0 votes
There seems to be an issue with the way Data Lab queries for FITS from the legacy survey. I am downloading FITS images for a lot of galaxies, and roughly 10% of them have weird dimensions and are not centered on any galaxy. I am certain these galaxies exists in the legacy survey as they show up on the skyviewer app. Here are two galaxies as an example, one can copy and paste these into the "Image cutout" tool for Legacy Survey DR7 with size = 0.05 degrees and Filter=All:

ra     dec     id(optional)

In doing this, the second galaxy returns the expected results-- galaxy is centered, dimensions are equal and square, and the amount of pixels + using the pix/arcsec scale translates back to 0.05 degrees FOV. In the first galaxy, the dimensions are uneven and the galaxy does not appear. Opening up the FITS and looking at the WCS, the image is close in coordinate space to the galaxy I input, but the exact coordinates are still slightly outside the FOV. I have more "problematic" galaxies in case having multiple examples may help identify the issue.

by | 301 views
Just wondering if anyone is taking a peek at this?

1 Answer

0 votes
Hi Derrick,

Apologies for the slow reply, but this problem should be fixed now.  There was an error both in the query used to find overlapping images and in the cutout tool used to extract the pixels.  If you continue to have problems or questions please let us know.
by datalab (20.7k points)

409 questions

426 answers


630 users

Welcome to Data Lab Help Desk, where you can ask questions and receive answers from other members of the community.