Stellar activity

The activity common object stores hyperparameters that are shared by stellar-activity models. A model can still decide that some of these quantities are dataset-specific; for example, the rotation period and decay timescale are usually shared, while the covariance amplitude is often fitted independently for each dataset.

Note

The activity common object only defines the available parameters, boundaries, priors, spaces, and fixed values. The model listed in the models section decides which parameters are used and whether they are common to all datasets or local to one dataset.

Model definition

• common object name: activity

• default common object for most Gaussian-process models of stellar activity

Shared stellar parameters

Several activity models can take the stellar rotation period or the activity decay timescale from the star_parameters common object instead of from activity.

use_stellar_rotation_period

• accepted values: True | False

• replaces Prot with rotation_period from the star_parameters common object.

use_stellar_activity_decay

• accepted values: True | False

• replaces Pdec with activity_decay from the star_parameters common object.

The same flags can be activated either in the activity common object or in the individual model definition. See the simultaneous photometry and spectroscopy GP example for a complete configuration.

Example

common:
  activity:
    boundaries:
      Prot: [10.0, 20.0]
      Pdec: [20.0, 1000.0]
      Oamp: [0.001, 1.0]
    priors:
      Prot: ['Gaussian', 14.0, 0.5]
  star:
    star_parameters:
      boundaries:
        rotation_period: [10.0, 20.0]
        activity_decay: [20.0, 1000.0]

Parameters

Only the parameters required by the selected models have to be defined in the YAML file.

Name	Parameter	Typical use
Prot	Stellar rotation period	Quasi-periodic, rotation, and multidimensional activity kernels
Pdec	Decay timescale of active regions	Quasi-periodic and exponential-sine periodic kernels
Pcyc	Long-timescale cycle or squared-exponential timescale	Quasi-periodic plus squared-exponential kernels
Oamp	Coherence scale of the periodic component	Quasi-periodic and exponential-sine periodic kernels
Hamp	Covariance amplitude	Dataset-specific amplitude in trained GP models
Camp	Secondary covariance amplitude	Cosine, derivative, or cycle component depending on the model
rot_sigma	Amplitude of a celerite2 rotation term	celerite2 rotation kernels
rot_Q0	Base quality factor of the celerite2 rotation term	celerite2 rotation kernels
rot_deltaQ	Difference between the rotation-mode quality factors	celerite2 rotation kernels
rot_fmix	Fractional amplitude of the secondary rotation mode	celerite2 rotation kernels
grn_period	Granulation SHO timescale	celerite2 granulation plus rotation
grn_sigma	Granulation SHO amplitude	celerite2 granulation plus rotation
grn_k*_period	Timescale of the k-th granulation SHO term	configurable celerite2 granulation/oscillation/rotation
grn_k*_sigma	Amplitude of the k-th granulation SHO term	configurable celerite2 granulation/oscillation/rotation
osc_k*_period	Timescale of the k-th oscillation SHO term	configurable celerite2 granulation/oscillation/rotation
osc_k*_sigma	Amplitude of the k-th oscillation SHO term	configurable celerite2 granulation/oscillation/rotation
osc_k*_Q0	Quality factor of the k-th oscillation SHO term	configurable celerite2 granulation/oscillation/rotation
sho_scale	SHO undamped period, or rho	SHO kernels
sho_decay	SHO damping timescale, or tau	SHO kernels
sho_sigma	SHO process amplitude	SHO kernels
matern32_scale	Matern-3/2 scale, also accepted as matern32_rho in some models	Matern-3/2 kernels
matern32_sigma	Matern-3/2 covariance amplitude	Single-output Matern-3/2 kernels
rot_amp	Coefficient of the derivative component of a latent GP	Multidimensional GP models
con_amp	Coefficient of the latent GP itself	Multidimensional GP models
cos_amp	Coefficient of the cosine latent component	Multidimensional quasi-periodic plus cosine kernel
cos_der	Coefficient of the derivative of the cosine component	Multidimensional quasi-periodic plus cosine kernel
cyc_amp	Coefficient of the squared-exponential cycle component	Multidimensional quasi-periodic plus squared-exponential kernels
cyc_der	Coefficient of the derivative of the squared-exponential cycle component	Multidimensional quasi-periodic plus squared-exponential kernel
matern32_multigp_sigma	Coefficient of the Matern-3/2 latent GP	Multidimensional Matern-3/2 kernel
matern32_multigp_sigma_deriv	Coefficient of the derivative of the Matern-3/2 latent GP	Multidimensional Matern-3/2 kernel
Vc, Vr, Lc, Bc, Br	Original Rajpaul-framework coefficients	Legacy gp_framework_quasiperiodic model