LCA, asking the tough questions
Those reading this blog regularly - and they are hundreds of thousands of course - will notice that some time has passed since my last post. Well, the attentive readers will also notice that something has happened during this time that was quite a surprise and overshadowed all the rest.
However, these readers can rest assured that this blog will never host a post titled How COVID-19 changed LCA forever or something on this line. Hell no.
But, but but but, I must also admit that these have been months of intense reflection, between one video call with screaming kids on the background and another, and when not surfing the web while juggling, I mean working, from home.
Because indeed, the pandemic and the lockdown have been an occasion for revisiting habits and personal attitude towards life and work (yes there is life beyond work), and the meaning of all of these.
While it’s beyond my ambitions to write about the meaning of life, I can write some words about the meaning of LCA. Because when you reconsider stuff like the way you have meetings with people and how much you go to the office, why not reconsidering stuff like your research domain?
So this is a post asking the tough questions for LCA, the questions that I - at least, but perhaps others too - am sometimes afraid to ask, and reluctant to answer.
On the impact of LCA
Is LCA really useful? Is LCA really making a change in this world? Are LCA studies truly supporting decisions? Or are LCA studies just made with the purpose of adding a green stamp on decisions that others have already made? Are we doing LCA for the sake of LCA? Or even worse for the sake of a publication? Or even even worse for the sake of somebody in particular - a client, a supervisor, a friend? And worse of all, can we say our LCA is not greenwashing?
On the need for LCA
When is LCA needed and when some other tools are instead a better fit? When I review sci papers I get sometimes the impression that LCA is used to answer research questions that have very little to do with a life cycle perspective. If there is no trade off between the life cycle stages of a product to be studied, or between different impacts in a life cycle, why should we use LCA at all? For which questions is LCA able to provide sufficiently accurate answers, e.g. regarding the modelling of future scenarios, emerging technologies, prospective questions, do we need the quantitative assessment framework of LCA at all or are we better off using other and perhaps (imagine a face screaming in fear Much-style) qualitative ones?
On LCA research
And related to the previous point, focusing on the research side of it, what is the point of writing sci articles on LCAs of all possible products? One paper on the LCA of red wine, one on the LCA or white wine, one on the LCA of rosé wine, one on vacuum cleaners, four on washing machines, seven on bioplastics, etc. etc. Should we go on until we have exhausted all possible products, and then start again as some are outdated, and thus continue forever? Which cases and problems are “big enough” to justify research that uses LCA? Should LCA research be only about having better methods and tools, and very complex models taking months to develop, and should we leave the case studies to consultants? Would we then get “disconnected” from the real world if we don’t apply LCA to case studies? What about the study of how LCA is used by practitioners, is this a first world problem?
On the freedom of LCA
Oh freedom (freedom) - Oh freedom - Oh, freedom (yeah) - Oh ya got to have freedom… When different LCAs of the same product return different results, does it mean the tool is useless? Is modelling freedom in LCA excessive, and does a high degree of modelling freedom jeopardise the tool’s credibility? How much freedom to model should we leave to LCA practitioners? Is it sufficient to have transparency and reproducibility and everything will be alright? Taking it to the extreme, if I can make an LCA model of whatever I want and how I want, but fully transparent and reproducible, will it be accepted as LCA? And to the other extreme: to what extent should we standardise LCA practice, and would it help? Can we find a set of rules so strict that two LCAs by two different people will look identical, and is this a reasonable purpose at all? Is there a middle way?
On LCA sophistication
We can make more and more complex LCA models, with more variables, more data, and producing more results and faster. But are we getting new answers? And do we get more certain answers? This is a well-known trade off in modelling in general. And indeed valid for LCA too, but how much are we aware of this when we develop new and more fancy LCA databases and models, and how explicitly is this addressed? How deeply do we discuss issues related to increasing uncertainty and sensitivity to modelling choices, and how much can they invalidate LCA results? Do we need a better statistical stance to LCA? Is LCA just a black-box number-spitting machine?
Wrapping up
This post has tested my English skills in reversing verb and subject to the extreme. Is the cat in the sac?
I asked a lot of questions in this post, and provided no answers. How convenient. While some of these questions are provocative on purpose, and some might never find an answer, all are probably worth considering to ensure a meaningful future in this domain of research and practice.
Meanwhile, I am looking forward to the day when I will see a paper with both “LCA” and “COVID-19” on the title. Because, really, don’t we miss Assessing the impact of the COVID pandemic by means of LCA or a LCA of COVID-19 masks? Of course we do.