Code and papers for the Siena Galaxy Atlas (SGA-2020).
License: MIT License
The Siena Galaxy Atlas (SGA-2020) delivers multicolor images and model fits for a diameter-limited sample of large galaxies in Legacy Survey imaging, which consists of deep grz optical imaging and unWISE W1-W4 mid-infrared imaging.
We gratefully acknowledge funding support for this work from the National Science Foundation under grants AST-1616414 and AST-1909374, and the U.S. Department of Energy, Office of Science, Office of High Energy Physics under Award Number DE-SC0020086.
The mask for NGC4395 is too small---reported by Eisenstein in [decam-chatter 10011].
Not sure how these made it through, but need to fix for SGA-2024.
https://sga.legacysurvey.org/group/PGC1411139
https://sga.legacysurvey.org/group/PGC155984
https://sga.legacysurvey.org/group/PGC131773
https://sga.legacysurvey.org/group/UGC02347
There are certainly others.
Feature: consider adding a definition of the possible "Galaxy" and "Group" entries next to the search bars on the SGA web interface (perhaps with a question mark help dot/bubble or additional grey text to supplement the main text).
Starter ticket for the SGA-2024.
Working outline:
Part 1:
Part 2:
Part 3:
Part 4:
The following galaxies are not really in the footprint (they have 1- or 2-band coverage)--
Might be worth looking into IC0207 - http://legacysurvey.org/viewer/?ra=32.414013&dec=-6.922181&zoom=14&layer=dr8&lslga, too.
From @ameisner:
A while ago I looked at the 300 largest galaxies by d25 in your old
leda-logd25-0.05.fits.gz file to try to customize those objects for my
unWISE masks. There are a couple of those objects that appeared to be
SDSS bright star artifacts and remain in your new file:
LSLGA_ID = 350073, d25 = 2.4434304
http://legacysurvey.org/viewer?ra=114.183498&dec=46.1580054&zoom=12&layer=sdss-co
LSLGA_ID = 433546, d25 = 5.8613820
http://legacysurvey.org/viewer?ra=134.08560&dec=21.887080&zoom=12&layer=sdss2
Both objects have the NEAR_BRIGHTSTAR bit set to False, which is mildly annoying (and probably means the magnitude-radius relation I'm using is not good enough).
...to bring out the tidal structure & other low surface brightness features, e.g.,
http://legacysurvey.org/viewer?ra=19.9127&dec=3.3386&zoom=11&layer=decals-dr7
Draco, Sextans, Leo I, Ursa Minor
Cross-referencing this issue--
legacysurvey/legacypipe#454
This pair of galaxies from DR8 made it into the final sample--
https://sga.legacysurvey.org/group/DR8-3225p122-20657_GROUP
When searching via the web portal (https://sga.legacysurvey.org), the per-galaxy "Elliptical Isophote Analysis: Photometry" table in the results page seems to be missing the correct z-band Curve of Growth Photometry: Integrated value, with the r-band value duplicated in its place.
E.g., for NGC1079 (https://sga.legacysurvey.org/group/NGC1079?index=1), the g, r, and z values are "11.557 | 10.811 | 10.811". (From the full FITS "ELLIPSE" table, the corresponding values are 11.557453, 10.811194, 10.304616.)
I noticed this for about 45 galaxies I checked by hand, so it seems like a consistent bug.
This looks great!
The galaxy diameter values are missing from the table on the home page and clicking the diameter heading results in traceback.
According to NED, this is GALEXASC J010318.95-044018.0 = APMUKS(BJ) B010046.87-045623.4, which unfortunately is not in LEDA--
https://ned.ipac.caltech.edu/cgi-bin/objsearch?search_type=Obj_id&objid=5042888&objname=5&img_stamp=YES&hconst=73.0&omegam=0.27&omegav=0.73&corr_z=1
I think I'm missing a factor of 2 in this part of the group-catalog generating code--
https://github.com/moustakas/SGA/blob/main/bin/SGA-build-parent#L3744
This error didn't matter (at all?) with the initial SGA catalog because they Hyperleda diameters tend to be much smaller than the D(26) diameters we have in-hand now. What happens is that in a galaxy group with a large central galaxy and a smaller companion on the periphery missing this factor of 2 can exclude the satellite.
from @ameisner:
moon_lev - AllWISE doesn't have all that much redundancy, especially
in W3 and W4 where there's typically just one sky pass, so there might
be swaths of sky where AllWISE catalog information in those bands is
corrupted
rchi2 - seems to be widely used in AllWISE selection criteria by
people I work with
w?gmag - there's also these 2MASS XSC based elliptical apertures,
which seem like they have some relevance
Many z-band images of nearby galaxies have some "fence" patterns throughout the galaxy. Below should you one example NGC2742:
Meanwhile, the sky background is over subtracted:
The region enclosed by the green curves are obviously (for me) darker than other sky regions!
