Repository navigation
Conversation
The atmodeller outgassing module now adds each supported condensate
(graphite, sulfur, water, sulfuric acid) when its elements have budgets
and T_magma lies inside its thermodynamic data range, instead of graphite
alone. A condensate that cannot be built is logged as a warning rather
than dropped silently.
Each solve writes {X}_cd_activity and {X}_cd_kg to the helpfile (zero
when the condensate was not solved for) and logs them. The columns are
registered as diagnostics, so helpfiles written before them still resume.
The new plot_saturated shows each condensate's activity, its condensed
mass while saturated, and the magma temperature and melt fraction.
The golden trajectory reference is not yet re-recorded for the eight new
columns, so test_the_committed_reference_covers_every_column_the_helpfile_writes
fails until it is recorded again with --record-golden.
The condensed mass panel now shows each saturated condensate per unit surface area (kg m-2, over 4 pi R_int^2), and a new panel estimates the thickness of a uniform layer of it from condensate_density. The densities are approximate ambient values with no temperature, pressure or porosity dependence, so the thickness is an estimate only.
atmodeller's condensed mass per element (e.g. graphite carbon) was added
to {e}_kg_solid, a column described as mass in the solid mantle and also
shared with volatile trapping. It now goes to a separate {e}_kg_condensed
reservoir for each volatile element.
The per-element closure check sums atm + liquid + solid + condensed,
desiccation keeps the condensed reservoir with the mantle ones so graphite
is not dropped from the budget when the atmosphere goes, and helpfiles
written before the column existed read it as zero on resume.
nichollsh
added
Enhancement
The schema parser and the static helpfile scanner now know
condensate_list, so the new {X}_cd_activity, {X}_cd_kg and
{e}_kg_condensed columns are parsed and attributed to the atmodeller
module. The wrapper and plot_saturated build those keys directly from
the loop variable so the scan can resolve them, the desiccation write to
{e}_kg_condensed is declared as a volatile-element template, and the
output reference page and JSON sidecar are regenerated.
The logged-activity test compares floats with pytest.approx.
…re on
The outgassing wrapper sets every {e}_kg_condensed to 0 before each chemistry solve, and atmodeller fills them only when include_condensates is on. The columns are always in the helpfile, and a backend or run without condensates now reports 0 by construction rather than by leaving them untouched.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #1037 +/- ##
==========================================
+ Coverage 94.44% 94.65% +0.21%
==========================================
Files 139 140 +1
Lines 23569 23717 +148
Branches 3944 3968 +24
==========================================
+ Hits 22259 22450 +191
+ Misses 882 874 -8
+ Partials 428 393 -35
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
The 13 new columns ({X}_cd_activity, {X}_cd_kg and {e}_kg_condensed) are added to golden_run.tsv. The run uses the dummy outgassing backend, so each is constant at 0; all 926 existing columns reproduce unchanged over the 56 rows.
…s parser From pandas 3.1.0 the default read_csv parser reads these values exactly, so the assert that it misses a last bit no longer holds. The test now shows that some of the values do not survive 16 significant digits, which holds whatever pandas does, and keeps the exact round-trip assert.
The check that over a fifth of the values do not survive 16 significant digits sits with the 11-digit check, before the write, as a count; the docstring says a quarter of the values need all 17 digits.
…atmodeller-graphite-tracking
plot_saturated: the plotted column is the condensed mass over 4 pi R_int^2 and the thickness that over the condensate density, the solver floor below saturation plots as zero, the plot is skipped without solved condensates, and the entry point reads the helpfile. atmodeller: a condensate joins only with its elements budgeted and T_magma inside its data range, and a skipped solve resets every condensate column to zero. Desiccation keeps graphite in C_kg_condensed alongside the solid mantle, scaled by the same share when escape took part of the held mass.
…racking # Conflicts: # src/proteus/outgas/atmodeller.py
nichollsh
marked this pull request as ready for review
October 23, 2026 07:32
With the flag off (the default) each chemistry solve re-equilibrates the whole
inventory, so a condensate such as graphite can re-vapourise as conditions
change. With it on, the mass already held in each {e}_kg_condensed reservoir
is withheld from the solve and added back on top of what the solve newly
condenses, together with each condensate's {X}_cd_kg, so a condensed layer can
only grow. {X}_cd_activity stays the solver's value for the inventory left in
the solve.
