SGA-2024 next steps

[moustakas]

Summarizing my current thinking regarding next steps for defining the SGA-2024 parent sample:

• [x] Build a parent-nocuts sample using a combination of HyperLeda, NED-LVS, and the Local Volume Database (LVD). A first version of this catalog has been generated (see #83), as summarized in the flowchart below. The current sample has nearly 5M entries, of which 4.1M (~82%) have approximate initial sizes or diameters based on a variety of sources (NED-BASIC; RC3; 2MASS; SDSS; ESO; LVD; or HyperLeda).
• [ ] Checks on parent-nocuts:
  • [ ] Identify large differences in coordinates---(RA_NED,DEC_NED) vs (RA_HYPERLEDA,DEC_HYPERLEDA) vs (RA_LVD,DEC_LVD)---which are almost always due to cross-identification errors. E.g., NGC7095 is actually ESO027-001. NED gets this right (and so parent-nocuts does, too), while HyperLeda is wrong.
  • [ ] Check the approximate geometry (diam, ba, pa) in parent-nocuts against HyperLeda's "homogenized" values (logd25, logr25, pa).
  • [ ] Resolve any additional close-coordinate "duplicates" by finding all groups of objects within 1-5 arcsec of one another. (Identical coordinate-duplicates have already been removed.)
• [ ] Build a parent sample by applying appropriate cuts on parent-nocuts. The most significant cut will be on diameter. E.g., cutting on DIAM>15/60 arcmin reduces the sample to ~1M (over the full sky).
  • [ ] Make sure this sample addresses #73 and also compare against the SGA-2020.
  • [ ] Develop an algorithm for choosing a primary object name. Currently there are three names in the catalog: OBJNAME_NED, OBJNAME_HYPERLEDA and OBJNAME_LVD.
  • [ ] Objects classified as GPair and GTriple (i.e., mergers) were removed while building parent-nocuts, but we need to make sure that our sample contains each individual object in the merging system. (Will probably need to do this using coordinates since object names are unreliable.)
  • [ ] Determine which objects should not trigger special handling in legacypipe, such as a subset of the "resolved" Milky Way dwarfs (e.g., Bootes V).
  • [ ] Compute the angular distance between each object and bright / medium Gaia stars.
• [ ] Determine which objects in parent have Legacy Surveys coverage in the north and, separately, the south in any band.
• [ ] Build a group catalog, parent-group.
• [ ] Generate preliminary coadds, sizes / geometry, and r-band diameters. Visually inspect all objects near bright / medium stars; all objects larger than one arcmin; and all objects in groups.

[moustakas]

@TriShotFlurry, I believe you left a comment on this thread that may have been deleted.

Thank you for the feedback regarding including objects classified by HyperLeda as object type 'g'. Since I opened this ticket in early September, I did indeed add ~60,000 objects with this classification to the parent sample.

However, it appears that the vast majority of these objects are either spurious or "parts of a galaxy", as shown in the gallery below. Fortunately I have an automated classifier which should allow me to pull out the "real" galaxy centers.

I'm also trying to stand up a visual inspection campaign and will be sure to loop you (and other members of the Legacy Surveys discussion group) when that's ready.

[TriShotFlurry]

I understand your point: I looked at the gallery, and there are some that 'appear' to be real galaxies, but the object types appear to be quite inaccurate. You can wish to identify galaxies with these object types not galaxies in NED but classified as galaxies in LEDA, if they appear to be real galaxies (without a duplicate entry). There are several 2MASX sources with object type 'g' that look worthy of being included in the sample anyway. so I think including 'g' is worthy indeed...

I implemented the HyperLEDA sample and went over some galaxies, there are lots of sources that appear to be too small to be incorporated into SGA-2024, consider removing these as well by software also.

[TriShotFlurry]

@moustakas Suggest adding a small addenda of galaxies, since I found browsing astronomical databases while constructing the SGA2024 designations catalog...

In fact, here's the worksheet https://docs.google.com/spreadsheets/d/15TsMKy-x-R5wQhikM_UsuoFv93OPh4CQ0Gt_QJAJgsc/edit?usp=sharing