Supplementary Material: Knobs and dials of retrieving JWST transmission spectra. II. Impacts of pipeline-level differences on retrieval posteriors

This is supplementary material to Schleich et al. (2025, submitted to A&A). The content of this repository (also listed in context.txt) is arranged as follows:

Retrieval (Data Reductions)

A collection of all data products, scripts, and plots associated with the atmospheric retrieval of the three transmission spectra of WASP-39 b used in this work.

Data

Ordered by input spectrum ("This Work", "Rustamkulov", and "Carter"), these subfolders contain:

• The atmospheric retrieval input spectrum ("*.spec")
• The TauREx input parameter file ("setups/CASE-NAME.par") and retrieval output ("results/CASE-NAME.h5")
• The output of the associated MultiNest sampling chains ("chains/CASE-NAME/")
• The slurm output log-file of all associated retrieval runs ("slurm/CASE-NAME.par.out")

Plots

This folder contains a collections of Python scripts used to generate the plots displayed in the paper, as well as supplementary plots for the individual retrieval runs.

Retrieval (Scatter)

Same as above, but for the 10 scattered instances of the spectrum produced in this work (SP-TW).

Data

Same as for "Retrieval" (i.e. subfolder and data structre). The individual scattered instances are store in "spectra/".

Plots
Plotting script used to generate the select posterior plot shown in the paper.

Eureka (WASP-39 b)

Data products

All Eureka! data products associated with the data reduction of the NIRSpec PRISM transit observation of WASP-39 b.

Auxiliary

Auxiliary files associated with the data reduction process:

• "eureka_parfiles/": Contains the parameter files used to perform the data reduction described in the Appendix C of the paper
• "nirspec_SUB512_mask.dat": Pixel locations identified on the detector sub-array as havin persistent / hot / dead pixels. Is used as an additional mask in stage 1
• "wasp39b_nirspec_prism_ap6_bg9_flagged_idxs.dat": Detector column values inferred in stage 4 as deviating from the expected photon noise by a factor of more than 1.75. These are excluded from the generation of the white light-curve, as well as from the binning of the finalised transmission spectrum