No, it's not a FIRST radio source, but what a cool green-winged seagull!

• by JeanTate

  zph 0.648±0.061 SDSS J152834.63+161042.8:

  

  Posted May 30, 2016 12:51 PM

• by ivywong scientist, admin

  Do you know roughly what the signal-to-noise these "wings" are relative to the r-band background? If real, this is likely another Voorwerpje.

  Posted May 31, 2016 6:51 AM

• by JeanTate in response to ivywong's comment.

  Do you know roughly what the signal-to-noise these "wings" are relative to the r-band background?

  The big caveat is this: "DEBLEND_NOPEAK DEBLENDED_AT_EDGE STATIONARY MOVED BINNED1 MANYPETRO NOPETRO NODEBLEND CHILD BLENDED", and for the E "wing" (SDSS J152834.84+161043.2): "TOO_FEW_GOOD_DETECTIONS DEBLEND_NOPEAK DEBLENDED_AT_EDGE STATIONARY MOVED BINNED1 DEBLENDED_AS_PSF MANYPETRO NOPETRO CHILD".

  There is no separate PO for the W "wing".

  Looking at just the g, r, and i bands:
  
  E "wing: 22.70±0.40, 20.83±0.12, 21.64±0.42
  
  body: 23.33±0.51, 21.65±0.19, 20.35±0.10
  

  So the r-band flux of the E wing is 0.82 mags greater than that of the body (trying to estimate the uncertainty of this would be foolish, for this BOTE calculation!). According to this SDSS DR12 page, the median r-band sky brightness is 20.86, with the lower/upper quartiles being 20.70 and 21.03, respectively. So, in the r-band, the E wing is at, or a tad above "the sky", while the body's is below it.

  If real, this is likely another Voorwerpje.

  A pretty extreme one! 😮

  Posted May 31, 2016 2:58 PM

• by ivywong scientist, admin

  yeah, you are up against the detection limit so it's a little hard to tell.

  Posted June 2, 2016 10:05 AM

• by JeanTate

  Ah, detection limits ...

  I'm sure my experience is no different than many of my fellow zooites: having finished classifying a field, you click on "Discuss", then bring up the SDSS image. Perhaps what you clicked as (a) host was fairly bright, in the WISE IR image, perhaps not so bright. Bright IR hosts may turn out to be obvious in SDSS, but often they are not; indeed, many are tiny smudges, or blobs, and when you click Examine, you see that there are warnings, sometimes many warnings. And the reported magnitudes, in any/all bands, are close-to-sky faint, with big errors.

  Yet they seem real! 😮 Soooo often there's a FIRST source, positionally ~coincident with both a WISE and an SDSS one; sometimes it's even only the SDSS z-band which has unambiguous detections!

  Of course, I know that there's a logical flaw in going from 'lots and lots of really faint SDSS sources are hosts of (compact) radio sources' to 'most really faint SDSS sources are real (galaxies)' ... I wonder if there's a way to analyze the former, to get a handle on what fractions/subsets of the latter are likely real?

  Posted June 4, 2016 7:00 PM

• by ivywong scientist, admin

  Short of downloading each field and picking up on the blobs that are about 5-sigma, I don't have a better suggestion. If you're keen to figure out the "realness" of a detection, perhaps align all your faint blobs and then co-add/stack (average) these images to see if you can an improvement in signal-to-noise. Real noise should cancel out and you'll find nothing but you will improve your signal-to-noise by sqrt(n) when you stack n frames.

  The only caveat is that they have to be in the same filter and you have to make sure the blobs are aligned well.

  Posted June 10, 2016 8:20 AM

• by JeanTate in response to ivywong's comment.

  Thanks @ivywong.

  I'm not actually all that interested in this particular object; rather, it's a possible example of what we've seen in quite a few posts here in RGZ Talk: faint features, possibly associated with FIRST and/or WISE sources. Some such are "outflows", some may be faint tidal tails, or faint EELRs; many are ~point sources (flagged as STAR by the automated SDSS photometric pipeline) or barely non-PSFs (flagged as zph ???).

  While I'm pretty confident that the FIRST, WISE, and SDSS astrometry is good (the coordinate systems are likely aligned to some fraction of an arcsec), offsets trouble me ... WISE/SDSS ones are relatively easy to understand (though it gets complicated when there's more than one SDSS source within 'the WISE resolution'); the WISE/FIRST and SDSS/FIRST ones not so much. Why? Lots of reasons; among them:

  • a FIRST source may have a host that is too faint to be detected by either WISE or SDSS
  • a FIRST source may be naturally offset from the associated WISE/SDSS one (e.g. a lobe, with no FIRST-detected core emission)
  • overlaps, a.k.a. chance alignments

  I appreciate that there are statistical arguments/analyses one can make, which can shed some light on these questions (in a general sense, not for any particular source), but they are beyond what any of us can do, now. That will change, of course, once RGZ is finished, and the results published ...

  Posted June 12, 2016 9:47 PM