LCA Databases

Key concepts

Life Cycle Assessment (LCA) is a standardized methodology (ISO 14040 and ISO 14044) for evaluating the environmental aspects and potential impacts associated with a product, process, or service throughout its entire life cycle. Often referred to as the "cradle-to-grave" approach, LCA tracks resources and emissions from raw material extraction, manufacturing, transportation, use, and end-of-life disposal or recycling.

Key steps

1. Goal and scope definition: clarify the purpose of the study, the system boundaries, the functional unit, and the level of detail required.
2. Life Cycle Inventory (LCI): collect and quantify inputs (energy, materials) and outputs (emissions, waste) for the system under study.
3. Life Cycle Impact Assessment (LCIA): translate LCI data into potential environmental impacts (e.g., global warming, acidification).
4. Interpretation: analyze the results, identify data limitations, and make recommendations.

Through these steps, LCA enables informed decision-making by highlighting the most significant contributors to environmental impacts and helping identify improvement opportunities.

Life Cycle Inventory (LCI)

• In a LCI, every process is modeled by tracking:
  • Technosphere flows (goods, services, and intermediate products exchanged among processes within the economic or industrial system).
  • Elementary flows (resources taken from the environment and emissions or wastes returned to the environment). Elementary flows in LCA are especially important because they are the basis for calculating potential environmental impacts (e.g., global warming potential, eutrophication, acidification) in a Life Cycle Impact Assessment (LCIA).
• When compiling an LCI for a product or service, you gather data on all flows in and out of each process.

---

Ecoinvent 3.10 attributional ('cut-off')

What is ecoinvent?

• ecoinvent is one of the most widely used life cycle inventory (LCI) databases worldwide.
• It is developed and maintained by the ecoinvent Association, based in Switzerland, in collaboration with a network of experts and research institutions.
• ecoinvent provides transparent, consistent, and high-quality LCI data covering a broad range of sectors: energy, materials, transportation, agriculture, chemicals, waste treatment, and more.

System models

• ecoinvent offers different system models to address different methodological needs:
  1. Attributional ("cut-off") model: assigns environmental burdens to the processes generating them, typically excluding burdens or credits from recycled materials that appear "beyond" the product's life cycle.
  2. Consequential model: uses different basic assumptions to assess the consequences of a change in an existing system, used for studies aiming to model potential future scenarios or "what-if" situations.

Database structure

• Datasets are typically compiled at the unit process level (i.e., a specific manufacturing step or operation).
• Each dataset includes inputs (materials, energy), outputs (products, emissions, waste), geography (country or region), technology descriptions, and time references.

What is the technosphere matrix?

• In ecoinvent (and in life cycle assessment in general), the technosphere matrix is the structured representation of all exchanges (flows) of goods and services between processes within human-made (technical) systems.

---

Specificities of Ecoinvent 3.10

Release context

• Ecoinvent 3.10 was released in late 2022.

Key updates and improvements

1. Expanded and updated datasets

   • Energy and fuels: inclusion of updated data for electricity mixes in various regions, refined data for conventional and renewable energy technologies, and improvements for fuels (e.g., natural gas, hydrogen production) to reflect evolving energy markets.
   • Mobility and transport: refined datasets for road, rail, and maritime transport, often incorporating newer vehicle technologies or more detailed emission factors.
   • Construction and materials: improved datasets for key construction materials (e.g., concrete, steel, insulation), ensuring more consistent and up-to-date inventory data.

2. Data quality enhancements

   • Cross-checking and harmonization of data sources, improving consistency in modeling assumptions across different sectors.
   • Refinement of background data (e.g., chemicals, intermediates) to correct or update emission factors and resource usage.

3. Regionalization

   • Continued efforts to make data more region-specific — improving geographical resolution where possible.
   • Helps users capture differences in energy mixes and technology efficiencies between countries or regions.

4. Documentation and transparency

   • Detailed documentation accompanying each dataset, describing data sources, assumptions, and system boundaries.
   • Efforts to resolve known data gaps or to correct errors identified in previous versions, leading to more robust and transparent datasets.

