Estimating the Resilience of Non-Stationary Systems

Code for the publication: T Smith, A Morr, C Schötz, and N Boers. “Estimating the Resilience of Non-Stationary Systems.” arXiv (2026). https://doi.org/10.48550/arXiv.2604.24345

Code was run using Python 3.12 and standard libraries (pandas, numpy, etc). 

Synthetic data and glacier data access scripts can be found here: https://doi.org/10.5281/zenodo.18031340

Python code included here for deseasoning/detrending functions, calculation of classic (autocorrelation, variance) early-warning signals of critical slowing down, and new regression-based formulation for estimating system stability.

Code to recreate all Figures can be found as python notebooks, alongside the data for each Figure.