Problems around database standards, nomenclature and interoperability have been plaguing the life cycle assessment (LCA) community since its very start. It’s not been easy to come to an alignment. After many initiatives and false starts, there may now be a solution to the confusion around LCA data standards: the Global LCA Access to Data (GLAD) initiative. How did we get there? Strap in and pay attention, because it’s been a bumpy ride.
Ever since people started to gather around the concept of LCA and since the first LCA studies emerged, in the late eighties, early nineties, people have been discussing LCA data standards for database development, preferably efficient ones.
Several developments have shaped this landscape:
- From 1992-1995, the Society for the Promotion of Life-cycle data (www.spold.org) tried to organise the data flows. The Society was founded in the Nordic countries by companies and researchers, and created the first Spold data format.
- In 1996, a Swedish coalition under de name SPINE proposed a very detailed format, describing how to store all data in a consistent way. It was a great idea, with the unfortunate drawback that all software developers would have to invest a lot of effort in following these rules.
- Around the same time, the UNEP-LC initiative agreed on nomenclature.
- In the late nineties, Ecoinvent extended the Spold format and named it Eco-spold. It also adopted the agreed nomenclature.
- In parallel, work was done on the ISO 14048 data format, intended to be a data exchange format. It looked promising, but in the final versions it was defined much less strictly, so it could not be used as a data exchange format anymore. It is now referred to as documentation format.
- Around 2000, it all seemed to come together. At SETAC in Lille, the six main software developers at that time came together to demonstrate how their software could import and export to and from Ecospold…. Problem solved! Of course, not really.
- The JRC announced it was going to develop a new format, and they also introduced a new nomenclature. To this day, I am still in the dark why this was done. Since that time, we have been beset by confusion and inefficiencies, as now we suddenly had two formats and two nomenclatures.
To sort out the mess and to put an end to many other unproductive discussions in the LCA community, the SETAC-UNEP Life Cycle Initiative initiated its first Pellstone style workshop in Shonan Japan , funded by the Japanese and US governments, not the JRC. 50 experts (or should we say trouble-makers) reached a remarkable high level of consensus: the Global Guidance principles on LCA.
Problem solved? Yes and no. Global Guidance is great, but it does not generate data and that is what we all need. An intergovernmental collaboration was initiated by the EU a few years later in 2013. It was partially inspired by the PEF pilots, as Europe would like to ensure that exporting countries would be able to provide data. After meetings in Brussels and Washington, the third meeting in Malaysia established an ambition to make the major databases interoperable. Three taskforces were set up, researching:
- System architecture that would allow for searching for and obtaining interoperable data
- Nomenclature: a standard list of substance names to identify resources and emissions
- Metadata: describing the required documentation fields that could be used by LCA practitioners to understand what the data was about, what it represents and for which year it is representative, etc.
The idea behind interoperability is that every database can be maintained independently. Countries should not judge the quality of other countries’ data, but the user should be able to identify if the data is “fit for purpose”. The metadata fields are to be used as selection criteria for the data.
The fourth meeting in Brazil established a really good common understanding. The data architecture was approved, and a call for tenders to develop the system was sent out. The figure below describes the architecture, developed by a group with Christoffer Krewer, Sangwon Suh and myself as co-chairs. It depicts a system where all databases are connected via an interface (API) to a central node, where users can search and download data.
At the same time, the nomenclature group, co-chaired by JRC and Ecoinvent, came up with a unified nomenclature. The metadata group is making good progress on the documentation fields, some required and some optional. Additionally, a Pellston workshop was organised recently in Spain, with the Flagship 1b project, booking good progress on global guidance and alignment.
Problem solved? This could very well be the solution, as we have now come to a real consensus among all parties, and the will to solve the issue seems to be there. Although the initiative started as an intergovernmental collaboration process, aiming at making the national databases interoperable, it has developed into Global LCA Access to Data (GLAD). GLAD does not only work with governments, but also with non-governmental and commercial players. This opening up could have far-reaching consequences, as there is no reason single LCA software users couldn’t become nodes, if they want to. This could develop into a real collaboration platform between LCA practitioners.
Here at PRé, we are of course following this development closely. We are already seeing so many users help each other with data in our LCI Discussion List. We also believe in the potential of people helping each other and initiating database initiatives on such a platform. If you would like to share your thoughts, or if you are interested in starting database developments, please send me an e-mail.