POSEIDON.high_res

Functions

airtovac(wlum)

vactoair(wlum)

read_hdf5(file_path)

read_high_res_data(data_dir[, names])

fit_uncertainties(flux[, n_components, initial_guess, ...])

blaze_correction(flux, filter_size[, Print])

prepare_high_res_data(data_dir, name, spectrum_type, ...)

sysrem(data_array, uncertainties[, niter])

SYSREM procedure adapted from https://github.com/stephtdouglas/PySysRem, originally used for detrending light curves.

fast_filter(flux, uncertainties[, niter, Print])

TODO: Add docstrings.

make_data_cube(data[, n_components])

PCA_rebuild(flux[, n_components])

fit_out_transit_spec(flux, transit_weight[, degree, ...])

get_RV_range(Kp_range, Vsys_range, phi)

cross_correlate(Kp_range, Vsys_range, RV_range, wl, ...)

plot_CCF_phase_RV(phi, RV_range, CCF_phase_RV[, ...])

find_nearest_idx(array, value)

Function that will find the index of the value in the array nearest a given value

plot_CCF_Kp_Vsys(Kp_range, Vsys_range, CCF_Kp_Vsys[, ...])

loglikelihood_PCA(V_sys, K_p, d_phi, a, wl, ...)

Perform the loglikelihood calculation using Principal Component Analysis (PCA). Based on M. Line 2019.

loglikelihood_sysrem(V_sys, K_p, d_phi, a, b, wl, ...)

Perform the loglikelihood calculation using SysRem. Based on N. Gibson 2021.

loglikelihood_high_res(wl, planet_spectrum, ...)

Return the loglikelihood given the observed flux, Keplerian velocity, and centered system velocity.

get_rot_kernel(V_sin_i, wl, W_conv)

Get rotational kernel given V sin(i) and wavelength grid of the forward model.

remove_outliers(wl_grid, flux)

transit_model(R_p, R_s, a, phi)

make_injection_data(data, data_dir, name, wl, ...[, ...])

Module Contents

POSEIDON.high_res.airtovac(wlum)
POSEIDON.high_res.vactoair(wlum)
POSEIDON.high_res.read_hdf5(file_path)
POSEIDON.high_res.read_high_res_data(data_dir, names=None)
POSEIDON.high_res.fit_uncertainties(flux, n_components=5, initial_guess=[0.1, 200], Print=True)
POSEIDON.high_res.blaze_correction(flux, filter_size, Print=True)
POSEIDON.high_res.prepare_high_res_data(data_dir, name, spectrum_type, method, flux, wl_grid, phi, uncertainties=None, transit_weight=None, V_bary=None, pca_ncomp=4, sysrem_niter=15)
POSEIDON.high_res.sysrem(data_array, uncertainties, niter=15)

SYSREM procedure adapted from https://github.com/stephtdouglas/PySysRem, originally used for detrending light curves.

Use this function in a high resolutional rerieval. nord: number of spectral order. nphi: number of time-resolved phases. npix: number of wavelengths per spectral order.

Parameters:

  • data_array (2D np.array of float) – Blaze-corrected data of a single order. Shape: (nphi x npix)

  • uncertainties (np.array of float) – Time and wavelength dependent uncertainties obtained from fit_uncertainties. Shape: (nphi x npix)

  • Niter (int) – Number of basis vectors to consider.

Returns:

2D Array representing the residuals data of a single order after filtering.

Shape: (nphi x npix)

U (2D np.array of float):

Basis vectors obtained from SYSREM of a single vector. Shape: (nphi x N_iter + 1)

Return type:

residuals (np.array of float)

POSEIDON.high_res.fast_filter(flux, uncertainties, niter=15, Print=True)

TODO: Add docstrings. Use this function in a high resolutional rerieval.

Args:

Returns:

The residuals data of a single order after filtering.

Shape: (nord x nphi x npix)

Us (3D np.array of float):

Basis vectors obtained from SYSREM of a single vector. Shape: (nord x nphi x N_iter+1)

Return type:

residuals (3D np.array of float)

POSEIDON.high_res.make_data_cube(data, n_components=4)
POSEIDON.high_res.PCA_rebuild(flux, n_components=5)
POSEIDON.high_res.fit_out_transit_spec(flux, transit_weight, degree=2, spec='median', Print=True)
POSEIDON.high_res.get_RV_range(Kp_range, Vsys_range, phi)
POSEIDON.high_res.cross_correlate(Kp_range, Vsys_range, RV_range, wl, planet_spectrum, data, Print=True)
POSEIDON.high_res.plot_CCF_phase_RV(phi, RV_range, CCF_phase_RV, label=None, save_file_path=None, cmap=cmr.ember)
POSEIDON.high_res.find_nearest_idx(array, value)

Function that will find the index of the value in the array nearest a given value

Input: array, number

Output: index of value in array closest to that number

POSEIDON.high_res.plot_CCF_Kp_Vsys(Kp_range, Vsys_range, CCF_Kp_Vsys, label=None, Kp_expected=None, Vsys_expected=None, RM_mask_size=10, plot_slice=False, save_file_path=None, cmap=cmr.ember)
POSEIDON.high_res.loglikelihood_PCA(V_sys, K_p, d_phi, a, wl, planet_spectrum, star_spectrum, data)

Perform the loglikelihood calculation using Principal Component Analysis (PCA). Based on M. Line 2019. nord: number of spectral order. nphi: number of time-resolved phases. npix: number of wavelengths per spectral order.

