#!/usr/bin/env python3

# -*- coding: utf-8 -*-

"""

Created on Sun Dec 25 22:45:55 2016

@author: throop

"""

import numpy as np

import matplotlib.pyplot as plt

import matplotlib

import astropy

from astropy.table import Table

import astropy.table as table

from astropy.coordinates import SkyCoord

import astropy.coordinates as coord

import math

import hbt

import spiceypy as sp

import astropy.units as u

import astropy.constants as c

import pickle

from astroquery.vo_conesearch import conesearch

import astroquery

from astroquery.vizier import Vizier

from matplotlib.patches import Ellipse

import os.path # For expanduser

# NB: In future, conesearch will move from astropy to astroquery -- see mailing list 17-Mar-2017.

# But it hasn't happened yet, so even if I wanted to be pro-active, I can't.

#==============================================================================

# A quick wrapper to query the Gaia catalog using Vizier

# From https://michaelmommert.wordpress.com/2017/02/13/accessing-the-gaia-and-pan-starrs-catalogs-using-python/

#==============================================================================

def gaia_query(ra_deg, dec_deg, rad_deg, maxmag=20,

catalog="I/337/gaia")[0]

#==============================================================================

# Return a list of NH MegaCam Gaia stars that match a given position and mag limit

#==============================================================================

# NB: After I finished this function, I realized that RA of catalog excerpt is

# different than RA of MU69 KEM encounter. So maybe this section is just academic.

def nh_gaia_query(ra_deg, dec_deg, rad_deg, maxmag=20, maxsources=10000):

return nh_gaia[is_good]

#==============================================================================

# Read the NH MegaCam-Gaia star catalog from disk. Use a pickle file if available.

#==============================================================================

# NB: After I finished this function, I realized that RA of catalog excerpt is

# different than RA of MU69 KEM encounter. So maybe this section is just academic.

def load_nh_gaia():

return gaia

#==============================================================================

# Start of main program

#==============================================================================

#==============================================================================

# Define kernels and files

#==============================================================================

#file_tm = '/Users/throop/git/NH_rings/kernels_nh_pluto_mu69.tm' # C/A time 11:00:00 (??) - older

file_tm_dayside = '/Users/throop/git/NH_rings/kernels_nh_pluto_mu69_tcm22.tm' # C/A time 07:00:00 - newer version

file_tm_nightside = '/Users/throop/git/NH_rings/kernels_nh_pluto_mu69_nightside.tm' # Sort of hacked this together

file_hd_pickle = '/Users/throop/git/NH_rings/cat_hd.pkl'

DO_FLYBY_DAYSIDE = True # For this plot, do night-side (testing) or day-side (default) flyby?

#==============================================================================

# Initialize setting

#==============================================================================

if (DO_FLYBY_DAYSIDE): # Dayside flyby works as intended

file_tm = file_tm_dayside

side_str = 'dayside'

else:

file_tm = file_tm_nightside # I have not really validated that nightside works. Results look dfft than expected.

side_str = 'nightside'

sp.furnsh(file_tm) # Start up SPICE

hour = 3600

day = 24 * hour

fs = 15 # General font size

if ('tcm22' in file_tm):

utc_ca = '2019 1 jan 07:00:00'

else:

utc_ca = '2019 1 jan 03:09:00' # MU69 C/A time. I got this from GV.

et_ca = sp.utc2et(utc_ca)

dt_tof = 180 # Time-of-flight uncertainty (halfwidth, seconds)

# Define the dates of the OpNavs. I have no time for these, but it probably doens't matter.

# Taken from MOL = Master Obs List spreadsheet 27-Mar-2017

utc_opnav = np.array(

] )

et_opnav = np.zeros(np.size(utc_opnav))

for i in range(np.size(utc_opnav)):

et_opnav[i] = sp.utc2et(utc_opnav[i])

hbt.figsize((15,10))

mag_limit = 8 # Plot stars brighter than this.

plot_tick_every = 360 # Plot a time-tick every __ seconds

DO_PLOT_TOF = False # In general, plotting TOF is not necessary -- it

# is a translation in position along track,

# and changes the occultation times but not positions.

DO_PLOT_LORRI = True # Plot limits of LORRI FOV?

DO_SCALEBAR_ARCSEC = False

DO_UNITS_TITLE_DAYS = False # Flag: Use Days or Hours for the units in the plot title

DO_UNITS_TICKS_DAYS = False

DO_PLOT_UNCERTAINTY_MU69 = False # Make a plot of errorbars in MU69 position?

DO_TIMES_OPNAV = False # Plot times just for OpNavs?