I didn't find the same problem in the g-band, r-band, and DECam z-band data.
This problem may make the photometry based on MzLS z-band data suspicious.
Is the DESI team aware of the problem and are there any solutions?
Otherwise I have to remove all the galaxies observed with MzLS from my sample. very upset.
Who knows how this is happening.
Bug: DIAM_REF is LEDA for all sources, even if they didn't come from HyperLeda.
#20 brought in a dependence on the ellipse-fitting code in legacyhalos. Remove that.
Adam Myers has written some nice masking tools here for asking questions like, "is this location within X half-light radii of the center of this galaxy."
This isn't really an "issue" but bookmarking the code here for now.
All of the morphological types are (incorrectly) set to "zero" (string), but especially morphtype in the openNGC catalog.
Something has gone wrong with how the links to the viewer are being constructed in the Django template. E.g., clicking on "Sky Viewer" here https://sga.legacysurvey.org/group/PGC1168221?index=1 sends you to this (wrong) url--
https://www.legacysurvey.org/viewer/?ra={{%20primary.group_ra%20%20%20%20%20%20%20}}&dec=0.5240977&zoom=15&layer=ls-dr9&sga
When we cull the sample for galaxies near bright stars we should take into account the flux ratio between the star and the galaxy, and be less aggressive with our masking. For example, a 10th magnitude star next to an 18th magnitude galaxy gets crushed. But next to an 8th magnitude galaxy, no big deal.
When generating the large-field mosaics, apodize the inverse variance images so that the edges aren't so prominent. Here's an example of NGC681 --
http://legacysurvey.org/viewer?ra=27.2981&dec=-10.4255&zoom=13&layer=decals-dr5&ccds5
A few galaxies (I have checked 400+ galaxies) have a weird bright strip across the galaxy. For example, check the center of https://portal.nersc.gov/project/cosmo/data/sga/2020/data/334/ESO108-023/ESO108-023-largegalaxy-image-grz.jpg. May I ask are they cosmic ray? It there any way to remove them?
In addition of searching around one RA, Dec location, it could be useful to have the option to search for a short list of galaxies by entering the list of RA, Dec values in a text box [ RA, Dec, [ID] ], or by uploading a temporary file. There are pros and cons, so this may not be a desired feature after all. Feel free to close this issue if you decide against it!
The parent NED sample is full of junk, e.g., from SDSS/DR6 because NED doesn't do any cuts on photometric quality flags. I think the path forward is to cross-correlate the sample with DR14 on Casjobs so we can clean these up.
We'll have to be careful, however, to not throw out big galaxies that SDSS fails on / shreds.
Here's an incompleteness in the Hyperleda catalog I just discovered--
http://legacysurvey.org/viewer-dev?ra=171.3932&dec=9.6677&zoom=16&layer=decals-dr7&lslga
The SDSS object has a logd25 listed (on the top-level summary page)
http://leda.univ-lyon1.fr/ledacat.cgi?SDSSJ112535.05%2B094004.1&ob=ra
but the 2MASS object does not
http://leda.univ-lyon1.fr/ledacat.cgi?2MASXJ11253331%2B0939462&ob=ra
even though they both have estimated SDSS diameters.
I'm communicating with the Hyperleda help desk to figure this out.
In light of what is written in https://www.legacysurvey.org/sga/sga2020/#known-issues (third item), I can't figure out if the total magnitudes [G,R,Z]_COG_PARAMS_MTOT are correct and can be used; if this is not the case, I would like to get some suggestions to obtain the total magnitudes for SGA galaxies in the r-band, at least.
Thank you very much
A fun plot for a proposal or paper would be to plot limiting surface brightness (e.g., in a 5 arcsec diameter aperture) vs filter wavelength and solid angle (in deg2).
This plot could include LS, SDSS, 2MASS, GALEX, WISE, unWISE+NEOWISE, DES, etc.
Dustin has been trying to process the last few hundred bricks for DR10, and I did a quick pass through the brick images which he generated to see what was up. There are a small number of large galaxies and star clusters that are causing failures, and I summarize them below. Could you take a quick look at the list below and the fits to see why these are producing issues for tractor runs?
Also, there are at least three additions to our masks for star clusters and PNe as well (Phoenix dwarf galaxy, Arp 2, and ESO 28-19). I have listed my very rough guess at the mask location and radius (for circular masks) for these three.
Here’s my summary of checking the locations on the DR10-early viewer:
Big Galaxies:
NGC 55 + IC1537 - SGA ellipse is not big enough and misses fluxes, not in Star Cluster list
NGC253 - SGA ellipse not big enough, not in Star Cluster list
NGC300 - SGA ellipse looks ok, not in Star Cluster list
Phoenix Dwarf Galaxy - SGA looks OK, not in Star Cluster list. Should we add a “Star Cluster" mask at position RA=27.779507, DEC=-44.446111, Radius=4 arcmin?
