ecoinvent data is available for environmental reporting, research, and sustainable decision-making.
ecoinvent v3.13 is here to empower businesses, researchers, and policymakers. A new sector joins the database, several existing sectors get deeper and more current environmental data, and we introduce a new feature to enable easier dataset comparisons.
ecoinvent data is available for environmental reporting, research, and sustainable decision-making.
In version 3.13, a new sector is introduced in the ecoinvent database: pharmaceuticals. This new sector aims to support the ongoing efforts of the pharmaceutical industry toward Life Cycle Assessment (LCA) harmonization. It was therefore developed in collaboration with the Pharma LCA Consortium and Boehringer Ingelheim, ensuring alignment with the first Product Category Rule (PCR) for pharmaceutical products (PAS2090). It covers the production of 61 newly regionalized components across the three key geographies of Europe (RER), North America (RNA), and Asia (RAS). The new datasets include critical sector-specific substances such as amino acids, excipients, growth media components, pharmaceutical-grade water, treatment of average pharmaceutical waste, and solvents recycling processes.
The agriculture, fishery, and animal husbandry sector focuses on several refinements through technical corrections to improve the quality of the data. In several Australian cereal datasets across different regions, a carbon imbalance was identified due to the incorrect assignment of embodied biogenic carbon. Specifically, the value was mistakenly allocated to the flow “Carbon dioxide, land transformation” instead of the correct flow “Carbon dioxide, non-fossil”. The amount of the elementary exchange “Carbon dioxide, non-fossil” has been updated to derive the emissions from burning residues. Also, the specific comments of these flows have been harmonized consistently.
Further to the carbon balance correction applied in v3.12, a carbon imbalance was identified in the Brazilian land use change, perennial crop datasets. The correction added “Carbon dioxide, to soil or biomass stock” as the balancing flow for carbon uptake, but the pre-existing “Carbon dioxide, in air” flow, which it was meant to replace, was not removed. This inconsistency in the v3.12 correction has not yet been fixed and results in double counting of the carbon uptake.
For this release, the batteries sector introduces 24 new datasets for battery modules and packs of different sizes (75 kWh and 100 kWh packs) for several battery chemistries (NCA, LFP, NMC111, NMC532, NMC622 and NMC811). These new datasets are based on the BatPaC model developed by the Argonne National Laboratory. Basing the inventories on the same underlying model allows for more consistency across the various datasets created.
The new battery pack datasets are intended to be used in electric vehicles. Their sizes, in total energy (kWh), have been chosen to reflect widely used battery sizes in current electric vehicles. This update increases the granularity of available battery packs in the database and allows users to choose more appropriate packs for mobility applications.
The sector also benefited from updates in other sectors:
Version 3.13 includes a set of technical corrections to construction-related datasets, addressing inconsistencies and unrealistic exchanges identified through user feedback, internal review, and peer-reviewed literature.
In the cement sector, the composition of the Swiss Portland cement (CEM I) dataset has been restored to a mass-balanced formulation consistent with EN 197-1, correcting an inconsistency introduced in version 3.10. For infrastructure datasets, machinery throughput and copper inputs in several building and packing datasets have been revised to reflect realistic operational assumptions and literature-based material intensities (Tokeda & Rouwette, 2024; Wang et al., 2022; Marinova et al., 2019). Finally, a mismatched land transformation flow in the perlite quarry operation dataset has been corrected, resolving a spurious negative result in the land use indicator (EF 3.1). Together, these corrections improve the reliability of LCA results in construction product chains, particularly for impact categories sensitive to mineral resource use and land transformation.
The chemicals and plastics sector in ecoinvent v3.13 is updated to further improve the technological and geographical representation of key chemical and plastic products. Formaldehyde production is expanded to distinguish the three main industrial production technologies at the global level and provide average production data for 23 countries, complemented by more geographically differentiated markets. Additionally, the geographical coverage of acetaldehyde and key polystyrene grades—general purpose (GPPS), high impact (HIPS), and expandable polystyrene (EPS)—is expanded, building on data updated or introduced in recent database versions. Finally, the release also continues the product portfolio expansion achieved in preceding versions, introducing twelve new products not previously represented in the database. Efforts to further expand the technological, geographical, and product coverage of chemicals and plastics will continue in future releases.
