Tutorial

[2]:

from ReflectX import *

Anatomy of the XArrays

Each spectrum is stored as an XArray. XArrays provide details and a reference dimension, in this case all spectra are referenced to the “wavelength” coordinate.

Each XArray contains a stellar spectrum, a planet albedo spectrum, a planet/star contrast ratio spectrum (called ‘fpfs’), and a planet flux spectrum (where planet flux = fpfs * star flux). Flux spectra are from the surface of the planet/star.

Each cooridnate and data variable is accessed through the relevant keyword.

The unit for each cooridinate and data variable is shown by toggling the paper icon next to the variable and accessed through dot or keyword notation

[38]:

path = '/Volumes/Oy/'
Teff = 3500
planet = 'Saturn'
CtoO = 0.5
model = LoadModel(path, Teff, planet, CtoO, teq = 150, phase = 90, clouds = {'fsed':0.1, 'kzz':1e9})

[39]:

model.fpfs.units, model['fpfs'].units

[39]:

('erg/cm**2/s/cm/', 'erg/cm**2/s/cm/')

[40]:

plt.semilogy(model.wavelength, model.fpfs)
plt.xlabel('$\\lambda$ [{}]'.format(model.wavelength.units))
plt.ylabel('$F_p / F_*$ [{}]'.format(model.fpfs.units))

[40]:

Text(0, 0.5, '$F_p / F_*$ [erg/cm**2/s/cm/]')

They also contain a set of attributes outlining every parameter used to generate the model, accessed via .attrs

[25]:

model.attrs

[25]:

{'ReflectX.Model': ' ReflectX Gas Giant Planet Reflected Light PICASO model spectra  for a world with r=0.8, m=0.3 with 3000K/0.2Rsun/logg5.0 star at 0.6 au  Spectra are R=60,000; to get a different gridding use picaso.justdoit.mean_regrid  Column [0] is wavelength in microns; columns labeled  "fpfs" are the planet/star flux ratio spectrum, columns labeled "PlanetFlux" are the planet spectrum only All flux in units of ergs cm$^{-2}$ s$^{-1}$ cm$^{-1}$  Created 2025-11-04 by Logan Pearce',
 'modeltype': 'GasGiant',
 'wavelength.range': '[0.3, 2]',
 'star.teff': '3000',
 'star.radius': '0.2',
 'star.radius.unit': 'Rsun',
 'star.logg': '5.0',
 'star.metallicity': '1',
 'star.star.model.database': 'phoenix',
 'star.mass': '0.16',
 'planet.custom.pt.profile.dataframe': 'none',
 'planet.gravity': 'none',
 'planet.gravity.unit': 'm/(s**2)',
 'planet.radius': 0.8,
 'planet.radius.unit': 'Rjup',
 'planet.mass': 0.3,
 'planet.mass.unit': 'Mjup',
 'planet.tint': '100',
 'planet.separation': 0.6044891940409404,
 'planet.separation.unit': 'au',
 'planet.log.metallicity': '1.0',
 'planet.ctoo': 0.5,
 'planet.tiovo': 'no',
 'planet.path.to.correlated.k-coefficient.files': 'picaso/reference/kcoeff_2020/',
 'planet.type': 'Neptune',
 'planet.teq': 150,
 'bottom.pressure': '2.5',
 'top.pressure': '-6',
 'nlevel': '91',
 'nstr_upper': '85',
 'nstr_deep': '89',
 'nofczns': '1',
 'rfacv': '0.5',
 'phase': 90,
 'ntangle': '6',
 'ngangle': '6',
 'cloud species': ['CaTiO3',
  'Cr',
  'H2O',
  'KCl',
  'Mg2SiO4',
  'MgSiO3',
  'MnS',
  'NH3',
  'Na2S',
  'ZnS'],
 'fsed': 0.1,
 'kzz': '1e+09'}

[49]:

plt.semilogy(model.wavelength, model.fpfs)
plt.xlabel('$\\lambda$ [{}]'.format(model.wavelength.units))
plt.ylabel('$F_p / F_*$ [{}]'.format(model.fpfs.units))
plt.title('Contrast Ratio at R = {} \n for {} Mjup planet with \n Teff={} K and C/O={}'.format(model.attrs['spectral.resolution'],
                                                                                          model.attrs['planet.mass'],
                                                                                         model.attrs['star.teff'],
                                                                                         model.attrs['planet.ctoo']))
plt.annotate('T_eq = {0} K \n Sep = {1:.1f} {2} \n Phase = {3} deg \n f_sed = {4} \n Kzz = {5}'.format(
    model.attrs['planet.teq'],
    model.attrs['planet.separation'],
    model.attrs['planet.separation.unit'],
    model.attrs['phase'],
    model.attrs['fsed'],
    model.attrs['kzz']
),
            xy = (0.1,0.1), xycoords='axes fraction', va='bottom', fontsize=15)

[49]:

Text(0.1, 0.1, 'T_eq = 150 K \n Sep = 0.6 au \n Phase = 90 deg \n f_sed = 0.1 \n Kzz = 1e+09')

Loading the full model set allows easy comparison among properties

[3]:

path = '/Volumes/Oy/' # Path to the location of the "ReflectX" folder:
Teff = 3500
planet = 'Saturn'
CtoO = 0.5
model = LoadModel(path, Teff, planet, CtoO)

[6]:

teqs = [75, 100, 150, 300]
phase = 90
fsed = 1
kzz = 1e9

fig,ax = plt.subplots()
for teq in teqs:
    key = 'teq{}/phase{}/fsed{}/kzz{:.0e}'.format(teq,phase,fsed,kzz)
    m = model[key]
    ax.plot(m.wavelength, m.albedo, label='{} K'.format(teq))
ax.set_xlabel('$\\lambda$ [{}]'.format(m.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[6]:

<matplotlib.legend.Legend at 0x3477156a0>

Yuck that’s hard to look at. You can regrid the spectrum onto a lower resolution to make comparison easier.

