Quick start: all-dummy run¶

This tutorial runs PROTEUS with all modules set to "dummy" backends. No external solvers (AGNI, SPIDER, SOCRATES) are needed; the run completes in under a minute and exercises the full coupling architecture. Use this to verify your installation and understand the code flow before moving to production runs.

Prerequisites¶

PROTEUS installed with bash install.sh (see the installation guide); the installer sets FWL_DATA in the proteus conda environment

No external solvers, spectral files, or EOS data are required.

The configuration file¶

PROTEUS ships with an all-dummy config at input/dummy.toml. The key settings are:

Planet: 1 M\(_\oplus\), starting fully molten (T\(_\mathrm{magma}\) = 4000 K, \(\Phi\) = 1). Volatile budget of 10,000 ppmw H, 1000 ppmw C, 500 ppmw N, 500 ppmw S.
Star: fixed solar luminosity (no evolution)
Orbit: 0.5 AU, no tidal heating
Interior structure: Noack & Lasbleis (2020) ¹ analytical scaling laws
Interior energetics: heat-capacity integrator with prescribed solidus (1700 K) and liquidus (2700 K)
Outgassing: melt-fraction-dependent partitioning; 10% of volatiles are always in the atmosphere (finite solubility floor), with the atmospheric fraction increasing as the mantle solidifies
Atmosphere: grey-body opacity (\(\gamma\) = 0.5)
Escape: disabled (rate = 0), so the run reaches solidification
Chemistry: parameterised vertical profiles (offline)

The simulation terminates when the global melt fraction drops below the solidification threshold.

Running the simulation¶

conda activate proteus
proteus start --offline -c input/dummy.toml

The --offline flag skips data downloads. The run takes less than a minute, including the plots.

Expected output¶

The run creates a directory inside output/ named with a timestamp (run_<timestamp>_xxxx, called <run folder> below). Check output/<run folder>/plots/plot_global_lin.png for a multi-panel overview. Your output should look similar to this. A run writes its plots in the default matplotlib style; the figures of the tutorial pages show the same plots in the PROTEUS visual language, in a light and a dark variant.

Dummy tutorial output — **All-dummy tutorial output.** (a) Heat fluxes: the interior and atmospheric fluxes track each other as the planet cools; absorbed stellar flux (ASF) is constant. (b) Surface partial pressures: H₂O dominates (~10⁴ bar), with CO₂, N₂, and SO₂ as minor species; pressures increase as solidification forces dissolved volatiles into the atmosphere. (c) Surface temperature: monotonic cooling to ~1700 K (solidus). (d) Atmospheric mole fractions: H₂O at ~95%, stable throughout. (e) Mantle evolution: melt fraction drops from 1 (fully molten) to ~0 (solidified) over ~23,000 yr; the rheological front (orange) moves toward the surface as the melt fraction drops. (f) Volatile partitioning: dissolved fraction decreases from ~90% to ~0% as the melt fraction drops, transferring volatiles from the interior to the atmosphere.

To regenerate these plots from your own output, pass the resolved config that the run wrote into its folder:

proteus plot -c output/<run folder>/init_coupler.toml all

input/dummy.toml sets params.out.path = "auto", so the plot command with that file would look for a new run folder.

Understanding the helpfile¶

Open runtime_helpfile.csv in the output directory to see the full time series. Key columns:

Column	Units	What to expect
`Time`	yr	Stays at 0 for the first 3 iterations (init stage), then advances to ~23,000 yr
`T_magma`	K	Decreases monotonically from 4000 to ~1700
`Phi_global`	1	Drops from 1.0 to ~0.003, triggering the solidification stop
`P_surf`	bar	Increases from ~7,000 to ~70,000 as volatiles outgas
`F_atm`	W m\(^{-2}\)	Net flux at the top of the atmosphere; decreases as the surface cools
`F_int`	W m\(^{-2}\)	Interior heat flux; the dummy interior takes the `F_atm` of the step before, so it is a few percent above `F_atm` while the flux falls
`M_planet`	kg	Constant throughout (mass conservation)

What to look for¶

Cooling and solidification: T_magma decreases smoothly from 4000 K. When it crosses the solidus (~1700 K), Phi_global approaches zero and the run terminates with "Planet solidified!!".
Outgassing: as the melt fraction drops, volatiles transfer from the interior to the atmosphere. P_surf increases and the dissolved fraction in panel (f) decreases. This is the core coupling feedback that the production modules (CALLIOPE, Aragog) compute with full thermodynamics.
Energy balance: the net atmospheric flux F_atm (grey line in panel a, with the OLR in red just above it) and the interior flux (orange dashed) track each other. F_int is slightly above F_atm because the dummy interior takes the F_atm of the step before, and the flux falls with time. The absorbed stellar flux (blue dashed) is constant because the star is fixed.
Mass conservation: M_planet should remain constant within rounding. No atmospheric escape occurs in this configuration.

Next steps¶

Vary the greenhouse effect: increase atmos_clim.dummy.gamma toward 1.0 to slow cooling (more opaque atmosphere traps more heat) or decrease it toward 0 for faster cooling (more transparent)
Enable escape: set escape.dummy.rate = 1e4 and params.stop.escape.enabled = true to see atmospheric mass loss
Change volatile inventory: increase H_budget to 50,000 ppmw for a thicker steam atmosphere, or decrease it to 1,000 ppmw for a thinner one (~11,000 bar at the end of the run, against ~74,000 bar here). The solidification time stays the same, because the grey atmosphere uses a fixed greenhouse factor (see the parameter grid tutorial)
Move to production modules: the Earth analogue tutorial uses Aragog, Zalmoxis, CALLIOPE, and AGNI for a quantitatively meaningful simulation

Noack, L. & Lasbleis, M., Parameterisations of interior properties of rocky planets, Astronomy & Astrophysics, 638, A129, 2020. SciX. ↩