This github repository contains all the code used to generate the results presented in ** Glacial abrupt climate change as a multi-scale phenomenon resulting from monostable excitable dynamics ** by Keno Riechers, Georg Gottwald and Niklas Boers (Journal of Climate, 2024)
https://doi.org/10.1175/JCLI-D-23-0308.1
The code was written using python 3.8.10. The file requirements.txt contains a list of packages which were installed for that python version at the time when the code was executed to produce the outputs for the publication. Not all packages indicates in the requirements.txt are necessary for executing the code.
Executing the file master.py successively produces the results presented in the publication, except for figures 2 and 3. Notice that a random number generator is used for some simulations. The results shown in the paper were generated without setting the seed of the generator. Hence, an exact reproduction of the results based on random numbers is not possible.
Under the points compute transient trajectories the user must specify how many trajectories will be generated. For the manuscript we generated 1000 trajectories. Please consider, that this took approximately one day on a cluster computer.
Similarly, under create fig05 the user may specify how many trajectories shall be generated. Additional trajectories serve for comparison and make the conclusions more robust.
Running master.py automatically creates a repository output, where the results (including figures) will be stored.
The directory proxy_data contains data that was published as supplementary material of other research articles. The corresponding research articles are indicated in the individual files as well as in the main text of our manuscript.
Two create figures 2 and 3 please run fig02.py and fig03.py. These figures will be stored in the main directory directly.