Lid seal¶
A crust that closes over the magma ocean can hold the volatiles dissolved in the melt; Volatiles held under a closed crust explains the model and the tutorial runs it. This page explains how to switch the seal on, what it needs, and which output columns to read. The settings are listed with their units and defaults in the config reference.
Switch it on¶
The seal needs the Aragog interior solver, because it reads the melt fraction and the radius of each mesh cell:
[interior_energetics]
module = "aragog"
[outgas.lid_seal]
enabled = true
delta_close = 1e4 # crust thickness at which the seal closes [m]
tau_leak = 1e5 # relaxation time of the held volatiles [yr]
The seal also needs:
outgas.module = "calliope";escape.reservoir = "outgas"(the default) or"bulk";- with the Zalmoxis structure,
interior_struct.zalmoxis.dry_mantle = true(the default).
It cannot be combined with params.stop.solid.freeze_volatiles = true, which also freezes the melt inventory, or with outgas.vapourise = true.
With the defaults tau_leak = inf and crust_rate = 0, no held mass leaves while the seal holds. The desiccation check counts the held mass. While the seal holds, a planet is therefore not desiccated if the held mass of any volatile element or noble gas is above outgas.mass_thresh. A planet is desiccated when it has lost its volatiles (definition). Set a finite tau_leak or a positive crust_rate to let the held volatiles reach the atmosphere.
Read the output¶
These output columns show what the seal does (full list):
seal_active[1]: 1 while the seal holds, else 0.seal_delta[m] andseal_phi_top[1]: the crust thickness and the melt fraction of the top mesh cell that the seal reads in the step. On the mesh,seal_deltais the depth of the top run of cells belowphi_solidwhile the seal is open, and belowphi_openwhile it holds and on the step where it opens, so it can jump on the first held step.seal_path[1]: the crust that a closed seal reads in the step: 1 for the crust on the mesh, 2 for the sub-grid lid; 0 while the seal is open, also whenseal_deltacomes from the sub-grid lid.{sp}_kg_sealed,{sp}_mol_sealedand{e}_kg_sealed: the held mass of each species and element.{sp}_kg_eq[kg]: the floor of each soluble species at the end of the step; 0 while the seal is open, on the step where it closes, withoutgas.calliope.solubility = false, and on held steps with no leak rate (tau_leakinfinite and no crust production) or nothing held.seal_t_close[yr]: the time the seal last closed.seal_Tmagma_close[K]: the magma temperature of the chemistry solve that closed the seal, at leastoutgas.T_floor.seal_phi_close[1]: the global melt fraction of the step that closed the seal, clipped to [0, 1].seal_xtal_kgandseal_tl_kg[kg]: under front trapping, the crystals formed since the seal closed and the melt trapped in them, which the reservoir of the floor counts; a remelt removes the youngest first. Both are 0 at closure and while the seal is open.seal_crust_rate[kg yr\(^{-1}\)] andseal_volc_capped[1]: the rate at which melt reaches the crust in the step, and whethercrust_rate dtwas cut to the melt mass.{e}_kg_leak_tauand{e}_kg_leak_volc[kg]: the mass of each element that leaked in the step by relaxation and by volcanism.seal_leak_kg_cum[kg]: the leaked mass since the start of the run.seal_open_kg[kg]: the held mass that returned to the melt in a step where the seal opened.
Resume a sealed run¶
proteus start -c <config.toml> --resume continues a run that stopped while the seal held. The resume stops with an error in these cases:
- the last row is sealed and
seal_Tmagma_closeis not finite or not above 0 K; - the last row holds sealed mass, but the seal is off in the config (resume with
outgas.lid_seal.enabled = true) orseal_activeis 0; - with front trapping,
seal_xtal_kgorseal_tl_kgis not finite or negative; - with front trapping, both are 0 while
Phi_globalis belowseal_phi_closeor either melt fraction is not finite; - with front trapping,
seal_t_closeis not above 0 yr.
A resume fills a missing column with 0. Validation checks these cases.