How do I plot galactic coordinates using matplotlib and astropy in python?

Question

I want to make a projection of millisecond pulsars on the galactic plane, much like this one from Sala et al. 2004:

I have tried several methods, and have not gotten anywhere. This is my current code, along with what it is producing:

import math as m
import numpy as np
import csv
import matplotlib.pyplot as plt
import pandas as pd
import astropy.coordinates as coord
import astropy.units as u
from astropy.io import ascii
from astropy.coordinates import SkyCoord

data = pd.read_csv('galacticwperiod.csv')

xarr = np.array(data.iloc[:,0])
yarr = np.array(data.iloc[:,1])
eq = SkyCoord(xarr[:], yarr[:], unit=u.deg)
gal = eq.galactic

#print(xarr)
#xarr = np.array(df.iloc[:,0])
#yarr = np.array(df.iloc[:,1])
#zarr = np.array(df.iloc[:,2])

#ra = coord.Angle(xarr[:], unit=u.hour)
#ra.degree
#ra = ra.wrap_at(180*u.degree)
#dec = coord.Angle(yarr[:], unit=u.deg)
#print(ra)

plt.figure(figsize=(6,5))
fig = plt.figure()
ax = fig.add_subplot(111, projection="aitoff")
plt.plot(gal.l.wrap_at(180*u.deg), gal.b.wrap_at(180*u.deg), linestyle='None')
ax.scatter(gal.l, gal.b, linestyle='None')


#ax.set_facecolor('xkcd:battleship grey')
#fig.patch.set_facecolor('white')
#ax.tick_params(axis='both', which='major', labelsize=10)
#ax.grid(color='b', linestyle='solid')
fig.show()
#plt.savefig('millisecondcoloraitoff.png', dpi=600)

Here are a few lines of the input file 'galacticwperiod.csv':

Gl,Gb
111.383,-54.849
305.898,-44.888
305.913,-44.877

This is the image it produces:

I am almost certain this is wrong because they are not distributed along the galactic plane, which they should be. The data I am using is from the ATNF Pulsar Catalog.

These are the sites I have already looked at, for reference:

https://stackoverflow.com/questions/33105898/astropy-matplotlib-and-plot-galactic-coordinates https://astropy4scipy2014.readthedocs.io/_static/notebooks/06_Celestial_Coordinates_solutions.html

Any help with this would be greatly appreciated.

Answer 1

This code reads coordinates as equatorial (ra, dec) and transforms them to galactic (l, b):

eq = SkyCoord(xarr[:], yarr[:], unit=u.deg)
gal = eq.galactic

The contents of 'galacticwperiod.csv' are already in galactic coordinates and should not be transformed. Something like this may give better results:

gal = SkyCoord(xarr[:], yarr[:], frame='galactic', unit=u.deg)

The other issue is that pyplot's geographic projections seem to expect angles in radians. This code:

plt.subplot(111, projection='aitoff')
plt.grid(True)
plt.scatter(gal.l.wrap_at('180d').radian, gal.b.radian)

produces this plot:

Answer 2

Looking at the plot you have, I notice that there is a concentration of dots at the "poles". The concentration should be along the equator, and particularly towards the centre of the galaxy.

I think you have your axes switched, so points along the galactic equator are being drawn along the vertical "prime meridian".

Answer 3

from astropy.io import fits    
from astropy import wcs    
import matplotlib.pyplot as plt

x = fits.open('f160w_noopt_drz.fits')['SCI']
y = wcs.WCS(x.header)

plt.subplots(figsize=(8, 8), subplot_kw={'projection':y})
plt.imshow(x.data, cmap='viridis', origin='lower', vmin=0, vmax=10)
plt.show()