Input Deck Reference#

Auto-generated from schema.lua. Each table below documents one section of the config table.

config.basis#

Key

Type

Default

Required

Description

nnodes

double

—

No

Number of DG nodes per cell.

config.bc#

fluid#

Key

Type

Default

Required

Description

bc_i

string

—

No

Inner fluid boundary condition. Options: ‘reflecting’, ‘outflow’, ‘surface’, ‘periodic’.

bc_o

string

—

No

Outer fluid boundary condition. Options: ‘reflecting’, ‘outflow’, ‘surface’, ‘periodic’.

surface_pressure_i

number

—

No

Prescribed pressure at an inner surface boundary. Defaults to zero.

surface_pressure_o

number

—

No

Prescribed pressure at an outer surface boundary. Defaults to zero.

radiation#

Key

Type

Default

Required

Description

bc_i

string

—

No

Inner radiation boundary condition. Options: ‘reflecting’, ‘interior’, ‘free_streaming’, ‘marshak’, ‘periodic’. ‘interior’ uses the interior-state physical flux, analogous to fluid outflow but not a no-incoming-characteristics condition.

bc_o

string

—

No

Outer radiation boundary condition. Options: ‘reflecting’, ‘interior’, ‘free_streaming’, ‘periodic’. ‘interior’ uses the interior-state physical flux, analogous to fluid outflow but not a no-incoming-characteristics condition.

marshak_incoming_energy_i

number

—

No

Incoming volumetric radiation energy for an inner Marshak boundary.

config.composition#

Key

Type

Default

Required

Description

ncomps

int

—

No

Number of composition species.

config.diagnostics#

optical_depth#

Key

Type

Default

Required

Description

enabled

bool

false

No

Enable Rosseland optical depth output integrated inward from the outer boundary.

photosphere#

Key

Type

Default

Required

Description

enabled

bool

false

No

Enable photosphere history tracking from the optical-depth profile.

tau

double

0.6666666666666666

No

Optical-depth threshold used by the photosphere tracker. Must be > 0.

shock#

Key

Type

Default

Required

Description

enabled

bool

false

No

Enable shock history tracking using the strongest velocity compression.

config.engine#

thermal#

Key

Type

Default

Required

Description

enabled

bool

—

No

Enable thermal energy injection engine.

energy

double

—

No

Total injected energy in erg.

mend

double

—

No

Mass coordinate of injection upper boundary in Msun. Must be > 0.

mode

string

—

No

Injection mode. Options: ‘direct’, ‘asymptotic’.

operator_split

bool

false

No

Apply engine as operator-split source.

tend

double

—

No

End time for energy injection. Must be > 0.

config.eos#

Key

Type

Default

Required

Description

gamma

double

1.4

No

Adiabatic index. Default: 1.4.

k

double

—

No

Polytropic constant K. Required for polytropic EOS.

n

double

—

No

Polytropic index n. Required for polytropic EOS.

type

string

—

No

Equation of state type. Options:. ‘ideal’, ‘paczynski’, ‘marshak’, ‘polytropic’.

config.fluid#

limiter#

Key

Type

Default

Required

Description

b_tvd

double

1.0

No

TVD parameter b. Used with minmod limiter.

characteristic

bool

false

No

Enable characteristic limiting.

enabled

bool

true

No

Enable slope limiter for fluid.

gamma_i

double

—

No

WENO central weight. Required for WENO limiter.

gamma_l

double

—

No

WENO left weight. Inferred from gamma_i if omitted.

gamma_r

double

—

No

WENO right weight. Inferred from gamma_i if omitted.

m_tvb

double

0.0

No

TVB parameter M. Used with minmod limiter.

tci_opt

bool

false

No

Enable troubled-cell indicator.

tci_val

double

—

No

Troubled-cell indicator threshold.

type

string

minmod

No

Limiter type. Options: ‘minmod’, ‘moment’, ‘weno [experimental]’.

weno_r

double

2.0

No

WENO smoothness exponent. Must be > 0.

config.gravity#

Key

Type

Default

Required

Description

gval

double

—

No

Acceleration [cm s^-2] for the constant gravity model. Must be > 0.

limiter_energy_correction

bool

false

No

Experimental. Return to the fluid the gravitational potential energy that the end-of-step limiter moves by relocating interior mesh nodes, as a uniform per-cell specific-energy shift clamped to the EOS internal-energy floor. Cancels the committed-state limiter mesh work exactly, but not the per-stage-limiting or temporal residuals, so some drift remains. Not robust in violent/cold-supersonic regimes; see the gravity docs.

model

string

—

No

Gravity model. Options: ‘constant’, ‘spherical’.

operator_split

bool

false

No

Apply gravity as an operator-split source. Uses a self-contained direct source (pointwise g and g*v) instead of the coupled weak form, giving up well-balancing and exact energy conservation. Incompatible with limiter_energy_correction.

config.heating#

nickel#

Key

Type

Default

Required

Description

enabled

bool

—

No

Enable Ni56 decay heating.

model

string

—

No

Nickel heating model. Options. ‘jeffery’ and ‘full_trapping’.

operator_split

bool

false

No

Apply nickel heating as operator-split source.

config.ionization#

Key

Type

Default

Required

Description

fn_degeneracy

string

—

No

Path to degeneracy factors atomic data file.

fn_ionization

string

—

No

Path to ionization atomic data file.

ncomps

int

—

No

Number of species for Saha solver.

solver

string

linear

No

Saha solver mode. Options: ‘linear’, ‘log’.

