Prerequisite knowledge & skills

Code knowledge. This package is implemented in the programming language Python. In order to use this properly, it is expected that you have a basic understanding of Python, using an Anaconda command prompt, as well as an environment manager like conda, or know how to install packages using pip. The workflow extensively uses the packages pyam,`` openscm``, scmdata, and pandas for operations. Examples are implemented using Jupyter Notebooks.

Understanding emissions input files This workflow uses the IAMC format used also by pyam for emissions input. The input emissions must follow the AR6 scenario submission template. To see which emissions are accepted as input for the climate assessment workflow, see the file “data/emissions_variable_list_climateruns.xlsx”.

Computing power. Running one scenario for all configurations of one simple climate model can easily take up to one hour on a personal computer. For running scenario sets bigger than about 10-20 scenarios, you require more computing power. While MAGICC7 and CICERO-SCM mostly benefit from more cores, FaIR especially requires more RAM.

Tuning parallellisation settings for your specific setup currently requires delving into the code, editing joblib settings, and re-installing (e.g. with pip install -e .).

This workflow has been tested on Windows 10 and Linux (Ubuntu).

Questions. For questions, feel free to contact kikstra@iiasa.ac.at or raise an issue at https://github.com/iiasa/climate-assessment/issues