The Importance Of LCA Data Standards And Alignment
LCA has existed for several decades now. But still, there are no clear standards for technical issues, such as substance nomenclature and data quality in LCA, among other. Every now and then, initiatives arise with a new passion to tackle this problem. Right now, several powerful initiatives are doing just that. This article explores the status quo of LCA data standards at this moment, and shows you the latest developments in changing it.
As a relatively young member of LCA community, born the same year that PRé was established, I took it for granted that a consensus on many technical issues would already have been reached. Working with data conversions and implementing databases and impact assessment methods in SimaPro, however, very quickly brought me down to earth. Even in 2016, access to substantial amounts of data is limited because there is no consensus on the technicalities. This really made it clear why interoperability is such a desirable adjective to describe LCA databases.
Past efforts to agree on dataset format or substance naming have not resulted in one widely accepted solution, but in several alternative solutions. The lack of harmonization all those years has led to quite a frustrating situation, where data from different sources are hard to combine in one study and much time is spent on conversions, which cause loss of information. However, now we are in a stage of dynamic development of LCA data (for instance, national databases and PEF data), the alignment initiatives are building momentum once again. This drive is fuelled by the consequences that not reaching a consensus may have – a proliferation of standards that aim to tackle the same issue. At worst, they are contradictory. In any case, they cause further fragmentation, push practitioners away even further and make access to data and data exchange even more complicated. We already see examples of new data formats created by countries entering the LCA community, which never made a choice to follow one solution.
What’s In A Name?
The differences and inconsistencies in substance naming in LCA may not be obvious when you are using datasets in software such as SimaPro. They usually show up at an earlier stage. owHoewever, the differences are significant and can lead to errors in database implementation. This problem would be nipped in the bud if the whole LCA community recognized and integrated a generic list of flows, something that the Global LCA Data Access network (GLAD) is advocating for. GLAD, an intergovernmental initiative, aims to align national and commercial databases. Their Nomenclature working group is striving to reach consensus on naming the flows.
Just few months ago, UNEP, which chairs the GLAD initiative, hired a Life Cycle Assessment Nomenclature Consultant to support the working group, consolidate its fruits into “a single mapping file for Life Cycle Inventory (LCI) datasets” and provide a critical review of existing nomenclature systems. Expected to be delivered in October this year, the mapping and review will be a strong foundation for harmonizing the nomenclature in LCA databases.
This is a positive development. But how will this affect initiatives using other nomenclatures? JRC’s Life Cycle Data Network, for example, welcomes data from any data developer as long as it fulfils entry-level requirements such as following the ILCD nomenclature.
Data Quality Has A Similar Problem
Inconsistently developed databases that do not have an indication of dataset quality can be hard to use and easy to question. We would prefer to have that the other way around. It is often mentioned how important high-quality data is for a robust LCA, but there is no consensus about quality criteria.
The goal of evaluating datasets according to concrete criteria would not be to label any developers as incompetent but rather to highlight the strong points of each database and to assure proper application of data in relation to the goal and scope of the study. As with nomenclature, the lack of alignment on data quality has led to several independently developed data quality systems.
Data Quality Standards Are Being Developed
The experience from those systems is now being aggregated by UNEP SETAC, which have flagship activity dedicated to Data and database management. The aim is to utilize the recommendations from Global Guidance Principles for LCA Databases in a more operational manner. According to the authors, one of their motivations is to answer requests in this area from countries setting up their national databases and looking up to more experienced LCA players for support.
A draft set of criteria have already been used for first assessments. After a round of feedback and improvements, the criteria will be applied to Australian, Chilean and Thai datasets. The longevity of this initiative will be assured by creating a network of database managers and trainers to support the implementation of quality criteria in the future.
Environmental Protection Agency Offers High-level Guidance On Data Quality
The issue of data quality criteria also caught the attention of the US Environmental Protection Agency (EPA), which is also involved in the UNEP/SETAC activities in the data quality field. In June 2016, EPA published its Guidance on Data Quality Assessment for Life Cycle Inventory Data with an updated set of data quality indicators at both the flow and process level. One of the reasons for improving the indicators was the fact that the results of quality assessments were not reproducible. EPA asked 12 LCA practitioners to assess the quality of the same dataset using different data quality systems and the consistency of results was described as very poor.
The Guidance document also included interesting remarks about future developments – the initiative is looking to create indicators for the model level, aggregate of data quality scores and score the match of elementary flows in the inventory and impact assessment method.
Slowly Working Toward Consensus
As has become clear, working towards a consensus on data nomenclature and quality standards is recurring issue. Data standardization would provide huge strides in efficiency, interoperatbility and quality assurance. Working with standardized data will make it easier for LCA practitioners to assess, trust and use data coming from different sources, regardless of the software used to do the LCAs. This much-needed consensus is not there yet. But this time, initiatives have decades of experience at their disposal.