Green Star (YSO Candidate)

Looks like a green star to me (which don’t exist if I’m correct), but it is a Young Stellar Object Candidate in SIMBAD through GAIA photometric alert. But even then why would it be green in regular Decaps DR2 bands?


Yup green stars don’t exist. Andrew posted something interesting on why they don’t and why they are actually blueish/white stars. This wiki page has some reasons for it. Will try find Andrew’s old post.

1 Like

I got this:


Ah yes. Also this, posted by Andrew on a similar topic a while ago.

"Hi there, this is a really interesting find.

What is so interesting about it is that the color is not typical of what you would expect for stars. However, when looking at the single exposures, this object is clearly there at many different points in time at the exact same location. This is in contrast to something transient like an asteroid, that might appear to have a weird color because it is only seen in one “band” (color) the one time it happened to be at that location, before whizzing along the sky to a different location.

I had a fun day tracking it down and it appears to be a failure mode of the first data processing pipeline that calibrates the data. This operates on the data before most astronomers (like me) start to play with it and analyze the data. Unfortunately, that pipeline seems to think that two stars, that are moderately faint, approximately equal brightness, and with a specific separation look like a cosmic ray. They don’t… but the algorithm gets confused and interpolates over them thinking that they are an artifact to be removed. Furthermore, this apparently happens most often in z-band (one of the color channels in the image) as opposed to the other bands, giving rise specific green looking star in the post-processed image, even though the star is totally normal.

So in summary, a real star(s) that are artificially being made to look like they have the wrong color. A super cool failure mode!"


Well, this is not a pair of adjacent stars. It is a single isolated “greenish” star. To be sure, I pulled the single exposure images for this source and other than the fact it is near the chip edge for a couple of exposures, there is nothing obviously wrong happening in z/g band that would make it appear abnormally green (large in r-band). I remain stumped at the moment.


Awesome, being stumped is the only way forward : D

(I assumed being a YSO candidate would somehow explain it, although entirely unsure how that would work other than some sort of emission nebula)

Thanks for checking & the feedback!

It is possible that a very young (Class I) YSO has enough Hα to appear green… that just looks very green…

1 Like

Gaia alert has it as a CV, which is consistent with its gaia colors and WISE variability. Not sure why the green though - WDs do have Hα, but never seen one appear green before.

I get the following from SIMBAD & VizieR:

SIMBAD > GAIA Alerts > YSO Candidate

VizieR > Various Catalogs > Variable Soure
> GAIA > Cataclysmic Variable
> GAIA > extremely low-mass WD candidate
> GAIA > White dwarf main catalog

RA, Dec 135.4764, -51.8763

It is also visible in skymapper, but doesn’t seem to be extraordinarily green there

Skymapper Object Page

I don’t know how to access the Decaps DR2 single exposures (site / no download / no FITS files?) but perhaps it being a variable source perhaps was just brighter during the r (?) band observations.

Much bluer in DECaPS DR1. Definitely a WD rather than a YSO.

You may be right that the r band exposure may have been during an outburst, hence the color. Not sure what time difference there is between different bands’ exposures for DECaPS.

1 Like

I get the same green object for DECaPS DR1 & DR2

Oh apologies, it seems I clicked on DECaPS2 r/i/y rather than DECaPS1

1 Like

CMD position is weird (large green dot). Though CVs tend to do that.


Awesome graph, really points to an odd outlier (or regular CV perhaps)!

I also noticed some TESS entries in VizieR but never learned the skill to compile a light curve graph from the data. I think that can reveal a lot more of the story of this object.

For single exposure access, you will very soon be able to just click the links on the LS viewer data tab (currently implemented on the dev side of the LS viewer website). They are all available from the DECaPS website here which is paralleled by a data model here. In the meantime, I can answer this quickly. Here are the bands that go into the coadd and when they were taken.

g band: 2016-03-14 01:45:15
g band: 2016-03-15 01:30:08
g band: 2016-03-17 00:25:16
g band: 2016-03-27 01:31:23
g band: 2017-04-28 00:33:45
r band: 2016-03-14 01:47:19
r band: 2016-03-15 01:32:13
r band: 2016-03-27 01:30:24
r band: 2017-01-20 06:36:10
z band: 2016-03-24 01:36:19
z band: 2016-03-25 01:12:14

1 Like

Here are the 2nd and 4th r-band exposures, which were taken ~10 months apart. Clearly the seeing on the 4th r-band exposure was worse, but at first glance, the object of interest does appear to be a bit of an outlier and might be a bit brighter in the 4th exposure than the 2nd.

The object of interest is centered, but the image is approximately a 180 degree rotation from the OPs postage stamp.

1 Like

Yeah definitely big change in brightness, relative to surrounding sources


1 Like

Then it would come down to g band: 2017-04-28 00:33:45; if that one is also much brighter than the g band exposures from 2016 it is ‘just’ a brightening of a variable star / CV.

But as those exposures are all part of the RGB image in the viewer this star wouldn’t be green since both r & g band exposures from 2017 would contribute equally to the RGB image with only z band missing the brighter exposure from 2017.

If g band: 2017-04-28 00:33:45 isn’t brighter relative to the 2016 g band exposures that would really restrain the flare-up episode considerably; 2016-03 not, 2017-01 yes, 2017-04 not.