Note

Download full notebook here

Using Full Output for Diagnostics

Sometimes it helps to have a bigger picture of what the full output is doing.

These steps will guide you through how to get out gain more intuition from your runs.

import warnings
warnings.filterwarnings('ignore')
import numpy as np
import pandas as pd
import astropy.units as u
#picaso
from picaso import justdoit as jdi
from picaso import justplotit as jpi

#plotting
from bokeh.io import output_notebook
output_notebook()
from bokeh.plotting import show

Loading BokehJS ...

We will use a cloudy Jupiter again to guide us through the exercise.

opa = jdi.opannection(wave_range=[0.3,1])

case1 = jdi.inputs()

case1.phase_angle(0)

case1.gravity(gravity=25, gravity_unit=u.Unit('m/(s**2)'))

case1.star(opa, 6000,0.0122,4.437)

case1.atmosphere(filename = jdi.jupiter_pt(), delim_whitespace=True)
case1.clouds(filename = jdi.jupiter_cld(), delim_whitespace=True)

Return PICASO Full Ouput

df = case1.spectrum(opa, full_output=True) #note the new last key
wno, alb, full_output = df['wavenumber'] , df['albedo'] , df['full_output']

Visualizing Full Output

Mixing Ratios

show(jpi.mixing_ratio(full_output))
#can also input any key word argument acceptable for bokeh.figure:
#show(jpi.mixing_ratio(full_output, plot_width=500, y_axis_type='linear',y_range=[10,1e-3]))

Cloud Profile

Depending on your wavelength grid, you might exceed Jupyter Notebook's data rage limit. You can fix this by initiating jupyter notebook with a higher data rate limit.

jupyter notebook --NotebookApp.iopub_data_rate_limit=10000000

show(jpi.cloud(full_output))

Pressure-Tempertaure Profile

show(jpi.pt(full_output))

Photon Attenuation Depth

This is a useful plot to see the interplay between scattering and absorbing sources. It should explain why you are getting bright versus dark reflectivity

show(jpi.photon_attenuation(full_output, at_tau=0.1, plot_width=500))

Compare it to the spectrum and you can see right away what is driving the overall shape of your spectrum

show(jpi.spectrum(wno,alb,plot_width=500))

Disecting Full Output

full_output.keys()

dict_keys(['weights', 'layer', 'wavenumber', 'wavenumber_unit', 'taugas', 'tauray', 'taucld', 'level', 'latitude', 'longitude', 'albedo_3d', 'reflected_unit'])

full_output['layer'].keys()

dict_keys(['pressure_unit', 'mixingratio_unit', 'temperature_unit', 'pressure', 'mixingratios', 'temperature', 'cloud'])

taugas = full_output['taugas'] #matrix that is nlevel x nwvno
taucld = full_output['taucld'] #matrix that is nlevel x nwvno
taugas = full_output['taugas'] #matrix that is nlevel x nwvno