DO_PLOT_WIDE = False # Make the plot extra-wide (cover full range of RA), or zoom on a narrow RA?

DO_PLOT_USNO = False

DO_LABEL_USNO = False

DO_PLOT_GAIA = False

DO_LABEL_GAIA = False

DO_PLOT_NH_GAIA = False

DO_LABEL_NH_GAIA = False

maglimit_plot = 100 # Plot stars brighter than this

# Define the start and end time for the plot. This is the main control to use.

case = 7

if (case == 1): # Outbound, 0.3 .. 96 hour

et_start = et_ca + 0.3*hour

et_end = et_ca + 96*hour

pad_ra_deg = 1 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 1

DO_LABEL_HD = True # Plot star IDs on chart

DO_PLOT_HD = True

hbt.figsize((15,10))

mag_limit = 8 # Plot stars brighter than this.

if (case == 2): # Inbound, -96h .. -0.1h

et_start = et_ca -96*hour

et_end = et_ca -0.1*hour

pad_ra_deg = 2 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 2

DO_PLOT_HD = True

DO_LABEL_HD = True

hbt.figsize((25,10))

mag_limit = 8 # Plot stars brighter than this.

if (case == 3): # Outbound, 2 .. 200 hour

et_start = et_ca + 2*hour

et_end = et_ca + 200*hour

pad_ra_deg = 0.5 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 0.5

DO_PLOT_HD = True

DO_LABEL_HD = True # Plot star IDs on chart

hbt.figsize((15,10))

mag_limit = 10 # Plot stars brighter than this.

if (case == 4): # Outbound, 2d .. 10 *** This one is key!! It has one good occultation.

et_start = et_ca + 2*day

et_end = et_ca + 20*day

pad_ra_deg = 0.01 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 0.01

DO_PLOT_HD = False

DO_LABEL_HD = DO_PLOT_HD # Plot star IDs on chart

DO_PLOT_USNO = True

DO_LABEL_USNO = DO_PLOT_USNO

plot_tick_every = 7200*5

hbt.figsize((15,10))

mag_limit = 12 # Plot stars brighter than this. HD is probably complete to 10 or so.

DO_PLOT_LORRI = False

if (case == 4.5): # Outbound, zoom on the good one!

et_start = et_ca + 3.0*day

et_end = et_ca + 20*day

pad_ra_deg = 0.005 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 0.005

DO_PLOT_HD = False

DO_LABEL_HD = DO_PLOT_HD # Plot star IDs on chart

DO_PLOT_USNO = True

DO_LABEL_USNO = DO_PLOT_USNO

plot_tick_every = 7200*5

hbt.figsize((15,10))

mag_limit = 12 # Plot stars brighter than this. HD is probably complete to 10 or so.

DO_PLOT_LORRI = False

if (case == 5): # Inbound, 2d .. 10

et_start = et_ca - 20*day

et_end = et_ca - 2*day

pad_ra_deg = 0.01 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 0.01

DO_PLOT_HD = False

DO_LABEL_HD = DO_PLOT_HD # Plot star IDs on chart

DO_PLOT_USNO = True

DO_LABEL_USNO = DO_PLOT_USNO

plot_tick_every = 7200*5

hbt.figsize((15,10))

mag_limit = 12 # Plot stars brighter than this. HD is probably complete to 10 or so.

DO_PLOT_LORRI = False

if (case == 6): # Inbound, K-40d .. K-14 (for Spencer one-off project on LORRI OpNav inbound occs, not for an MT)

# Using K-40 since it is one that JS did calculations for in his 22-Mar-2017 email.

# This is the routine I am using for search for LORRI inbound stellar occultations that would happen

# by serendipity during OpNav. The idea is to use USNO and/or the 'NH_GAIA' catalog. The latter is

# the special catalog prepared for me by Stephen Gwyn y JJ Kavalers.

# et_start = et_ca - 40*day

# et_end = et_ca - 14*day

et_start = np.amin(et_opnav)

et_end = np.amax(et_opnav)

pad_ra_deg = 0.003 # 0.019 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 0.0035

if (DO_PLOT_WIDE):

pad_ra_deg = 0.030

pad_dec_deg = 0.019

# Set the flags for which stars to plot. In general the code just supports one at a time.

DO_PLOT_HD = False

DO_LABEL_HD = DO_PLOT_HD # Plot star IDs on chart

DO_PLOT_USNO = False

DO_LABEL_USNO = DO_PLOT_USNO

DO_PLOT_GAIA = False

DO_LABEL_GAIA = False # DO_PLOT_GAIA

DO_PLOT_NH_GAIA = True

DO_LABEL_NH_GAIA = DO_PLOT_NH_GAIA

DO_TIMES_OPNAV = True # If True, set the plot time limits based on OpNav times.

