ngcs

Quick and simple way to access NGC galaxy images with python.

For any of the 4221 NGC galaxies of the Siena Galaxy Atlas 2020 (SGA-2020), ngcs gives:

• its metadata: position, redshift, size (at the 26 mag/arcsec² isophote), g, r, z magnitudes and colours;
• its image from the DESI Legacy Surveys (DR10) or Pan-STARRS1, in nanomaggies per pixel, with its WCS;
• the same image without the foreground stars (Gaia DR3), masked and filled;
• the galaxy alone: stars removed, sky subtracted, nothing around (0 outside the galaxy), cropped and centred: e.g. to simulate the galaxy at another redshift.

Installation

pip install -e .            # numpy, scipy, astropy, pandas, pyarrow, requests
pip install -e ".[plot]"    # + matplotlib, for the show() methods

Usage

import ngcs

galimg = ngcs.GalaxyImage.from_name("NGC 6643")      # 'NGC6643', 'ngc 6643', or 'M74' (resolved by Sesame)
image = galimg.get_image()                           # the Legacy Surveys cutout (an ngcs.Image)
image_nostars = galimg.get_image(rm_stars="auto")    # without the Gaia stars
galaxy = galimg.get_galaxy()                         # the galaxy alone
metadata = galimg.get_metadata()                     # name, ra, dec, redshift, sma, diameter, mag_g/r/z, g-r...

galaxy.data, galaxy.pixel_scale, galaxy.meta["center_pixel"]   # array [nmgy/pixel], arcsec, (x, y)
galimg.show()                                        # diagnostic plot

An ngcs.Image holds data (2d, or a (nbands, ny, nx) cube), wcs, bands, pixel_scale and meta, and has band(), magnitude(), show(), to_fits() and Image.from_fits().

Options

ngcs.GalaxyImage.from_name("NGC 6643", survey="panstarrs", bands="gri", margin=2)  # field = margin x D26
galimg.get_image(rm_stars="sky")                      # fill the stars with a fake sky
galimg.get_image(rm_stars="interpolate")              # ... or by interpolation
galimg.get_image(rm_stars="auto", radius_scale=1.5)   # larger star masks
galimg.get_galaxy(isophote=25)                        # tighter galaxy [AB mag/arcsec²]
ngcs.GalaxyImage.from_image("my_image.fits", metadata={"ra": ..., "dec": ..., "sma": ...})  # any image
ngcs.get_catalog()                                    # the whole catalog (pandas.DataFrame)

How it works

• Stars: the Gaia DR3 sources (VizieR TAP) with a parallax or proper motion detected at more than 5 sigma: the HII regions, clusters and nucleus of the galaxy are also Gaia sources, but they do not move. Each star is masked within a radius growing with its brightness (2.5" at G = 18, x 10^0.1 per magnitude, between 2" and 25"), grown to the connected pixels more than 3 sky sigma away from the local background (halos, spikes, bleed trails).
• Galaxy alone: the connected region around the galaxy centre brighter than 26 mag/arcsec² (r band, smoothed), holes filled, grown by 2", limited to an ellipse of 1.2 x the SGA size with the axis ratio and orientation of the galaxy (image moments).
• Cache: every download (cutouts, Gaia) is kept in ~/.cache/ngcs (or $NGCS_CACHE): a second call is instantaneous. ngcs.io.clear_cache() empties it.

Faint stars (G ≳ 19) without a significant parallax or proper motion, and background galaxies seen through the disk, stay in the images.

Tests

pip install -e ".[tests]"
pytest                  # offline tests (synthetic images)
pytest --remote-data    # + the tests that download images and Gaia stars

Credits

SGA-2020: Moustakas et al. 2023, ApJS 269, 3. Legacy Surveys: Dey et al. 2019. Pan-STARRS1: Chambers et al. 2016. Gaia DR3: Gaia Collaboration, Vallenari et al. 2023.