Parameters:

  • V_sys (float) – The system velocity (km/s) at which we do the loglikelihood calculation.

  • K_p (float) – The Keplerian velocity (km/s) at which we do the loglikelihood calculation.

  • d_phi (float) – Phase offset.

  • a (float) – Scale parameter a.

  • wl (np.array of float) – Wavelength at which the model is defined.

  • planet_spectrum (np.array of float) – Observed flux of the planet (at distance = 1 pc). Should be broadened in the same way as the star_spectrum.

  • star_spectrum (np.array of float) – Observed flux of the star (at distance = 1 pc). Should be broadened in the same way as the star_spectrum.

  • data (dict) –

    Data dictionary read using utility.read_high_res_data. All values should be prepared beforehand. Has the following key-value pairss:

    reasiduals (3D np.array of float):

    3D Array representing the top principal components removed data. Shape: (nord x nphi x npix)

    flux (3D np.array of float):

    3D Array representing the raw flux of data. Shape: (nord x nphi x npix)

    V_bary (np.array of float):

    Array of time-resolved Earth-star velocity. We have absorbed V_sys into V_bary, so V_sys = V_sys_literature + d_V_sys. Shape: (nphi, )

    phi (np.array of float):

    Array of time-resolved phases. Shpae (nphi, )

Returns:

Log-likelihood value. CCF_sum (float):

Cross-correlation value.

Return type:

loglikelihood_sum (float)

POSEIDON.high_res.loglikelihood_sysrem(V_sys, K_p, d_phi, a, b, wl, planet_spectrum, data, star_spectrum=None)

Perform the loglikelihood calculation using SysRem. Based on N. Gibson 2021. nord: number of spectral order. nphi: number of time-resolved phases. npix: number of wavelengths per spectral order.

Parameters:

  • V_sys (float) – The system velocity (km/s) at which we do the loglikelihood calculation.

  • K_p (float) – The Keplerian velocity (km/s) at which we do the loglikelihood calculation.

  • d_phi (float) – Phase offset.

  • a (float) – Scale parameter for spectrum.

  • b (float or None) – Scale parameter for noise. If None, calculate loglikelihood using nulled b (b that maximizes logL).

  • wl (np.array of float) – Wavelength at which the model is defined.

  • planet_spectrum (np.array of float) – Observed flux of the planet (at distance = 1 pc). Should be broadened in the same way as the star_spectrum.

  • data (dict) –

    Data dictionary read using utility.read_high_res_data. All values should be prepared beforehand. Has the following key-value pairs:

    residuals (3D np.array of float):

    3D Array representing the residuals data after filtering. Shape: (nord x nphi x npix)

    Bs (list of 2D np.array of float):

    A list of basis vectors returned by fast_filter. Shape: (nord x nphi x npix)

    phi (np.array of float):

    Array of time-resolved phases. Shape (nphi, )

    transit_weight (np.array of float):

    Transit model of the planet. A value of 1 is out of transit, and 0 is full transit. Shape (nphi, )

    uncertainties (None or 3D np.array of float):

    Time and wavelength dependent uncertainties obtained from fit_uncertainties. Can be None only if b is also None. Shape: (nord x nphi x npix)

Returns:

Loglikelihood value.

Return type:

Loglikelihood (float)

POSEIDON.high_res.loglikelihood_high_res(wl, planet_spectrum, star_spectrum, data, spectrum_type, method, high_res_params, high_res_param_names)

Return the loglikelihood given the observed flux, Keplerian velocity, and centered system velocity. Should only use this function in a high resolutional rerieval.

Parameters:

  • wl (np.array of float) – Wavelength at which the model is defined.

  • planet_spectrum (np.array of float) – Observed flux of the planet (at distance = 1 pc). Should be broadened in the same way as the star_spectrum.

  • star_spectrum (np.array of float) – Observed flux of the star (at distance = 1 pc). Should be broadened in the same way as the star_spectrum.

  • data (dict) – Data dictionary read using utility.read_high_res_data. All values should be prepared beforehand. For more information, see code spec for loglikelihood_PCA or loglikelihood_sysrem.

  • model (dict) – Dictionary containing properties of the model.

Returns:

Loglikelihood calculated based on which filtering method (specified in data[‘method’]).

Return type:

loglikelihood (float)

POSEIDON.high_res.get_rot_kernel(V_sin_i, wl, W_conv)

Get rotational kernel given V sin(i) and wavelength grid of the forward model.

Parameters:

  • V_sin_i (float) – Projected rotational velocity of a star. The component of the rotational velocity along the line of sight, which is esponsible for broadening.

  • wl (np.array of float) – Wavelength grid of the forward model.

  • W_conv (int) – Width of the rotational kernel.

POSEIDON.high_res.remove_outliers(wl_grid, flux)
POSEIDON.high_res.transit_model(R_p, R_s, a, phi)
POSEIDON.high_res.make_injection_data(data, data_dir, name, wl, planet_spectrum, K_p, V_sys, method, a=None, continuum=None, W_conv=None, star_spectrum=None)