# If False, uset them based on ET values that are defined in the code header.

plot_tick_every = 24*60*60

hbt.figsize((15,10))

mag_limit = 12 # Plot stars brighter than this. XXX This parameter is not really used.

maglimit_plot = 100 # Plot only stars brighter than this XXX This one is the one to use.

DO_PLOT_LORRI = False

DO_SCALEBAR_ARCSEC = True

DO_UNITS_TITLE_DAYS = True

DO_UNITS_TICKS_DAYS = True

DO_PLOT_UNCERTAINTY_MU69 = True

if (case == 7): # Inbound, Outer core. For checking whether it is worth doing a the LORRI

# Using K-40 since it is one that JS did calculations for in his 22-Mar-2017 email.

# This is the routine I am using for search for LORRI inbound stellar occultations that would happen

# by serendipity during OpNav. The idea is to use USNO and/or the 'NH_GAIA' catalog. The latter is

# the special catalog prepared for me by Stephen Gwyn y JJ Kavalers.

# et_start = et_ca - 40*day

# et_end = et_ca - 14*day

et_start = et_ca - 14*day

et_end = et_ca - 7*day

pad_ra_deg = 0.003 # 0.019 # Add additional padding at edge of plots, RA = x dir. Degrees.

pad_dec_deg = 0.0035

if (DO_PLOT_WIDE):

pad_ra_deg = 0.030

pad_dec_deg = 0.019

# Set the flags for which stars to plot. In general the code just supports one at a time.

DO_PLOT_HD = False

DO_LABEL_HD = DO_PLOT_HD # Plot star IDs on chart

DO_PLOT_USNO = False

DO_LABEL_USNO = DO_PLOT_USNO

DO_PLOT_GAIA = False

DO_LABEL_GAIA = False # DO_PLOT_GAIA

DO_PLOT_NH_GAIA = True

DO_LABEL_NH_GAIA = DO_PLOT_NH_GAIA

DO_TIMES_OPNAV = False # If True, set the plot time limits based on OpNav times.

# If False, uset them based on ET values that are defined in the code header.

plot_tick_every = 24*60*60

hbt.figsize((15,10))

mag_limit = 12 # Plot stars brighter than this. XXX This parameter is not really used.

maglimit_plot = 100 # Plot only stars brighter than this XXX This one is the one to use.

DO_PLOT_LORRI = False

DO_SCALEBAR_ARCSEC = True

DO_UNITS_TITLE_DAYS = True

DO_UNITS_TICKS_DAYS = True

DO_PLOT_UNCERTAINTY_MU69 = True

# DO_LABEL_USNO = False

# DO_LABEL_GAIA = False

fov_lorri = 0.3*hbt.d2r # Radians. LORRI width.

num_dt = 21 # Number of timesteps. 4000 is way too many -- 200 should be fine.

# Define plot colors

color_kbo = 'purple'

color_kbo_radius = 'pink'

color_kbo_center = 'red'

color_lorri = 'lightgreen'

color_roche = 'yellow'

color_stars = 'green'

color_gaia = 'crimson'

color_tof = 'blue'

color_uncertainty_mu69 = 'blue'

hbt.set_fontsize(fs)

try:

lun = open(file_hd_pickle, 'rb')

print("Loading file: " + file_hd_pickle)

hd = pickle.load(lun)

lun.close()

except IOError: # If Pickle file not found, then go ahead and read the catalog from scratch

# Read HD into a table. Its positions are 1900. Coords are radians.

print ("Reading HD catalog...")

hd = read_hd_all()

num_stars = np.size(ra_1950)

# Precess stars from 1900 to 2000

print("Precessing to J2000")

mx = sp.pxform('B1950', 'J2000', et[0]) # SPICE knows about 1950, but not 1900!

ra_1950 = t['RA']

dec_1950 = t['Dec']

ra_2000 = []

dec_2000 = []

pt_1950 = []

# Loop over every star and precess it individually. This loop is a real bottleneck.

for i in range(num_stars):

pt_1900 = sp.radrec(1, ra_1950[i], dec_1950[i])

pt_1950 = sp.mxv(mx, pt_1900)

pt_2000 = sp.mxv(mx, pt_1950) # SPICE will not precess 1900 -> 2000. So we apply 1950 -> 2000, do it 2x.

