Plasma Configuration

The plasma configuration gives TARDIS the necessary information to calculate the plasma state (see Plasma):

type	object
properties
initial_t_inner	initial temperature of the inner boundary black body. If set to -1 K will result in automatic calculation of boundary
	type	quantity
	default	-1 K
initial_t_rad	initial radiative temperature in all cells. If set to -1 K will result in automtatic calculation of the initial temperatures
	type	quantity
	default	-1 K
disable_electron_scattering	disable electron scattering process in montecarlo part - non-physical only for tests
	type	boolean
	default	False
disable_line_scattering	disable line scattering process in montecarlo part - non-physical only for tests
	type	boolean
	default	False
ionization	ionization treatment mode
	type	string
	enum	nebular, lte
excitation	excitation treatment mode
	type	string
	enum	lte, dilute-lte
radiative_rates_type	radiative rates treatment mode
	type	string
	enum	dilute-blackbody, detailed, blackbody
line_interaction_type	line interaction mode
	type	string
	enum	scatter, downbranch, macroatom
w_epsilon	w to use when j_blues get numerically 0. - avoids numerical complications
	type	number
	default	1e-10
delta_treatment	In the saha calculation set delta equals to the number given in this configuration item. if set to None (default), normal delta treatment (as described in Mazzali & Lucy 1993) will be applied
	type	number
nlte	type	object
	default
	properties
	species	Species that are requested to be NLTE treated in the format [‘Si 2’, ‘Ca 1’, etc.]
		type	array
		default
	coronal_approximation	set all jblues=0.0
		type	boolean
		default	False
	classical_nebular	sets all beta_sobolevs to 1
		type	boolean
		default	False
	additionalProperties	False
continuum_interaction	type	object
	default
	properties
	species	Species that are requested to be treated with continuum interactios (radiative/collisional ionization and recombination) in the format [‘Si II’, ‘Ca I’, etc.]
		type	array
		default
	enable_adiabatic_cooling	enables adiabatic cooling of the electron gas
		type	boolean
		default	False
	enable_two_photon_decay	enables two photon decay processes
		type	boolean
		default	False
	additionalProperties	False
helium_treatment	none to treat He as the other elements. recomb-nlte to treat with NLTE approximation.
	type	string
	enum	none, recomb-nlte, numerical-nlte
	default	none
heating_rate_data_file	Path to file containing heating rate/light curve data.
	type	string
	default	none
link_t_rad_t_electron	Value used for estimating the electron temperature from radiation temperature.
	type	number
	default	0.9
nlte_ionization_species	List of species treated with nlte ionization. In the format [“H I”, “He II”] etc.
	type	array
	default
nlte_excitation_species	List of species treated with nlte excitation. In the format [“H I”, “He II”] etc.
	type	array
	default
nlte_solver	Selects NLTE population equation solver approach.
	type	string
	enum	root, lu
	default	root
additionalProperties	False

inital_t_inner is initial temperature (will be updated for most modes of TARDIS — see convergence section) of the black-body on the inner boundary. initial_t_rad is the initial radiation temperature (will be updated for most modes of TARDIS - see convergence section). For debugging purposes and to compare to synapps calculations one can disable the electron scattering. TARDIS will issue a warning that this is not physical. There are currently two plasma_type options available: nebular and lte, which tell TARDIS how to run the ionization equilibrium and level population calculations (see Plasma for more information). The radiative rates describe how to calculate the \(J_\textrm{blue}\) needed for the NLTE treatment calculations and Macro Atom calculations. There are three options for radiative_rates_type:

1) blackbody, in which \(J_\textrm{blue} = \textrm{Blackbody}(T_\textrm{rad})\)

2) dilute-blackbody in which \(J_\textrm{blue} = W \times \textrm{Blackbody}(T_\textrm{rad})\)

3) detailed in which the \(J_\textrm{blue}\) are calculated using an estimator (this is described in ../../../physics/montecarlo/estimators).

TARDIS currently supports three different kinds of line interaction: scatter — a resonance scattering implementation, macroatom — the most complex form of line interaction described in Macro Atom and downbranch a simplified version of macroatom in which only downward transitions are allowed (see Line Interaction Treatments).

Finally, w_epsilon describes the dilution factor to use to calculate \(J_\textrm{blue}\) that are 0, which causes problems with the code (so \(J_\textrm{blue}\) are set to a very small number).

Continuum Interaction

plasma:
    link_t_rad_t_electron: 1.0
    continuum_interaction:
        species:
            - H I
            - H II
            - He I
            - He II
        enable_adiabatic_cooling: True

This will add continuum interactions for all specified species. Setting \(T_\textrm{rad} = T_\textrm{electron}\) through link_t_rad_t_electron: 1.0 is recommended to enforce LTE (unless the simulation uses NLTE treatment). enable_adiabatic_cooling enables adiabatic cooling.

NLTE

nlte:
    coronal_approximation: True
    classical_nebular: False

The NLTE configuration currently allows setting coronal_approximation, which sets all \(J_\textrm{blue}\) to 0. This is useful for debugging with chianti for example. Furthermore, one can enable ‘classical_nebular’ to set all \(\beta_\textrm{Sobolev}\) to 1. Both options are used for checking with other codes and should not be enabled in normal operations.

NLTE Ionization

plasma:
    nlte_ionization_species: [H I, H II, He I, He II]
    nlte_solver: root

This option allows the user to specify which species should be included in the NLTE ionization treatment. Note that the species must be present in the continuum interaction species as well. Here, nlte_solver can be set to root or lu. root is the default and uses a root solver to calculate the NLTE populations. lu uses an iterative LU decomposition scheme to calculate the NLTE populations.

Note

lu iterates over the solutions up to a set tolerance. This tolerance is currently hard-coded to 1e-3. This can be changed in the code by changing the NLTE_POPULATION_SOLVER_TOLERANCE constant in tardis/plasma/properties/nlte_rate_equation_solver.py. Furthermore, the maximum number of iterations is set to 1000. This can be changed in the code by changing the NLTE_POPULATION_SOLVER_MAX_ITERATIONS constant in tardis/plasma/properties/nlte_rate_equation_solver.py.

Warning

lu is generally faster than root but does not solve explicitly for the electron density. Therefore, it is not recommended to use lu for simulations where the electron density is important (e.g. for simulations where NLTE excitation is important).