What uncertainty does the Intergovernmental Panel on Climate Change (IPCC) give for fuel activity data in industry: one default or a table?

Both. IPCC Volume 2 Chapter 1 says to assume plus or minus 5 percent for fossil fuel data if nothing better is known. Chapter 2, Table 2.15, gives ranges by sector: 3-5% for other industry from surveys in well developed statistical systems, and 10-15% in less developed ones. They are ranges for national statistics.

The default

• Chapter 1, section 1.5.2: if no further data are available, assume plus or minus 5 percent for fossil fuel combustion data. That is 0.95 to 1.05 times the point estimate, as a 95 percent interval.

Table 2.15 by sector

Chapter 2, Table 2.15 gives activity data uncertainty from surveys and from extrapolation, first for well developed and then for less developed statistical systems. Each row below reads: surveys, extrapolation, then surveys, extrapolation.

• Electricity and heat production: less than 1% and 3-5%, then 1-2% and 5-10%.
• Commercial, institutional and residential combustion: 3-5% and 5-10%, then 10-15% and 15-25%.
• Industrial combustion, energy intensive industries: 2-3% and 3-5%, then 2-3% and 5-10%.
• Industrial combustion, others: 3-5% and 5-10%, then 10-15% and 15-20%.
• Biomass in small sources: 10-30% and 20-40%, then 30-60% and 60-100%.

What IPCC says around the table

• Statistics of fuel combusted at large sources, from direct measurement or obligatory reporting, are likely to be within 3 percent of the central estimate, and likely more accurate for some energy intensive industries.
• Countries with good data collection and quality control may keep random error in total recorded energy use to about 2-3 percent of the annual figure. Percentage errors for individual energy uses can be much larger.
• Systematic and random error combined are probably about plus or minus 5 percent for most developed countries and, where energy data systems are less developed, could be considerably larger, probably about plus or minus 10 percent. Informal activities may raise it to as much as 50 percent in some sectors for some countries.
• It is good practice to develop country-specific uncertainties, using expert judgement or statistical analysis, where possible.

How the default and the table fit

The 5 percent is the fallback when nothing else is known. Table 2.15 is more specific and, in IPCC's words, may be used when reporting uncertainties. Neither is a measured error for your plant. IPCC does not use Scopes, so these are ranges for fuel data in national inventories.

Using it at one plant

My reading: a plant that reads a calibrated meter or weighs every delivery has better evidence than a national range and can argue for a narrower one, with the instrument accuracy on file. A plant that only has estimates should not claim less than the table shows. Whoever relies on the figure, a buyer or a verifier, decides what it accepts.

What each range means on 20,000 litres of diesel

• 20,000 litres x 2.645 kilograms (kg) of carbon dioxide (CO2) per litre = 52,900 kg, or 52.9 tonnes
• Plus or minus 3% (IPCC: fuel statistics at large sources likely within 3 percent): 19,400 to 20,600 litres = 51.31 to 54.49 tonnes
• Plus or minus 5% (the default): 19,000 to 21,000 litres = 50.26 to 55.55 tonnes
• Plus or minus 10% (low end of a less developed system, industry others, surveys): 18,000 to 22,000 litres = 47.61 to 58.19 tonnes

Example only: IPCC wrote these ranges for national statistics, and the emission factor has its own range on top of the fuel quantity.

What Senseible does here

Senseible reads fuel invoices and electricity bills into monthly records, so the litres or tonnes behind each figure stay traceable to a bill.

• Calculating GHG inventory uncertainty
• Uncertainty on biomass fuel tonnes
• IPCC 95% range around default fuel factors
• Kinds of uncertainty and year-on-year change
• Does meter uncertainty count in CBAM materiality?

Sources: IPCC 2006 Guidelines, Volume 2, Chapter 2 Stationary Combustion (section 2.4.2, Table 2.15), IPCC 2006 Guidelines, Volume 2, Chapter 1 Introduction (section 1.5.2)

Last reviewed 2026-11-05. Table 2.15 is credited in the chapter to the IPCC Good Practice Guidance of 2000. The ranges are for national statistics; a plant should use its own instrument accuracy where it can show it.