How Darwin Uses ecoinvent

Product Data Processing

When you enter product data (e.g., "150 tonnes of paper"):

1. Product matching: Your product is matched to ecoinvent activities
2. Life cycle inventory: All inputs and emissions are calculated
3. Pressure extraction: Environmental pressures are aggregated
4. Geographic refinement: Region-specific data is applied when available

Commodity Data Processing

For commodity data (e.g., "50 tonnes of copper"):

1. Commodity mapping: Direct match to ecoinvent elementary flows
2. Production processes: Extraction and processing included
3. Supply chain: Upstream impacts captured

Product Categories

Examples of ecoinvent coverage:

---

Agribalyse

What is Agribalyse?

• Agribalyse is a national Life Cycle Inventory (LCI) database focused on agricultural and food products, developed and maintained by ADEME (Agence de la transition ecologique) in collaboration with INRAE, the French agricultural research institute, and a network of technical institutes and industry experts.
• It is specifically designed to assess the environmental impacts of French agricultural production systems and food products, from primary production to the farm gate or the point of sale.
• Agribalyse provides transparent, sector-specific LCI data covering a broad range of food categories: cereals, fruits and vegetables, dairy products, meat, fish, processed foods, beverages, and more.

System models

• Agribalyse is built on an attributional approach, consistent with the cut-off model used in ecoinvent. Environmental burdens are assigned to the processes generating them, following the physical allocation principles recommended for agri-food LCA.
• Background processes (energy, transport, packaging materials, etc.) are modeled using ecoinvent datasets, ensuring consistency and interoperability between the two databases.

Database structure

• Datasets are structured at the unit process level, covering specific agricultural operations (e.g., wheat cultivation in France, milk production in Brittany) or food transformation steps.
• Each dataset includes inputs (seeds, fertilizers, pesticides, water, energy), outputs (agricultural products, emissions to air, water and soil, waste), geography (French regions or national average), technology descriptions, and time references.
• Agribalyse is organized around two main layers:
  • Agricultural products: raw commodities at the farm gate (e.g., raw milk, fresh tomatoes, wheat grain).
  • Food products: processed and packaged products available at the point of sale (e.g., pasteurized milk, canned tomatoes, bread).

Interoperability with ecoinvent

• Agribalyse uses the same elementary flow nomenclature as ecoinvent, enabling direct combination of both databases within a single LCA study.
• Background system data (e.g., electricity mixes, transport, chemicals) are sourced from ecoinvent, making Agribalyse particularly well-suited for studies where French or European agri-food supply chains are central.

Specificities of the Agribalyse version used in darwin

Version context

• Darwin uses Agribalyse 3.1, released in 2022, which is the most comprehensive update of the database to date.

Key updates and improvements

1. Expanded product coverage

   • Significant increase in the number of datasets: Agribalyse 3.1 covers more than 2,500 food products, including a greatly expanded range of processed and packaged items representative of French consumption patterns.
   • New datasets were added for aquaculture, organic agriculture, and several underrepresented food categories (e.g., legumes, nuts, spices).

2. Improved agricultural modeling

   • Updated agronomic data reflecting current French farming practices, including revised fertilizer application rates, pesticide use, and yield data sourced from national agricultural statistics.
   • Better representation of soil carbon dynamics and direct/indirect N2O emissions from agricultural soils.

3. Refined food transformation processes

   • More detailed and up-to-date datasets for industrial food processing operations (e.g., milling, pasteurization, fermentation, freezing), incorporating recent energy consumption benchmarks.

4. Enhanced regionalization

   • Improved geographical resolution for key agricultural products, distinguishing between production regions within France where data availability allows.
   • National average datasets remain available for products with insufficient regional data.

5. Updated background system

   • Background processes are aligned with ecoinvent 3.8, ensuring consistency in the modeling of energy, transport, and ancillary material flows.

6. Documentation and transparency

   • Each dataset is accompanied by detailed documentation describing data sources, modeling assumptions, allocation choices, and system boundaries.
   • Known data gaps and uncertainty estimates are explicitly reported, supporting robust sensitivity analyses.