Cabin-scale analysis for airborne transmission in microgravity: code and build sources
Year: 2026
Type: software
Abstract: Companion code and build sources for the paper "What microgravity actually changes about airborne transmission: a gravitational hierarchy". The analysis asks, of each term in the governing equations of airborne transport, how much it changes when gravity falls from 1 g to the residual acceleration of an orbiting vehicle, across three nested scales: the droplet, the exhaled jet, and the cabin. The filtration architecture of the International Space Station US Laboratory module is constrained from flight documentation rather than left as a free parameter, which turns out to govern the magnitude of the predicted effect more than gravity does. Single source of truth. One self-contained Python script computes every numerical value reported in the paper and emits it as LaTeX: fourteen table fragments, of which twelve are included by the manuscript and the supplementary information through \input{}; a file of fifty macros for the values quoted in running text, including those of the abstract; and the four figures. No number in the text, the tables or the figures is typed by hand, so the paper cannot drift from the code. Two exceptions are declared in the manuscript itself: the intrapulmonary deposition reductions of Section 3.6, quoted from Darquenne and Prisk (2008), and values explicitly attributed to other published work. Two of the fourteen fragments, Rfactor.tex and Hsens.tex, are emitted but not included, Figure 3 having superseded them; they are kept so that the tabulated form remains available to a reader who prefers numbers to curves. Verification. The deposit carries its own checker. verify.py scans for absolute paths and credentials, confirms that the expected files are present and match MANIFEST.sha256, records the interpreter and library versions, re-runs the analysis in a temporary directory, compares the regenerated console output and LaTeX fragments byte for byte against the deposited references, checks eleven key numbers of the paper against the values quoted in the abstract and Section 3, verifies that every \input{} target exists, and exits non-zero on any failure. Running it from an empty copy of this archive reports no failure and no warning. verify_report_offline.txt records that run carried out with all network access blocked, as evidence that the deposit is self-contained. Reproducing. Run python3 cabin_scale_analysis.py > out.txt and diff the result against console_output.txt; expected difference: none. Then two passes of pdflatex on manuscript.tex and on supplementary.tex; expected result: 24 and 4 pages, no error, no warning, no overfull box, no unresolved citation. The script writes tables/ and figures/ beside itself rather than in the current directory, so it can be run from anywhere, and it makes no network call and reads no file outside its own directory. Requires Python 3.7 or later with numpy and scipy; matplotlib is needed only for the figures, and without it the script still writes every table and value and says so. Exact versions of the reference run are pinned in requirements.txt and environment.yml. The figures are set in DejaVu Sans, which ships with matplotlib, so they render identically on any machine. Scope. No measurement, no simulation, no experimental dataset. The analysis is analytical throughout; it introduces no turbulence model and fits no parameter beyond the puff spreading rate, whose sensitivity is reported. It states which terms of the governing equations change and by how much; it does not predict a trajectory, a concentration field, or where deposition occurs. Ballistic impaction driven by the exhaled jet is identified as the open question and is left explicitly to a geometry-resolved simulation. The hardware values that constrain the filtration and ventilation terms are read from conference papers cited in the manuscript bibliography; no third-party dataset is redistributed here. Licence and versioning. MIT. The manuscript sources are included so that the build can be verified; should a copyright transfer later restrict their redistribution, a new version will be published without manuscript.tex and supplementary.tex, and everything needed to verify the numbers will remain. Version 1.0.0 corresponds to the manuscript as submitted. A correction that changes no value will be 1.0.1; any change to a reported value will be 1.1.0. Cite the version DOI of the release used, not the concept DOI.
Author: Chopard-Vilhem, Théo
Institution: Iuliu Hațieganu University of Medicine and Pharmacy
Language: English
Subfield: Pulmonary and Respiratory Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7221193215