APOD: Centaurus A: A Peculiar Island of Stars (2023 May 03)

[MarkE.]

Yes the mirrored star patterns are what got me questioning what i was looking at, if indeed they are stars?

There are nearby(line of sight) galaxies and other objects(?) too that look suspiciously similar to one another but the image isn't the clearest when zoomed in, which makes it difficult to determine exactly what we are looking at!?

...It's possibly nothing, but it caught the eye, so i thought I'd throw it out there, along with a little confusion it seems!

[VictorBorun]

The dots look to me like stars, and the smear(s) like background spiral galaxies.

but background dwarf galaxies with bright cores can look like dull distant stars of Milky Way, can not they

Visually, maybe. They are easy to distinguish if we're really interested, however. All you need to do is run a star extractor app on the image (better if it's not a JPEG, but even that usually works). Stars all have similar PSFs, quite different from galaxies or any other extended objects. So you get out a list of stars and a list of nonstellar objects. If you're looking for lensed structures, you'd typically be looking for nonstellar objects clustered around a fairly nearby galaxy cluster. Then you'd want spectroscopic data to determine if those objects all had the same redshift.

PSF like point spread function?
But would that work with a distant dwarf galaxy?
Let me think. If a distant Milky Way star and a dwarf far behind Cen A are of the same colour and temperature and look the same in a picture, they must have the same angle area of the radiating surface which the telescope and camera smears to the same spot of their resolving power.
So if the star is 50 kly away and the dwarf galaxy is 50 Mly away, the dwarf's bright point-like core must have 1 million stars. Why not?

[Chris Peterson]

but background dwarf galaxies with bright cores can look like dull distant stars of Milky Way, can not they

Visually, maybe. They are easy to distinguish if we're really interested, however. All you need to do is run a star extractor app on the image (better if it's not a JPEG, but even that usually works). Stars all have similar PSFs, quite different from galaxies or any other extended objects. So you get out a list of stars and a list of nonstellar objects. If you're looking for lensed structures, you'd typically be looking for nonstellar objects clustered around a fairly nearby galaxy cluster. Then you'd want spectroscopic data to determine if those objects all had the same redshift.

PSF like point spread function?
But would that work with a distant dwarf galaxy?
Let me think. If a distant Milky Way star and a dwarf far behind Cen A are of the same colour and temperature and look the same in a picture, they must have the same angle area of the radiating surface which the telescope and camera smears to the same spot of their resolving power.
So if the star is 50 kly away and the dwarf galaxy is 50 Mly away, the dwarf's bright point-like core must have 1 million stars. Why not?

It would be extremely unlikely for a distant galactic core to have the same Moffat/Gaussian PSF shape that the other stars in the image have. Not impossible, but not something encountered very often. Distinguishing stars from extended objects, even ones that superficially resemble point sources, is a pretty reliable process.

[VictorBorun]

Visually, maybe. They are easy to distinguish if we're really interested, however. All you need to do is run a star extractor app on the image (better if it's not a JPEG, but even that usually works). Stars all have similar PSFs, quite different from galaxies or any other extended objects. So you get out a list of stars and a list of nonstellar objects. If you're looking for lensed structures, you'd typically be looking for nonstellar objects clustered around a fairly nearby galaxy cluster. Then you'd want spectroscopic data to determine if those objects all had the same redshift.

PSF like point spread function?
But would that work with a distant dwarf galaxy?
Let me think. If a distant Milky Way star and a dwarf far behind Cen A are of the same colour and temperature and look the same in a picture, they must have the same angle area of the radiating surface which the telescope and camera smears to the same spot of their resolving power.
So if the star is 50 kly away and the dwarf galaxy is 50 Mly away, the dwarf's bright point-like core must have 1 million stars. Why not?

It would be extremely unlikely for a distant galactic core to have the same Moffat/Gaussian PSF shape that the other stars in the image have. Not impossible, but not something encountered very often. Distinguishing stars from extended objects, even ones that superficially resemble point sources, is a pretty reliable process.

telescope+camera Takahashi CCA250 (250/f5) - ZWO 6200MM pixels: 4960 x 7440
How do I calculate the angle size of a pixel in this picture?
To test if a dwarf galaxy core at 50 Mly distance containing 1 million stars can not in fact be just a point smeared only by the point spread function ?

[Chris Peterson]

PSF like point spread function?
But would that work with a distant dwarf galaxy?
Let me think. If a distant Milky Way star and a dwarf far behind Cen A are of the same colour and temperature and look the same in a picture, they must have the same angle area of the radiating surface which the telescope and camera smears to the same spot of their resolving power.
So if the star is 50 kly away and the dwarf galaxy is 50 Mly away, the dwarf's bright point-like core must have 1 million stars. Why not?

It would be extremely unlikely for a distant galactic core to have the same Moffat/Gaussian PSF shape that the other stars in the image have. Not impossible, but not something encountered very often. Distinguishing stars from extended objects, even ones that superficially resemble point sources, is a pretty reliable process.

telescope+camera Takahashi CCA250 (250/f5) - ZWO 6200MM pixels: 4960 x 7440
How do I calculate the angle size of a pixel in this picture?
To test if a dwarf galaxy core at 50 Mly distance containing 1 million stars can not in fact be just a point smeared only by the point spread function ?

Well, if I solve the image (against the Gaia DR3 database) it finds 4871 sources that it calculates the PSFs for, and an image scale of 0.78 arcsec/pixel. I could export the extraction data to a spreadsheet and sort on one of the PSF characteristics, like FWHM or eccentricity or skew to group out the sources that aren't apparently stellar, then go and examine those individually. My results would be better if I worked with just one channel (not the color) and in an uncompressed, high dynamic range format like FITS (not JPEG). As I noted, the probability of a galaxy looking exactly like a star is small. With proper high dynamic range data and good S/N, a distant galaxy will still show a profile in the data, even though the core will be the only thing bright enough to see visually and it may appear stellar visually.

[VictorBorun]

It would be extremely unlikely for a distant galactic core to have the same Moffat/Gaussian PSF shape that the other stars in the image have. Not impossible, but not something encountered very often. Distinguishing stars from extended objects, even ones that superficially resemble point sources, is a pretty reliable process.

telescope+camera Takahashi CCA250 (250/f5) - ZWO 6200MM pixels: 4960 x 7440
How do I calculate the angle size of a pixel in this picture?
To test if a dwarf galaxy core at 50 Mly distance containing 1 million stars can not in fact be just a point smeared only by the point spread function ?

Well, if I solve the image (against the Gaia DR3 database) it finds 4871 sources that it calculates the PSFs for, and an image scale of 0.78 arcsec/pixel. I could export the extraction data to a spreadsheet and sort on one of the PSF characteristics, like FWHM or eccentricity or skew to group out the sources that aren't apparently stellar, then go and examine those individually. My results would be better if I worked with just one channel (not the color) and in an uncompressed, high dynamic range format like FITS (not JPEG). As I noted, the probability of a galaxy looking exactly like a star is small. With proper high dynamic range data and good S/N, a distant galaxy will still show a profile in the data, even though the core will be the only thing bright enough to see visually and it may appear stellar visually.

Let me try the other way around.
Consider the compact dwarf M60-UCD1 at 54 Mly with mass of 140 MSun. It is visually a disk with a hazy perimeter. Were there similar dense dwarfs of 1 MSun mass, their radius would be 5 times smaller which would still be visibly not a point.
So I accept it: a well seen star can be told from a well seen galaxy, even a compact dwarf type