C/2023 A3 with HDR: artifacts or data?

andrea tasselli:
No artefacts there, those are veritable structures (jets) from the nucleus of the comet. So kudos to you for bringing them out.

Wow, thank you so much!
I did not expect to be able to capture such details! 😳

Depends on which structures you refer to. The jet streams from the nucleus has been present in most images I've seen so far. Those are real. If you mean the spiked pattern surrounding the blown core, that's probably another story.

Those seem to be the effect of not being careful enough when applying your script. I would encourage you to carefully use something like HDRMT, wavelets and masks to avoid it. The nucleus should be nothing but a bright dot if these are applied correctly, so I reckon you could get a lot more detail from this image )

Yes, I meant the spiked pattern; I should have been clearer.

I've also seen many beautiful pictures with jet streams, and that's precisely why I tried using the HDR script: to see if I could reveal some on my image.
And I was happy to see some, indeed. But then it also revealed or created this spiked structure around the core, which troubled me.

It seems somehow aligned with the stream direction, but also too good to be true.

I'll try to reprocess and give HDRMT a try!
Thanks a lot for your answer!

The "spike" are real, I used to get them too (on other comets, not this one as it is still out of reach for me). If I had the raw I can easily enough test it.

Yeah, that one would do. But I won't be able to give it a go before later this evening.

I'm sure if they are in the data they will come out.

Jean-David Gadina:
I uploaded the files on my Google Drive:
https://drive.google.com/drive/folders/1cyJ01hf77Z4KE3fpcmQGxv4mIMSM5Lpi?usp=share_link

Are these images non-linear? They seem to be according to the history log in PI, but if so it is over-exposed. Probably due to the high gain used, to a certain degree the exposure length but I can't say for sure whether 15s would be to much on a F/2 system or not. The way the comet is looking though I can totally see how you could get a spiked coma by applying HDR techniques, the coma has been saturated too much to get any kind of detail back from the core/nucleus.

If this is the un-stretched (linear) image I'm afraid the nucleus and inner coma is gone:

Are these images non-linear? They seem to be according to the history log in PI, but if so it is over-exposed. Probably due to the high gain used, to a certain degree the exposure length but I can't say for sure whether 15s would be to much on a F/2 system or not. The way the comet is looking though I can totally see how you could get a spiked coma by applying HDR techniques, the coma has been saturated too much to get any kind of detail back from the core/nucleus.

If this is the un-stretched (linear) image I'm afraid the nucleus and inner coma is gone:

I haven't applied any stretch.
This is my second time using the HyperStar and the first comet I try to stack.
I'm still trying to figure out the correct settings. Going from f10 to f2 is quite wild.
I should have reduced the gain - I used 252, as for my other shots at f10.

I also uploaded a single frame in the same folder if you want to see it:
https://drive.google.com/drive/folders/1cyJ01hf77Z4KE3fpcmQGxv4mIMSM5Lpi?usp=share_link

That's great, single exposures are usually the best when determining certain things. It is definetly blown out:

It will still make a good image, but I would mask off the brightest parts of the comet (coma) if applying the HDR techniques on this image. This way you get the nice jets from the tail section, and can moderately stretch the image so that it looks nice with the anti-tail as well 

That is definetly a big change, but using the histogram while imaging can be very helpful in that regard.

Thanks a lot for these suggestions!
I will try reprocessing it as such.

I usually look at the histogram in NINA before shooting, but not this time.
I think I got too excited! The weather has been so bad here that it was only the second night I could try and the first one with the telescope.

But I should definitely have done that and adjusted the gain accordingly.
I'll definitely remember that for the next one!

The image was out of focus. Not sure I can do much about that.

Yes, some mis-collimation is present in the image I would say. Or so it looks like to me at least. But that's not the cause of the issue at hand, nothing more than what BXT can handle somewhat ok. But one cannot repair over exposed stars/cores. You can however work around them a bit, the stars become very white though. Something like this perhaps, a bit of star residue streaking across but not too bad:

Slightly strange looking with that ball of light being very prominent


A bit better IMO with a little less separation of the over-exposed area.


There were some very nasty gradients there though, even after working on the background, especially to the upper left and slightly to the right in the image.

Jean-David Gadina:

Yes, some mis-collimation is present in the image I would say. Or so it looks like to me at least. But that's not the cause of the issue at hand, nothing more than what BXT can handle somewhat ok. But one cannot repair over exposed stars/cores. You can however work around them a bit, the stars become very white though.

There were some very nasty gradients there though, even after working on the background, especially to the upper left and slightly to the right in the image.

Thank you so much for taking the time to do this!
I really appreciate it.

I still need to check the collimation on the HyperStar. I'll add this to the list.
About gradients, I had several clouds passing over during the capture, unfortunately.
Maybe I could try integrating fewer frames, discarding the more blurry ones.

Absolutely, no problem at all!

That would explain it, I dealt with the same thing as well and had to discard a lot of frames. I kept some sub-par frames where the clouds didn't interfer with the comet as long as they weren't huge. But WBPP and some careful processing handled it well.