taylorbell57 / bell_ebm Goto Github PK

It seems to differ from the time of eclipse from batman (Kreidberg 2015)

This, combined with simultaneous wavelength phasecurves (Issue #6), can allow you to simultaneously find the albedo and temperature map of a planet when fitting the model to a real lightcurve (or multiple phasecurves e.g. TESS and Spitzer). The albedo map would affect the heat deposited in each region. It wouldn't be self-consistent (e.g. temperature wont affect albedo), just a fairly low order spherical harmonic map or something like that.

Good for a v2 release

Speed up orbital calculations when running models by pre-computing distances and true anomaly to avoid duplication of effort.

This could allow for the future possibility of using the code for inverting phasecurves of eccentric planets. That's not overly likely though, given that the computation time would need to be ~1 millisecond at most. That's not impossible though - current runs including plotting take ~10 seconds and I'm burning in the model way longer than needed for inverting noisy Spitzer phasecurves.

This will likely be for v2.0 - will allow for contributions from a rock/water surface as well as the atmosphere. Will require radiation transfer between layers which will be more difficult. The atmosphere will also have vWind, while rock wouldn't but water could, so the layers might rotate with respect to each other. Maybe ignore that to begin with though.

If you run planet = ebm.Planet(argp=45.), then you find planet.ssp() always returns [45, 0].

Need to allow easy access to parameters like heat capacity and radius while propagating the changes through the rest of the dependent parameters.

Just add another dimension to the arrays when computing lightcurves, and sum over that dimension if doing band-integrated flux. This could lead to different opacities for SWD and OLR which would be a nice future addition

Use a map object instead of a temperature array inside the code, or at least return the map object from the run_model function instead of the temperature array. This will save people from doing weird things with the slipping maps due to the advection method currently used in the main branch.