Research Plan · Environment & Measurement · 2026–2030

Unified Environmental Footprint:
one number is not enough

Carbon has a footprint metric. Water is beginning to have one. Soil humus, mineral depletion, biodiversity loss and the capacity of a system to recover after fire have almost none that reach an individual decision. This programme develops a single multidimensional footprint model across those axes, and derives two outputs from it: the EcoScore for products, and a personal indicator showing how far a household is from halving.

Active · Model in development · White paper WP-2026-01 in preparation
The Problem

Six dimensions, one decision

6
footprint dimensions the model carries, against the one or two in common use
9
planetary boundaries the model is anchored to, so that a target has a reference
50%
the halving target a personal indicator has to make legible at household level

A person choosing between two products, or a household deciding what to change first, meets an environmental claim expressed in a single quantity — usually carbon. That quantity is well measured and genuinely useful, and it is also one axis among several that move independently. A choice that lowers carbon can raise water use; a choice that lowers both can degrade soil or narrow biodiversity. Presenting one axis as the answer produces confident decisions on incomplete grounds.

The obstacle is not a shortage of science on the individual axes. Life-cycle inventories, water footprint accounting, soil organic carbon factors and characterisation factors for biodiversity loss all exist. The obstacle is that they are not commensurable, and any attempt to combine them requires weighting choices that are usually made silently by whoever builds the index.

The research gap: Existing multi-criteria environmental assessment either aggregates dimensions into a single score using undisclosed weights, or refuses to aggregate and leaves the reader with an uninterpretable table. No framework treats the weighting itself as the object the user should see and vary — and none reaches the individual or household scale where the decision is actually made.

The Model

The dimensions and why each is included

Carbon
The best-established axis, retained as the reference against which the others are calibrated rather than as the headline. Its maturity is what makes the incommensurability of the remaining axes visible.
Water
Consumptive and degradative water use, weighted by local scarcity, because a litre in a water-stressed basin and a litre in a water-rich one are not the same quantity. Connects directly to the institute's freshwater programme.
Humus depletion
Loss of soil organic matter under a given land use and management regime. Included because it is a stock, not a flux: a decision can be carbon-neutral in a single year while drawing down a stock that takes decades to rebuild.
Mineral depletion
Reframed from the conventional abiotic depletion potential. Extracted material is not lost; it is transformed. The proposed measure is therefore the time until processed material returns to a state in which nature can bind it again — plastic back to carbon, aluminium back to something bauxite-like — rather than a stock-remaining ratio.
Biodiversity loss
Species loss attributable to land occupation and transformation, using regionalised characterisation factors. Included because it is the axis least substitutable by technology: a lost species is not recoverable by process improvement.
Renewal capacity through fire
Taken in three senses at once: pyrophilic ecosystems for which fire is part of the renewal cycle rather than a loss; the material life cycle in which combustion yields biochar or ash with different binding futures; and resilience, measured as time to recovery. The novel and least developed of the six.
Hypotheses

What we are testing

H1 — Weighting changes the ranking, and users can see it
For a substantial share of realistic product and consumption comparisons, the ranking of options reverses under defensible alternative weightings of the six dimensions. If this holds, a single-score footprint is not a summary of the evidence but one choice among several, and the weighting has to be exposed as an interface rather than fixed by the index author. The model fails this test if rankings turn out to be robust to weighting, in which case a single score is defensible after all.
H2 — Return time is a better mineral measure than stock remaining
Time-to-rebindable-state discriminates between material choices more usefully than abiotic depletion potential, because it captures the difference between a material that re-enters biogeochemical cycles within years and one that persists for centuries at identical extraction volumes. Test: rank a set of materials under both measures and compare each ranking against independently assessed end-of-life outcomes.
H3 — Boundary anchoring makes a halving target meaningful
A personal reduction target expressed against planetary boundaries produces different priorities than the same target expressed as a percentage of current personal consumption. Halving an axis already far below its boundary and halving one far above it are not equivalent acts, and an indicator that treats them alike misdirects effort. Outcome: the divergence between boundary-anchored and baseline-anchored priority orderings for matched households.
H4 — Distributed sensing can detect footprint, not only report it
A distributed sensor network can identify local environmental change from aggregated low-cost measurements, in the way an immune system identifies a local anomaly without a central observer. If this holds, footprint measurement moves from declared inventory data toward observed data, which is the harder and more valuable form. EquoraVault is the prototype for this test.
H5 — Uncertainty must be visible or the model misleads
The six dimensions differ by orders of magnitude in evidential quality: carbon and water inventories are far better established than humus, biodiversity and fire-renewal estimates. Presenting them at equal visual weight transfers unwarranted confidence to the weakest axis. The model therefore encodes per-dimension uncertainty in its own output, and this hypothesis tests whether doing so changes how users read a comparison.
Methodology

