JGCRI / gcam-core

GCAM -- The Global Change Analysis Model
http://jgcri.github.io/gcam-doc/
Other
263 stars 159 forks source link

Downscaling electricity generation by technology to the basin level in GCAM 7 #424

Open Zaid2995 opened 1 month ago

Zaid2995 commented 1 month ago

Hello all,

I am currently in the process of downscaling electricity generation (by technology) to the basin level in GCAM V7. The overall objective is to indicate to the model the form of electricity generation (RE/non-RE) each basin may experience in the future. I would also like to gain insights related to the shifting of basins in the case of deteriorating water levels in the basins. The methodology I have adopted is as follows:

  1. Shares of water withdrawal/consumption for electricity generation are provided across regions and basins (from 2005 data) within the CSVs in the model.
  2. Level 1 chunk L1233.Elec_water outputs the amount of electricity generated by region, sector, fuel, technology and cooling technology.
  3. Multiplying the above two would provide us with the electricity generated at the region, basin, sector, fuel, technology and cooling technology level.

To reflect this new basin-level structure, basins have to be added across the electricity_water structure. The electricity_water structure is a bit challenging to alter since there are many tables that are generated in the L2233.electricity_water chunk that do not appear in the outputs folder and are only visible in the debug mode of the batch level of the electricity_water chunk. I overcame this challenge by creating those tables separately and adding them exogenously to the model (only for one region since these tables are too large to be replicated manually), but the model outputs an error "No Discrete Choice Function defined for US, electricity, PacArctic". This makes sense since I haven't made alterations to the electricity sector. I have only changed the electricity_water related files. I would like to get some advice on how to: a) access those tables which are "hidden" so that the changes can be implemented for all the regions in the model, b) and whether I have correctly identified the issues which are obstructing the restructuring I am doing.

@pkyle @pralitp @realxinzhao