SharpCap Forums

Cygnus is coming 'round again, and it is best to start early. Last night after two days of rain, the sky cleared around midnight. Tonight's work picks up where I left off last Fall.

The first objective is to see how much signal there is in each of my narrowband filters: Ha, OIII, SII, each 7nm bandwidth. The second objective is to measure its Photon Flux Density (PFD) using AstroImageJ, a useful tool for Photometry. The PFD calculation factors out my kit's light-gathering and light-sensing properties. It can then by used in a calculator to estimate how many frames to capture to achieve a particular signal-to-noise ratio for any telescope and camera, not just my own kit.

The attached "contact sheet" shows a single frame in each filter for three objects: M27, Sh2-86, and Sh2-105 (Crescent Nebula). The images are NOT calibrated. Next steps are calibration, AstroImageJ, and data reduction to calculate PFD for each object and filter. All this must be done before investing multiple session capturing actual data.

Brian

Cool to see this
And it also confirms what I did read about the fact that OIII is very useful for planetary nebulae

Menno

Menno,

Yes, OIII is quite strong in M27. Here is a better image. You can see the "shock waves" in the outer shell.

Brian

Brian

Very interesting post. I surmise that once you have processed each of the Ha, OIII and SII frames in AIJ you will be able to estimate the number of frames to give a balanced SNR across the 3 filters. This will avoid 'wasted time' in capturing more of a particular filter that is actually needed - I will coin the phrase 'intelligent imaging'.

It looks like OIII is the key to accessing M27's outer shell.

Dave

Hi Dave,

Balancing the SNR across the three filters ensures that each receives equal weight in the final image. Certainly that is one possibility, but it is not necessarily the one that depicts reality. If you want reality, then you should capture an equal number of frames. However, depicting reality is not a common goal in the AP community. Take for instance the PacMan nebula. Most images you see at AstroBin and elsewhere show a strong OIII component (cyan color). However, OIII is quite weak in the nebula.

Most people approach AP by capturing an equal number of frames, and then use Astro Pixel Processor or PixInsight to boost the weak channels. That works, but the downside is that you are also boosting noise. My approach is to capture more frames in the weak channels, and then to rely on the magic of stacking to naturally boost them, therefore yielding a less noisy image. The PFD analysis identifies the weak channels, and tells you that you need to capture more of those, whereas for the strong channels, you need to capture fewer. So yes, it is "smart" in that sense. It helps you optimize your time at the telescope.

Brian

Thanks for the expanded explanation Brian.

Dave

Hi Brian,

Yes useful to see those sighting shots. The SII is very weak - it would I imagine take an awful lot of SII imaging to intelligently 'balance up'. Hmmm and I have just bought an SII ...

Reality is a slippery concept though in the sense that it depends which aspect of it you want to represent. For the planetaries if it is "what it really looks like" then I would imagine that the single green/ blue doubly charged oxygen OIII line is about all could ever be seen 'naturally' ? i.e the Ha, NII and SII reds are deep and relatively way off the eyes natural peak of sensitivity as well as being weaker signals.

But trying to depict the chemical composition is another valid take on reality - which is why I quite like the Hubble palette approach and trying to balance things up and then just taking the resulting view as a rather qualitative guide to what elements are there --and also how hot it is - i.e.

Whereas HA emission originates from the decay of UV excited H atoms from 3 down to n = 2 I understand that emission from O III, NII (and I guess also SII) comes from an entirely different process that is dependent upon the collision frequency with free electrons and that is therefore highly temperature-dependent. So although there is - in reality- much more hydrogen than any of these other elements - the oxygen signal dominates where it is very hot (> 10,000 K) - which I guess fits a planetary --- but not in more diffuse cooler emission nebula.

I am guessing also therefore that in most planetary nebulae that flash of red (e.g on M27 and M97) that we amplify up and makes it look attractive is probably more due to NII than it is to HA ? My filter is 7 nm so it wouldn't distinguish but maybe there are pictures using the expensive 3 nM Astrodon ones so it must be known?

currently a rare clear night and imaging galaxies...

Tim

Tim,

Here is an indispensable website for planetary nebulae:
https://web.williams.edu/Astronomy/research/PN/nebulae/

I've attached a snapshot of the spectra for M27 and M97 with annotations. M27 has strong NII emission lines that account for over 50% of the signal seen in my image using a 7nm Ha filter. To isolate Ha, one must use a 3nm filter. M97 also has NII lines, but they are not as strong.

Brian

Tim,

Creating the contact sheet is a small 30-minute investment that helps decide the following:
1. Do I have a chance at tri-color? If yes, then which channel(s) require more integration time to bring them in balance?
2. Do I have a chance at bi-color? Many times this is a viable option. I am partial to the HOO Palette.
3. There is nothing wrong with monochrome!

Regarding choice #3. I am going to try LHaRGB with Sh2-86. The nebula will be red, but star color will be accurate. One thing that I dislike about conventional NB imaging: ugly star colors.

Also, keep in mind that you don't necessarily need to perfectly balance the channels through stacking. You can still boost the channels any remaining amount in PixInsight or APP. Put another way, if PFD analysis says that you need 10 hours of SII for every hour of Ha, then compromise! Choose 3:1 and then make up the difference in APP or PI.

Brian

Brian,

Many thanks indeed for linking that website. Now I can replace supposition with fact. So in fact the red fringe in M97 is mainly due to HA rather than NII emission.

A lot of interesting information and possible inferences from those spectra..

For example it is interesting to note that the ratio of the (thermal collision frequency determined) OIII and NII peak heights to that of the (ionising radiation determined) HA peak is greater in M27 than it is in M97 despite M27 having the cooler central star at 85,000 K as compared to an estimated 124,000 K in M97.

I suppose that if both nebulae were the same size, density and age etc one might expect it to be the other way around with the one having the hotter central star generating relatively more OIII and NII photons relative to HA? M97 and M27 do in fact seem to be of a similar age (6-10K years) and size 1.82 and 1.44 light years in radius. However M27 is reported to comprise about 4 times as much mass as M97 (0.56 solar mass versus 0.13) in a slightly smaller volume so it is must be about 5-6 times denser which -all facts taken together - is consistent with M27 being overall about 3X brighter than M97 and also exhibiting the higher ratio of OIII to HA emission.

Tim