d, ra_2000_i, dec_2000_i = sp.recrad(pt_2000)

dec_2000.append(dec_2000_i)

ra_2000.append(ra_2000_i)

hd['RA_2000'] = ra_2000

hd['Dec_2000'] = dec_2000

# Now save this pickle file, so we never have to run this routine again

lun = open(file_hd_pickle, 'wb')

pickle.dump(hd, lun)

lun.close()

print("Wrote: " + file_hd_pickle)

# Use the virtual observatory to get some UCAC2 stellar positions

# crval is a two-element array of [RA, Dec], in degrees

# Use conesearch.list_catalogs() to get list

encounter_phase = 'Inbound' if (et_start < et_ca) else 'Outbound'

# Set up the ET array

if DO_TIMES_OPNAV:

et = et_opnav

else:

et = hbt.frange(et_start, et_end, num_dt)

index_asymptote = 0 if (encounter_phase == 'Outbound') else 1

# We make one call to SPICE to look up the asymptote position

# Most SPICE calls are later, but we need to do this one now, ahead of order.

(st,lt) = sp.spkezr('MU69', et[index_asymptote], 'J2000', 'LT', 'New Horizons')

(junk, ra_asymptote, dec_asymptote) = sp.recrad(st[0:3])

crval = np.array([ra_asymptote, dec_asymptote]) * hbt.r2d

radius_search = 0.15 # degrees # We only use these catalog for very fine searches, so narrow is OK.

DO_PLOT_GSC = False # Goes to about v=12

if DO_PLOT_GSC:

name_cat = u'The HST Guide Star Catalog, Version 1.1 (Lasker+ 1992) 1' # works, but 1' errors; investigating

stars = conesearch.conesearch(crval, radius_search, cache=False, catalog_db = name_cat)

ra_stars = np.array(stars.array['RAJ2000'])*hbt.d2r # Convert to radians

dec_stars = np.array(stars.array['DEJ2000'])*hbt.d2r # Convert to radians

mag_stars = np.array(stars.array['Pmag'])

if DO_PLOT_USNO:

name_cat = u'The USNO-A2.0 Catalogue (Monet+ 1998) 1'

stars = conesearch.conesearch(crval, radius_search, cache=False, catalog_db = name_cat)

ra_stars = np.array(stars.array['RAJ2000'])*hbt.d2r # Convert to radians

id_stars = np.array(stars.array['USNO-A2.0']) # ID

id_stars = id_stars.astype('U') # Convert from byte string to Unicode, ugh.

dec_stars = np.array(stars.array['DEJ2000'])*hbt.d2r # Convert to radians

mag_b_stars = np.array(stars.array['Bmag'])

mag_r_stars = np.array(stars.array['Rmag'])

usno = Table([id_stars, ra_stars, dec_stars, mag_b_stars, mag_r_stars],

names = ['ID', 'RA_2000', 'Dec_2000', 'Bmag', 'Rmag'])

if DO_PLOT_GAIA:

stars = gaia_query(crval[0], crval[1], radius_search, maxmag=20, maxsources=10000)

ra_stars = np.array(stars['RA_ICRS'])*hbt.d2r # Convert to radians

id_stars = np.array(stars['Source'])

dec_stars = np.array(stars['DE_ICRS'])*hbt.d2r # Convert to radians

mag_stars = np.array(stars['__Gmag_'])

gaia = Table([id_stars, ra_stars, dec_stars, mag_stars],

names = ['ID', 'RA_2000', 'Dec_2000', 'mag'])

if DO_PLOT_NH_GAIA:

stars = nh_gaia_query(crval[0], crval[1], radius_search, maxmag=20, maxsources=10000)

ra_stars = np.array(stars['RA'])*hbt.d2r # Convert to radians

# id_stars = np.array(stars['Source'])

dec_stars = np.array(stars['Dec'])*hbt.d2r # Convert to radians

# Do a bit of maniplation to take G mag, or B mag, or mean, depending on what we have

mag_stars = np.array(stars['g'] + stars['r'])

is_both = (stars['r'] * stars['g']) > 0.0

mag_stars[is_both] /= 2

# Remove a very small number of stars that have clearly spurious magnitudes

is_error = (mag_stars > 50)

mag_stars[is_error] = 20

nh_gaia = Table([ra_stars, dec_stars, mag_stars],

names = ['RA', 'Dec', 'mag'])

#==============================================================================

# Look up NH position (ie, KBO position)

#==============================================================================

print("Looking up NH position...")

dt = et[1] - et[0] # Timestep

ra_kbo = []

dec_kbo = []

dist_kbo = []

ra_kbo_tof_minus = []

ra_kbo_tof_plus = []

dec_kbo_tof_minus = []

dec_kbo_tof_plus = []

for et_i in et:

