ecoinvent Forum Archive

Hello,

I'm making use of the correspondence file provided for mapping activities from ecoinvent 3.6 to 3.7. One of the columns is labeled "replacement amount" and this value is different from 1 for most datasets that are marked as changed in some way between the two versions.

How is this number to be interpreted? In some activities it seems as though the replacement amount is the amount of the new activity that should be used for a unit of the old activity. This suggests that the reference flows have changed. For instance, in "ammonium nitrate production", the reference product changed from "ammonium nitrate, as N" to "ammonium nitrate" which is significant. However, there are multiple activities that have the same change and they do not have the same replacement amount. The four "ammonium nitrate production" activities all have the same stoichiometric change but different replacement amounts. Why?

In other cases it appears that the reference flows did not change, yet the replacement amount is not 1. For instance when a process is split by geography, the replacement amount seems to indicate the fraction that is allocated to the new geography. It seems hazardous to represent completely different kinds of information in the same table column- given that the spreadsheet already has nearly fifty columns, why not just create a new one?

Thanks

Hi Brandon,

The correspondence file and the replacement amounts it contains are meant to help users answer the following question: if I am using 1 unit of product x from dataset y in version 3.6, what should I use in version 3.7 if I want to get an equivalent functional unit?

To illustrate this, here are the steps for the example of "ammonium nitrate production":

1. First, select the dataset that you were using in v3.6. This requires filtering in the columns activityName, geography and product name under eiv3.6.

2. Let's assume you were using in v3.6 the dataset "ammonium nitrate production", "RoW", "ammonium nitrate, as N". In the cut-off system model, there are 3 rows that contain that dataset. That's because in v3.7, there are 2 new geographies that were added, which both are contained in the RoW from v3.6. Note that the RoW in v3.7 represents a different geography than the RoW in v3.6 for this particular activity.

3. Based on the datasets listed under eiv3.7, to get the same functional unit as in v3.6, the user should now use 0.5 kg from the CN dataset, plus 0.65 kg from the RNA dataset and 1.7 kg from the new RoW dataset. This is considered to be the most appropriate replacement for 1 kg of "ammonium nitrate, as N" supplied by the RoW "ammonium nitrate production" dataset in v3.6

If in your study you are actually modelling a process taking place in the USA for example, and you know that your input of "ammonium nitrate" is entirely supplied by a producer located in the USA, then, when switching to v3.7, you might want to use only the dataset for RNA (Region North America) instead of the default mix suggested in the correspondence file.

In this case, if you were using 0.5 kg of "ammonium nitrate, as N" input back in v3.6, in v3.7 you would need instead 0.5*(0.5 + 0.65 + 1.7) = 1.425 kg of the new product "ammonium nitrate", supplied by the RNA dataset.

In the case of ammonium nitrate, the replacement amounts thus reflect both the geographical splitting as well as the scaling to account for the change in N content. This is a rather special case, very few products require scaling between v3.6 and v3.7. On the other hand, splitting of a geography is much more common. For these cases (where there's no scaling of the product), the replacement shares sum up to 1 for a given dataset in v3.6.

Another thing, you might want to already switch to version 3.7.1, which has been published yesterday (more information here: https://www.ecoinvent.org/database/ecoinvent-371/new-data-in-ecoinvent-371/new-data-in-ecoinvent-371.html). The correspondence file for v3.6 to v3.7.1 is identical to the one for v3.6 to v3.7.

Best regards,

David FitzGerald

Junior Project Manager

ecoinvent