Overlaying off-disk contours using a screen

This example shows how to use a screen to overlay off-disk contours from one map onto another map.

When overlaying contours from one data set onto another data set, there is almost always some small difference in the coordinate frame of the two data sets, namely in the observation time and the observer location. With the slight shift in perspective, any off-disk contours are not plotted by default due to the lack of knowledge of where the contour exists in 3D space. (In contrast, on-disk contours do not have this issue because the default assumption is that on-disk contours lie on the surface of the Sun.) To change the behavior for off-disk contours, we can use a screen to specify an assumption for the 3D location of 2D off-disk points.

See Reprojecting Using a Spherical Screen for a different use case for screens.

import matplotlib.pyplot as plt

import astropy.units as u

import sunpy.data.sample
import sunpy.map
from sunpy.coordinates.screens import SphericalScreen

We use two AIA images that were not taken at exactly the same time. We crop the maps to an active region at the eastern limb.

aia171_map = sunpy.map.Map(sunpy.data.sample.AIA_171_IMAGE)
aia193_map = sunpy.map.Map(sunpy.data.sample.AIA_193_IMAGE)
aia171_map = aia171_map.submap((0, 500)*u.pix, top_right=(250, 750)*u.pix)
aia193_map = aia193_map.submap((0, 500)*u.pix, top_right=(250, 750)*u.pix)

First, let’s plot the AIA 171 contours on their native map. The contours extend off disk due to the coronal loops.

fig = plt.figure()
ax = fig.add_subplot(projection=aia171_map)
aia171_map.plot(axes=ax)
aia171_map.draw_contours(axes=ax, levels=[1000, 2500] * u.DN, colors="green")

<matplotlib.contour.QuadContourSet object at 0x741cbd183bc0>

Next, let’s try overlaying those same AIA 171 contours on the AIA 193 map. Note that none of the off-disk contours appear. Again, this is because it is not well defined where to plot such contours from the vantage point of the AIA 193 map.

fig = plt.figure()
ax = fig.add_subplot(projection=aia193_map)
aia193_map.plot(axes=ax, title="Default overlay behavior")
aia171_map.draw_contours(axes=ax, levels=[1000, 2500] * u.DN, colors="green")

<matplotlib.contour.QuadContourSet object at 0x741cb94f44a0>

Finally, let’s overlay those contours again, but this time using the SphericalScreen context manager. This screen assumes that off-disk points lie on the inside of a spherical screen. Note that the off-disk contours are now visible.

fig = plt.figure()
ax = fig.add_subplot(projection=aia193_map)
aia193_map.plot(axes=ax, title="Using SphericalScreen")
with SphericalScreen(aia171_map.observer_coordinate, only_off_disk=True):
    aia171_map.draw_contours(axes=ax, levels=[1000, 2500] * u.DN, colors="green")

plt.show()