python - Brightway2 : import Simapro dataset with water resource locations

Question

I'm importing a Simapro dataset and I plan to resolve the unlinked exchanges by applying a strategy that translate them into existing ones in "biosphere3" database.

The problem is that some water exchanges have specific locations like :

Water, well, BR (resource:in water)
Water, unspecified natural origin, OCE (resource:in water)

example of simapro exchanges and their translation

I would like to keep the specific resources/emissions locations and potentially use them with regionalized LCIA methods like Impact World+.

I know how to specifiy location to technosphere dataset but not to a biosphere dataset. How could I do that ? With "bw2-regional" (https://brightway2-regional.readthedocs.io/index.html) ?

question from:https://stackoverflow.com/questions/65926862/brightway2-import-simapro-dataset-with-water-resource-locations

Answer 1

I have tried something similar in the past. I can confirm, that you can perform regional LCIA calculations using the brightway2 package (without using bw2-regional package). I have done so for a regional water stress method. However, this entails quite a bit of work. In the following a rough description of what I did (feel free to ask more specific questions):

1. Import regional lcia method

As biosphere3 does not contain regional (water) flows, I first imported my regional lcia method. I created my own importer, but you can probably also create one based on the LCIAImporter class, which will allow you to use the add_missing_cfs() function to add new biosphere flows for all the regional characterisation factors of you method. For biopshere flows you can specify the location by adding a new 'location' key to the dictionary object to store the regional identifier.

2. Import simapro database

Only after your biosphere contains all the regional flow you want to keep in your simapro database, you can try to create a new strategy that links the flows of your database to those you created in your biosphere. Note that the SimaProCSVImporter performs many strategies that will change the biosphere flows in you LCI, which would have to be altered (but I suspect this is the part you already worked on).

Alternatively, you could also import you database using the current brightway2 functions. Afterwards you could try to re-link the biosphere flows you want to regionalise to those new ones you added to your biosphere. The wurst package could come in handy.

bw2-regional

With regard to the bw2-regional package I have no experience working with it, but I think it allows for much more complex regional LCA calculations than what you might be used to from simapro.