View Full Version here: : Seeing and what it can do to your images
29-02-2012, 09:49 PM
There have been some discussions around seeing and how it can affect the clarity of images so I thought this would be a nice illustration of what can happen if your seeing is bad even at medium focal lengths.
Both images taken with Starfire 152 EDF at F7.5 using Starlightxpress SXVH9 CCD, same site (Canberra ACT 780m ASL), same photographer, same mount, with only basic processing and esposure times were similar too, plus careful attention to focus was used (the imager is anal about this) in both cases.
Image on the Left was taken in very poor seeing
Image on the right was taken in quite good seeing
Even at the relatively modest focal length of 1140mm - the difference is remarkable!
Here is the comparison (http://www.pbase.com/strongmanmike2002/image/141800216/original)
So seeing is everything (as is focus) as far as I am concerned, here in Newcastle the left hand image is easy to take most nights but the right hand image is a rarity indeed (if ever) :sadeyes:
Nice comparison Mike - all too easy for us to gloss over these essentials while worrying about collecting enough data.
You know, those pictures are showing some real potential. This might be an object you should perhaps spend some more time on.... :whistle:
29-02-2012, 09:58 PM
Sorry Mike , while I entirely agree, you have not discovered fire here.
The Pro's have been putting fast optics on mountain-tops for decades........ I suspect it was not because they liked the fresh mountain air :)
29-02-2012, 10:08 PM
:rolleyes: Never thought I had but unlike Rob, clearly you seemed to have missed the point of the post Peter.
29-02-2012, 10:11 PM
Ta ta ta daaa! (http://www.pbase.com/strongmanmike2002/image/97061387/original)
29-02-2012, 10:15 PM
Great comparison there Mike, very nice. Do you have the FWHM values for the two images, that would be interesting to know?
29-02-2012, 10:21 PM
No unfortunately not, all I remember is that the night of the left hand image I limited it to 2min exposures I think to try and by pass some of the crap seeing (in vain of course) and that I was swearing a bit :P The right hand image was 6 X 5min Lum and then from memory something happened, maybe cloud or an equipment failure I can't remember :question: I didn't get any colour for that one.
29-02-2012, 10:30 PM
Please don't get me wrong.
Dr Alan Holmes (aka SBIG) wrote a good piece (yep on SBIG's website somewhere) on how amateur gear could rival Pro mag limits *only if* their seeing was better (he'd just got back from Paranal)
...but if you are not saying even in modest apertures (less than about 150mm) seeing effects CCD images most nights.......
...well...I guess I have missed the point.
29-02-2012, 10:34 PM
Not sure why all the cryptic code is necessary :shrug: I really don't think the point of my post is that complicated
29-02-2012, 10:44 PM
106mm aperture same object 2 nights apart. Both focused with focusmax
Just give up and go inside when it's that bad.
29-02-2012, 11:13 PM
You have given a graphic example of an obvious truth....and that's great.:thumbsup:
Chris has (kindly) shown an similar example of (as I had shown on a commercial web site....that I won't point to in respect to the other Mike...! :) ) well, about a decade ago (OK, I'm starting to feel like a grumpy old bugger :rolleyes: )
Yes..... Seeing affects FWHM's ....so I'm now puzzled (I ask myself "why" more fequently in my dotage :) )
I don't see any other point. Yes, seeing can be awful and degrade images from otherwise perfect optics.
Was there something else I missed ?
I'd be interested to know if/how people quantitate their seeing while setting up an imaging run. To be honest I'm not sure I'd recognise terrible seeing (you dont always have a great image from your equipment to compare with).
Does anyone have a quantitative rather than visual estimate way of doing this?
01-03-2012, 12:00 AM
Hi Peter ,
Thanks for posting those 2 examples.
Your Starfire 152mm has a Dawes limit of 0.76 arc seconds.
Seeing would normally be only about 2 arc secs.
I think top seeing conditions would seldom be better than 1 arc sec.
I have noticed that seeing is much better at higher altitudes
compared with sea level.
It would be interesting if you re-did the pic again at high altitude.
01-03-2012, 12:16 AM
Great example Mike, currently my seeing is really bad at the moment so it is a good comparison not the worry so much about my images.