config.mesh#

Key

Type

Default

Required

Description

geometry

string

—

No

Domain geometry. Options: ‘planar’, ‘spherical’.

grid_type

string

—

No

Grid spacing type. Options: ‘uniform’, ‘logarithmic’.

nx

double

—

No

Number of grid cells. Must be > 0.

xl

double

—

No

Left boundary of the domain.

xr

double

—

No

Right boundary of the domain. Must be > xl.

config.opacity#

Key

Type

Default

Required

Description

filename

string

—

No

Path to tabular opacity HDF5 file.

kP

double

—

No

Planck mean opacity (cm^2/g). Required for constant and powerlaw.

kP_offset

double

0.0

No

Planck opacity additive offset. Powerlaw only.

kR

double

—

No

Rosseland mean opacity (cm^2/g). Required for constant and powerlaw.

kR_offset

double

0.0

No

Rosseland opacity additive offset. Powerlaw only.

rho_exp

double

—

No

Density exponent for powerlaw opacity.

t_exp

double

—

No

Temperature exponent for powerlaw opacity.

type

string

—

No

Opacity model. Options: ‘tabular’, ‘constant’, ‘powerlaw’.

floors#

Key

Type

Default

Required

Description

core_planck

double

0.24

No

Core Planck floor. Used with core_envelope model.

core_rosseland

double

0.24

No

Core Rosseland floor. Used with core_envelope model.

env_planck

double

0.01

No

Envelope Planck floor. Used with core_envelope model.

env_rosseland

double

0.01

No

Envelope Rosseland floor. Used with core_envelope model.

planck

double

0.001

No

Planck floor value. Used with constant floor model.

rosseland

double

0.001

No

Rosseland floor value. Used with constant floor model.

type

string

core_envelope

No

Opacity floor model. Options: ‘core_envelope’, ‘constant’.

config.output#

Key

Type

Default

Required

Description

basename

string

No

Optional prefix for HDF5 output filenames; prepended to the problem name.

dt_fixed

double

—

No

Fixed timestep override. Optional.

dt_growth_frac

double

1.05

No

Initial timestep growth factor. Must be > 1.

dt_hdf5

double

—

No

Time interval between HDF5 outputs. Default: t_end / 100.

dt_init

double

1.000000e-16

No

Initial timestep. Must be > 0.

ncycle_out

double

100

No

Number of cycles between stdout output.

history#

Key

Type

Default

Required

Description

dt

double

—

No

Time interval between history writes. Default: dt_hdf5 / 10.

fn

string

athelas.hst

No

History output filename.

config.physics#

Key

Type

Default

Required

Description

composition

bool

false

No

Enable multi-species composition.

engine

bool

false

No

Enable energy injection engine.

gravity

bool

false

No

Enable gravitational source terms.

heating

bool

false

No

Enable nuclear heating sources.

ionization

bool

false

No

Enable Saha ionization. Requires composition = true.

radiation

bool

false

No

Enable radiation transport.

config.problem#

Key

Type

Default

Required

Description

cfl

double

—

No

CFL double for timestep control. Must be > 0.

name

string

—

No

Unique identifier for this simulation problem.

params

—

—

No

Problem-specific parameters. Validated by the problem generator.

config.radiation#

Key

Type

Default

Required

Description

ap_coefficient

double

0.0

No

Coefficient C in the AP LLF damping factor 1 / (1 + C tau). C = 0 recovers standard LLF (no damping); C > 0 enables the optical-depth correction. Must be >= 0.

discretization

string

—

No

Spatial discretization of the transport term. Options: ‘implicit’ or ‘explicit’.

limiter#

Key

Type

Default

Required

Description

b_tvd

double

1.0

No

TVD parameter b. Used with minmod limiter.

characteristic

bool

false

No

Enable characteristic limiting. Currently unsupported for radiation.

enabled

bool

true

No

Enable slope limiter for radiation.

gamma_i

double

—

No

WENO central weight. Required for WENO limiter.

gamma_l

double

—

No

WENO left weight. Inferred from gamma_i if omitted.

gamma_r

double

—

No

WENO right weight. Inferred from gamma_i if omitted.

m_tvb

double

0.0

No

TVB parameter M. Used with minmod limiter.

tci_opt

bool

false

No

Enable troubled-cell indicator.

tci_val

double

—

No

Troubled-cell indicator threshold.

type

string

minmod

No

Limiter type. Options: ‘minmod’, ‘moment’, ‘weno [experimental]’.

weno_r

double

2.0

No

WENO smoothness exponent. Must be > 0.

newton#

Key

Type

Default

Required

Description

atol

double

0.0

No

Absolute residual tolerance for the implicit transport Newton iteration. Accepts the solve when the residual norm falls below this value regardless of relative reduction. 0 disables.

max_iter

int

10

No

Maximum Newton iterations for implicit transport solve.

tol

double

1.000000e-08

No

Convergence tolerance for implicit transport Newton iteration.

timestep#

Key

Type

Default

Required

Description

energy_change_scale

double

1.000000e-10

No

Specific radiation energy scale used in the implicit transport relative-change timestep controller.

max_change_f

double

—

No

Maximum allowed absolute change in radiation reduced flux. Timestep control for implicit transport.

max_fractional_change_e

double

—

No

Maximum allowed fractional change in radiation energy. Timestep control for implicit transport.

config.time#

Key

Type

Default

Required

Description

integrator

string

—

No

Time integration method. E.g. ‘IMEX_SSPRK11’, ‘IMEX_ARK32_ESDIRK’.

nlim

double

-1

No

Maximum number of cycles. -1 for unlimited.

t_end

double

—

No

End time of the simulation.