Configuration

SPARTA requires two configuration files to be modified:

• constants.py Defines the instrument modes, required reduction steps, reference files, and image processing parameters.

• planet.cfg Contains the light curve fitting parameters.

Data reduction parameters

constants.py

• INSTRUMENT Specifies the instrument to be used (now supports NIRSpec, NIRCam, MIRI_IMAGE, MIRI_LRS).

• SUBARRAY Defines the SUBARRAY in use (e.g., SUB2048, SUB512, SUB256, SUB128, FULL).

• FILTER Defines the filter in use (e.g., for NIRSpec/PRISM, use CLEAR; for NIRCam, use F322W2 or F444W; for MIRI_IMAGE, use F1500W etc.). Additional filters are under development.

• REFERENCE_FILES Paths to the calibration reference files (e.g., flat fields, dark). The master directory for all the reference files can be set up by REF_DIR.

  Users must obtain suitable reference files for the selected instrument and subarray from the JWST CRDS archive and specify the desired file versions.

• SKIP_REF/FLAT/SUPERBIAS/EMICORR Flags to skip specific reference file corrections during processing. Set to True or False as needed.

  For MIRI_IMAGE, these steps are skipped by default. We encourage users to experiment with the emicorr step and compare the resulting outputs.

Light curve fitting parameters

planet.cfg

The configuration file specifies both the inference settings and the light curve fitting parameters. Each row defines a parameter with columns:

• The first column is the parameter name.

• The second column is the parameter type: free, fixed, independent.

• PriorType can be:

  • U = Uniform (PriorPar1 = lower bound, PriorPar2 = upper bound)

  • LU = Log-Uniform (PriorPar1 = lower log bound, PriorPar2 = upper log bound)

  • N = Normal (PriorPar1 = mean, PriorPar2 = standard deviation)

Parameter	Description
fitting_method	Sampling engine (e.g., emcee for MCMC or dynesty for nested sampling).
mcmc_nwalkers	Number of walkers for emcee.
mcmc_burnin	Number of burn-in steps (discarded).
mcmc_production	Number of production steps (kept).
dynesty_nlive	Live points for nested sampling.
dynesty_bound	Bounding strategy (none, single, multi, balls, cubes).
dynesty_sampling	NS sampling method (unit, rwalk, slice, rslice, hslice).
dynesty_dlogz	Convergence threshold (smaller = stricter).
joint_fit	If True, fit multiple datasets simultaneously.
start_wave	Lower wavelength bound (nm).
end_wave	Upper wavelength bound (nm).
exclude	Integrations to exclude. Should be a list of lists, e.g., [[start1, end1], [start2, end2]…].
data_path	Path to the light-curve data from ap_extract.py.
lc_savepath	Output path for modeled light curve.
best_fit_savepath	Output path for best-fit parameters.
rp	Planet-to-star radius ratio (Rp/Rs)
fp	Planet-to-star flux ratio (secondary-eclipse depth).
per	Orbital period (days).
t0	Mid-transit epoch (BJD).
inc	Orbital inclination (deg).
b	Orbital impact parameter. Please ensure commenting out inc if b is used, and vice versa.
a_star	Scaled semi-major axis (a/Rs).
sqrt_ecosw	Eccentricity parameterization term. \(\sqrt{e}cos\omega\)
sqrt_esinw	Eccentricity parameterization term. \(\sqrt{e}sin\omega\)
t_secondary	Secondary-eclipse mid-time (BJD).
Rs	Stellar radius.
limb_dark	LD law (uniform, linear, quadratic, kipping2013. Only those supported by batman are allowed).
q1 and q2	Limb-darkening coefficient
Fstar	Stellar baseline flux normalization.
m	Linear slope.
A	Exponential ramp amplitude term.
tau	Exponential ramp time scale.
x_coeff	Linear decorrelation vs. detector x-position.
y_coeff	Linear decorrelation vs. detector y-position.
error_factor	Multiplicative inflation of uncertainties.