# Nominal position

(st,lt) = sp.spkezr('MU69', et_i, 'J2000', 'LT', 'New Horizons')

(dist_i, ra_i, dec_i) = sp.recrad(st[0:3])

ra_kbo.append(ra_i) # MU69 RA/Dec, in radians

dec_kbo.append(dec_i)

dist_kbo.append(dist_i)

# TOF uncertainty -- negative

(st,lt) = sp.spkezr('MU69', et_i-dt_tof, 'J2000', 'LT+S', 'New Horizons')

(dist_i, ra_i, dec_i) = sp.recrad(st[0:3])

ra_kbo_tof_minus.append(ra_i) # MU69 RA/Dec, in radians

dec_kbo_tof_minus.append(dec_i)

# TOF uncertainty -- positive

(st,lt) = sp.spkezr('MU69', et_i+dt_tof, 'J2000', 'LT+S', 'New Horizons')

(dist_i, ra_i, dec_i) = sp.recrad(st[0:3])

ra_kbo_tof_plus.append(ra_i)

dec_kbo_tof_plus.append(dec_i)

# Convert all of these from lists to NP arrays

dec_kbo = np.array(dec_kbo)

ra_kbo = np.array(ra_kbo)

dec_kbo_tof_plus = np.array(dec_kbo_tof_plus)

dec_kbo_tof_minus = np.array(dec_kbo_tof_minus)

ra_kbo_tof_plus = np.array(ra_kbo_tof_plus)

ra_kbo_tof_minus = np.array(ra_kbo_tof_minus)

ra = hd['RA_2000']*hbt.r2d # Look up stellar coordinates - be sure to use J2000

dec = hd['Dec_2000']*hbt.r2d

dist_kbo = np.array(dist_kbo)*u.km # NH-MU69 distance

# Compute the size of the Hill sphere

rho_kbo = 2.5*u.gram/u.cm**3

r_kbo = 16.5 * u.km

m_kbo = 4/3 * math.pi * r_kbo**3 * rho_kbo

a_kbo = 43*u.au

a_hill = (a_kbo * (m_kbo / c.M_sun/3)**(1/3)).to('km')

# Compute angular size of Hill sphere

# In general the Hill sphere is just too large to plot... LORRI covers it only a couple days after C/A

angle_hill = (a_hill / dist_kbo).to('').value # Convert to radians, and drop the units

angle_hill = np.clip(angle_hill, -math.pi, math.pi) # Clip it (same as IDL > or < operator)

# Compute angular size of KBO itself

angle_kbo = (r_kbo / dist_kbo).to('').value # Radians

# Compute angular size of Roche limit

r_roche = 2.5 * r_kbo

angle_roche = (r_roche / dist_kbo).to('').value # Radians

# Set the x and y limits for the plot

xlim = hbt.mm(ra_kbo*hbt.r2d) # Plotting limits, in deg

ylim = hbt.mm(dec_kbo*hbt.r2d)

xlim = np.array(xlim) + pad_ra_deg * np.array([-1,1]) # Pad the plot by the specified amount

ylim = np.array(ylim) + pad_dec_deg * np.array([-1,1])

# Filter the stars. Set a flag for each one we want to plot, based on mag and position.

is_good = np.logical_and(

dec < ylim[1]) ) )

# Generate human-readable strings for the plot

if (DO_UNITS_TITLE_DAYS):

t_start_relative_str = "K{:+.0f}d".format((et_start - et_ca)/day)

t_end_relative_str = "K{:+.0f}d".format((et_end - et_ca)/day)

else:

t_start_relative_str = "K{:+}h".format((et_start - et_ca)/hour)

t_end_relative_str = "K{:+}h".format((et_end - et_ca)/hour)

#%%

#==============================================================================

# Make plot #1, which is in a fixed star field, with MU69 moving across it.

#==============================================================================

# Draw the main trajectory

hbt.figsize((21,12)) # Medium-sized figure

fig, ax = plt.subplots()

DO_PLOT_TRAJECTORY = True

if DO_PLOT_TRAJECTORY:

ax.plot(ra_kbo*hbt.r2d, dec_kbo*hbt.r2d, color=color_kbo_center)

# Plot the time-of-flight uncertainty.

# Concl: TOF error does not change the position of any of these occultations. It just changes the time

# at which we see them. This wasn't obvious to me, but these two curves lay over each other identically.

if DO_PLOT_TOF:

ax.plot(ra_kbo_tof_minus*hbt.r2d, dec_kbo_tof_minus*hbt.r2d, color=color_tof, marker = '+', ms=5)

ax.plot(ra_kbo_tof_plus*hbt.r2d, dec_kbo_tof_plus*hbt.r2d, color=color_tof, marker = '+', ms=10)

