Companion code for the newsletter A Journey Through Climate Modelling: climate models built from the ground up, one piece of physics at a time.

Climate models can feel like black boxes. Here we open them up. We start with the simplest models (heat conduction in soil, a basic energy balance) and add one process at a time, so you can see how a single equation grows into a full climate model.

How the series works

Every post follows the same template, and every post has a runnable model in this repository:

1. The idea: the physical process and why it matters for climate
2. The equations: derived step by step
3. The code: a minimal implementation in Julia
4. The results: plots and animations
5. The limits: what the model cannot explain, which motivates the next one

No expertise needed: curiosity, basic maths and a bit of code reading are enough.

Quick start

git clone https://github.com/hadronnez/AJourneyThroughClimateModelling.git
cd AJourneyThroughClimateModelling
julia --project=. -e 'using Pkg; Pkg.instantiate()'
julia include("run_experiment.jl")

A run is controlled by three YAML files in cfgs/:

File	Controls
exp_cfg.yaml	experiment id and what to save
sim_cfg.yaml	grid, physics, initial and boundary conditions
plot_cfg.yaml	figures: language, style, which plots and animations

Change a parameter, run again, and compare.

Outputs

Each run writes to experiments/<id>/:

• cfgs/: the three configs exactly as used (defaults filled in). Copy them back into cfgs/ to reproduce the run.
• data/: results in NetCDF
• plots/: figures and animations
• log/: run log

Follow along

New posts and the code that goes with them appear on the Substack. Questions and suggestions are welcome as issues. …]()