Emerging International Consensus On A Truly Global Network of LCA Databases
An intergovernmental process has been initiated to solve the problems around the lack of availability of life cycle data once and for all.
Last April, a Washington, DC meeting of experts and policy makers made great progress towards developing a vision. I was one of the experts present, invited by the US EPA. Here is what is emerging.
Prelude To The Washington Meeting
Countries from all over the world are seeing the need to solve the problem of a lack of truly global LCA databases, as all countries experience great difficulty implementing lifecycle-based policies. Different approaches are being developed to resolve this problem, such as the Japanese EPD scheme and the European PEF pilot. This process started back in 2011, when 50 experts gathered in the Japanese village of Shonan for one full week on invitation of the UNEP-SETAC Lifecycle initiative, to work out a global consensus on databases. The result was the Global Guidance Principles for LCA Databases, a report that is still relevant today.
The most contentious issue at the time was determining how this future ‘global database’ would be managed (see also Chapter 6 of the Global Guidance Principles). Some countries advocated something resembling like an open source model, while others definitely did not want this. I was rapporteur in the subcommittee tackling this question, and we had to resort to very diplomatic language. The Shonan group described three possible futures: a state-controlled model, a completely open-source model such as Wikipedia, and a compromise model, not unlike ecoinvent. We could not even number these scenarios as 1, 2 and 3, as that would reveal a preference, so we resorted to naming them scenario L, C and I. We developed policy recommendations that could be used by any country to increase the chance that their preferred future becomes reality.
Our Vision For Global LCA Database Development
The meeting in Washington, DC stood in sharp contrast to that Shonan meeting. Thanks, in part, to excellent facilitators Patrick Tallerico and Troy Hawkins, we were all surprisingly constructive and we painted the vision summarised in the graph below.
There are approximately 60,000 LCI processes available to the LCA community at large. UNEP has found them in over 20 databases and half of them use the ecospold format. Beyond that, we all agreed there are still large untapped data resources in national databases and statistics. The most important source of data is the emission registry system that contains mandatory annual emission reports from companies.
While it easy to agree we all want more data, it was more difficult to agree on who determines what is good and bad data. Before we knew it, we were arguing at length about this question. The solution that emerged is to have the gathering and reviewing of data done by separate groups:
- We will collaborate on getting as much data as possible into a network of ’raw’ unverified data. Two strict requirements apply:
Data should be disaggregated as much as possible (into unit processes)
Data should have good characterisations and descriptions (metadata: data about the data)
- Each country or database provider is free to select, review and verify from this pool of raw data according to its policy or commercial objectives. This can only be done if the metadata is complete and reliable.
The benefit of this split is that we do not need to agree on one LCI methodology and one policy objective. France, for example, can determine its own selection, just like Japan or ecoinvent or any other organisation that wants to manage and build a database.
Although it is unclear yet which countries would like to engage and fund this system, for the first time we have a shared vision to strive towards. This vision creates the clarity the LCA community has been lacking for so long.