Fixing the inconsistencies between US LCI and OpenLCA impact methods

Question

Let me know if this issue has been brought up before. There appears to be a huge disconnect between how flows have been defined in many of the US LCI unit processes. There are many flows (I have mostly been looking at energy resource flows) that are not in the impact methods database. The result is that if, for example, the crude oil resource flow in the unit process is not in the CML impact assessment method, the impact method will report a ZERO value for depletion of fossil fuels. In fact, this happens a lot. This is very dangerous. I would guess that this is more related to VERY inconsistent data creation in the US LCI datasets. Nevertheless, it is a huge problem, rendering the use of impact assessment virtually useless for this database. I have seen the same thing for GWP100 numbers.

My quick fix has been to go through the inventory for a given system, and individually look for missing flows in the impact method, and enter them myself.

Are there plans for addressing this problem? Indeed, I teach two LCA related courses and could put some student-power to work on this issue, but would like thoughts from the community about how best to do this. My intuition is that the problem needs to be fixed by renaming flows in the US LCI database. But the issues may go both ways.

I really like OpenLCA and the idea of free, transport data such as US LCI. And I would love to help address these issues.

Answer 1

Hi, thank you very much indeed for your post. We have thought of providing the US LCI database for openLCA, for quite some time, but did not do so since there are, as you say, really a number of issues that we felt would need to be fixed before. Of course the database is available in EcoSpold format and can as such be imported. I think it would be excellent if we could collaborate in refactoring the database and in finally making it available for openLCA. Please contact me directly and we can discuss details. Thanks a lot again!
Andreas

Comments

Hi,
I sometimes use US-LCI datasets in modelling; it`s particularly useful for certain raw materials if they come froma US source as well as for US transport. I agree that it would be nice to have the US-LCI database easily available for openLCA. I wanted to add here that my preference is generally to retain flownames as they appear in the original database and extend the list of flownames that are captured by the LCIA method. Then if I need to make an assumption in the translation - like an energy content for "crude oil" to capture it in the Cumulative Energy Demand IA method, it`s very visible with the other Characterisation Factors.
Chris