Forward Models (taurex.model)

Base

Class for forward spectral modeling.

class ForwardModel(name: str)

Bases: Loggable, Fittable, Writeable, Citable

A base class for producing forward models.

add_contribution(contrib: object) → None

Add a contribution to the forward model.

build() → None

Build forward model.

citations() → List[str]

Citations for forward model.

Will either return a list of string or a list of Citation objects if pybtex is installed.

compute_error(samples: Callable[[], float], wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, binner: Binner | None = None) → Tuple[Dict[str, ndarray[tuple[int, ...], dtype[float64]]], Dict[str, ndarray[tuple[int, ...], dtype[float64]]]]

Error of the model and its components given a sample function.

defaultBinner() → Binner

property derivedParameters: Dict[str, Tuple[str, str, Callable[[], float], bool]]

Returns a dictionary of derived parameters

property fittingParameters: Dict[str, Tuple[str, str, Callable[[], float], Callable[[float], None], Literal['linear', 'log'], bool, Tuple[float, float]]]

Returns a dictionary of fitting parameters

generate_profiles() → Dict[str, ndarray[tuple[int, ...], dtype[float64]]]

Generate profiles to store.

Must return a dictionary of profiles you want to store after modeling

initialize_profiles() → None

Initialize profiles.

classmethod input_keywords() → Tuple[str, ...]

Input keywords for forward model.

model(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]] | None, Dict | T | None]

Computes the forward model for a wngrid

model_contrib(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], Dict[str, Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], Any | None]]]

Computes the forward model for a wngrid with each contribution.

model_full_contrib(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], Dict[str, List[Tuple[str, ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], Any | None]]]]

Computes the forward model for a wngrid for each contribution

Considers each contribution has subcomponents as well.

write(output: OutputGroup) → OutputGroup

Write forward model to output group.

Basic Model

Basic forward model.

class OneDForwardModel(name: str, planet: Planet | None = None, star: Star | None = None, pressure_profile: PressureProfile | None = None, temperature_profile: TemperatureProfile | None = None, chemistry: Chemistry | None = None, nlayers: int | None = 100, atm_min_pressure: float | None = 0.0001, atm_max_pressure: float | None = 1000000.0, contributions: List[object] | None = None)

Bases: SimpleForwardModel

A forward model for a 1D atmosphere.

Automatically sets up the following profiles:

• Temperature

• Pressure

• Chemistry

• Planet

• Star

• Contributions

Must implement the following methods:

• path_integral()

  • Main integration function that must

  return the absorption and optical depth for the given wngrid.

class SimpleForwardModel(name: str, planet: Planet | None = None, star: Star | None = None, pressure_profile: PressureProfile | None = None, temperature_profile: TemperatureProfile | None = None, chemistry: Chemistry | None = None, nlayers: int | None = 100, atm_min_pressure: float | None = 0.0001, atm_max_pressure: float | None = 1000000.0, contributions: List[object] | None = None)

Bases: ForwardModel

A ‘simple’ 1D base model.

It in the sense that its just a fairly standard single profiles model. It will handle settingup and initializing, building collecting parameters from given profiles for you. The only method that requires implementation is:

• path_integral()

WARN = True

property altitudeProfile: ndarray[tuple[int, ...], dtype[float64]]

Atmospheric altitude profile in m.

auto_grid() → None

Automatically sets the native grid.

build() → None

Build the forward model.

Must be called at least once before running model()

Will automatically be called by model() if not called before.

property chemistry: Chemistry

Chemistry model.

citations() → str

Citations for model.

collect_derived_parameters()

Combine all derived parameters from all components.

collect_fitting_parameters() → None

Combine all fitting parameters from components.

compute_error(samples: Callable[[], float], wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, binner: Binner | None = None) → Tuple[Dict[str, ndarray[tuple[int, ...], dtype[float64]]], Dict[str, ndarray[tuple[int, ...], dtype[float64]]]]

Computes standard deviations from samples.

