Inconsistencies in biosphere between 3.2 and 3.3 (ElementaryExchanges.xml)

[aleksandra-kim]

Original report by Adrian Haas (Bitbucket: haasad, GitHub: haasad).

---

I have mentioned this some days ago in a comment and now experienced the same thing in another place.

I made a short notebook to illustrate this.

• There are some strange biosphere exchanges (social, unspecified), that appear to be unused
• There is one biosphere exchange from 3.2 that is missing in 3.3 (Carbon dioxide, non-fossil, from calcination)
• One biosphere exchange (Water, unspecified natural origin) is missing in 'ecoinvent elementary flows 3.3.xml', but is a valid flow in cutoff 3.3.

As far as I understand it, this is caused by ecoinvent, but I still think a workaround in brightway would be nice. I could create a pull-request that contains the following: Either
a) manually update bw2io/importers/ecospold2_biosphere.py and bw2io/data/lci/ecoinvent 33 new biosphere.json with the two missing processes or
b) make bw2io/importers/ecospold2_biosphere.py load both xml files and return the union of the two and dynamically create bw2io/data/lci/ecoinvent 33 new biosphere.json

In case of b) this could be reused for the next version of ecoinvent. I'm not sure what the effect of such a fix would be for the LICA methods, maybe they also need to be adapted.

[aleksandra-kim]

Original comment by Chris Mutel (Bitbucket: cmutel, GitHub: cmutel).

---

Could you make a pull request with option (a)? I am already nervous about the size of the data files in this package, and it seems silly to add megabytes for a few flows.

As for LCIA methods, I don't think we need to worry - there weren't any LCIA CFs for "Carbon dioxide, non-fossil, from calcination" in 3.2.

[aleksandra-kim]

Original comment by Chris Mutel (Bitbucket: cmutel, GitHub: cmutel).

---

Fixed in 98a679c