Water use (ReCiPe) calculation with Ecoinvent

Question

Hi,

I'm working in openLCA with both Ecoinvent 3.10 and Agribalyse databases, using the ReCiPe 2016 midpoint (H) impact assessment method. I have some difficulties understanding how the Water use indicator is calculated with Ecoinvent. In Agribalyse, all water-related inputs and outputs are taken into account, and their difference (with the relevant characterization factors) is used to calculate the final value of the indicator. This way of working is straightforward. However, in Ecoinvent, it seems that the only elementary flow used for the calculation of the indicator is the emission of water in the air. Other water-related processes and flows do not show up in the impact analysis. As such, I have no clue how to model other kinds of water consumptions (water dumped into the sea/sewers, or integrated into products).

Any insights on how to properly model water uses other than evaporation in OpenLCA with Ecoinvent?

Thanks

Answer 1

It's true that the water use indicator in the original ecoinvent LCIA methods implementations is not very intuitive. However, it also gives correct unregionalised values. In cases like AWARE inside EF 3.1, the result will be not very useful and simply the net water use times the world average impact factor of 42.95, because it is unregionalised. Or in the ecoinvent implementation for ReCiPe it is simply a factor of 1.0 and the net m3 volume.

The reason why ecoinvent models the net water use as emission to air and not as input water flows minus output water flows has to do with possible distortions of the mass balance when allocation is applied. The positive water emission to air values are (in principle) already the net water use that remains if one would subtract the inputs and outputs. So the water emission to air is the net balance value, since the water that is not inside the product and the water that is not equal (balanced) in the input and output flows will be the water evaporated. Imagine a process with no water inside the product and with no water evaporation at all, then the net water mass flow would be zero. ecoinvent writes: "This quantity represents the water that leaves the ecosystem without being available for its usual function".

It is also possible to add all fresh water inputs (positive characterization factor) and all fresh water outputs (negative characterization factor) to a new impact category, which yields nearly the same amounts.

You can find a short note from ecoinvent in their methods manual in chapter 6.5.3:

https://19913970.fs1.hubspotusercontent-na1.net/hubfs/19913970/Knowledge%20Base/Database/Releases/LCIA%20implementation%203.10%20(1).pdf

You can find this link and document on their support website:

https://support.ecoinvent.org/ecoinvent-version-3.10

Comments

Thank you for this exhaustive answer!