[14]:

fig,ax = plt.subplots()
for teq in teqs:
    key = 'teq{}/phase{}/fsed{}/kzz{:.0e}'.format(teq,phase,fsed,kzz)
    m = model[key]
    wav,alb = MeanRegrid(m.wavelength, m.albedo, R = 1000)
    ax.plot(wav, alb, label='{} K'.format(teq))
ax.set_xlabel('$\\lambda$ [{}]'.format(m.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[14]:

<matplotlib.legend.Legend at 0x36568a390>

[19]:

teq = 150
phase = 90
fseds = [0.1,1,3,10]
kzz = 1e9

fig,ax = plt.subplots()
for fsed in fseds:
    key = 'teq{}/phase{}/fsed{}/kzz{:.0e}'.format(teq,phase,str(fsed).replace('.',''),kzz)
    m = model[key]
    wav,alb = MeanRegrid(m.wavelength, m.albedo, R = 1000)
    ax.plot(wav, alb, label='f_sed = {}'.format(fsed))
ax.set_xlabel('$\\lambda$ [{}]'.format(m.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[19]:

<matplotlib.legend.Legend at 0x366fb6de0>

[18]:

teq = 150
phase = 90
fseds = 1
kzzs = [1e9,1e11]

fig,ax = plt.subplots()
for kzz in kzzs:
    key = 'teq{}/phase{}/fsed{}/kzz{:.0e}'.format(teq,phase,fsed,kzz)
    m = model[key]
    wav,alb = MeanRegrid(m.wavelength, m.albedo, R = 1000)
    ax.plot(wav, alb, label='Kzz = {:.0e}'.format(kzz))
ax.set_xlabel('$\\lambda$ [{}]'.format(m.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[18]:

<matplotlib.legend.Legend at 0x1720b6de0>

[22]:

teq = 150
phases = [0,45,90,120]
fseds = 1
kzz = 1e9

fig,ax = plt.subplots()
for phase in phases:
    key = 'teq{}/phase{}/fsed{}/kzz{:.0e}'.format(teq,phase,str(fsed).replace('.',''),kzz)
    m = model[key]
    wav,alb = MeanRegrid(m.wavelength, m.albedo, R = 1000)
    ax.plot(wav, alb, label='{}$^\\circ$'.format(phase))
ax.set_xlabel('$\\lambda$ [{}]'.format(m.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[22]:

<matplotlib.legend.Legend at 0x34086ad20>

Loading single XArrays allows comparison across star/planet types

[4]:

path = '/Volumes/Oy/'
Teffs = [3000, 3500, 4000, 5000, 7000]
planet = 'Saturn'
CtoO = 0.5

fig,ax = plt.subplots()
for Teff in Teffs:
    model = LoadModel(path, Teff, planet, CtoO, teq = 150, phase = 90, clouds = {'fsed':1, 'kzz':1e9})
    wav,alb = MeanRegrid(model.wavelength, model.albedo, R = 1000)
    ax.plot(wav, alb, label='{} K'.format(Teff))
ax.set_xlabel('$\\lambda$ [{}]'.format(model.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[4]:

<matplotlib.legend.Legend at 0x33891b6b0>

[7]:

path = '/Volumes/Oy/'
Teff = 3500
planets = ['Neptune','SuperNeptune','Saturn','Jupiter','SuperJupiter']
CtoO = 0.5

fig,ax = plt.subplots()
for planet in planets:
    model = LoadModel(path, Teff, planet, CtoO, teq = 150, phase = 90, clouds = {'fsed':1, 'kzz':1e9})
    wav,alb = MeanRegrid(model.wavelength, model.albedo, R = 1000)
    ax.plot(wav, alb, label='{}'.format(planet))
ax.set_xlabel('$\\lambda$ [{}]'.format(model.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[7]:

<matplotlib.legend.Legend at 0x35028ad20>

[9]:

path = '/Volumes/Oy/'
Teff = 3500
planets = 'Saturn'
CtoOs = [0.5,1.0,1.5]

fig,ax = plt.subplots()
for CtoO in CtoOs:
    model = LoadModel(path, Teff, planet, CtoO, teq = 150, phase = 90, clouds = {'fsed':1, 'kzz':1e9})
    wav,alb = MeanRegrid(model.wavelength, model.albedo, R = 1000)
    ax.plot(wav, alb, label='C/O = {}'.format(CtoO))
ax.set_xlabel('$\\lambda$ [{}]'.format(model.wavelength.units))
ax.set_ylabel('Albedo')
plt.legend()

[9]:

<matplotlib.legend.Legend at 0x317da5c40>

Tutorial

Loading a full model

Loading individual spectrum XArrays

Anatomy of the XArrays

Loading the full model set allows easy comparison among properties

Loading single XArrays allows comparison across star/planet types

Filters