This is the full method, in order. It is written so that a reviewer can disagree with a specific step rather than with the whole result, which is the only way an impact claim can be argued about usefully.
The order matters more than any individual step. Most bad impact numbers are produced by starting at step ten and working backwards to find a justification.
What is actually wrong at the site: no reliable power, a disposal cost, unsafe water, a regulatory deadline. Stated in the operator's terms, not in technology terms.
What happens today, and what would happen instead if nothing changed. This is the step that decides whether any claim is possible at all. No baseline, no avoided-emissions claim.
What is being installed, on what feedstock, at what scale, in what operating mode. Power and heat generation, or a fuel configuration. Never both at once, because they draw on the same gas.
The quantity a meter or a laboratory can produce: carbon conversion, gas composition, throughput, hours, char yield, electrical output. These are measurements rather than estimates, and each has a data sheet.
The causal chain from that output to the effect on the ground, with every step named. If a step is assumed rather than evidenced, it is marked as assumed.
The output measured where it actually matters: at the site, on that feedstock, over a defined operating period rather than a campaign.
Exhaust emissions, tar and condensate, ash and char handling, feedstock transport, process energy, auxiliary electricity and maintenance load all come off the top before any benefit is stated.
Carbon goes to gas or to char, not both. District heat and electricity from the same gas are not two savings. Char retention and power generation cannot be added together.
Every figure carries its tolerance, its feedstock, its conditions and its basis. A point value for an uncertain quantity is a form of false precision, and it is treated as one.
With a boundary, a baseline, measured outputs, negative effects and uncertainty in place, a number can be published. Until then, the pathway is described and the missing measurement is named.
An avoided emission is not measured in a pipe. It is the difference between what happened and what would have happened instead, and the second half of that sentence has to be written down before the first half can be claimed.
We show a pathway only if it names what it replaces. Every pathway on this site states its baseline next to its output, and where that baseline is not yet established for a given site, the pathway is marked as deployment dependent rather than counted.
Method reference: WBCSD, Avoided emissions Implementation hub, which defines the reference scenario as the most likely situation, and how emissions would evolve, if the solution were not used.
Methane is a faster and sharper warming agent than carbon dioxide over twenty years and a smaller one over a hundred. Quoting one figure without saying which horizon it belongs to is how a climate claim becomes misleading, so we name the horizon every time.
The warming figures themselves, with both time horizons and the inventory rule that decides what is countable, are set out in the evidence centre with their sources.
See the figures →A credible impact number for this platform requires a defined system boundary, a named feedstock and site, a baseline for every avoided claim, operating conditions and hours, a full mass and energy balance, certified stack emissions data, and a life cycle assessment.
We hold some of those things and not others, and the ones we hold are not on the same footing as the ones we do not. Publishing a total now would produce a number that does not survive its first question, and it would take the credibility of the measured figures with it.
What we do publish. Measured outputs with their conditions. Published external evidence, cited. Pathways with their missing measurements named. Negative effects with their handling measures. And the roadmap of what is measured next. That is a smaller claim than a headline total, and it is the one that holds.
Measured on a pilot unit. A figure taken from a named campaign on an earlier, smaller X-series unit, with the feedstock, the conditions and the data sheet stated. This is the only kind of number we treat as our own result.
Published evidence. A fact established outside Zero-X, cited inline to its source. It describes the world, not our machine, and is never presented as a Zero-X result.
Not yet shown. Technically plausible, relevant to the pathway, and not yet demonstrated on Jenja or an earlier pilot unit. The missing measurement is always named next to it.
Deployment dependent. The effect exists only at a specific site with a specific baseline. Without a named baseline there is no claim, which is the counting rule below.
No measurement exists. We state this rather than substitute an estimate, a supplier figure or a comparable from another design.
The measurement exists and the data sheet has not been released. The figure is named with its campaign and feedstock, and flagged as pending.
Published evidence is the third state: a fact from outside Zero-X, cited to its author and year, describing the world rather than the machine.