Research and measurement approach

Dimension harmonisation
Each axis is built from named, published inventories and characterisation factors, with its own uncertainty range documented rather than absorbed into a composite. Where an axis has no established factor — fire-renewal capacity in particular — the modelled derivation and its assumptions are stated explicitly and flagged as the weakest link in the model.
Weighting as an interface
Rather than fixing weights, the model is delivered as a re-weightable instrument in which the user sets the relative importance of the six axes and watches the ranking respond. This has already been prototyped publicly on iterators.org, where a beverage comparison lets readers move the footprint and health weights and see the map rearrange.
Boundary anchoring
Each dimension is expressed against its corresponding planetary boundary, so that a personal or product figure has a reference point beyond the population average. This is what converts a halving target from a relative gesture into a statement about a limit.
Observed data via EquoraVault
The distributed measurement layer supplies observed environmental data to sit alongside inventory data, and serves as the test bed for H4. Deployment begins at Equora Spaces and the Living15 NanoLab in Budapest VII., where baseline measurement can be established before the network is extended.
EcoScore derivation
The product-level output, developed for Impact presenT, applies the model to manufactured items so that a purchase decision can be read across all six axes. The derivation from the underlying model is published, so that a score can be recomputed by anyone who disagrees with the weighting.
Personal halving indicator
The individual-level output, developed with GoHalve, shows how far a household stands from halving on each axis and which change moves the most. The indicator reports per-axis distance rather than a single composite, because a composite would conceal exactly the trade-off the model exists to expose.
See the EquoraVault measurement programme →
Timeline

Research roadmap

2026 Q2–Q3
Model specification. Six dimensions defined with named sources and per-axis uncertainty. Re-weightable prototype published on iterators.org. White paper WP-2026-01, Halving Toward a Boundary, prepared for Zenodo with a DOI.
2026 Q4
Weighting sensitivity study. H1 tested across a set of realistic comparisons. Mineral return-time measure specified and applied to a first material set for H2.
2027
Boundary anchoring and sensing. H3 tested with matched households. EquoraVault deployment extended for H4. EcoScore v1 derived for a first product range with Impact presenT.
2028–2030
Publication and release. Personal halving indicator released with GoHalve. Methodology submitted to a venue in industrial ecology or environmental assessment. Model and weighting engine released openly so that other groups can substitute their own factors.
What this programme does not claim

Boundary of the work

The model does not produce a single authoritative environmental number, and it is designed so that it cannot be mistaken for one. Its output is a comparison whose shape depends on stated weights, and its central claim is that this dependence is a property of the subject rather than a defect of the instrument.

Nor does it claim equal confidence across the six axes. Carbon and water rest on decades of inventory work; humus, biodiversity and fire-renewal do not. Any use of the model that hides that asymmetry misrepresents it.

Method & Scope

How this page was produced

Trilith Method™ · Research Card
Programme status
Active · Model in development · White paper WP-2026-01 in preparation
Machine contributionI
Literature discovery and synthesis, candidate identification, computational modelling, and first drafts of this page. Volume and speed are the machine's contribution; none of it is treated as verified on its own.
VerificationII
Claims traced to primary sources rather than to summaries of them. Contested claims run through assert–refute–adjudicate across independent model families. Errors found after publication are corrected on this page with their date.
Human governanceIII
Problem selection, the evidentiary bar, and the decision to publish rest with Pölö (László Papp), EQUORA Institute, who holds editorial responsibility for this page.
Evidence basis
Stated in the Methodology section, per component. Where a source is a preprint, a pilot study, a single trial or a modelled estimate, the page says so at the point of use.
Version
v1.0 · 2026-07-29
Method
Trilith Method™ framework · how the method is being tested
What this page is

This is a research plan rather than a result. It sets out what the programme intends to test, on what basis, and what would count as failure. Parts of it will turn out to be wrong; where that happens, the correction is recorded here with its date instead of being quietly removed.

Work by third parties is attributed to its sources and described at the confidence its evidence supports. Nothing on this page should be read as professional advice in the programme's domain, and nothing here has been peer reviewed unless a specific publication is cited as such.

Research Lead

Principal investigator

Pölö (László Papp) — Founder, EQUORA Institute. Researchers in industrial ecology, soil science, biodiversity assessment or fire ecology — particularly on the fire-renewal axis, which is the least developed of the six — are welcome to get in touch: lpapp@equora.institute