A sensitivity analysis of interstellar ice chemistry in astrochemical models

Paper: "A sensitivity analysis of interstellar ice chemistry in astrochemical models" DOI: This repository contains data used in the creation of the figures in the paper shown above. The code to read and analyze the data is available at the GitHub repository, using 03_data_analysis/analysisTools.py, Models.from_single_hdf. The data consists of two directories: varying_all and varying_reactions. varying_all varies all the parameters, which means binding energies, diffusion barriers, desorption prefactors, diffusion prefactors and reaction energy barriers. varying_reactions only varies the reaction energy barriers, and was used for Fig. D3 in the paper. Description of the files: MC_parameter_runs.csv: The parameters of the chemical network used on every filestep. Each row corresponds to a different sampled network, and each column to a different parameter indicated by the headers (for varying_all, varying_reactions only has the "LH" columns): Ediff: diffusion barrier Ebind: binding energy diffprefac: diffusion prefactor desprefac: desorption prefactor ... + ... + LH -> ... + ...: reaction energy barrier *.h5: Model calculations for each set of physical conditions. Physical conditions are derived from the filepath as follows: {temperature}_{number_density}_{zeta}_{radfield}.h5 These files are written such that they contain a couple of keys: 'abundances_columns': list of strings that are the column names. This is things like "H2", "#CO", "@H2O" but also "Time", "Density", etc. 'nominal': numpy array of physical conditions and abundances at each timestep, using the nominal network 'rates_columns': columns for the reactions. These are strings for every single reaction, but also "Time", "Density", etc. 'nominal_rates': reaction rates at every timestep. '0' to '999': 1000 numpy arrays of the physical conditions and abundances at each timestep, using the sampled networks described by the rows in MC_parameter_runs.csv.