search

pik-piam / remind2

The remind2 package contains the REMIND-specific routines for data and model output manipulation.
0 stars 39 forks source link

Add basicmode to reportEmi #607

Closed gabriel-abrahao closed 1 month ago

gabriel-abrahao commented 1 month ago

The new MAGICC7_AR6 climate implementation requires nicely formatted emissions data.

reportEmi currently needs a lot of sector-specific information to provide the breakdown of CO2 emissions. In REMIND, module 15_climate runs before some of the modules that generate this information, making it not work on a GDX that was generated in the postsolve in the first iteration.

This PR adds a basicmode optional argument to reportEmi, that circumvents the need for sector-specific information by calculating Energy and Industrial Processes CO2 emissions via the difference between total and LUC emissions, and reports only the emissions variables needed for the climate assessment. Since the function is a bit messy at the time, it's hard to detach these from the current code, so there's quite a bit of code repeating here. However, the only out of this might be a complete refactor of reportEmi. Putting this functionality in a separate function might be worse than this solution, as it wouldn't be maintained as closely.

gabriel-abrahao commented 1 month ago

@tonnrueter buildLibrary() is complaining about a million linter problems, including many things that definitely were there before. Would you mind taking a look?

tonnrueter commented 1 month ago

With the current state of reportEmi major adjustments like ours do not seem advisable. Therefore, Gabriel & I agreed to instead implement a dedicated reportEmiForClimateAssessment function that extracts the quantities needed for climate assessment in between GAMS iteration from fulldata_postsolve.gdx. Although we risk to miss bug fixes etc to reportEmi, this seems the most practicable way to go foward.

fschreyer commented 1 month ago

The emissions equations in the REMIND core and the emissions reporting in reportEmi are unfortunately quite complex and not ideally structured. We consider doing a general refactoring once the time allows it. Let me know if you have specific questions on the current implementation.

fbenke-pik commented 1 month ago

@fschreyer the sooner the better. let me know if i can help somehow.

fschreyer commented 1 month ago

@fschreyer the sooner the better. let me know if i can help somehow.

Thanks, good to know. However, this will be a bigger process where we first need to get conceptually clear which kind of structure we want depending on current and future developments and need to coordinate across the group. I could lead this but I won't have time to start this process before spring 2025.