Introduction

This code combines a variety of data sources to compute fair national emissions allocations, studies variability in these allocations and compares them with NDC estimates and cost-optimal scenario projections. We plan to make the code more accessible in terms of commenting and cleaning up old code over time. The output data of this code is publicly available on Zenodo:

Main features

The main notebook you will be using is Main.ipynb in the main directory. Each cell in that notebook loads in a particular class and then goes through all of its methods.

• The first one being called is the class in the file class_datareading.py. This class gathers all data from external sources, such as population, GDP, historical emissions, etc. on a country level, but also computes global pathways into the future. It does not do any effort sharing. The output is a file called xr_dataread.nc, which is your friend in case you need any country-specific data for debugging, for example. The main uncertainties / work from our side is around generating those future pathways under various circumstances (peak temperature, risk, non-CO2 assumptions, etc.).
• The second class being called is the one from class_allocation.py. That one does all the effort-sharing. It loads in xr_dataread.nc and its methods are named after the allocation rule being computed. The class requires an input: the region (ISO3) that is focused on (called self.FocusRegion in class_allocation.py). So in Main.ipynb, you see that there is a loop over all countries and even some country groups (e.g., G20) in the world. The output are files in the form of xr_alloc_XYZ.nc where XYZ is the ISO3 code for a country, and they are saved to K:/ECEMF/T5.2/Allocations/.
• The third and fourth classes (class_allocation_combinedapproaches and class_tempalign) are for future research.
• The fifth class (class_policyscens) is for reading cost-optimal scenarios and NDCs
• The final class (class_variancedecomp) conducts the variance decomposition (Sobol analysis)

One final important notebook is Aggregator.ipynb. This script aggregates the country-individual files xr_alloc_XYZ.nc. One output is the aggregation of those files into a single file for a single year, for example xr_alloc_2030.nc. Useful for analysis purposes. Input variables can be changed in input.yml.

Installation instructions (currently aimed at direct colleagues)

It is recommended to clone the code to the K:\ directory for best performance.

git clone https://github.com/imagepbl/EffortSharing

Setting up the environment

conda

We recommend to install the Conda environment also on the K:\ drive to increase performance. Create folders environments\effortsharing_env on K:\. Create a new conda environment with a --prefix in that folder. Activate it and update it the content based on the file environment.yml.

cd .\environments\effortsharing_env\
conda create --prefix .
conda activate K:\environments\effortsharing_env
conda env update --name K:\environments\effortsharing_env --file K:\<FILEPATH>\environment.yml

pip

pip install -r /path/to/requirements_pip.txt

Setting up required folder structure

You will need to make a folder+subfolder in the K:/ drive as follows: K:/Data/Data_effortsharing/DataUpdate_ongoing. There all the data files will appear. Most output files that are generated here are in netcdf (or xarray) format.

cd K:
mkdir Data\Data_effortsharing\DataUpdate_ongoing\Allocations
mkdir Data\Data_effortsharing\DataUpdate_ongoing\Allocations_CO2_excl