We want to inform you of two significant issues that have been identified in version 3.10 of the ecoinvent database. These issues affect certain impact assessment indicators across all system models. If you are still using earlier versions of ecoinvent (version 3.9.1 or earlier), the issues below are not relevant and do not affect your work.
Summary of issues
- Coke Production Emissions (China and Rest-of-World): The first issue involves incorrect emissions values for two unit processes that model coke production. Emissions of “Cobalt II” and “Anthracene” are significantly overestimated, leading to inflated impact scores for indicators related to ecotoxicity, human toxicity, and human health. This affects a substantial part of the database due to the widespread use of coke in product supply chains. However, indicators outside of these specific impact categories remain unaffected.
- Brazilian Electricity Market Production Volumes: The second issue pertains to production volumes in Brazilian electricity markets, where the volumes are three orders of magnitude too small. This underestimates the contribution of Brazilian markets to the Latin American market groups (RLA), which particularly impacts biogenic climate change and climate change related to land use indicators. Only a small number of datasets (less than 0.5%) are affected, though the impact is noticeable in RLA market groups and around ten additional datasets.
For more details, please take a look at the Known Issues section on the ecoinvent website.
Should you have any questions or concerns, feel free to reach out to our support team at support@simapro.se or PRés support team at support@simapro.com