If you have any questions concerning the functioning of the USEtox model, you may send us a mail at and we shall try to the best of our ability to answer you as soon as we can.
USEtox FAQ
1. How can the human toxicity Characterization Factors (CFs) best be reported regarding the carcinogenic and non-carcinogenic effects?
Provide a set of 3 CFs: "carcinogenic", "non-carcinogenic" and "total impact", which is the sum of the former two impacts
2. Are the CF factors only for the "continental scale"?
Yes, the continental scale is the box where all the emission shall occur. The global box is only there to capture the potential impact generated by transboundary emissions of persistent and long range chemicals at a global scale. The CFs for an emission at continental scale already includes the cumulative impacts generated within the continental and the global scale. This latter should therefore not be used as an emission box.
3. How can the CFs for the different compartments be used in LCA related to Ecoinvent?
Urban air: use Urban air CFs to assess emission to "high population density"
Continental rural air: use Continental rural air CFs to assess emissions to "low population density", "lower stratosphere/upper troposphere"
Calculate a new weighted CF based on 50% urban + 50% Continental rural air CFs to assess "unspecified" emissions
Continental freshwater: use Continental freshwater CFs for the water emission subcategories "lake", "river", "ground", and "unspecified"
Continental agricultural soil: Use Continental agricultural soil CFs to asses emissions to "agriculture".
Natural soil: Use Natural soil CFs to asses emissions to "forestry", "unspecified" or "industrial"
Coastal water: Use Coastal Water CFs to assess emissions into ocean
4. Is toxicity to seawater species considered?
The current version of USEtoxTM only models impacts in the freshwater environment The reason only freshwater impacts are investigated is that mainly toxicity tests on freshwater species are available.
5. Does USEtoxTM model fine dust (PM)?
USEtox does not include PM for this moment. More sophisticated atmospheric chemical fate models better describe the fate of PM in the air, including the formation of secondary PM from NH3, NOx, SO2 emissions. The ILCD recommends using the RiskPoll model for midpoint characterization factors and Van Zelm et al (2008; included in ReCiPe) for endpoint.
6. Which substance is used as reference to calculate the CFs?
USEtoxTM does not make use of a reference substance, but expresses the characterization factors in terms of Comparative Toxic Units (CTU). For human toxicity it represents the likelihood of one cancer or non-cancer case per kg of chemical emitted. Similarly for ecotoxicity it represents one PDF*m2*yr / kg
7. Can I use the interim characterization factors in my LCA?
Yes, the so-called “interim” USEtoxTM characterisation factors should always be used together with the “recommended” factors, as otherwise the substances concerned would be characterised with zero impact as no characterisation factor is applied to their emissions. The flag “Interim” means a higher uncertainty of the characterization factor compared to the “recommended ones”, because not all the minima requirements are met for its calculation. Therefore, when an emission characterised with interim factors is dominating the overall impact, it implies that these latter have to be interpreted as having a lower level of confidence. A sensitivity study might by performed by applying only the recommended factors to see if and how the results (and the conclusions) change.
8. Why is there no damage matrix for human health in the USEtox model?
The damage matrix links the number of incidences (i.e. cancer or non-cancer cases) to the potential consequences of a chronic toxicological effect. It accounts for severity of the diseases/incidents considered. As there is a high uncertainty related to the severity and there is a lack of evidence for significant differences when combining dose-response slope and severity of
disability, the damage matrix and differences between cancer and non-cancer have been left
out for the moment.
9. Have the resulting CFs been implemented in LCA software?
To our knowledge, all characterisation factors have been implemented in GaBi, SimaPro, and OpenLCA. We have no complete overview of all LCA software and their status of implementation for USEtoxTM CFs, hence there might be other software that also has the factors included.
10. When can we expect an update of the model and the databases?
As the USEtoxTM model supplies a stable method, only a yearly update in May is foreseen with small improvements. The version number can be found on the sheet “Version” of the model.
Every six months there will be an update of individual substances in the database and consequently of individual CFs.
The latest available versions will always appear on the website.
11. When are updates expected for the interim (flagged) substances?
Metals – End of 2011
Dissociating substances - End of 2011
Organometallics – uncertain, expected in 2012
Amphiphilics - uncertain, expected in 2012