Gas characterization

For each producing node, MARIO multiplies the matched gas accounts by their characterization factors and sums them:

\[E_{\mathrm{GHG},j} = \sum_{g \in G} \mathrm{GWP}_g E_{g,j}.\]

The same operation applies to direct final-demand emissions in EY and to the native environmental blocks of a SUT. Existing environmental coefficients and footprint matrices receive the weighted row too. Missing footprint matrices are calculated when requested.

Only matching accounts enter the result. Missing gases are not estimated. All matched accounts must share a unit; the method does not convert their magnitudes. For example, kilograms of each gas produce a result interpreted as kilograms of CO2-equivalent.

Built-in GHG profiles

A built-in profile connects source account labels to factors. The profile is selected from source/name metadata unless explicitly supplied. Inspect it with database profiles.

The API defaults to AR6 and a 100-year horizon. Those arguments select factors where the profile stores multiple reports or horizons. A scalar factor is unchanged. In particular, an already characterized source CO2eq total keeps its original characterization; a new output label does not recharacterize it.

BONSAI’s recipe includes a factor of 0.00772 for Accelerated CO2, Air | Air when present. This is a BONSAI modelling convention checked for AR6/GWP100, not a general IPCC factor for CO2. The conventional factors and the available alternatives are documented in db.calc_ghg().

Supply your own factors

Pass exact account labels and the factors appropriate to your assessment. For example, only if these are your source labels, this mapping represents the listed conventional AR6/GWP100 gas factors:

factors = {
    "CO2 - combustion - air": 1.0,
    "CH4 - combustion - air": 29.8,
    "N2O - combustion - air": 273.0,
}
characterized = db.calc_ghg(
    gwp=factors, label="GHG selected combustion accounts", inplace=False,
)

This example covers those three accounts only. Include other relevant source accounts explicitly. With gwp=..., report and horizon arguments do not modify the supplied factors; preserve the factor mapping with your results.

Supply-chain footprints

On an IOT, direct emission coefficients e describe emissions per unit of output. The per-unit footprint includes the supply chain:

\[f = e\,(I-z)^{-1}.\]

MARIO can obtain the same result by solving a linear system rather than materializing the inverse. See computation settings. For a SUT, use the native activity and commodity quantities, including fc for commodity footprints; see the matrix catalogue.

A multiplier is not a national total. Production-based and consumption-based totals also require the appropriate activity or demand aggregation and treatment of direct final-demand emissions. The comparison example makes those choices explicit.

Cross-database comparisons

Record the source release and reference year, account coverage, actual factors, units and accounting boundary. The Explorer’s published comparison recipes can differ from calc_ghg() defaults: selecting the same report name alone does not guarantee identical account coverage or factors.

Return to Calculate greenhouse gas emissions for the operational steps.