TPDDTEC v4.0 — how clean cookstove credits are calculated

The full name is Technologies and Practices to Displace Decentralised Thermal Energy Consumption (“TPDDTEC”), currently at version 4.0 under Gold Standard for the Global Goals.^[1] It is one of the most widely used methodologies for clean-cookstove and related-technology carbon projects globally, and — as of 2026 — one of the methodologies whose baseline construction and quantification approach has been the subject of extended debate in the voluntary carbon market.

This article walks through the substantive machinery of the methodology: what it measures, how baselines are constructed, what conservative factors apply, and what verification produces. The intended reader is someone doing due diligence on a specific cookstove project or evaluating cookstove credits at portfolio level. The treatment is technical.

What the methodology covers

TPDDTEC applies to projects that displace decentralised thermal energy consumption. In practice, this means projects that replace less-efficient thermal-energy technology and practice with more-efficient technology and practice at the household, institutional (schools, prisons, kitchens), or commercial scale.^[2] Improved biomass cookstoves are the archetypal application, but the methodology also covers:

Solar cooking devices;
LPG stove distribution, with constraints on the counterfactual and treatment of upstream emissions;
Efficient charcoal-producing kilns feeding into cooking supply chains;
Institutional cooking systems (large kitchens serving schools, prisons, hospitals).

The methodology explicitly excludes fossil-fuel technologies whose lifecycle emissions exceed the counterfactual baseline. It is designed for decentralised applications; industrial-scale thermal energy is out of scope.

The core arithmetic: what is being counted

At its arithmetical core, TPDDTEC v4.0 computes the emissions reduction attributable to a project as:

ER = (BE − PE − LE) × cf_1 × cf_2 × … × cf_n

Where:

BE is the baseline emissions (what would have been emitted without the project);
PE is the project scenario emissions (what is actually emitted with the project in place);
LE is the leakage emissions (emissions displaced to elsewhere by the project);
cf_i are conservative discount factors applied to reflect specific measurement or scientific uncertainties.

Each element is technically demanding to establish. What follows is the substantive machinery of each.

Constructing the baseline

The baseline is the counterfactual: what would households have done, and with what emissions consequences, in the absence of the project intervention. Under TPDDTEC v4.0, the baseline must be established through primary data collection in the project area, not through regional averages or literature values.^[3] Three parameters anchor the baseline:

Baseline fuel consumption. Household fuel use is measured through a Kitchen Performance Test (KPT), a standardised protocol in which weighed fuel is supplied to a household over a defined period (typically 3–5 days), and consumption is measured directly.^[4] The KPT captures actual usage patterns — meal frequency, meal size, cooking duration, seasonal variation — rather than laboratory-idealised measurements. Sample sizes are specified by the methodology; sampling must be representative of the project population.

Baseline stove efficiency. The thermal efficiency of the baseline technology (open fires, three-stone fires, or traditional stoves being replaced) is measured through a Water Boiling Test (WBT), a controlled protocol in which the stove’s ability to bring water to boil and hold it at boil is measured under specified conditions.^[5] Baseline efficiency for traditional biomass cooking typically falls in the 10–15% range — reflecting that only that fraction of the fuel’s energy content reaches the cooking pot.

Fraction of non-renewable biomass (fNRB). This is the single most consequential parameter in the entire methodology. Not all biomass combustion has the same carbon-cycle consequence. Biomass harvested at or below the rate of forest regrowth is, in principle, carbon-neutral over the harvest cycle — the CO₂ released in combustion is eventually reabsorbed by regrowth. Biomass harvested at rates exceeding regrowth is functionally part of a deforestation flux, and its combustion contributes net CO₂ to the atmosphere at a rate equal to the fraction of harvest that is non-renewable.

The fNRB factor is expressed as a country- or region-specific fraction between 0 and 1. Where fNRB is high (Madagascar’s national fNRB is above 0.9, reflecting that the country has lost over 30% of its tree cover since 2000^[6]), avoided biomass combustion translates almost tonne-for-tonne into avoided net CO₂. Where fNRB is low, avoided combustion has a much smaller net effect.

TPDDTEC v4.0 requires fNRB values to be drawn from officially sanctioned sources: the CDM Standardised Baseline for the country, national data submitted to the UNFCCC, or peer-reviewed literature. Values obtained through less-rigorous methods are not accepted. This is a material tightening on earlier methodology versions, in which fNRB choices had sometimes been contested.

The project scenario

Once the baseline is established, the project scenario is quantified through analogous measurements against the improved technology.

Project stove efficiency. The project technology’s WBT efficiency is measured under the same protocols as the baseline. Improved biomass stoves typically achieve efficiencies of 30–45% depending on design; LPG stoves achieve materially higher. The efficiency delta is what drives the fuel-savings claim.

