This repository contains the code and data associated with the paper: Medeiros, L. P., Sorenson, D. K., Johnson, B. J., Palkovacs, E. P., and Munch, S. B. (2025). Revealing unseen dynamical regimes of ecosystems from population time-series data. Proceedings of the National Academy of Sciences.
In this study, we introduce an approach that integrates population time-series data and information on an environmental driver into a Gaussian Process regression model with time-delay embedding (GP-EDM) to predict previously unseen dynamical regimes of ecosystems.
The repository contains 4 folders: code, data, figs, and tables. Under code/scripts we provide the R scripts to reproduce Figures 1-5 and SI figures. Empirical data sets are located under data/microbial_experiment and data/lake_plankton. All figures and tables from our study can be found under figs and tables.