Differences between openLCA and ecoinvent IPCC 2021 methods

Question

Hi,

The characterization factors for carbon dioxide vary between openLCA 2.3.1 and ecoinvent 3.9.1 methods. In openLCA 2.3.1 version, Carbon dioxide, to soil or biomass stock is +1 whereas in previous versions and for ecoinvent 3.9.1 it is -1. Shouldn't it be -1, since this represents carbon uptake due to changes in land carbon stocks? This may result in drastically different results for bio-based systems. Please advise!

openLCA 2.3.1:

ecoinvent 3.9.1:

Answer 1

Hi, good question, this deserves indeed explanation, we have set impact directions in "our" categories, and with these set, the direction of the flow in the category does not need to be considered in the factor, but only the amount (for an output category, such as climate change, flows on the input side always are deducted from the LCIA result, while outputs add to the result). This does not need to be reflected with a negative sign. A negative sign is only needed if a flow on the same side as the impact category reduces the impact. Note that this works in openLCA 2.0.3 only (and future versions).

We kept the ecoinvent methods untouched therefore they still need the negative sign.

I hope this was clear? Thank you and best wishes,

Andreas

edit: I was incorrect for carbon dioxide to soil, please see comments below

Comments

Thank you for the response! I ran a test process with the three following flows as outputs: Carbon dioxide, to soil or biomass stock; Carbon dioxide, land transformation; Carbon dioxide, fossil. If I run results using the ecoinvent method, I get 1 kg CO2e. If I run results using the openLCa method, I get 3 kg CO2e. In this case the ecoinvent method is correct because Carbon dioxide, to soil or biomass stock is -1 and the other two flows are +1.

To get the correct results using the openLCA method, does the Carbon dioxide, to soil or biomass stock flow need to be added on the input side of the process? This means that we need to build models differently if we are using the openLCA or ecoinvent methods. Is that correct?

---

I think you are right and the 'carbon dioxide to soil' is actually a case where the same direction leads to a reduction of the impact, and we will need to update the method once more, since here a -1 is correct.. Thank you for your comment and sorry for the need to update the package once more..
Update: the new method package is now in Nexus, here: https://nexus.openlca.org/database/openLCA%20LCIA%20methods