search

rffscghg / MimiGIVE.jl

MIT License
9 stars 9 forks source link

From Forum: Deterministic SCC Result #70

Open lrennels opened 5 days ago

lrennels commented 5 days ago

moved from forum

https://forum.mimiframework.org/t/first-time-user-calibration-of-scc-estimates/313

I am a first-time user and trying to get acquainted with both Julia and MimiGIVE. I was able to install Juia, add packages and run the default model. However, when I run the default model, I got result that seemed far too low. I was expecting to get something closer to the $185/ton that is found the SCC explorer tool. Any tips for calibrating my model to make sure it’s running correctly? I was following sections 2a (fully coupled model) and 4a (deterministic scc calculation) on the Github repository.

scc = MimiGIVE.compute_scc(year=2020) 26.166523152489944

scc = MimiGIVE.compute_scc(year=2030) 38.34888259268095

lrennels commented 5 days ago

Good question! Our deterministic SCC calculation calculates the SCC based on one set of parameters, but it is not a central value of the full distribution, since many of the key uncertainty parameters, like the socioeconomic pathway and the climate model parameter set, do not have a "central" run. What you're doing is correct, and your confusion is fair! In order to get the reported value (within the standard error) you'll need to run a Monte Carlo. The deterministic calculation can be helpful for testing, debugging, etc. but will not give a sense for the central tendency of the full distribution.