Overplotting the position of the Venus transit

How to accurately plot the position of Venus as it transitted in front of the Sun as observed by SDO/AIA.

import matplotlib.pyplot as plt

import astropy.units as u
from astropy.coordinates import SkyCoord
from astropy.coordinates import solar_system_ephemeris

from sunpy.net import Fido, attrs as a
import sunpy.map
from sunpy.coordinates import get_body_heliographic_stonyhurst

Let’s download an image of the Venus transit.

result = Fido.search(a.Time('2012/06/06 04:07:25', '2012/06/06 04:07:35'),
                     a.Instrument('aia'),
                     a.Wavelength(1600*u.angstrom))
files = Fido.fetch(result)
aiamap = sunpy.map.Map(files[0])

Out:

Files Downloaded:   0%|          | 0/1 [00:00<?, ?file/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:   0%|          | 0.00/7.54M [00:00<?, ?B/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:   1%|          | 47.1k/7.54M [00:00<00:22, 336kB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:   2%|1         | 120k/7.54M [00:00<00:19, 375kB/s] 

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:   3%|2         | 214k/7.54M [00:00<00:16, 432kB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:   5%|4         | 349k/7.54M [00:00<00:13, 517kB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:   7%|6         | 520k/7.54M [00:00<00:11, 624kB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  10%|9         | 718k/7.54M [00:00<00:09, 748kB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  12%|#2        | 941k/7.54M [00:00<00:07, 888kB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  16%|#5        | 1.19M/7.54M [00:01<00:06, 1.04MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  20%|#9        | 1.48M/7.54M [00:01<00:04, 1.22MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  23%|##2       | 1.70M/7.54M [00:01<00:04, 1.28MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  27%|##7       | 2.06M/7.54M [00:01<00:03, 1.53MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  33%|###2      | 2.48M/7.54M [00:01<00:02, 1.88MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  38%|###7      | 2.83M/7.54M [00:01<00:02, 2.06MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  43%|####2     | 3.21M/7.54M [00:01<00:01, 2.32MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  48%|####8     | 3.64M/7.54M [00:02<00:01, 2.70MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  54%|#####4    | 4.10M/7.54M [00:02<00:01, 3.08MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  61%|######    | 4.58M/7.54M [00:02<00:00, 3.31MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  68%|######7   | 5.09M/7.54M [00:02<00:00, 3.62MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  76%|#######5  | 5.70M/7.54M [00:02<00:00, 4.12MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  84%|########3 | 6.30M/7.54M [00:02<00:00, 4.42MB/s]

aia_lev1_1600a_2012_06_06t04_07_29_12z_image_lev1.fits:  92%|#########2| 6.96M/7.54M [00:02<00:00, 4.92MB/s]

                                                                                                            
Files Downloaded: 100%|##########| 1/1 [00:03<00:00,  3.65s/file]
Files Downloaded: 100%|##########| 1/1 [00:03<00:00,  3.65s/file]
/home/docs/checkouts/readthedocs.org/user_builds/sunpy/conda/stable/lib/python3.7/asyncio/base_events.py:623: ResourceWarning: unclosed event loop <_UnixSelectorEventLoop running=False closed=False debug=False>
  source=self)

For this example, we require high-precision ephemeris information. The built-in ephemeris provided by astropy is not accurate enough. This call requires jplephem to be installed. This will also trigger a download of about ~10 MB.

solar_system_ephemeris.set('de432s')

Out:

<ScienceState solar_system_ephemeris: 'de432s'>

Now we get the position of Venus and convert it into the SDO/AIA coordinates. The apparent position of Venus accounts for the time it takes for light to travel from Venus to SDO.

venus = get_body_heliographic_stonyhurst('venus', aiamap.date, observer=aiamap.observer_coordinate)
venus_hpc = venus.transform_to(aiamap.coordinate_frame)

Out:

INFO: Apparent body location accounts for 144.14 seconds of light travel time [sunpy.coordinates.ephemeris]

Let’s crop the image with Venus at its center.

fov = 100 * u.arcsec
top_right = SkyCoord(venus_hpc.Tx + fov, venus_hpc.Ty + fov, frame=aiamap.coordinate_frame)
bottom_left = SkyCoord(venus_hpc.Tx - fov, venus_hpc.Ty - fov, frame=aiamap.coordinate_frame)
smap = aiamap.submap(top_right, bottom_left)

Let’s plot the results.

ax = plt.subplot(projection=smap)
smap.plot()
smap.draw_limb()
ax.grid(False)
ax.plot_coord(venus_hpc, 'x', color='deepskyblue', label='Venus')
plt.legend()
plt.show()