01-03-2012, 04:00 AM
Peter, I dont think you have missed anything:D,
I, for one, who has not a set up like you or Mike, can use the comparison as a guide. There are so many factors that can affect AP and the pics give an ( slight)insight as to what to expect when imaging in good and bad seeing....
01-03-2012, 07:36 AM
The longer the focal length the more the seeing will wreak havoc. I did not know what was happening with my scope last year during a week of particularly bad seeing.
It would go out of focus from one focus subexpsoure to the next without changing the focus at all.
It is also highlighted when you use a small chipped camera as that creates a digital zoom effect and magnifies the seeing more.
So the moral of the story is widefield and large chips to be less affected by the seeing (also larger pixel sizes rather than smaller).
01-03-2012, 08:12 AM
The difference in the two shots is quite striking. Sometime it's just not worth doing long FL work. Stick to lenses when the jetstream's up ;)
01-03-2012, 08:20 AM
Greg this is purely an illusion of course, the detail in the image is affected exactly the same by the seeing and it is really only the choice of file size you choose to display that causes this apparent difference. When you have a wide field compared to a narrower field at the same arc sec/pix and then display them at the same file resolution the affects of the seeing on the two images will look identical.
01-03-2012, 08:30 AM
Awe Mike ...... now you have just completely confuzzled me ......
back to google I go to figure out what you have said....:eyepop:maybe Wolfram Alpha will be able to decyfer what you said there....:thumbsup:
:D I'll:shrug:maybe:help:nah:question: forget about it......
One day I'll get there:lol:
01-03-2012, 08:49 AM
Yeh?...sorry :question: :lol:
01-03-2012, 09:26 AM
I just was trying to be funny in my response but failed....
I just did not understand the reply you made to Greg.......It's a bit beyond my scope of knowledge.....:lol:
01-03-2012, 09:47 AM
I initially thought ... What tha? ... Mike (and you know I loves ya mate) is stating the bleeding obvious. But then I realised that we (people in this forum) are not all experienced imagers and seeing seeing (hehe) illustrated like this is beneficial.
My advice to any up & coming imager who doesn't already do so is to monitor FWHM on all their subs!! While FWHM does depend on the star field (number of bright stars vs fainter stars & exposure duration), it is an excellent indicator of seeing. I check every sub using CCDInspector and I've come to know the best and worst seeing conditions at my site. Typically I expect to see FMHM ranging from 1.5 to 3.75 arcsecs (and much worse sometimes), with the average "good" night around 2.5. I automatically discard ANY sub greater than 4 (even at F5.5) and if most subs are significantly less than that I lower that threshold because I strive for fairly uniform FWHM in my image stacks. If I'm imaging at F11.7 (1.06 arcsec/pix) then the threshold is also lower.
As Mike's example shows, and considering I obsess over detail, there's simply no point imaging beyond a certain threshold - especially for Lum and Ha. Of course this threshold will vary depending on your site and your image ratio so if your rig has a ratio of say greater than 4 arcsec/pix (the super WFers among us) then lucky you - you'll be able to image even when the seeing sucks!!!
01-03-2012, 09:55 AM
Well, yes...... but..... unless the sampling is adequate you can never get very high resolution images with a short focal length.
Under-sampling will easily mask any seeing blur....with images looking pretty much the same from night to night, no matter how good or bad the seeing is.
At longer FL's you end up in the over-sampled region, hence seeing makes obvious differences to star sizes and resolution.
01-03-2012, 10:03 AM
And that was my point exactly :thumbsup:
Love you too :love:
01-03-2012, 11:01 AM
ATM any clear sky is good sky.
01-03-2012, 12:55 PM
I think I understand what Mike is saying here about focal legnth. If you take the same camera say my QHY9 with 5.4nm pixels and put it on my scope at 800mm focal legnth I get an image scale of 1.39 arc seconds per pixel.
Now If I were to take the same camera and put it on a 2000mm focal legnth scope the image scale would be 0.55 arc seconds per pixel.
now say the seeing is around 2 arc seconds. Because the sensor size has not changed the same star with the same seeing will appear as more pixels in the 2000mm scope (4 pixels as apposed to around 1.6 with perfect guiding)
01-03-2012, 02:54 PM
Ok, this is probably cracking a nut with a steam-rolller, but the mouse-over below....illustrates my under-sampling point pretty well...