Parameters:

• samples – A callable function that returns a weight for each sample

• wngrid – Wavenumber grid, default is to use native grid

• binner – A Binner object to bin the spectrum

property densityProfile: ndarray[tuple[int, ...], dtype[float64]]

Atmospheric density profile in m-3.

generate_profiles() → None

Generate profiles to store.

initialize_profiles() → None

Initializes all profiles.

model(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]] | None, Dict | T | None]

Runs the forward model.

Parameters:

• wngrid – Wavenumber grid, default is to use native grid

• cutoff_grid – Run model only on wngrid given, default is True

Returns:

• native_grid – Native wavenumber grid, clipped if wngrid passed

• depth – Resulting depth

• tau – Optical depth.

• extra (None) – Empty

model_contrib(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], Dict[str, Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], Any | None]]]

Models each contribution seperately.

model_full_contrib(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], Dict[str, List[Tuple[str, ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], Any | None]]]]

Model each sub-component of each contribution.

Like model_contrib() except all components for each contribution are modelled

e.g. Absorption will be modelled seperately for each molecule in the atmosphere.

property nLayers: int

Number of layers.

property nativeWavenumberGrid: ndarray[tuple[int, ...], dtype[float64]]

Best grid for given cross-sections.

Searches through active molecules to determine the native wavenumber grid

Returns:

wngrid – Native grid

Return type:

array

Raises:

InvalidModelException – If no active molecules in atmosphere

path_integral(wngrid: ndarray[tuple[int, ...], dtype[float64]], return_contrib: bool | None) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Main integration function.

Must return the absorption and optical depth for the given wngrid.

property planet: Planet

Planet model.

property pressure: PressureProfile

Pressure model.

property pressureProfile: ndarray[tuple[int, ...], dtype[float64]]

Central atmospheric pressure profile in Pa.

set_native_grid(spectral_grid: Quantity | ndarray[tuple[int, ...], dtype[float64]]) → None

Sets the native grid.

Parameters:

wngrid – Wavenumber grid

property star: Star

Star model.

property temperature: TemperatureProfile

Temperature model.

property temperatureProfile: ndarray[tuple[int, ...], dtype[float64]]

Atmospheric temperature profile in K.

write(output: OutputGroup) → OutputGroup

Write forward model to output group.

Will also write all components to the output group.

Transmission

Transit forward model.

class TransmissionModel(planet: Planet | None = None, star: Star | None = None, pressure_profile: PressureProfile | None = None, temperature_profile: TemperatureProfile | None = None, chemistry: Chemistry | None = None, nlayers: int | None = 100, atm_min_pressure: float | None = 0.0001, atm_max_pressure: float | None = 1000000.0, contributions: List[object] | None = None, new_path_method: bool | None = False)

Bases: OneDForwardModel

A forward model for transits.

compute_absorption(tau: ndarray[tuple[int, ...], dtype[float64]], dz: ndarray[tuple[int, ...], dtype[float64]]) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Compute final absorption and optical depth.

compute_path_length() → List[ndarray[tuple[int, ...], dtype[float64]]]

Compute path length for each layer, new method.

compute_path_length_old(dz: ndarray[tuple[int, ...], dtype[float64]]) → List[ndarray[tuple[int, ...], dtype[float64]]]

Compute path length for each layer.

classmethod input_keywords() → Tuple[str, ...]

Input keywords for this class.

path_integral(wngrid: ndarray[tuple[int, ...], dtype[float64]], return_contrib: bool | None = False) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Compute path integral.

Calculates the absorption and optical depth for each layer assuming hemispherical geometry.

