Measuring resolutions

From CoolWiki
Revision as of 22:32, 31 July 2020 by Rebull (talk | contribs)
Jump to navigationJump to search

Since resolution issues are going to be more important to us in the infrared, let’s focus on those bands for most of our by-hand (as opposed to by-Google) work. For the M8 (Lagoon Nebula) project, our infrared (through submm) inventory is:

  • 2MASS
  • Spitzer (SEIP, GLIMPSE)
  • WISE
  • SCUBA (partial coverage)
  • Herschel (partial coverage)
  • AKARI (only have images for FIS, not IRC)
  • MSX
  • IRAS

You should aim for, at minimum, empirically finding the resolution of 2MASS, Spitzer, and WISE. Use Finder Chart (or IRSA Viewer) to retrieve images of our region from at least 2MASS, Spitzer, and WISE. (If you want to do more, you will need also to use IRSA Viewer, rather than just Finder Chart.)

Q1.1 : Retrieve images of our area. For the images that it returns, what is the size of each pixel for each survey? (Option #1 to do this: Make the image big enough in your view of it that you can see pixels, and measure the size of it using ruler tools (not a real ruler). Option #2 to do this: look in the FITS header and find a useful keyword.) Try at least one image from each of the surveys.

Q1.2 : You will need to Google for this one. What is the original native pixel size for these surveys? Finder Chart gives you images that come straight from the original surveys, so they should match the original native pixel size for each survey.

Q1.3 : Are there any images you've retrieved that have “run off the edge” of a stored tile? (Hint: yes.) Which ones? For the surveys where you have run off the tile rather dramatically (at least 2MASS), you can use Skyview to get a larger image. The four most important parameter choices Skyview gives you are:

  • center position
  • survey (wavelength)
  • image size in pixels
  • image size in degrees

Skyview will happily and without complaint or warning resample and regrid the pixels to whatever scale you want. What do you need to do to get ‘native pixel’ resolution out of Skyview? You should have the information from earlier questions to figure out how many pixels you need to cover our region, so go and do the math, and ask Skyview to give you a full-sized image of your desired size. Note that you can request more than one survey at a time, but Skyview will use the same parameters for each of them. Alternatively, you can use IRSA Viewer to request 2MASS images from the 6 degree (note 6d not 6x) images.

Q1.4 : Did you do the calculations right? Here's how to check. Look at the sources in the 2MASS image you retrieved from Finder Chart (which you know is native px size) and compare it to the sources in the image you retrieved from Skyview. Have you lost information? (To see what this looks like, try to make it lose information deliberately by asking for much larger pixels.)

Q1.5 : Skyview attempts to knit tiles together, but sometimes you can see the original tile boundaries, and it looks like a patchwork quilt. Do you see this here?

Q1.6 : For at least one frame from each of a few of the surveys we picked, from either your Finder Chart or Skyview images (assuming you are confident you have native pixel resolution), go and measure the sizes of 3 to 5 ‘typical’ isolated point sources in these images. What kinds of sizes are you getting for each survey? (It is going to be hard to find 'typical' in IRAS; do what you can.) Changing the color table/stretch is useful for telling if the image is slightly asymmetric (implying a barely resolved companion) or saturated or other things.

Skyview won't give you Spitzer images, because Spitzer isn't an all-sky survey. But there are lots of large images available at IRSA from Spitzer. SEIP = Spitzer Enhanced Imaging Products, but this too works in tiles, and the request you give Finder Chart or IRSA Viewer may run off the edges of some of those tiles. There are data there, just not in the tile that the IRSA tools may be pulling for you. To find individual sources in regions off the tile it gives you, ask for a smaller region.

Q1.7: The IAU-compliant names of sources are based on positions. Many of the catalogs and papers that we have list some sort of unique ID within the survey, but its ‘real’ name is the position-based name, which is typically included in the catalogs if not all the journal articles (the journal articles are supposed to use position-based names, but they don’t always). People often assign and use internal source IDs in papers because it’s easier to say “source 346” in conversations with collaborators rather than the full phone number that might look like 18033652-2423108. But, why is it that IRAS sources are given as, e.g., "IRAS 18006-2422" and 2MASS sources are given as, e.g., “2MASS 18033652-2423108”?