they combine disparately methodized LCA data. this is explicitly against good practice. the fact that they found outrageous disparities got them great headlines and impressive graphs, but the underlying science is questionable at best. I could go deeper but this is the thousand foot view.
of course it’s not. Meta analyzes fly in the face of the guidance for LCAs. it’s just not good science.
since I’m already being tasked to address this again, it’s worth pointing out that poore and nemecek didn’t even gather the LCA data themselves. they, themselves, actually cite other meta-analyzes of LCA data. those meta-analyzes do recognize that they are violating best practices in the text themselves, and just go ahead and do it anyway. egregiously, poore and nemecek Don’t even acknowledge this faux pas and pass off their “findings” as sound investigation.
To elaborate and give a few exmples, LCA data is highly specific to a single production process, and might cover entirely different things.
There’s a huge difference between “one liter of paint from prepared from pigment and solvent” and “Me driving over to get a house sanded and cleaned, then repainted, per square meter of wall”. But both are LCA’s for painting, but the latter will be much higher.
It can go the other way too. There are also lots of sub-processes that have negative costs. Putting up a new streetlight has a environmental higher cost than replacing one, because replacing one gives you an old streetlight to recycle. You can’t just create a pile of “streetlight LCA data” and take the average.
They can even be very time-specific. If I’m sitting on a giant mountain of gravel, I can give you an LCA for your zen garden that’s much lower than last year when I had to import gravel from Norway.
Looking at chocolate here, they include lots of land-use-change, which is caused by cocoa farmers expanding and turning trees into cocoa farms. But that’s only because they’re expanding. The next harvest won’t have that change.
can you expand on that for those of us who don’t wanna Google and confirmation bias ourselves?
they combine disparately methodized LCA data. this is explicitly against good practice. the fact that they found outrageous disparities got them great headlines and impressive graphs, but the underlying science is questionable at best. I could go deeper but this is the thousand foot view.
It’s a meta analysis, so I’m not sure it would be possible to get identical methodologies for all data sets.
of course it’s not. Meta analyzes fly in the face of the guidance for LCAs. it’s just not good science.
since I’m already being tasked to address this again, it’s worth pointing out that poore and nemecek didn’t even gather the LCA data themselves. they, themselves, actually cite other meta-analyzes of LCA data. those meta-analyzes do recognize that they are violating best practices in the text themselves, and just go ahead and do it anyway. egregiously, poore and nemecek Don’t even acknowledge this faux pas and pass off their “findings” as sound investigation.
To elaborate and give a few exmples, LCA data is highly specific to a single production process, and might cover entirely different things.
There’s a huge difference between “one liter of paint from prepared from pigment and solvent” and “Me driving over to get a house sanded and cleaned, then repainted, per square meter of wall”. But both are LCA’s for painting, but the latter will be much higher.
It can go the other way too. There are also lots of sub-processes that have negative costs. Putting up a new streetlight has a environmental higher cost than replacing one, because replacing one gives you an old streetlight to recycle. You can’t just create a pile of “streetlight LCA data” and take the average.
They can even be very time-specific. If I’m sitting on a giant mountain of gravel, I can give you an LCA for your zen garden that’s much lower than last year when I had to import gravel from Norway.
Looking at chocolate here, they include lots of land-use-change, which is caused by cocoa farmers expanding and turning trees into cocoa farms. But that’s only because they’re expanding. The next harvest won’t have that change.