Creating a visualization with ImageAnimatorWCS

This example shows how to create a simple visualization using ImageAnimatorWCS.

# Start by importing the necessary modules.
from itertools import product

import astropy.wcs
import astropy.units as u

from sunpy.time import parse_time
from sunpy.visualization.animator import ImageAnimatorWCS
from import AIA_171_IMAGE, AIA_193_IMAGE

To showcase how to visualize a sequence of 2D images using ImageAnimatorWCS, we will use images from our sample data. The problem with this is that they are not part of a continuous dataset. To overcome this we wil do two things. Create a stacked array of the images and create a WCS header. The easiest method for the array is to create a MapSequence.

# Here we only use two files but you could pass in a larger selection of files.
map_sequence =, AIA_193_IMAGE, sequence=True)
# Now we can just cast the sequence away into a NumPy array.
sequence_array = map_sequence.as_array()

Now we need to create the WCS header that ImageAnimatorWCS will need. To create the new header we can use the stored meta information from the map_sequence.

# This dictionary comprehension is extracting the three basic keywords we need
# to create a astropy.wcs.WCS header: 'CTYPE','CUNIT' and 'CDELT'
# from the meta information stored in the 'map_sequence'.
wcs_input_dict = {f'{key}{n+1}': map_sequence.all_meta()[0].get(f'{key}{n}')
                  for n, key in product([1, 2], ['CTYPE', 'CUNIT', 'CDELT'])}

# Now we need to get the time difference between the two observations.
t0, t1 = map(parse_time, [k['date-obs'] for k in map_sequence.all_meta()])
time_diff = (t1 - t0).to(u.s)
wcs_input_dict.update({'CTYPE1': 'TIME', 'CUNIT1':, 'CDELT1': time_diff.value})

# We can now just pass this into astropy.wcs.WCS to create our WCS header.
wcs = astropy.wcs.WCS(wcs_input_dict)

# Now the resulting WCS object will look like:


WCS Transformation

This transformation has 3 pixel and 3 world dimensions

Array shape (Numpy order): None

Pixel Dim  Data size  Bounds
        0       None  None
        1       None  None
        2       None  None

World Dim  Physical Type                   Units
        0  time                            s
        1  custom:pos.helioprojective.lon  deg
        2  deg

Correlation between pixel and world axes:

             Pixel Dim
World Dim    0    1    2
        0  yes   no   no
        1   no  yes  yes
        2   no  yes  yes

Now we can create the animation. ImageAnimatorWCS assumes the last two axes are the two that form the image. However, they are the first two in this case, so we change this by passing in the image_axes keyword.

wcs_anim = ImageAnimatorWCS(sequence_array, wcs=wcs, vmax=1000, image_axes=[0, 1])

You might notice that the animation could do with having the axes look neater. As we are using a WCS object, we can use wcsaxes to change the properties of the visualization. The documentation is here and details each possible method. In this example, we will just showcase some of these possible methods.

# We have to recreate the visualization since we displayed it earlier.
wcs_anim = ImageAnimatorWCS(sequence_array, wcs=wcs, vmax=1000, image_axes=[0, 1])

# We need to access the underlying WCSAxes object, so we can change the
# properties of our visualization. The methods on the ``axes.coords`` attribute
# of ``wcs_anim`` allows us to set various properties.
# We assign them to a easy to remember variable name.
# Note the order of assignment out from ``wcs_anim.axes.coords``.

time, solar_y, solar_x = wcs_anim.axes.coords

# Now we can label the X and Y axes.
solar_x.set_axislabel('Solar X (arsec)')
solar_y.set_axislabel('Solar Y (arsec)')

# Move the axis labels to avoid the slider.

# Now we can change the spacing and we have to use `astropy.units` here.
# We are setting the spacing to be quite small here and this will cause overlap.
solar_x.set_ticks(spacing=1*u.arcmin, color='black')
solar_y.set_ticks(spacing=1*u.arcmin, color='black')

# We can make sure the ticks do not overlap.
# Set this one to false, so we can see the difference.

Total running time of the script: ( 0 minutes 2.191 seconds)

Gallery generated by Sphinx-Gallery