Downloading and plotting LASCO C3 data

How to download SOHO/LASCO C3 data with Fido and make a plot.

import matplotlib.pyplot as plt

from import Fido, attrs as a
from import read_file

In order to download the required data, we use, a downloader client. We define two search variables: a timerange and the instrument.

timerange = a.Time('2002/05/24 11:06', '2002/05/24 11:07')
instrument = a.Instrument('LASCO')
detector = a.Detector('C3')
result =, instrument)

Let’s inspect the result


The following shows how to download the results. If we don’t provide a path it will download the file into the sunpy data directory. The output provides the path of the downloaded files.

downloaded_files = Fido.fetch(result)

The downloaded file lacks the correct meta. We want to open the file and access both the data and the header information.

data, header = read_file(downloaded_files[0])[0]

# Add the missing meta information to the header
header['CUNIT1'] = 'arcsec'
header['CUNIT2'] = 'arcsec'

With this fix we can load it into a map and plot the results. Please note that there is no plot displayed below as this example is skipped due to timeouts that can occur when you try to download LASCO C3 data.

lascomap =, header)
fig = plt.figure()

Total running time of the script: ( 0 minutes 0.000 seconds)

Gallery generated by Sphinx-Gallery