Reproducibility package for "Testing scalar–tensor dark energy with gravitational-wave propagation: reconstruction and calibration limits"

Abstract

Reproducibility package accompanying the manuscript “Testing scalar–tensor dark energy with gravitational-wave propagation: reconstruction and calibration limits,” by Nikhil Goenka. This record contains the scientific software, synthetic inputs, and numerical results supporting the manuscript. The study examines when expansion and gravitational-wave propagation information can test canonical scalar–tensor explanations of dark energy, with particular attention to source calibration, statistical identifiability, and functional resolution. Scientific contents The package includes Python implementations of canonical kinetic and on-trajectory potential reconstruction, finite-window consistency tests, calibration-dependent exclusion-power bounds, and matched-history comparisons. It also contains the illustrative Chevallier–Polarski–Linder (CPL) calculations based on published DESI DR2 parameter summaries, conditional canonical realizations, Planck-function excursion bounds, and deterministic background and reconstruction checks. Supporting files include input configurations, fixed random seeds, synthetic observations, covariance matrices, full-precision reconstruction coefficients, saved numerical outputs, scientific tests, figure-generation scripts, dependency information, and reproduction instructions. Scope The results are theoretical and computational, evaluated under the assumptions specified in the manuscript. No observational catalogue is analyzed and no observational likelihood is evaluated or refitted. Published DESI DR2 parameter summaries are used only to define illustrative expansion histories. The results do not constitute an observational exclusion or establish complete physical viability of the reconstructed models. The accompanying manuscript explains the assumptions, limitations, and AI-assisted research and verification workflow. Reproduction and provenance Consult the top-level README and code/REPRODUCIBILITY_MAP.md for environment setup, file-level provenance, and the mapping between manuscript results and the relevant programs and data. The documentation distinguishes quick verification, figure regeneration, and longer numerical reruns. Full-precision inputs and saved outputs are included where required to reproduce the reported numerical comparisons. The internal checksum manifest identifies the packaged files. Paths cited in the manuscript are relative to the root of the extracted reproducibility archive. Licensing Original software, scripts, and tests are licensed under the MIT License. Original generated data, numerical outputs, and documentation are licensed under Creative Commons Attribution 4.0 International (CC BY 4.0), to the extent applicable rights exist. Third-party material retains its original terms. The archive’s LICENSE_SCOPE.md identifies the applicable license for each component; MIT and CC BY 4.0 are not interchangeable alternatives for every file. Complete license texts and attribution information are included in the archive.