ET EnviroTruth Environmental data, explained
Policy · Data · Method

How to read an emissions dataset without getting fooled.

An emissions figure is only as good as the inventory behind it. Five checks to run before a greenhouse gas number goes into an argument.

Faded ledger paper with a faint grid dissolving into a deep teal wash

Start with the scope

An emissions total always answers a narrower question than the headline suggests. Check which gases are counted (carbon dioxide alone, or the full basket of greenhouse gases expressed as CO2 equivalent), whether land use, land-use change and forestry is included, and whether the figure is territorial (released inside the border) or consumption-based (embodied in what residents buy). The same country can look like a leader or a laggard depending on how those three switches are set.

Check the base year and the units

Targets are written against a base year, and the choice matters. Canada's Kyoto commitment was measured against 1990; its later Copenhagen target used 2005, a year in which emissions were already far higher, as the Canada and Kyoto timeline shows. A cut "below 2005" and a cut "below 1990" are not the same promise. Units add a second trap: CO2 equivalent depends on which global warming potential values were applied, and national inventories have moved from the IPCC's AR4 factors to its AR5 factors, which shifts methane-heavy sectors such as agriculture and waste.

Find the revision history

National inventories are recalculated every year. When methods improve, the whole back series is revised, so a figure quoted in an older news story may not match the current submission for the same year. The annual inventory submissions published by the UNFCCC carry the current series together with the methodology notes that explain each change. Setting a new figure beside an old one taken from a different edition is one of the most common ways a trend gets invented or erased.

Separate measurement from modelling

Most inventory numbers are not read off a smokestack. They are activity data (fuel sold, cattle counted, tonnes landfilled) multiplied by emission factors, and the published uncertainty ranges are often wide for agriculture and waste. A headline that reports a 2 % change in a sector whose stated uncertainty is many times larger is reporting noise. The same reasoning sits behind the long temperature series on the climate data pages, where the trend line is drawn with the year-to-year scatter left in.

The habit travels

The same questions work on any explainer, from a carbon budget to a guide on safe online casinos Canada readers might come across in a search. Who compiled the material, what was actually checked, against which rules, and how recently? A National Post explainer on that subject can be read the same way: look for whether each claim is tied to a named regulator, whether the page states when it was last reviewed, and whether licensing facts are kept apart from opinion. The subject changes; the method for testing a claim does not.

A short checklist

These are the questions applied to every dataset cited on this site; the sourcing rules themselves are set out on the About page.

  • Which gases, which sectors, territorial or consumption-based?
  • Which base year, and which global warming potential values?
  • Which edition of the inventory, and has the series been revised since?
  • Measured or estimated, and how wide is the stated uncertainty?
  • Is the comparison like for like across years, countries and editions?

Sources

  • UNFCCC, National inventory submissions of Annex I Parties.
  • Environment and Climate Change Canada, National Inventory Report: Greenhouse Gas Sources and Sinks in Canada.
  • IPCC, 2006 Guidelines for National Greenhouse Gas Inventories and 2019 Refinement.
  • IPCC, Fifth Assessment Report, Working Group I, Chapter 8 (global warming potentials).