Replication data for: "Predicting forced responses of probability distributions via the fluctuation-dissipation theorem and generative modeling"

This dataset contains all data used to generate the main figures in the manuscript “Predicting forced responses of probability distributions via the fluctuation-dissipation theorem and generative modeling” by L.T. Giorgini, F. Falasca, and A.N. Souza (PNAS, 2025). The data include simulation outputs and processed results for the scalar stochastic climate model, the slow-fast triad model (ENSO), the six-dimensional barotropic model for atmospheric regime transitions, and the two-dimensional Navier-Stokes turbulence case. Data for each figure are provided in HDF5 (.h5) format with naming conventions matching those referenced in the manuscript and code. Variable descriptions are included in the accompanying README file. These data enable complete reproduction of all main text figures and facilitate independent re-analysis and verification of the study’s key findings. Code for generating, processing, and plotting the data is available in the associated GitHub repository: https://github.com/ludogiorgi/ClustGen