Different result for default scenario than standard LCA

[brandchristina]

Hi there,

if I calculate a scenario LCA the result for the default scenario is significantly different than the result in the standard LCA (10,6 g CO2-Eq. vs. 14,0 g CO2-Eq.). This should not be the case and I don't know which is the correct result. I think this relates to issue #592.

My problem occurs in the stable and in the development version of the ab. I already updated bw and ab but it doesn't fix the problem.

[marc-vdm]

Could you share the database File>Export Database (excel export) and the scenario file?

(and what to put into the calculation setup.)

That way we could replicate the problem.

[brandchristina]

I hope with the following information you can replicate my problem.

This is my scenario table: scenario_table_Hackervarianten.xlsx

These are the two LCA Setups for the standard and the scenario LCA:

[marc-vdm]

Thanks, I have both good and bad news, but it's mostly bad. The good news is that you're not doing anything wrong. AB seems to be doing weird things here. The bad news is that I don't understand yet what the problem is. This will require a lot more time to understand I'm afraid. Anyway, you sharing these files is really helpful for us to better understand the problem.

I'm suspecting that AB might be messing with the parameter values in the 'standard LCA' in a way that we can't easily see. I saw that when setting both Anteil_Anbauhacker and Anteil_LkwHacker to 1 in a scenario, the result is very close to the 'standard' result.

~~Anyway, as a intermediate fix, I'd suggest to trust the default result over the standard one, as you're explicitly setting values your want that way.~~ More up-to-data advice below

[brandchristina]

Thank you very much @marc-vdm for the analysis and the fast answer!!

[marc-vdm]

Hi @brandchristina, I've been looking further into the issue and have to come back from my previous advice.

Something in the whole parameters code seems very broken to me when looking at results. It looks to me like the default column in a scenarios file will be ignored (but also not always?). And instead the parameters will be used as 'default' values (which I don't think is supposed to happen).

Additionally, there is somehow a disconnect between the results of a standard LCA and the default column when using parameters.

My new advice is the following: Trust the standard LCA and the non-default scenarios. I'm afraid something is not right with the default column somehow.

I'm sorry for the inconvenience.

[brandchristina]

Hi @marc-vdm thank you for taking the time looking at this issue. Since I am unfortunately not able to understand what is happening there, I have to trust you which of the results are the true values. Thanks very much for sharing your new advice!

[marc-vdm]

Since I am unfortunately not able to understand what is happening there

Unfortunately I am not either haha. However, I mostly trust the above options. Again, I'm sorry for the inconvenience.

Personally, I have a little more faith in non-parameterized scenarios, though that requires a different work-flow that might not be suitable to how you are working.

[marc-vdm]

Closing this as resolved in #1025