# Draw the LORRI FOVs -- 0.3 x 0.3 deg square

if DO_PLOT_LORRI:

dec_kbo_plus_lorri = dec_kbo + (fov_lorri/2) * np.sqrt(2) # All radians

dec_kbo_minus_lorri = dec_kbo - (fov_lorri/2) * np.sqrt(2)

ax.fill_between(ra_kbo*hbt.r2d, (dec_kbo_plus_lorri)*hbt.r2d, (dec_kbo_minus_lorri)*hbt.r2d,

color=color_lorri, alpha=0.5, label = 'LORRI FOV')

# Draw the Hill sphere

ax.plot(ra_kbo*hbt.r2d, (dec_kbo - angle_hill)*hbt.r2d, linestyle = '--')

ax.plot(ra_kbo*hbt.r2d, (dec_kbo + angle_hill)*hbt.r2d, linestyle = '--')

# Draw the Roche radius

ax.fill_between(ra_kbo*hbt.r2d, (dec_kbo+angle_roche)*hbt.r2d, (dec_kbo-angle_roche)*hbt.r2d,

color=color_roche, alpha=0.5, label = 'MU69 Roche radius = {} km'.format(r_roche.to('km').value))

# Draw the MU69 radius

ax.fill_between(ra_kbo*hbt.r2d, (dec_kbo+angle_kbo)*hbt.r2d, (dec_kbo-angle_kbo)*hbt.r2d,

color=color_kbo_radius, alpha=1, label = 'MU69 radius = {} km'.format(r_kbo.to('km').value))

# Draw '+' symbols every hour

for i,et_i in enumerate(et):

if (np.mod(et_i - et[0], plot_tick_every)) < dt:

if (i == 0): # Pass the info for this to plt.label()... but only for the first call!

kwargs = {'label' : 'MU69 position, SPICE, tcm22'}

else:

kwargs = {}

ax.plot(ra_kbo[i]*hbt.r2d, dec_kbo[i]*hbt.r2d, marker='+',

markersize=20, color='black', **kwargs )

ax.text(ra_kbo[0]*hbt.r2d, dec_kbo[0]*hbt.r2d, t_start_relative_str + ' ' ,

fontsize=12, clip_on=True, horizontalalignment='center') # Why does plt.text() kill my plot??

ax.text(ra_kbo[-1]*hbt.r2d, dec_kbo[-1]*hbt.r2d, ' ' + t_end_relative_str,

fontsize=12, clip_on=True, horizontalalignment='center') # Why does plt.text() kill my plot??

# Plot the HD stars

if DO_PLOT_HD:

ax.plot(hd['RA_2000'][is_good]*hbt.r2d, hd['Dec_2000'][is_good]*hbt.r2d, linestyle='none', marker='.',

label = 'HD, V < {}'.format(mag_limit), color=color_stars, clip_on = True)

# Label the HD stars

if (DO_LABEL_HD):

for i in range(np.size(is_good)):

if (is_good[i]):

string_hd = hd['ID'][i]

else:

string_hd = ''

ax.text(hd['RA_2000'][i]*hbt.r2d, hd['Dec_2000'][i]*hbt.r2d,

' {} {:.1f} {}'.format(hd['Type'][i], hd['Ptg'][i], string_hd),

fontsize = 8, clip_on = True)

# Plot the USNO stars

#plt.plot(usno['RA_2000'][is_good]*hbt.r2d, hd['Dec_2000'][is_good]*hbt.r2d, linestyle='none', marker='.',

# label = 'HD star', color=color_stars)

if DO_PLOT_USNO:

ax.plot(usno['RA_2000']*hbt.r2d, usno['Dec_2000']*hbt.r2d, linestyle='none', marker='.',

label = 'USNO positions', color=color_stars)

name_catalog = 'USNO'

if DO_LABEL_USNO:

for i in range(np.size(usno)):

ax.text(usno['RA_2000'][i]*hbt.r2d, usno['Dec_2000'][i]*hbt.r2d,

' B={:.1f}, R={:.1f} {}'.format(usno['Bmag'][i], usno['Rmag'][i], usno['ID'][i]),

fontsize = 8, clip_on = True)

