Huge oxygen target next to Andromeda galaxy

[Paul M]

The original image of this feature took my breath away. But then after absorbing it, I decided that it's maybe not such a big deal. It struck me that such feature might befound all over the sky, given the same processing treatment. 

The enhacement of the feature is, maybe, so focussed and intense that it gives it more gravity than it deserves.

I am looking forward to further amateur imges that not only show it but put it into context. 

[Marvin Jenkins]

I was interested by the Brey Falls YT video where he points out another concentrated O3 area in the andromeda galaxy itself.

The O3 object would be the size of a dwarf galaxy surely? Although uninformed as we all are in respect of these particular objects it’s size would point to a foreground object?

M

[vlaiv]

The more I look at these images - the more I'm convinced that this is processing artifact and not the real thing.

To reiterate - image that I've found online in the first post:

versus screen capture from the you tube:

Just try to match Ha patterns in the image and you'll see that there are vast patches of what should be Ha - missing on one or the other image.

How can this be?

[ONIKKINEN]

Interesting, never thought to doubt it when i first saw the video a while back when all this was announced. Just kind of thought these guys must know what they are doing but it doesn't look great here wit the examples.

[vlaiv]

15 minutes ago, ONIKKINEN said:

Interesting, never thought to doubt it when i first saw the video a while back when all this was announced. Just kind of thought these guys must know what they are doing but it doesn't look great here wit the examples.

It could legit thing, and that is why I'd like to see more data that is also more convincing.

It could be the same data they used - but without fancy processing like that "continuum subtraction" - just good old noisy monochromatic data that shows that something is indeed there.

Or third party confirmation that has not been processed in the same way of course.

We can see this all over the place - people try to push the data beyond what it really shows by using "fancy" / "shiny" new algorithms that often employ AI and "do wonders". I'm not completely against the use of such things to produce nice image - but not for anything that is remotely scientific.

[symmetal]

I have a set of 2" Astronomik 6nm MaxFR Ha, OIII & SII filters (f1.8 to f2.2) to use with the RASA 11 (f2.2, 620mm) and ASI6200MM so can give it a try, when it comes around, if it's confirmed to be a genuine feature. 🙂

Alan

[ShinyTwelve]

Why doesn't somebody ask the professionals for their opinion, such as Mark McCaughrean the Senior Scientific Advisor to ESA?

[vlaiv]

Here is brief paper with very brief description of process (no continuum subtraction is mentioned):

https://iopscience.iop.org/article/10.3847/2515-5172/acaf7e?fbclid=IwAR12grzwnrY-GHKDBsR9sp9s3sMGPR-hQb2-TBZbi3cNsNfbCJW-98wkG6M

It looks like some professional astronomers were included in the paper.

(well I got my black and white images there ).

 

[Gfamily]

5 minutes ago, ShinyTwelve said:

Why doesn't somebody ask the professionals for their opinion, such as Mark McCaughrean the Senior Scientific Advisor to ESA?

One of the points when this was originally identified, is that this is a wide angle (almost 10° wide) field image taken using narrow band filters for long exposures. 

In the old phrase; "Pick two of the  three" - there are not that many professional instruments that can manage such widefield imaging (without having to take multiple images and mosaicking them) - it is very hard to get auth to allocate multiple hours of time on any target.

[vlaiv]

Ok, I'm also not overly happy with those black and white images:

I see two troubling features that I'd like to try to explain before I accept this.

1.  Features in lower image that are of equal brightness as detected feature - but not related to it. There seems to be some sort of signal on both ends of M31 of similar intensity as OIII feature - in fact whole half of M31 seems to be "swimming" in this signal in lower image. Top image has too narrow field of view to be able to see if it is present there as well.

2. Difference in stretch and brightness of the feature. I'd expect two images to show the same amount of stretch for same / similar brightness of the feature.

Feature looks to be about the same brightness in the image (dark versus light - feature vs background) - however, top image is significantly less stretched than the bottom. In top image there is clear gap between M31 and M110. In fact this "bridge" between two images is fainter than the OIII feature (less dark in inverted image).

