Virtual Observatory

The COATpy wrapping of pyvolib exposes a number of different virtual observatory tools. We will go over each of these in turn.

Name Position Resolver

Names of things

Presently the “names” of things have to be URL encoded or the tool will not resolve. Use IC+348 or IC348 instead of IC 348. Programmatically this looks like:

urllib.quote_plus("IC 348")

This will be fixed in an upcoming update.

Sesame is a basic name resolver that returns the position of objects as a tuple. You can specify which service you wish to use:

from coatpy import Sesame

simbad = Sesame(opt='S')

ned = Sesame(opt='N')

ned.resolve('M51')

# the services do not agree; doing this will produce an exception.
assert simbad.resolve('IC348') == ned.resolve('IC348')

Catalog Search

The ConeSearch tool is a standardized interface to one of hundreds of catalogs. The necessary parameters are:

• RA: decimal degrees
• DEC: decimal degrees
• SR: cone radius in decimal degrees

An optional parameter VERB varies the verbosity of the query response.

Verbs

Only VERB=1 is standardized across all catalogs. A VERB=1 query returns only the positions of objects found the queried catalog. VERB=2 returns the “default” set of columns chosen by the archive. VERB=3 returns all columns. Not all services respond to VERB parameters.

As with the web queries you can wrap up the parameters in a dictionary:

from coatpy import ConeSearch, Sesame

simbad = Sesame(opt='S')

params = {}
params['RA'] = simbad.resolve("IC348")[0]
params['DEC'] = simbad.resolve("IC348")[1]
params['SR'] = 1./60.
params['VERB'] = 3

Discovering catalogs

At this point the system breaks down a little bit as we are not providing a programmatic tool for discovering catalogs. You have to use an online Directory to search the catalogs you want. Here is an example Directory search. Once a catalog is found, you want to copy the “accessURL” from the search table or open the “Full Record” and open the “Simple Cone Search” option. Here is the catalog used in this example.

enter the query URL for the target catalog (see sidebar), and create a catalog handler ic348cxo that wraps the query tool. We use the getRaw function of the ConeSearch to retrieve the raw string result that we then dump into a file and extract back the data using astropy:

from astropy.table import Table
url = "http://vizier.u-strasbg.fr/viz-bin/votable/-A?-source=J/AJ/122/866&"
ic348cxo = ConeSearch(url)

with open('ic348cxo.xml','wb') as f:
    f.write(ic348cxo.getRaw(**params))

data = Table.read(f.name, format='votable')

Image Search

Siap is the tool for querying image services is:

from coatpy import Siap

The Siap acronym stands for Simple Image Access Protocol and as with ConeSearch the inputs are fairly simple:

• RA: Decimal Degrees
• DEC: Decimal Degrees
• SIZE: An image size in degrees
• FORMAT: [Optional] default: image/fits (or image/jpeg)

Note

Various image archives have appended other query parameters onto the SIAP standard to provide additional initial filtering of the images returned. A good example is the IRSA 2MASS service. The Siap tool currently throws Exceptions on extra parameters.

Exercise: find HST images of a young cluster

In this exercise we will go through the steps necessary for downloading images with Siap from the Hubble Legacy Archive.

The first step is to set up the query, which is nearly identical to the previous examples.

import urllib2
from coatpy import Siap, Sesame
import vo

simbad = Sesame(opt="S")

url = " http://hla.stsci.edu/cgi-bin/acsSIAP.cgi?strict=1"

hla = Siap(url)

params = {}
position = simbad.resolve('IC348')
params.update(zip(('RA', 'DEC'), position))
params['SIZE'] = 1./60.
params['FORMAT'] = 'image/fits'

with open('hla_ic348_images.xml', 'wb') as f:
    f.write(hla.getRaw(**params))

The main difference between a catalog and an image query is that an image query results in a series of pointers to the image files that can be examined and filtered before the reference files are downloaded. We use the vo module to parse the results, extracting the table of images that satisfy our query.

vot = vo.table.parse_single_table('hla_ic348_images.xml')
image_table = vot.array

The returned array contains 66 fields!

print(image_table.dtype.names)

If all we want is the actual data then the important column is URL while the filename column is also useful.

# just grab the first image returned
image = image_table[0]
image_url = image['URL']
filename = image['filename'] + '.fits'

with open(filename, 'wb') as image_file:
    image_handler = urllib2.urlopen(image_url)
    image_file.write(image_handler.read())
    image_handler.close()

Continue this exercise by examining the resulting image.

Click to Show/Hide Solution

We use astropy.io.fits to open the image file and examine its contents.

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

hdulist = fits.open(filename)

# check for multiple FITS extensions and their contents
# in this case the "PRIMARY" header is empty
for hdu in hdulist:
    print('name: {0}  type: {1}'.format(hdu.name, type(hdu.data)))

fig = plt.figure(figsize=(15, 7))
f1 = aplpy.FITSFigure(filename, hdu=1, subplot=[0.1,0.1,0.3,0.65], figure=fig)
f1.set_tick_labels_font(size='x-small')
f1.set_axis_labels_font(size='small')
f1.show_grayscale()

f2 = aplpy.FITSFigure(filename, hdu=2, subplot=[0.4,0.1,0.3,0.65], figure=fig)
f2.hide_xaxis_label()
f2.hide_xtick_labels()
f2.hide_yaxis_label()
f2.hide_ytick_labels()
f2.show_colorscale()

f3 = aplpy.FITSFigure(filename, hdu=3, subplot=[0.7,0.1,0.3,0.65], figure=fig)
f3.hide_xaxis_label()
f3.hide_xtick_labels()
f3.hide_yaxis_label()
f3.hide_ytick_labels()
f3.show_colorscale(cmap='spring')