Data Buttons: Quick Mosaics for You and Me

Introduction

data_buttons offers a number of tools to query archives, download image data for a given source and mosaic it into a final image.

The biggest feature of this module is the Data ButtonTM, which allows you an easy way to put in an object (and optionally an image size), and get a mosaic out. No muss, no fuss.

Installation

  • Clone the data_buttons repository with git clone https://github.com/thomaswilliamsastro/data_buttons into a folder of your choice
  • Navigate to the directory that contains setup.py
  • Install using pip install -r requirements.txt -e .

Current capabilities

data_buttons currently works with GALEX (data_buttons.galex), SDSS (data_buttons.sdss), 2MASS (data_buttons.two_mass), WISE (data_buttons.wise), and HST (data_buttons.hst) data. In the case of GALEX, the mosaic is weighted by the exposure time. For HST data, outputs are produced for each proposal ID. More telescopes and data coming soon!

Warning

I am not well-versed in HST data, and that module should be considered in its infancy. Your mileage may vary, so be aware!

data_buttons also offers a module with image-related tools (data_buttons.tools). Currently, this allows you to create a mosaic of pre-downloaded files, without going through the whole Data Button. It also offers a tool to calculate the sigma-clipped background median of an image (for subtracting off a constant background), or a slightly fancier polynomial fitting tool. More useful tools coming soon!

Example: Creating a GALEX Mosaic of M51

Using the Data ButtonTM, we’ll create a GALEX mosaic of the classic poster child galaxy, M51.

import os
from data_button import galex

os.chdir('directory')
galex.galex_button('M51')

And that’s all there is to it! The program will query the necessary databases, download the data and mosaic it for you. It’ll put the final files (in this case, M51/GALEX/M51_NUV.fits and M51/GALEX/M51_FUV.fits), into the directory you specify in the os.chdir() call.

Advanced Example: HST Imaging of the Crab Nebula

As a more involved example, we’ll create a 3-colour mosaic of WFPC2 observations of the Crab Nebula (M1). This will involve the (slightly more) involved HST button.

The WFPC2 observations we’ll use come from proposal ID 8222 (PI J. Hester). Because the Crab has a lot of archival HST observations, we’ll just pare down to this particular proposal on this particular instrument.

import os
from data_button import hst, tools

os.chdir('directory')

galaxy = 'M1'
    prop_id = '8222'
    instrument = 'WFPC2/WFC'

    # Create the mosaics of M1.

    hst.hst_button(galaxy,
                   instruments=instrument,
                   prop_ids=prop_id,
                   skymethod='globalmin',
                   verbose=True,
                   )

Generally, I’ve found the default skymethod in the button (‘globalmin+match’) to be fine, but since these observations have very little overlap it causes weird issues with the background matching, so we’ll perform our own subtraction later. This will put mosaicked files into M1/HST/8222.

We’ll now optimize the size of these files for later work. In some cases, this can cut the file size down by a factor of 2, so it’s worth doing! Since there’s only a single observation per field for the continuum, the HST button won’t have done any cosmic ray masking so we’ll not use this, but it leaves us with 3 narrow band filters (perfect for a 3-colour mosaic!). We’ll then perform a median background subtraction to get rid of any remaining background issues.

# Optimize the size for the narrow filters, and perform a median
# background subtraction.

filters = ['F502N', 'F631N', 'F673N']
folder = galaxy+'/HST/'+prop_id+'/'
instrument = instrument.replace('/','_')

for hst_filter in filters:

    filename = folder+galaxy+'_'+instrument+'_'+hst_filter+'_'+prop_id

    hdu_trimmed = tools.optimize_size(filename+'.fits')

    # Subtract background median.

    hdu_trimmed.data -= tools.background_median(hdu_trimmed)

    hdu_trimmed.writeto(filename+'_trim.fits',
                        overwrite=True)

The final, trimmed and background subtracted files will be in the same folder as the original mosaics. Playing around a bit in DS9 we can make a nice looking picture.

_images/M1.png