ecoinvent offers Life Cycle Impact Assessment (LCIA) scores (or “carbon footprints” or “emission factors”) following IPCC 2021.
The main method is also simply called “IPCC 2021”. Additionally, the method “IPCC 2021 (incl. biogenic CO2)” was introduced. It contains a characterization of biogenic carbon dioxide uptake and release characterized with -1/+1, as suggested in some standards. The two methods are complementary, and the latter is not meant to be used as a stand-alone method. Implementation details can be found in the LCIA Implementation report. The focus here is on how to use the scores for different impact categories that ecoinvent provides to be in line with standards and guidelines.
Available Impact Categories and Indicators
The indicator that is most often used is global warming potential 100 (GWP100). Most impact categories offered are quantified for this indicator (Figure 1). The main impact category is “climate change”, but sub-categories group impacts as follows:
- total (all relevant elementary flows assessed)
- fossil (fossil elementary flows assessed)
- aircraft emissions (fossil aircraft emissions assessed)
- direct land use change (land use change related elementary flows assessed)
- biogenic (biogenic/non-fossil elementary flows assessed)
Additionally, emissions and removals are given separately for “direct land use change” and “biogenic” sub-categories. Furthermore, it is made explicit in impact categories whether biogenic carbon dioxide is included or not (“excl./incl. biogenic CO2”) where relevant (“fossil” by definition excludes “biogenic” and land use change related carbon dioxide flows are separated from biogenic ones in modeling). Finally, for the main sub-categories there are impact categories available that include short-lived climate forcers (SLCFs). These are carbon monoxide, nitric oxide, and volatile organic compounds (VOCs). AR6 does not provide metrics for these anymore, which is why CFs from the IPCC 2013 implementation are used here (see LCIA Implementation report for details).
Other indicators than GWP100 are only offered for the “total” sub-category. These indicators are global warming potential 20 and 500 (GWP20 and GWP500) and global temperature potential 50 and 100 (GTP50 and GTP100).
Figure 1. Available impact categories and indicators and mapping to standards; (x) = not explicitly mentioned in the standard but recommended to be included in the analysis.
General Recommendations
What could probably be termed the “standard” indicator for assessing “climate change” is the impact category/indicator combination “climate change: total (excl. biogenic CO2)”/“GWP100”. However, it usually makes sense to compare different categories and indicators for further insights. The Life Cycle Initiative hosted by UN Environment has published recommendations on greenhouse gas emissions and climate change impacts in their first global guidance for LCIA indicators report. These recommendations relate to AR5/IPCC 2013 as this was the report available back then. The recommendations regarding indicators/time horizons are:
- Using GWP 100 as the indicator for the shorter-term climate change impact category.
- Using GTP100 as a proxy for long-term impacts because it is an instantaneous indicator targeting potential temperature rise 100 years in the future (because GTP50 leads to similar conclusions as GWP100).
Furthermore, it is recommended to perform a sensitivity analysis including short-lived climate forcers (SLCFs, called near-term climate forcers NTCFs in AR5).
Recommendations Regarding Biogenic Carbon Dioxide and Standards and Guidelines
First and foremost, we recommend using “IPCC 2021” where biogenic carbon dioxide is characterized with 0/0 whenever possible. However, some standards and guidelines—for example, the ISO 14067 and EN 15804 standards—demand characterizing biogenic carbon dioxide uptake and release with -1/+1. ecoinvent has implemented this in the “IPCC 2021 (incl. biogenic CO2)” method, which complements “IPCC 2021” with impact categories including biogenic carbon dioxide. These impact categories are “climate change: total” without and with SLCFs and “climate change: biogenic” including biogenic carbon dioxide for the total as well as the emissions and removals separately to provide more information and transparency.
Impact categories from “IPCC 2021” and “IPCC 2021 (incl. biogenic CO2)” can be mapped to standards and guidelines (Figure 1). Further carbon information like the carbon content of products might be needed to comply with standards and guidelines, but this information does not come from LCIA methods. We provide a list with carbon contents of all products in the “Files” section in ecoQuery.
Scores including biogenic carbon dioxide are subject to potential distortions by allocation and there is the risk of overestimating uptake. They should be handled with care, especially if they are negative, and they should never be used as a stand-alone score. Another approach to assess biogenic uptake is to simply check the biogenic dioxide bound in the product based on its non-fossil carbon content. This can be compared to the difference of the scores excluding and including biogenic carbon dioxide as – in theory – they should be equal.
Users assume full responsibility for their application and interpretation.
Additional Content
How IPCC Data Is Used to Calculate LCIA Scores
“Carbon footprint”, “Carbon accounting”, “Emission factor”, and “LCIA scores” are all the same and rely on publications by the Intergovernmental Panel on Climate Change (IPCC) by the United Nations. The panel regularly releases Assessment Reports (ARs) containing emissions metrics for Global Warming Potential (GWP) and Global Temperature Change Potential (GTP). These numbers are implemented as characterization factors (CFs) in “IPCC” methods for the ecoinvent database.
To give an example: In AR6 released in 2021, fossil methane has a CF of 29.8 kg CO2-equivalents / kg methane for the indicator “global warming potential 100 years (GWP100)”. These numbers are applied as CFs to all relevant “elementary flows” (= greenhouse gas emissions) in an impact category (“climate change”) and an impact assessment score (or result) can be calculated. For example, a carbon footprint of 34 kg CO2-equivalents; or in LCIA terms: a score for the impact category “climate change” of 34 kg CO2-equivalents.
Implementation details can be found in the LCIA Implementation report.
Challenges in Modeling Biogenic Carbon and Allocation Correction
The relevant elementary flows in the ecoinvent database for assessing biogenic carbon dioxide are “Carbon dioxide, in air” (uptake) and “Carbon dioxide, non-fossil” (releases). Carbon uptake happens in biomass growth processes, such as in agricultural and forestry processes. We have checked biogenic carbon properties and balances in these processes to ensure the amounts for “Carbon dioxide, in air” relate to what comes out as carbon stored in products. Furthermore, we have checked biogenic carbon balances of processes with allocation in relevant sectors, such as agriculture, forestry, and pulp and paper. This allows implementing automated carbon allocation correction (see below). However, this is not applied to the full database and other sectors might suffer from incomplete properties and carbon imbalances. Hence, the further down supply chains, the higher the risk that scores including biogenic carbon dioxide are distorted.
Carbon Allocation Correction
Elementary flows are part of the carbon balance, and they are allocated to products by economic allocation. This means that the carbon uptake (“Carbon dioxide, in air”) also gets allocated by economic allocation. This can lead to distortions in the balance. If we assume an activity producing logs and wood chips is 50:50, but carbon uptake is allocated 90:10 because logs are nine times the price of wood chips, there will be too much carbon uptake in the allocated logs activity and too little in the allocated wood chips activity. The elementary flow “Carbon dioxide, non-fossil, resource correction” corrects the difference of the distorted result to the amount before allocation.