Emission

Radiative transfer modeeling of eclipses.

class EmissionModel(planet: Planet | None = None, star: Star | None = None, pressure_profile: PressureProfile | None = None, temperature_profile: TemperatureProfile | None = None, chemistry: Chemistry | None = None, nlayers: int | None = 100, atm_min_pressure: float | None = 0.0001, atm_max_pressure: float | None = 1000000.0, contributions: List[object] | None = None, ngauss: int | None = 4)

Bases: OneDForwardModel

A forward model for eclipses.

property albedoterm: float

Albedo term.

compute_final_flux(f_total: ndarray[tuple[int, ...], dtype[float64]]) → ndarray[tuple[int, ...], dtype[float64]]

Compute final flux value.

For emission this converts the flux emitted to:

\[F = \frac{F_{total}}{\F_{star}} \left(\frac{R_{planet}}{R_{star}}\right)^2\]

Parameters:

f_total (np.ndarray) – Flux emitted by the planet

evaluate_emission(wngrid: ndarray[tuple[int, ...], dtype[float64]], return_contrib: bool) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Evaluate emission flux on quadratures.

evaluate_emission_ktables(wngrid: ndarray[tuple[int, ...], dtype[float64]], return_contrib: bool) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Evaluate emission flux on quadratures using ktables.

classmethod input_keywords() → Tuple[str, ...]

Input keywords for this class.

property logBolometricFlux: float

log10 Flux integrated over all wavelengths (W m-2).

partial_model(wngrid: ndarray[tuple[int, ...], dtype[float64]] | None = None, cutoff_grid: bool | None = True) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Evaluate the flux and quandratues but do not compute the final flux.

path_integral(wngrid: ndarray[tuple[int, ...], dtype[float64]], return_contrib: bool) → Tuple[ndarray[tuple[int, ...], dtype[float64]], ndarray[tuple[int, ...], dtype[float64]]]

Evaluate emission flux and integrate quadratures for flux.

set_num_gauss(value: int, coeffs: ndarray[tuple[int, ...], dtype[float64]] | None = None) → None

Set number of gaussian quadrature points.

Parameters:

• value (int) – Number of gaussian quadrature points

• coeffs (np.ndarray, optional) – Coefficients for each gaussian quadrature point

set_quadratures(mu: ndarray[tuple[int, ...], dtype[float64]], weight: ndarray[tuple[int, ...], dtype[float64]], coeffs: ndarray[tuple[int, ...], dtype[float64]] | None = None)

Set quadrature points.

Parameters:

• mu (np.ndarray) – Quadrature points

• weight (np.ndarray) – Quadrature weights

• coeffs (np.ndarray, optional) – Coefficients for each gaussian quadrature point

property usingKTables: bool

Whether ktables are being used.

write(output: OutputGroup) → OutputGroup

Write model to output group.

contribute_ktau_emission(start_k, end_k, density_offset, sigma, density, path, weights, ngrid, layer, ngauss)

contribute_ktau_emission_numba(start_k, end_k, density_offset, sigma, density, path, weights, ngrid, layer, ngauss)

contribute_ktau_emission_numpy(start_k, end_k, density_offset, sigma, density, path, weights, ngrid, layer, ngauss)

Direct Image

Direct imaging model.

class DirectImageModel(planet: Planet | None = None, star: Star | None = None, pressure_profile: PressureProfile | None = None, temperature_profile: TemperatureProfile | None = None, chemistry: Chemistry | None = None, nlayers: int | None = 100, atm_min_pressure: float | None = 0.0001, atm_max_pressure: float | None = 1000000.0, contributions: List[object] | None = None, ngauss: int | None = 4)

Bases: EmissionModel

A forward model for direct imaging of exo-planets.

compute_final_flux(f_total: ndarray[tuple[int, ...], dtype[float64]]) → ndarray[tuple[int, ...], dtype[float64]]

Compute the final flux.

This is the emission flux that is observed at the telescope directly from an exo-planet.

classmethod input_keywords() → Tuple[str, ...]

Input keywords for this class.

compute_direct_image_final_flux(f_total, planet_radius, star_distance)