In bottom image - it is the opposite - bridge between M31 and M110 is very saturated (deep dark in inverted version) in comparison to the feature.

This is impossible if both images show pure OIII and are "normally" stretched - even if stretch is non linear - as long as it preserves order of values (larger values stay larger).

I don't think that images presented are pure OIII images and I believe they have been manipulated in some way resulting in two different relative brightness (one has bridge brighter than the feature and with other image it is the other way around) - not something that I would expect from regular feature.

 

[ShinyTwelve]

10 minutes ago, Gfamily said:

One of the points when this was originally identified, is that this is a wide angle (almost 10° wide) field image taken using narrow band filters for long exposures. 

In the old phrase; "Pick two of the  three" - there are not that many professional instruments that can manage such widefield imaging (without having to take multiple images and mosaicking them) - it is very hard to get auth to allocate multiple hours of time on any target.

What part of "ask them their opinion" was hard for you to take onboard?

[Elp]

9 hours ago, Gfamily said:

 there are not that many professional instruments that can manage such widefield imaging

Its not that wide at all (well not compared to a long FL scope). Both my first attempt through a telescope and my second attempt through a Samyang 135 both capture some if not all of that region around the bright blue star. Many people own a SY135 so it's not like it hasn't been imaged lots of times before. The dubious thing is how such a thing has been extracted from amateur imaging data, hence @vlaiv's query as to verification from a third party, you know like how all scientific discoveries are verified.

[Gfamily]

Here's a post by a professional astronomer about it. 

https://www.scientificamerican.com/article/a-recently-discovered-gas-cloud-near-andromeda-stumps-astronomers/

As described, this was initially detected by one team, and verified separately by another observer. 

A follow up paper 

https://iopscience.iop.org/article/10.3847/2515-5172/acaf7e#artAbst

Edited April 13, 2023 by Gfamily

[Elp]

Interesting read, and am aware four people were involved even though credit only went to the three. 160 hours is a lot of imaging time, and the data still needed math to reveal it properly. I'm sure a lot more people will attempt it this coming season.

Edited April 13, 2023 by Elp

[vlaiv]

1 minute ago, Elp said:

Interesting read, and am aware four people were involved even though credit only went to the three. 160 hours is a lot of imaging time, and the data still needed math to reveal it properly.

This math is what I'm worried about.

I haven't seen it mentioned in the paper, and there is no description of what they did - just a brief mention in the video, and from that - I can only conclude that it might be the wrong approach and what is causing the artifact (say part of IFN is being rendered like OIII because of that).

I thought about possibility of using continuum removal, and only place I see it possibly working is when using another NB filter with slightly wider band. Maybe using 6.5nm and 4nm. But then, what is the point of using wider band in the first place when the same imaging time could be done with narrower band filter.

Wider band filter will simply produce worse SNR image, and mixing worse SNR with good SNR data can degrade the whole thing.

Then there is mention of using blue to do it, but why not green as well? Both blue and green filters are designed to pass OIII.

 

[Mr Spock]

10 hours ago, ShinyTwelve said:

What part of "ask them their opinion" was hard for you to take onboard?

Please remember this is a friendly forum. Let’s try to keep it that way. 

[The Lazy Astronomer]

17 minutes ago, vlaiv said:

I haven't seen it mentioned in the paper, and there is no description of what they did

Yes - I have to admit I found this quite strange when I was looking for info on how they got to the final image. I wouldn't expect post processing details, but I would've thought (for a scientific paper) details of the pixelmath operations performed to the Oiii image should be included.

[michael.h.f.wilkinson]

I have spoken to a couple of colleagues in the astronomy department, and they think it is a legitimate detection. The fact that the same structure appears in data from different scopes in the same location is highly unlikely to be the result of a processing artefact. There have been more such detections (think of the Squid Nebula, and several very faint PNs) in O-III by amateurs willing to spend loads of time on a single target.  It does of course help if others can confirm the detection.