ecoinvent version 3.13 updates the energy sector’s electricity market datasets with the consistent approach applied in ecoinvent versions 3.6 through 3.12. Consistent with v3.12, Enerdata serves as the primary data source for electricity generation across most regions, with exceptions where detailed national data are available and, for European countries, where the Eurostat database from the European Commission is used. The electricity market mixes generally represent the year 2023, except for Brazil (BR), Canada (CA), China (CN), India (IN), Switzerland (CH), and the United States (US), which are updated to 2024 based on comprehensive national statistics. For India, the update corresponds to the fiscal year of April 2023-March 2024. Additionally, the European electricity residual mix markets in version 3.13 represent the year 2025 using statistics from AIB.
Furthermore, technical corrections were carried out for “electricity production, lignite” and “treatment of coal gas, in power plant” for selected countries. Moreover, harmonization efforts were carried out to update the documentation and meta-information of electricity markets, transformation, and import datasets.
The new ecoinvent version 3.13 in the fuels sector expands the geographical and technological coverage of calcined petroleum coke (CPC) production, introducing datasets for six key geographies: Brazil, China, Germany, India, Oman, and the United States. The update also improves the technological representativeness of CPC production by modeling the predominant shaft calcination technology in China, while rotary kiln technology is represented in the other five geographies, replacing the previously available Chinese dataset based on rotary-kiln data from an Indian facility. In addition, six corresponding CPC market datasets are introduced, providing geographically differentiated supply for downstream users; the existing Chinese market is replaced, while a global market and five new regional markets are added.
For the v3.13 release, the metals sector substantially expands the coverage of the lithium supply chain, introducing 32 new transforming activity and market datasets across Argentina, Australia, Chile, China, and the United States. The update covers both brine- and spodumene-based production routes, from raw material extraction and concentration to the production of lithium carbonate and lithium hydroxide monohydrate. The new datasets are based on recent, site-specific literature and provide greater geographical and technological differentiation of lithium production in the ecoinvent database.
The update also introduces an explicit distinction between battery-grade and technical-grade lithium carbonate, reflecting their different purity requirements and end-use applications. Downstream datasets in the chemicals and batteries sectors have consequently been updated to use these more specific lithium products. Together, these changes improve the representation of lithium-intensive value chains and provide users with more geographically and technologically differentiated data for modeling applications such as electric mobility, grid storage, and battery supply chains. Part of the new lithium data was developed in partnership with BRGM (Bureau de Recherches Géologiques et Minières), including datasets for Chinese spodumene-based production and selected South American brine production sites.
The aluminium value chain also benefits from updates in the fuels sector. The expansion and regionalization of calcined petroleum coke (CPC) production and market datasets improve the geographical representativeness of CPC inputs to aluminium anode production. Previously, CPC inputs were linked to the Chinese market as a proxy, regardless of the geography of the anode production dataset. They are now linked to the newly introduced CPC markets, reflecting regional supply mixes and improving the representativeness of downstream aluminium datasets.
Two additional technical corrections were implemented to address issues identified through user feedback and internal review, concerning uncertainty parameterization in an aluminium hydroxide dataset and an incomplete consequential market for aluminium cast alloy.
The pulp and paper sector has been enhanced with new and updated data on carton and paper production in Europe.
Corrugated board production data was updated with 2024 production-year data, provided by FEFCO (European Federation of Corrugated Board Manufacturers) in collaboration with RISE (Research Institutes of Sweden).
Kraft paper and paper sack production datasets were likewise updated with 2024 data, and a new dataset, “kraft paper production, for bags and wrapping”, was introduced to represent machine-glazed and machine-finished kraft paper used in applications such as grocery bags and wrapping. This data was provided by CEPI Eurokraft and Eurosac in collaboration with RISE.
In addition, new datasets on cartonboard production and conversion were introduced, covering both virgin and recycled cartonboard as well as the conversion of cartonboard into folding cartons. Unlike the previous cartonboard datasets, which combined production and conversion into a single activity, these new datasets separate the two processes, so cartonboard production and folding carton conversion are now modeled independently. The data for cartonboard production were provided by Pro Carton in collaboration with RISE.
The ecoinvent release v3.13 strengthens the coverage of Tier 4 textile data (raw material production & primary processing) for cotton and flax fibre across main producing regions.
