When deciding to get started with Life Cycle Assessment you probably have come across the quote: “You can’t manage what you can’t measure”. Therefore, your aim is to quantify the environmental impacts of your company or of one or several products with the objective to develop effective management strategies to reduce these impacts. Companies now utilise a range of environmental methods, such as Life Cycle Assessment (LCA), Material Flow Cost Accounting (MFCA), Carbon or Water Footprinting, Environmental Performance Monitoring, and more general Material and Energy Flow Analysis.
Performing such assessments can be a challenging task, when you do it the first time, even with the application of specialized and professional software tools. Instead of totally starting from the scratch professional software tools are very helpful to build on the expertise and knowledge of others to fit the specifics of a company’s products and processes. Experience shows that when starting with LCA 10-20 % of the user’s effort is needed for setting up the life cycle model of the production system and 80-90% of the time is used for collecting, finding, identifying, validating data.
What does this mean?
After defining goal and scope for your LCA project you get started with identifying which processes belong to your core or foreground system. For these processes, it is required or at least recommended to use primary data from the production system. For goal and scope settings that include the identification of reduction and improvement potentials in a specific product and service system primary and specific process data can always be considered preferable over secondary or generic data.
Certainly all LCA practitioners will agree that there are limitations in collecting or accessing primary data for a Life Cycle Model. This especially applies for data from life cycle phases which are not under the operational control of the organization initiating the LCA project, e.g. suppliers or downstream activities, occurring after the use phase, e.g. end of life treatment or recycling. One way to get secondary data is by asking suppliers or end of life service providers for inventory data related to the products and services that form part of a product’s life cycle. This approach still leaves data gaps open. To close these remaining data gaps life cycle inventory data bases – mostly referred to as LCA databases – become the important data source. If we want to be more accurate in using the expressions “LCI database” or “LCA database”, LCI databases provide Life Cycle Inventory datasets, while LCA databases include in addition Life Cycle Impact Assessments methods. In common language the term LCA databases is used more, without having this distinction in mind.
LCI databases provide data on energy, material and emission flows associated with a specific component, process or product. These datasets allow LCA practitioners in industry and research to fill the gaps in their own inventories. Especially in research projects without the involvement of a specific company and even more limited access to primary data the researchers are enabled to populate models from scratch. Using mostly generic data in an initial stage of creating a life cycle model can also help to set the right priorities for the data collection to avoid collecting the most detailed and accurate data on the least contributing processes of the life cycle model. There are a range of LCI and LCA databases available on the market.
LCI databases can be available as international, regional or national collection of datasets. There are a range of regional and national databases such as the USLCI, European Life-Cycle Database (ELCD), or the Australian version (AusLCI). These options can be sufficient if a company’s full supply chain (processes, products, components) occur in a single country. However, supply chains are predominantly transboundary and global—it’s unlikely that all components used in the manufacture of a product will come from a single nation. To transparently and fully capture the regional differences across the supply chain, it’s recommended you use regional specific LCI dataset, these can be found in global databases covering a wide range of regions or in regional or local databases.
There are two key market-leading international LCA databases: ecoinvent and GaBi Databases that both strive to cover more and more regions, partly by integrating regional database, partly by putting effort in data collection with regional partners. There are two reasons for taking this effort:
- Meeting the demand for regionalized data from the emerging LC communities around the globe.
- Meeting the demand for regionalized data of industries operating global supply chains.
At ifu Hamburg we work with both databases: ecoinvent v3.2 (with access to ecoinvent 2.2 as well) can be used in the LCA-Software packages Umberto NXT LCA or Umberto NXT Universal ; GaBi Professional and more than 20 GaBi extension databases are also available.
With both options accessible, how can you decide which database to work with?
ecoinvent and GaBi Databases are both market-leading, internationally-recognised LCA databases, so either would be a recommended choice. Your decision probably depends on your desired application, the specific data requirements of your processes, and the database you’ve used previously (if you have done so). Although the features and strengths of both are very similar, it’s useful to outline the key points of each, and any differences—this may determine the most suitable LCA database for you. What are their similarities?
- Relevance: Both databases provide the most comprehensive range of LCI datasets in the LCA field. ecoinvent is now on its third version with over 10,300 LCI datasets; GaBi Databases similarly has more than 8,000 profiles of materials and processes. Both largely cover the relevant areas including energy supply agriculture, transport, biofuels and biomaterials, bulk and specialty chemicals, construction, wood and waste treatment. Both provide access to commonly used and up-to-date LCIA methods.
- Up-to-date: ecoinvent and GaBi Databases are both updated (with new or amended data) on a yearly basis, energy mixes, technology and supply chain data are available.
- Data quality: Both provide high-quality datasets which have been compiled using primary industrial data collection. ecoinvent has data consistently supplied and reviewed by international research institutes and LCA consultants; GaBi Databases are continually reviewed by the global verification company DEKRA.
- Global/international: provide comprehensive global and regional datasets allowing users to more accurately reflect their integrated supply chains in their flow models. This global supply chain approach allows these databases to cover a wide range of markets and products.
As you can see, the key features and strengths of each are very similar. What are a few of the differences?
- Unit vs Process level data: ecoinvent provides data at both unit process (UPR) and system process (LCI) levels meaning the user can disaggregate a product’s process chain into its smaller units if required. Most datasets in GaBi Database are provided as aggregated process data (without a possibility to disaggregate these by the user).
- System models: in ecoinvent 3 the datasets are provided using 3 different system models: the Consequential System Model, the Cut-off System Model and the Allocation at the Point of Substitution System Model. The different system support different goal and scope settings of an LCA study. In GaBi database no distinction of system models exists.
- Sector-specific Extensions: a distinguishing feature of GaBi Databases is that it provides sector-specific GaBi Extension data bundles, which offer an advanced option for LCA professionals by covering a broader spectrum of data. In addition, the GaBi-Data team also offers a data-on-demand service.
- Quicker Updates on LCIA methods: observations from the last years show that the makers of the GaBi database are quicker in providing the latest developments of LCIA methods.
To get a better insight into the database content in order to select as your main LCI database, both database providers offer to search the content online free of charge. GaBi dataset content you can search online directly without registration via the GaBi database search. For ecoinvent you have to register as a guest and then get access to the search on the meta-information. It has become a common practice to select one, either GaBi databases or ecoinvent, as a main LCI database and complement it in case of remaining data gaps with the other. In Umberto NXT LCA and Umberto NXT Universal, both ecoinvent v3 and GaBi databases can be used in parallel. This gives users the freedom to select the most appropriate database to use on a case-by-case basis depending on their specific requirements.
The most important question to answer when deciding on data to use in a LCA project: Are the data I use “fit for purpose”? Goal and Scope or short “purpose” differ. That’s the reason why at ifu Hamburg we do not give an easy and quick answer on “which database is better?”. We always want to understand the purpose first, before entering in a conversation guiding the decision to one or the other database.
What users can expect from GaBi databases and ecoinvent in the future I asked Dr. Martin Baitz from Thinkstep and Dr. Gregor Wernet from the ecoinvent center. For their willingness to collaborate and answer my questions, I would like to thank both of them!
What’s missing? Right, the users’ perspective! We definitely would also like to give the word to users of LCA databases: I asked three experienced LCA practitioners about their tips and experiences related to LCA data they could share with readers of this article new to the LC Community. Thanks a lot to Angela Fisher from GE in the US, Enrico Benetto from LIST in Luxembourg and Gaël Fick from IRT-MP2 in France!
Read the interviews with LCA data users.
Your comments, your own experiences and your questions are highly welcome.