NGC4406 + NGC4402 - SGA ellipses look small, not in Star Cluster list
NGC4472 - SGA ellipse looks small, not in Star Cluster list
NGC5236 - not in SGA, not in Star Cluster list
IC5152 - SGA ellipse looks ok but wrong axis ratio, not in Star Cluster list
Star Clusters:
NGC104 - not in SGA, in Star Cluster list, but crowded field with incomplete coverage; Near SMC
NGC121 - not in SGA, in Star Cluster list, but crowded field with incomplete coverage; Near SMC
NGC5139 - not in SGA, in Star Cluster list, in a single isolated DECam pointing - drop!
NGC6752 - not in SGA, in Star Cluster list, in a crowded field
Arp 2 - not in SGA, not in Star Cluster list. Perhaps add a mask at position ra=292.1836 dec=-30.3529 radius=3 arcmin (https://www.legacysurvey.org/viewer?ra=292.1836&dec=-30.3529&layer=ls-dr10-early&zoom=13&sga&GCs-PNe)
ESO 28-19 - not in SGA, not in Star Cluster list. Perhaps add a mask at position ra=6.1922&dec=-72.7938 radius=100 arcsec (https://www.legacysurvey.org/viewer?ra=6.1922&dec=-72.7938&layer=ls-dr10-early&zoom=12&sga&GCs-PNe)
Not much we can do about these other than measure their fluxes from the Legacy Survey data themselves:
Issue reported by Si-Yue Yu:
I am recently using the SGA images to understand the galaxy structure. However, I found that the WISE images
are a bit abnormal. For example, the attached NGC0895_W1.png is the W1 image of NGC0895
([NGC0895-image-W1.fits.fz](https://portal.nersc.gov/project/cosmo/data/sga/2020/data/035/NGC0895/NGC0895-image-W1.fits.fz)).
There are some black (negative value) regions around the galaxy. It looks like the background is somehow
over-subtracted. The attached NGC0895_W1_model.png is its W1 model image
([NGC0895-largegalaxy-model-W1.fits.fz](https://portal.nersc.gov/project/cosmo/data/sga/2020/data/035/NGC0895/NGC0895-largegalaxy-model-W1.fits.fz)). '
Instead, there are many small black holes around the galaxy.
This ticket can track galaxies in SGA-2020 used for DR9 that should be re-fit with different starting parameters:
https://www.legacysurvey.org/viewer?ra=55.11&dec=-28.68&layer=ls-dr9&zoom=14&sga
Objects near stars can have very wacky surface-brightness profiles--
https://sga.legacysurvey.org/group/PGC2110473
https://sga.legacysurvey.org/group/UGC09630
We no longer do the sample selection in this notebook
https://github.com/moustakas/LSLGA/blob/master/doc/build-lslga-parent.ipynb
but there's some useful stuff in there.
Clean that up and also add more documentation so the analysis steps are clearly laid out.
NGC2366 is substantially under-masked and NGC0598 less so.
Totally missing from even the parent Hyperleda catalog!
http://legacysurvey.org/viewer-dev?ra=194.7586&dec=28.1159&layer=dr8&zoom=15&lslga
http://leda.univ-lyon1.fr/ledacat.cgi?o=NGC4858
https://portal.nersc.gov/project/cosmo/temp/ioannis/LSLGA-html-DR9-test2/194/NGC4860/NGC4860.html
Dear John,
I am trying to acces the SGA through LS website, I get the following error : " Your connection was interrupted. A network change was detected. ERR_NETWORK_CHANGED ". (I am trying to connect from Munich Germany, in case it helps)
Maybe there is there maintenance ongoing ?
Thank you and best wishes
Lots of now-obsolete notebooks; clean them up.
We need to define the sample based on the overlap of the parent NED sample I generated and the final Legacy Surveys / DESI footprint. We will also have to decide what constitutes "large," although I'm thinking 10-15 arcsec as a minimum size.
From this sample we can then select the sample which currently have grz photometry in DR6/DR7. A notebook along these lines is here.
UGC00212 needs adjusting--
http://legacysurvey.org/viewer-dev?ra=5.5955&dec=-1.3034&layer=dr8&zoom=15&lslga
Sculptor is in this DES footprint. It’s radius is listed as muchtoo small, see below, which is probably why those bricks are timing out in DR8. Even though it has lots of resolved sources, we could probably be treating it just fine like the other LSLGA galaxies with one extended source and whichever point sources are detected within that footprint.
Fornax is the other such Local Group galaxy in our footprint, but it’s listed size is more reasonable.
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.