# Plot the Gaia stars

if DO_PLOT_GAIA:

ax.plot(gaia['RA_2000']*hbt.r2d, gaia['Dec_2000']*hbt.r2d, linestyle='none', marker='.',

label = 'Gaia positions', color=color_gaia, alpha = 0.5)

name_catalog = 'Gaia'

if DO_LABEL_GAIA:

for i in range(np.size(gaia)):

ax.text(gaia['RA_2000'][i]*hbt.r2d, gaia['Dec_2000'][i]*hbt.r2d,

' {:.1f} {}'.format(gaia['mag'][i], gaia['ID'][i]),

fontsize = 8, clip_on = True)

if DO_PLOT_NH_GAIA:

ax.plot(nh_gaia['RA']*hbt.r2d, nh_gaia['Dec']*hbt.r2d, linestyle='none', marker='.',

label = 'Gaia-MegaCam positions', color=color_gaia, alpha = 0.5)

name_catalog = 'Gaia-MegaCam'

if DO_LABEL_NH_GAIA:

for i in range(np.size(nh_gaia)):

ax.text(nh_gaia['RA'][i]*hbt.r2d, nh_gaia['Dec'][i]*hbt.r2d,

' {:.1f}'.format(nh_gaia['mag'][i]),

fontsize = 8, clip_on = True)

# Plot a scalebar if requested

if (DO_SCALEBAR_ARCSEC):

d2as = 60. * 60.

as2d = 1/d2as

y0 = ylim[0]

dy = ylim[1] - ylim[0]

delta_pos_usno_as = 0.2

delta_pos_mu69_as = 0.5 # This is positional uncertainty now, from Earth. I want to map this into

# Get the distance to Pluto for start and end times

y0_scalebar = ylim[0] + 0.1*(ylim[1]-ylim[0])

x0_scalebar = xlim[0] + 0.1*(xlim[1]-xlim[0])

x1_scalebar = x0_scalebar + 1 * as2d

ax.hlines(y0 + dy * 0.1, x0_scalebar, x0_scalebar + 1*as2d)

ax.hlines(y0 + dy * 0.15, x0_scalebar, x0_scalebar + delta_pos_usno_as*as2d)

# plt.hlines(y0 + dy * 0.2, x0_scalebar, x0_scalebar + delta_pos_mu69_as*as2d)

ax.text(x0_scalebar, y0_scalebar, " = 1\" = {:.0f} km (at {}) = {:.0f} km (at {})".format(

(dist_kbo[ 0].value)*1*hbt.as2r,

t_start_relative_str,

(dist_kbo[-1].value)*1*hbt.as2r,

t_end_relative_str ),

bbox={'facecolor':'white', 'alpha':0.5, 'pad':10})

# ax.figtext(0.23, 0.23, "USNO accuracy = {}\"".format(delta_pos_usno_as))

# plt.figtext(0.23, 0.265, "MU69 uncertainty = {}\"".format(delta_pos_mu69_as))

# Plot error ellipses for the position of MU69. These are from JS's email to me, which are in turn from Marc Buie's

# slides as of 2-Mar-2017. I am assuming that everything is equatorial.

if (DO_PLOT_UNCERTAINTY_MU69):

dpos_x = 6413*u.km # Uncertainty in X position, km, from JS email 22-Mar-17. Halfwidth

dpos_y = 366*u.km # Uncertainty in Y position, km

angle = 3 # Rotation angle of ellipse, in degrees. This is a guess, just to make

# it vaguely align with what is in plots of Buie (from JS email).

# Plot at start of period

width = (dpos_x/dist_kbo[0]).value*hbt.r2d*2/math.cos(dec_kbo[0])

height = (dpos_y/dist_kbo[0]).value*hbt.r2d*2

xy = (ra_kbo[0]*hbt.r2d, dec_kbo[0]*hbt.r2d) # Get uncertainty in x and y, and convert to deg

ell = matplotlib.patches.Ellipse(xy=xy, width=width, height=height, angle = angle, alpha=0.1,

color=color_uncertainty_mu69, label = 'MU69 3$\sigma$ pos, Buie')

ax.add_patch(ell)

# Plot at end of period

width = (dpos_x/dist_kbo[-1]).value*hbt.r2d*2/math.cos(dec_kbo[-1])

height = (dpos_y/dist_kbo[-1]).value*hbt.r2d*2

xy = (ra_kbo[-1]*hbt.r2d, dec_kbo[-1]*hbt.r2d) # Get uncertainty in x and y, and convert to deg

ell = matplotlib.patches.Ellipse(xy=xy, width=width, height=height, angle = angle, alpha=0.1,

color=color_uncertainty_mu69)

ax.add_patch(ell)

