Validation: src/zephyrus/hydrostatic.py
This page tracks the @pytest.mark.reference_pinned tests that anchor the hydrostatic branch of zephyrus.hydrostatic.
| Test id | Reference | Scope |
|---|---|---|
tests/test_hydrostatic.py::test_volkov_flat_factor_against_published_correction |
Volkov et al. (2011), Phys. Fluids 23, 066601, Eq. (9) and their printed c(lambda) table | The drifting-Maxwellian flux ratio reduces to unity at zero bulk velocity and reproduces the printed linear coefficient across lambda 1 to 106. The test also demonstrates that this correction and the flat kinetic factor the branch applies have opposite slopes in lambda, so applying both would double-count. |
tests/test_hydrostatic.py::test_yelle_figure1_mars_hydrogen_flux |
Yelle (2024), Icarus 416, 116099, Figure 1 (fully specified Mars model) | The branch reproduces the diffusion-limited hydrogen plateau at \(2.4 \times 10^{8}\) cm^-2 s^-1 (40% tolerance: the binary H-CO2 coefficient source of the original calculation is not pinned in the paper), the saturation above 200 K, and the Jeans-limited collapse at 100 K checked as a regime. |
Notes
The escape-temperature identities (the lambda = 2 neutral criterion and the plasma value at half of it, after Chatterjee & Pierrehumbert 2026, their Eq. (34)) and the stoichiometric mass conservation of the element mapping are asserted as physics invariants in the same file.
Anchor type
Published benchmark (printed coefficient table and a published model figure) plus the analytical rest limit.
Date of last comparison against the sources: 2026-08-20.