When do you give up on Oiii or Sii in narrowband imaging?

9 replies454 views
Tony Gondola avatar
When imaging emission nebulae Ha is usually very strong but sometimes it's hard to tell if there's enough significate signal on Oiii or Sii to make the effort to pull out. What are you're methods for figuring this out?
Engaging
ScottF avatar
I usually look up the object here or elsewhere and see what high quality images look like and go from there. Otherwise it depends on my opportunity for clear skies. If the forecast is dismal, which is common for my area, I don’t try SII.
Himanshu Pandey avatar
I've been imaging the crescent nebula the past few days which is strong in H and O but very weak in S.  Despite that and because I don't know any better my thought is to go after a similar amount of integration with S as with H and O on the basis that I'll get better color variation on the stars in the final image if the star fields are approximately the same level in each channel. 

I'm also imaging the same object with OSC and dual narrowband and it looks fantastic.
Tony Gondola avatar
Himanshu Pandey:
I've been imaging the crescent nebula the past few days which is strong in H and O but very weak in S.  Despite that and because I don't know any better my thought is to go after a similar amount of integration with S as with H and O on the basis that I'll get better color variation on the stars in the final image if the star fields are approximately the same level in each channel. 

I'm also imaging the same object with OSC and dual narrowband and it looks fantastic.

I usually run RGB for stars when doing narrow band anyway so it won't effect star color. I'm really trying to find a breakpoint where it's just not worth the integration time. One thought I had would be to measure the SNR of a single long frame or  short stack, say 10 min. and decide from there.
Riccardo Civati avatar
usually SII is very low respect Halpha, i use also SII but is very faint.  In post processing i use the dynamic narrowband to figure out both OIII and SII.
First of all, i check the images on astrobin ( top picture nomination /top picture/apod) to see how it is done 

My last M27 was done with SII OIII Haplha. SII is really weak and completely overlaps with Alpha and I had to come up with some techniques to get it to extrapolate
andrea tasselli avatar
Use a narrowband survey unless a spectrogram is somehow available, which for my preferential targets (PNs) might be/usually is.
Himanshu Pandey avatar
Tony Gondola:
Himanshu Pandey:
I've been imaging the crescent nebula the past few days which is strong in H and O but very weak in S.  Despite that and because I don't know any better my thought is to go after a similar amount of integration with S as with H and O on the basis that I'll get better color variation on the stars in the final image if the star fields are approximately the same level in each channel. 

I'm also imaging the same object with OSC and dual narrowband and it looks fantastic.

I usually run RGB for stars when doing narrow band anyway so it won't effect star color. I'm really trying to find a breakpoint where it's just not worth the integration time. One thought I had would be to measure the SNR of a single long frame or  short stack, say 10 min. and decide from there.

That is a great way get started on a subjective determination based on your own tastes.   Good luck!
Daniel Cimbora avatar
Like others, I look at images here in AB to see what others have pulled out with narrowband filters. 

This is also a great resource for NB strength of common objects:  Link

For less commonly imaged objects, if I collect data for an hour and can’t see any signal on the stacked images for a particular filter, then I conclude that it’s probably not worth it unless I’m willing to commit many dozens of hours.
Well Written Helpful Insightful Concise
Tony Gondola avatar
Daniel Cimbora:
Like others, I look at images here in AB to see what others have pulled out with narrowband filters. 

This is also a great resource for NB strength of common objects:  Link

For less commonly imaged objects, if I collect data for an hour and can’t see any signal on the stacked images for a particular filter, then I conclude that it’s probably not worth it unless I’m willing to commit many dozens of hours.

That site is a great resource!
Jan Erik Vallestad avatar
Usually I spend a reasonable amount on each target, so what I tend to do is to stack each dataset as I go along. I'll quickly spot what's there and guesstimate how much I'd need to add in order to gain the SNR I want, based on what I want to bring forth. I do this for every new session most of the time, unless I've already decided that I need a lot more than I can get during one session.

EDIT: I very, very rarely bother doing Sii though. It's just not my cup of tea I've figured. But the same thing applies to any wavelength for that matter.