Change units to fractions?

[hippke]

Currently many units in [km]. Change to RP/R{\odot}?

[reneheller]

Yes, using the stellar radius as a base unit makes sense. Or otherwise, uncertainties in the stellar radius would propagate into all the other parameters that depend on the stellar radius.

[hippke]

Problems with fractional units:

• Pandora coordinate system is in stellar radii where 1 stellar radius == 1 transit duration. The transit duration is defined as per_planet / pi * arcsin(sqrt((r_planet / 2 + R_star) ** 2) / a_planet) which includes r_planet and R_star as well as a_planet in units of [km] in order to get a time axis in [days] which we need to compare to the original data.
• Similarly, moon orbital period must be calculated. It is defined as (2 * pi * sqrt((a_moon * 1000.0) ** 3 / (G * (M_planet + M_moon)))) / 60 / 60 / 24 which includes a_moon, M_planet, M_moon in physical units.

I see no elegant way out of this. Let's go through the code together during our Hackathlon?

[hippke]

Done:

• r_planet in units of R_star
• r_moon in units of R_star
• a_moon removed as free parameter (only calculated and used internally)
• per_moon as new free parameter, in units of days ==> @reneheller Ich weiß du hast dir SI-Einheiten gewünscht, aber ich würde trotzdem vorschlagen, dass wir bei den freien Parametern die der User eingibt für per_planet und per_moon bei Tagen bleiben (nicht Sekunden). Das ist intuitiver. Intern, also im Code, ist jetzt alles in SI Einheiten. Im Cornerplot sollten wir auch "Tage" zeigen und nicht "Sekunden.
• R_star bleibt freier Parameter, von mir aus in [metern]