Emission factor databases: frequent questions on sources, updates, and use

Emission factors are the core parameter of any carbon inventory — yet different databases and vintages disagree, and practitioners rightly ask which to trust. Here are the five most frequent questions, unpacked.

What exactly is an emission factor, and how are they classified?

An emission factor converts activity data (say, kWh of electricity) into emissions, derived from fuel composition, process efficiency, or measurement. Two broad families matter in practice: direct combustion factors (natural gas, diesel) and indirect factors for electricity and heat, which depend on grid mix or heat-source type.

Common classification axes: by source — national guidance (IPCC defaults), regional databases (such as China’s product carbon-footprint factor library), industry research factors, and company-measured factors; by accounting stage — upstream extraction/transport/processing, core combustion, downstream waste treatment (values differ a lot; match your boundary); by currency — static factors versus annually updated dynamic ones (grid factors after 2026 reflect rising wind and solar shares much better). For organizational inventories, national or industry factors come first; product footprints need finer upstream factors. And check the factor’s gas coverage — a CO2-only factor understates sectors with meaningful CH4 or N2O.

Databases disagree — which do I pick?

IPCC defaults, domestic official factors, and commercial LCA databases can differ by 30%+, because IPCC uses global or regional averages, domestic factors track the local energy structure, and commercial sets embed finer measurements or model defaults. Three tests: purpose match — domestic compliance reporting requires the factors your regulator publishes; export-product footprints sometimes must use the customer’s specified international database. Vintage — energy mixes shift yearly (2026 grid factors are far below 2020’s as renewables grow); use factors close to the emission year. Completeness — some factors include CH4 and N2O, some don’t; livestock and refrigerant-heavy sectors cannot afford the omission. And beware “newest is best”: a 2026-published factor may describe 2024 data. Prefer factors within about two years of publication lag, avoiding sudden jumps from methodology revisions.

How often do factors update, and how do I stay current?

Update rhythms vary widely: national grid factors annually (published mid-following-year), IPCC defaults roughly every 5–7 years, and some fuel heating-value factors barely at all. Finer regional grid factors are expected around 2026 as renewable-quota schemes advance. Practical routine: audit your factor-library version annually and swap in updates for the accounting year; set update alerts on regulator sites if you maintain inventory templates; and when a process-specific factor is missing, borrow the closest IPCC default with a documented source-and-uncertainty note. Best timing for the annual sweep is December–January, since most national factors publish within 3–6 months of year-end — don’t wait for an auditor to catch a stale factor.

What are the classic pitfalls?

Unit slips — factors come per tonne, kg, kWh, or m³; normalize activity-data units before input or risk order-of-magnitude errors. Net vs. gross heating value — the basis changes results by roughly 10%; domestic practice usually uses net, some IPCC factors use gross — confirm and convert. Ignoring stated uncertainty — factors carry ± ranges that formal reports should carry through; from 2026, several international standards require uncertainty analysis for validity. The working checklist: match units (tonne of coal vs. tonne of standard coal), verify non-CO2 gas coverage, prefer localized factors over global defaults, and archive factor provenance for audit. When unsure, spend the half-day in the literature rather than assuming — one wrong factor can send the whole inventory back for rework.

Questions & answers

Where do I get reliable factor libraries? Domestically, the environment ministry’s or provincial energy-statistics recommended factors; internationally, IPCC guidelines or recognized industry databases — always check version and vintage.

Why do databases differ so much? Different statistical scopes, geographies, and assumptions; IPCC’s global averages versus localized factors commonly differ 20–50%.

Must I restate after a factor update? Completed annual inventories stand; new accounting must use the latest version. Annual updates prevent cross-generation data breaks.

Can a company develop its own factors? Yes, with measured data or process models plus third-party validation — though a weak self-measured factor is worse than a peer-reviewed one.

Do grid factors fall every year? The trend is down as clean energy grows, but individual years can tick up on hydro variability or coal backstop — follow the official releases.

How do I judge sector fit? Read the factor’s stated applicable industry and technology boundary; if no exact match exists, take the closest process class and note the uncertainty.

Do factors expire? No fixed expiry, but beyond five years they are generally considered stale — usable for historical comparison, not for current accounting.