changed at no capital cost, would move our whole-life number most — and who owns it today?

Destiny — how it holds

What would have to be true for a specifier here to turn to the service life before the headline, out of habit, ten years from now?
Who outside this company has an interest in disagreeing with our numbers, and how would we invite them in before a regulator does?
What is the first sign we would see that a boundary had quietly moved — and who would notice it first?

WORKS CITED

Anderson, R. C. (1998). Mid-Course Correction: Toward a Sustainable Enterprise. Chelsea Green.

Anderson, R. C. and White, R. (2009). Confessions of a Radical Industrialist. St. Martin's Press.

Ember. European Electricity Review and the associated European power sector carbon-intensity series. Ember, annual.

European Committee for Standardization (2019). EN 15804:2012+A2:2019 — Sustainability of construction works: Environmental product declarations, core rules for the product category of construction products. CEN.

Interface, Inc. (2019). Interface Announces Mission Zero Success, Commits to Climate Take Back. Company announcement, November 2019.

Interface, Inc. (2020). Interface Achieves Carbon Negative Milestone with the Launch of Innovative Products to Decarbonize the Built Environment. Company announcement, October 2020.

Interface, Inc. Environmental Product Declaration: CQuest™ Bio Modular Carpet Tile. Declared to ISO 14025, ISO 21930:2017 and EN 15804+A2; TRACI 2.2 and EF 3.1 results tables.

Interface, Inc. (2024). Interface Commits to Absolute Emission Reduction Efforts. Announcement of the discontinuation of the Carbon Neutral Floors™ and Carbon Neutral Enterprise programmes, effective 30 April 2024.

International Organization for Standardization (2006). ISO 14040:2006 — Environmental management: Life cycle assessment — Principles and framework, with Amendment 1:2020. ISO.

International Organization for Standardization (2006). ISO 14044:2006 — Environmental management: Life cycle assessment — Requirements and guidelines, with Amendment 1:2017 and Amendment 2:2020. ISO.

International Organization for Standardization (2017). ISO 21930:2017 — Sustainability in buildings and civil engineering works: Core rules for environmental product declarations of construction products and services. ISO.

Ørsted A/S (2017). DONG Energy completes the divestment of its upstream oil and gas business to INEOS. Company announcement, 29 September 2017.

Ørsted A/S. Cutting scope 1 and 2 emissions in our operations and the associated decarbonisation disclosures. Ørsted, current.

Schlömer, S. et al. (2014). "Annex III: Technology-specific cost and performance parameters." In Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press.

United States Environmental Protection Agency. Greenhouse Gas Equivalencies Calculator: Calculations and References. US EPA, current.

West, T. A. P. et al. (2023). "Action needed to make carbon offsets from forest conservation work for climate change mitigation." Science, 381(6660), 873–877.

World Business Council for Sustainable Development (2023). Guidance on Avoided Emissions: Helping Business Drive Solutions for a Net-Zero Future. WBCSD.

Note on figures. Every module figure for the worked product is read from the CQuest™ Bio environmental product declaration named above; every derived figure — the service-life correction, the module shares, the six boundaries, the biogenic balance, the implied abatement cost, the counterfactual spread and the covenant adjustments — is computed in lib/verify/IV_01.py and reproducible with python3 lib/verify.py IV.01. Where a figure is a stated scenario rather than a measurement, it is labelled as one in the text and in the module.