Transit Duration Parameterization?

[eas342]

I noticed that the orbits.transit parameterizes the lightcurve in terms of duration.
https://jax.exoplanet.codes/en/latest/autoapi/jaxoplanet/orbits/transit/
Traditionally, I have seen transiting planets parameterized in terms of semi-major axis divided by the stellar radius, a/R*. Is there a reason duration might be better to sample? I'm wondering how one might deal with a non-circular orbit though, which has a non-trivial transit duration.

[soichiro-hattori]

Hi @eas342,

I think you should be able to parameterize your orbit with a/R* if you use the Keplerian orbits!!
The first part of the Quickstart tutorial might be what you're looking for?

[eas342]

Thank you @soichiro-hattori !

Ah, so for this a/R* parameterization and lightcurves, one should use from jaxoplanet.experimental.starry.light_curves import light_curve instead of from jaxoplanet.light_curves import limb_dark_light_curve? Does it do limb darkening and is it ready to be used, despite saying "experimental"?

[soichiro-hattori]

Sorry for a late response @eas342.

Ah, so for this a/R* parameterization and lightcurves, one should use from jaxoplanet.experimental.starry.light_curves import light_curve instead of from jaxoplanet.light_curves import limb_dark_light_curve?

You should be able to still use from jaxoplanet.light_curves.limb_dark import light_curve! The Occultation light curve of limb-darkened star section of the Quickstart tutorial shows an example!

Does it do limb darkening and is it ready to be used, despite saying "experimental"?

It does LD and this part is actually not in the "experimental" starry part. Also, @lgrcia can chime in more about this but the starry part should be pretty mature and ready to be moved out of "experimental"!

[eas342]

Super, thank you @soichiro-hattori ! Based on the information you sent, I made this short tutorial for myself, to match the Quick Start tutorial of the commonly-used batman package:
https://lkreidberg.github.io/batman/docs/html/quickstart.html

from jaxoplanet.orbits.keplerian import System, Central
import astropy.units as u
import astropy.constants as const
from jaxoplanet.units import unit_registry as ureg
from jaxoplanet.light_curves.limb_dark import light_curve
import numpy as np
import jax.numpy as jnp
from matplotlib import pyplot as plt

period = 1.0 * u.day## days
rp = 0.1 ## in Rp/R*
a = 15. ## in a/R*
inc = 87. * ureg.deg ## in degrees

# Use Kepler's 3rd law to calculate a mass given the a/R* and period

mass = (2. * np.pi / (period))**2 * (a * u.Rsun)**3 / const.G
#unitless_prefac = prefac.to(u.Msun).value

star = Central(
    radius=1.0 * ureg.R_sun,
    mass = mass.to(u.Msun),
)

system = System(star).add_body(
    period = period,
    radius=rp * ureg.R_sun,
    mass=0.0 * ureg.M_sun,
    inclination=inc,
)

#planet = system.bodies[0]

# checking the parameters of the system
#system

u = [0.1, 0.3]
time = jnp.linspace(-0.05, 0.05, 1000)

flux = 1.0 + light_curve(system, u)(time)


plt.plot(time,flux)
plt.xlabel("Time from central transit")
plt.ylabel("Relative flux")

Looks like the lightcurves agree with Batman to 0.1 ppm

Do you think this might be useful enough I should put it in the docs and do a pull request?