The held mass is also beyond reach elsewhere: unreachable_mass (trapped plus
held condensed) replaces locked_solid_mass in the escape reservoir, the escape
floor and the desiccation gate, so escape cannot strip a held layer and a
planet left with only a held crust registers as desiccated. The config is
passed through escapable_mass, limit_escape_step, calc_new_elements and
escapable_inventory as an optional argument; without it only trapped mass is
excluded, as before.
plot_saturated now draws a condensed mass above 1 kg at any activity, so a
held layer stays visible after the remaining inventory falls below
saturation. The config and output references are regenerated, the field is
documented in all_options.toml, and the v2 to v3 migration lists it as
neutral.
![github-code-quality[bot]](https://avatars.githubusercontent.com/u/9919?s=60&v=4){kind=small}
| ): | ||
| raise RuntimeError('solve failed') | ||
|
|
||
| assert hf_row['C_kg_total'] == pytest.approx(1.0e20, rel=1e-12) |
The per-solve atmodeller summaries (species and active elements, solve inputs, total pressure, surface pressure and MMW) and SPIDER's choice of P-S phase boundaries now log at DEBUG, so a coupled run's INFO log carries one atmodeller line per condensate on each step. That line reads "Condensate C: activity=0.250, condensed mass=3.00e-06 kg"; a missing or non-finite value prints as n/a rather than failing the float format. Tests check the condensate line and the DEBUG level of the summaries, the n/a line for an unavailable activity and mass, and that a run's derived P-S curves are passed to SPIDER in place of the melting_dir pair, at DEBUG.
input/planets/mercury.toml: Mercury mass, orbit and core radius fraction, SPIDER structure and energetics with the Wolf & Bower (2018) EOS and melting curves, atmodeller outgassing with condensates and prevent_desaturation, solar noble gases, AGNI and ZEPHYRUS. It is the reference config for fO2 x C grids, which set outgas.fO2_shift_IW and planet.elements.C_budget per case.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Tracks the saturated (condensed) phases from the atmodeller outgassing chemistry in the helpfile, keeps them in their own reservoir rather than the solid mantle, plots them, and adds an option to stop condensates from re-vapourising. Closes #1036.
Condensates in the chemistry
condensate_listinutils/constants.py) when its elements have budgets andT_magmalies inside its thermodynamic data range.Helpfile columns
{X}_cd_activity(1 = saturated) and{X}_cd_kgfor each condensate, written on every solve, zero when it is not solved for. Registered as diagnostics, so older helpfiles still resume.{e}_kg_condensedfor H, O, C, N, S: the element mass held in condensates. The outgassing wrapper zeroes them before every solve; only atmodeller withinclude_condensatesfills them.New key -
outgas.prevent_desaturation(defaultfalse)false: each solve re-equilibrates the whole inventory, so a condensate can re-vapourise.true: the mass already in each condensed reservoir is withheld from the solve and added back on top of what the solve newly condenses, together with{X}_cd_kg, so a condensed layer only grows.New plot -
plot_saturatedcondensate_density; T_magma and melt fraction.Other
prevent_desaturationadded toinput/all_options.tomland classified neutral for the v2 to v3 migration.Validation of changes
Linux (Fedora 44, kernel 7.2), Python 3.12, pandas 3.1.0, atmodeller 1.0.2 (local checkout at 18eabab), AGNI 1.13.2 on Julia 1.13.1.
C_kg_condensed(7.57e20 kg, equal toC_cd_kg) withC_kg_solid = 0, and atm + liquid + solid + condensed equalsC_kg_totalwithin 5e-6, the solver tolerance.prevent_desaturationend to end. Two real atmodeller steps at IW-5 with 1000 ppmw C, 1800 K then 2400 K: withfalsethe 4.29e19 kg layer from 1800 K re-vapourises at 2400 K (activity 0.30); withtrueit stays at 4.29e19 kg (activity of the remaining inventory 0.12). Total carbon is conserved and the carbon reservoirs close within 3e-5 in both.plot_saturatedon every step, with graphite saturation matching fixed-temperature atmodeller sweeps through the same wrapper.prevent_desaturation(withheld from the solve, only growing, restored when the solve raises, invisible to escape and to the desiccation gate), andplot_saturated(column and thickness against mass, area and density, floor masking, held layers, skip cases, entry point). Changed lines insrc/are fully covered in a local coverage run of the affected suites.mainhere; the branch adds none. The test-quality check reports the same 80 violations asmain.prevent_desaturation.Checklist
mainremain)all_options.toml)