#==============================================================================

# Plot a circle showing where rings would be, if they were there

#==============================================================================

# NB: There is a small bug here somewhere. Some stars are plotted *outside* the ring

# on this plot, but inside the ring on the other plot, or v/v.

# My guess is that this has to do with cos(dec)

DO_PLOT_RING_MU69 = True

if DO_PLOT_RING_MU69:

dpos_x = 3000*u.km # Radius of the rings to draw

dpos_y = 3000*u.km

angle = 0 # Rotation angle of ellipse, in degrees

# Plot ring at start of period

width = (dpos_y/dist_kbo[0]).value*hbt.r2d*2/math.cos(dec_kbo[0])

height = (dpos_y/dist_kbo[0]).value*hbt.r2d*2

xy = (ra_kbo[0]*hbt.r2d, dec_kbo[0]*hbt.r2d) # Get uncertainty in x and y, and convert to deg

ell = matplotlib.patches.Ellipse(xy=xy, width=width, height=height, angle = angle, alpha=0.5,

facecolor='none', edgecolor = 'grey', linewidth=3)

ax.add_patch(ell)

# Plot ring at end of period

width = (dpos_y/dist_kbo[-1]).value*hbt.r2d*2/math.cos(dec_kbo[-1])

height = (dpos_y/dist_kbo[-1]).value*hbt.r2d*2

xy = (ra_kbo[-1]*hbt.r2d, dec_kbo[-1]*hbt.r2d) # Get uncertainty in x and y, and convert to deg

ell = matplotlib.patches.Ellipse(xy=xy, width=width, height=height, angle = angle, alpha=0.5,

edgecolor='grey', facecolor='none', linewidth=3,

label = 'MU69 ring, radius = {:.0f} km'.format(dpos_x.to('km').value))

ax.add_patch(ell)

#==============================================================================

# Finalize and display the plot

#==============================================================================

if (DO_TIMES_OPNAV):

title_str = '{} .. {}, {} OpNav visits, {} flyby'.format(

t_start_relative_str, t_end_relative_str, np.size(et_opnav),

side_str)

else:

title_str = '{} .. {}, {} flyby'.format(t_start_relative_str, t_end_relative_str, side_str)

plt.title(title_str)

plt.xlim(xlim)

plt.ylim(ylim)

# Write the image to disk

dir_out = os.path.expanduser('~') + '/git/NH_rings/out/'

file_out = ('LORRI_occs_MU69_inbound_' + name_catalog +

(['', '_wide'][DO_PLOT_WIDE]) + '_' + side_str + '.png' )

plt.legend(loc = 'upper left')

plt.savefig(dir_out + file_out)

print("Wrote: " + dir_out + file_out)

plt.show()

#%%

#==============================================================================

# Make plot #2, which is with MU69 at center, and rings at a fixed scale, and stars moving across / into the ring.

#==============================================================================

# Stars pass from the outside toward the inside... that is, at the end,

# more and more stars are inside the ring, which makes sense.

hbt.figsize((18,18)) # Make a large figure here

radius_ring = 3000*u.km

radius_plot = 3400 # Radius, in km

color_stars = 'blue'

fig, ax = plt.subplots()

# Plot ring around MU69

if (DO_FLYBY_DAYSIDE): # Dayside flyby (nominal)

ax.set_xlim(radius_plot * np.array([-3,1.1])) # Set x range of plot. 0 is the x position of MU69

ax.set_ylim(radius_plot * np.array([-1.5,1.5]))

else: # Nightside flyby

ax.set_xlim(radius_plot * np.array([-1.1,3]))

ax.set_ylim(radius_plot * np.array([-1.5,1.5]))

# Grab RA and Dec for all stars

if (DO_PLOT_GAIA):

ra_stars = gaia['RA_2000']

dec_stars = gaia['Dec_2000']

mag_stars = gaia['mag']

name_catalog = 'Gaia'

if (DO_PLOT_NH_GAIA):

ra_stars = nh_gaia['RA']

dec_stars = nh_gaia['Dec']

mag_stars = nh_gaia['mag']

if (DO_PLOT_USNO):

ra_stars = usno['RA_2000']

dec_stars = usno['Dec_2000']

mag_stars = usno['Bmag']

name_catalog = 'USNO'

for i,et_i in enumerate(et): # Plot for every timestep

d_ra_ang = (ra_stars - ra_kbo[i])/np.cos(dec_stars)

d_dec_ang = dec_stars - dec_kbo[i]

d_ra_km_proj = d_ra_ang * dist_kbo[i]

d_dec_km_proj = d_dec_ang * dist_kbo[i]

ax.plot(d_ra_km_proj[mag_stars < maglimit_plot],

d_dec_km_proj[mag_stars < maglimit_plot],

marker = '.', linestyle='none', color = color_stars, markersize=1)

# Label each star at its final position (at inner edge)