New cotton data covers conventional and organic seed-cotton cultivation at the field level, as well as subsequent ginning, which separates the fibre (lint) from the cottonseed. Conventional production is represented for Brazil, China, India, Türkiye, and the United States, accounting for almost three-quarters of global production. Organic production is represented for Brazil, India, Peru, Türkiye, Tanzania, and the United States, covering almost three-quarters of global organic production. This project forms part of an ongoing collaboration with Textile Exchange to improve the availability and quality of textile data for the apparel and footwear industry. It is the first in a series of new datasets aimed at addressing critical gaps for textile materials.
Furthermore, the global datasets for seed-cotton cultivation and ginning (for both conventional and organic production) have been updated as weighted averages based on the new regional datasets.
Coverage of the European flax supply chain has also been extended with the publication of a dataset modelling the rescutching of short flax fibres. Rescutching is an important step in the flax-linen cottonization pathway since it prepares short flax fibres by improving their uniformity and cleanliness as required for subsequent textile applications. The dataset is based on primary data collected by our partners, the Alliance for European Flax-Linen & Hemp. The European flax-linen supply chain in ecoinvent is representative of flax fibre products certified under Masters of FLAX FIBRE™. As of 2026, Western Europe remains the dominant global producer of fibre flax, with France, Belgium and the Netherlands growing and processing threeßquarters of global production.
The updates of the waste management and recycling sector include some minor technical corrections and the introduction of two new waste streams with their respective primary and secondary treatment processes. These changes are part of the effort to enhance the reliability and coverage of the waste inventories.
In version 3.12, the EN15804 inventory indicators “Hazardous waste disposed (HWD)” and “Non-hazardous waste disposed (NHWD)” were added to many datasets to achieve consistency. However, wrong tagging of some datasets led to double counting of these indicators in some waste streams. The issue was amended in version 3.13, achieving internal consistency.
A few datasets concerning the disposal of residues generated during coal mining have been corrected. An underestimation was identified in the conversion of contained phosphorus to phosphate emissions, and in some cases also in the conversion of nitrogen and sulfur content to nitrate and sulfate emissions; this has been corrected in version 3.13.
Finally, the waste sector now includes primary and secondary end-of-life treatment activities for two newly introduced waste flows: waste solvent, treated via incineration, and wastewater from pharmaceutical production, treated via wastewater treatment. (For the corresponding recycling and recovery processes for solvents, please refer to the newly introduced pharma sector.)
For v3.13, we added the GHG breakdown as in the GHG Protocol for the IPCC 2021 method. Furthermore, the IMPACT World+, footprint version method was updated from v2.1 to v2.2.1. Additionally, several minor fixes and improvements were implemented. Finally, characterization factors for new elementary flows were added. For more details, consult the Report of Changes.
As with every release, the existing supply chains in the database were thoroughly reviewed to reflect the current reality of global and regional product flows. As a result, links between demanding and supplying activities have been updated, and new regional market datasets have been introduced.
All products and elementary exchanges with mass in ecoinvent version 3 come with at least six properties: dry mass, wet mass, water in wet mass, water content, carbon content fossil, and nonfossil. Additionally, every single product in the database has a price that can be used, among other things, for economic allocation.
The ecoinvent database is the only major LCI database supporting four system models: Allocation cut-off by classification, Allocation at the Point of Substitution, Consequential, and Allocation, cut-off, EN15804. In addition, for all version 3 users preferring to have access to unlinked unit processes, ecoinvent offers an undefined version on its website.
The documentation for v3.13, including the Database Overview File, Report of Changes, Correspondence File, and LCIA Implementation Report, can be viewed and downloaded on our Knowledge Base.
More information about v3.13’s files and supporting documents can be found on our support page, Files in ecoQuery.
The Database Overview file describes the contents of the database. The following information is contained in the file:
Activity overview(s)
The lists of datasets contained in each of the four system models and the list of unlinked and unallocated datasets.
ecoinvent Geographies
The geographies used in the ecoinvent database. Each geography is assigned a classification, and information regarding how geographies overlap is provided.
LCIA Methods
The list of the LCIA methods for which ecoinvent calculates impact scores. The version of the methods is indicated, and the original source used for defining the characterization factors is provided.
We will host two webinars this month to provide a technical introduction to version 3.13. Sign up now to learn exciting details from each sector. We will finish each webinar with a Q&A session.