[vlaiv]

1 hour ago, michael.h.f.wilkinson said:

I have spoken to a couple of colleagues in the astronomy department, and they think it is a legitimate detection. The fact that the same structure appears in data from different scopes in the same location is highly unlikely to be the result of a processing artefact. There have been more such detections (think of the Squid Nebula, and several very faint PNs) in O-III by amateurs willing to spend loads of time on a single target.  It does of course help if others can confirm the detection.

What do you make of this supposed continuum subtraction method (I know we don't have much data - but for the sake of argument, let's go with what we know - blue filter was somehow subtracted from OIII to remove residual light that made it thru the filter and is not OIII).

[ollypenrice]

Science being a culture of doubt, I'll embrace doubt.  I must say that this object has the potential to be an artifact, to my eye, because it follows the outline of M31. This means it might, at least in principle, have been generated by internal reflection.  On the other hand, if it is associated with M31 and doesn't lie much closer, it would probably have good reason to follow M31's shape.

A control experiment might be to carry out the same capture and processing on another large, bright galaxy and see if the phenomenon repeated itself.

Olly

[michael.h.f.wilkinson]

1 hour ago, vlaiv said:

What do you make of this supposed continuum subtraction method (I know we don't have much data - but for the sake of argument, let's go with what we know - blue filter was somehow subtracted from OIII to remove residual light that made it thru the filter and is not OIII).

What I imagine could be done is to use a broader band filter (one or more) to to estimate the background continuum contribution (e.g. by scaling it to the correct level using least-absolute-deviates linear fitting) That should work reasonably well. Ideally you would block the O-III line, but even without that, estimating the background continuum contribution boils down to solving a large set of linear equations.

[vlaiv]

In the paper, it says that estimated surface brightness is:

4 ± 2 × 10⁻¹⁸ erg cm⁻² s⁻¹ arcsec⁻²

How am I supposed to read that? From 0 to 4? It might not be even there with 0 surface brightness?

Anyway, if it is 4 x 10^-18, what would that be in magnitudes (if anyone has handy calculator)?

[vlaiv]

2 minutes ago, michael.h.f.wilkinson said:

What I imagine could be done is to use a broader band filter (one or more) to to estimate the background continuum contribution (e.g. by scaling it to the correct level using least-absolute-deviates linear fitting) That should work reasonably well. Ideally you would block the O-III line, but even without that, estimating the background continuum contribution boils down to solving a large set of linear equations.

Yes, a bit broader OIII is much better choice than blue filter.

And if I may another question - what do you think of signal strength inversion in presented images?

 

[michael.h.f.wilkinson]

10 minutes ago, vlaiv said:

Yes, a bit broader OIII is much better choice than blue filter.

And if I may another question - what do you think of signal strength inversion in presented images?

 

They may have been taken through slightly different filters. There seems to be more continuum emission in the lower one, especially in the outer reaches of M31

[vlaiv]

5 minutes ago, vlaiv said:

In the paper, it says that estimated surface brightness is:

4 ± 2 × 10⁻¹⁸ erg cm⁻² s⁻¹ arcsec⁻²

How am I supposed to read that? From 0 to 4? It might not be even there with 0 surface brightness?

Anyway, if it is 4 x 10^-18, what would that be in magnitudes (if anyone has handy calculator)?

Maybe we can go about it the other way.

erg is 1e-7J

So total amount of energy is 4 x 10e-25 joules per cm per second per arcsec squared.

Energy of a single photon is hc / lambda, so let's try to crunch the numbers

hc = 1.98644586...×10⁻²⁵

OIII wavelength is 500.7x10^-9 = 0.5007x10^-6

If we divide the two we get ~4 x 10^-19 joules

To get number of photons per cm squared per second - we need to divide those two, so 4 x 10^-25 / 4 x 10^-19 = 10^-6

That is one photon per cm squared per million seconds, or about one photon per hour per arc second squared for perfect 4" telescope (100% transmission and sensor QE and no atmosphere attenuation).

Ok, I'll buy into premise that one wants hundred of hours to detect it. At least we have a baseline for anyone attempting to capture it - to calculate needed exposure.