Adoption rate. A stove distributed is not a stove used. TPDDTEC v4.0 requires adoption to be measured through periodic usage surveys and — increasingly — through direct sensor monitoring on a sampled subset of installations. Adoption rate is applied as a multiplier that discounts the theoretical fuel savings by the fraction of households actually using the project stove.

Stove stacking. Many households use multiple stoves in parallel: the improved stove for some meals, the traditional stove for others. TPDDTEC v4.0 requires stacking behaviour to be measured and the reductions discounted accordingly. Stacking is the single most common cause of overestimated real-world reductions in project designs that ignore it.

Drop-off. Stoves break, are lost, or are abandoned. The methodology requires drop-off to be assumed (typically declining year-on-year) unless the project can evidence continued use through ongoing sampling and verification. Uncorroborated assumptions of continued use in later project years are the second most common cause of overestimation.

Conservative discount factors

TPDDTEC v4.0 applies multiple conservative factors that reduce the arithmetically-computed reduction to a defensible claim. These include:

Uncertainty discount on parameters measured with significant confidence intervals;
Sampling adjustment for parameters estimated from samples rather than complete enumeration;
fNRB conservatism applied where the fNRB estimation methodology has known uncertainty;
Leakage assumptions applied when household fuel demand may shift to other households, increasing overall consumption locally.

The cumulative effect of these factors is material: methodology-compliant credits typically claim 60–75% of the arithmetic reduction, with the balance held back to accommodate irreducible measurement uncertainty. Programmes and projects that claim near 100% of arithmetic reduction are, in TPDDTEC v4.0 terms, structurally under-conservative.

Verification

Verification is performed at defined intervals — typically annually or biannually — by an accredited independent third-party verifier accredited by the Gold Standard.^[7] The verifier:

Reviews the project’s monitoring data;
Sample-checks against field reality (visiting a defined subset of installations);
Examines the conservatism of assumptions;
Cross-checks reported fuel use, stacking rates, and drop-off assumptions against evidence;
Issues a verification statement determining what fraction of claimed reductions may be issued as credits.

The verification statement is the operational instrument through which claimed reductions become tradable credits. Where verification identifies overclaims, credits are correspondingly reduced or withheld. Verifier accreditation is itself audited by the Gold Standard.

Where TPDDTEC v4.0 differs from earlier versions

Version 4.0, in force from 2023, tightened several parameters relative to v3.x:

Stricter fNRB source requirements (as noted above);
More conservative stacking assumptions where empirical measurement is not available;
Reduced default WBT-to-KPT translation factors, correcting a known systematic overstatement of real-world fuel savings from laboratory efficiency measurements;
Enhanced monitoring frequency requirements, particularly for larger projects;
Explicit provisions for third-party sensor monitoring as a corroborating data stream, where projects wish to deploy it.

These changes were consequential. Projects transitioning from v3.x to v4.0 have generally seen claimed reductions decline, in some cases materially — a signal that the market has corrected in a way that voluntary carbon integrity observers regard as broadly appropriate.^[8]

Why the methodology matters for buyers

For a buyer, the important question about a cookstove credit is not “is it from an approved methodology” — it is “how conservatively has that methodology been implemented on this specific project?” Two projects operating under the same methodology, with the same nominal stove and geography, can produce credit inventories that differ by 30–50% depending on:

Whether adoption rates are measured or assumed;
Whether stacking is measured or ignored;
How fNRB is sourced;
Whether sensor monitoring corroborates usage claims;
How discount factors are applied.

Interrogating these features on a project-by-project basis is the substantive due-diligence work. Reliance on programme-level or methodology-level statements alone does not substitute for it.

Where SaniTap sits

SaniTap’s clean cookstove project in Madagascar operates under Gold Standard TPDDTEC v4.0. Distinctive features:

fNRB. Madagascar’s country-level fNRB, drawn from officially sanctioned sources, is above 0.9 — reflecting the severity of the country’s deforestation flux. This is not a project-specific advantage; it is a country-specific reality that translates into a high fraction of avoided combustion becoming avoided net CO₂.
Adoption monitoring. Adoption is measured through periodic household surveys plus third-party sensor monitoring on a randomised sample of installations. The sensor deployment is beyond the methodology minimum and provides independently corroborated usage data.
Stacking. Stove stacking is measured empirically rather than assumed. Reductions are correspondingly discounted.
Distribution scale. 120,000 stoves distributed as of the most recent reporting period, across urban, peri-urban, and rural communities in Madagascar.

For buyers evaluating cookstove credits at a rigour level appropriate to modern integrity frameworks (ICVCM CCPs, CORSIA Phase 2), the project’s project design documents, monitoring reports, and verification statements are available on request via our commercial team.