No amount of bad seeing is going to phase a 10 arc sec per pixel image, but it certainly will blur-out a 1 arc sec per pixel image.
the link is here (http://www.atscope.com.au/BRO/gallery38ro.html)
01-03-2012, 06:27 PM
Yes know all about good seeing versus bad seeing. Except at native puny focal lengths seeing is nothing. When you get some focal length then seeing is everything.
Good seeing (http://paulhaese.net/Saturn29March20111316UT.html)
Bad seeing. (http://paulhaese.net/Saturn14January20111912UT.html)
Good seeing makes great planetary imagers. Bad seeing and no one wants to know you. Know all about this good seeing bad seeing thing Mikie. ;)
That's about 0.1 arc second per pixel.
Yeah, I figured you had education in mind Mike.
I found the seeing simulation this page helpful early on, particularly from point of view of visual observing, but applies to imaging too of course.
01-03-2012, 07:23 PM
Very cool link - plenty of info there. :thumbsup:
01-03-2012, 07:52 PM
It's all about degrees Paul. To my eye at least, the effects of the seeing on your planetary images looks no more damaging from an aesthetics comparative viewpoint than it was with the Cen A shots like I linked too :shrug:. Specifically comparing your two Saturns in fact suggests less of a detrimental effect even..? This would likely be because you were still able to remove the worst seeing affected fraction of a second subs from the poor seeing session but I wasn't able to do this because my subs were much longer.
But in the end, it is true, seeing can be a pain for all but the shortest of focal lengths....viva la Marc Aragnou :P
01-03-2012, 08:14 PM
Thank you Mike.
To people new at this like myself, a post like this is very informative and helpful.... a real "thousand words" of information in your 2 photos.
Interesting post Mike,as I image at 2000mm with a small pixel camera seeing affects me every night. I often start a imaging session with poor seeing then shift to better then it may deteriorate then improve again. A quick glance at the dec. guiding graph tells all if the corrections are greater than R.A. If the seeing is poor but not very poor I switch to a colour filter and reduce the star sizes in image processing the RGB. On those great seeing nights take all the Luminance I can.
Clear skies Ken.
01-03-2012, 09:23 PM
You're right I can remove a lot of the problems with frame selection and with gathering high speed frames. A better example of what I am trying to say is the following images. One is 8.5/10 and the other just barely 5/10. That is when it can be chalk and cheese. Remembering of course there are many nights I look and the seeing is 5/10 so I don't bother setting up. At least with nights of average seeing one can get reasonable data for DSO imaging. Maybe not luminence but certainly the RGB.
Great seeing (http://paulhaese.net/Jupiter%2010%20may.html)
Average seeing (http://paulhaese.net/14september2011Jupiter.html).
Long and the short of it though and for those wanting to learn about seeing, Mikes point is well made. Luminence should only be gathered near the meridian and during good seeing. Likewise planetary imaging should not be conducted in poor seeing. If you are using one shot colour, well image when the seeing is best if you can, but if you have constant cloud, image whether the seeing is good or bad.
01-03-2012, 09:26 PM
Mike - Thanks for taking the time to clearly illustrate this difference - amazing to see the number of stars lost to bad seeing.
Have to disagree with the 'puny' focal length comments (Paul's) though - good/bad seeing is good/bad seeing at any focal length...its differences are obviously marked as FL increases as stated.
01-03-2012, 09:29 PM
Yeh I thought so too, no need to complicate it, most people don't get to see a direct comparison like this, plus given many people have seen my high res deep sky work, here is proof that it isn't always possible...as I am discovering with my new AG12, not because of the scope but the seeing :sadeyes:
01-03-2012, 09:31 PM
Yeh! that's a hell of a difference mate :eyepop: :lol:
We are all in the same boat when the seeing doesn't cooperate :thumbsup:
I hope the comparison was helpful to people anyway :)
01-03-2012, 09:48 PM
Another great comparison - the difference is jaw dropping!
Good seeing or bad seeing, getting out and making some pictures is the whole point of astronomy to me.
When the seeing is bad then you can make pictures that are not great but still good and when the seeing is excellent, the picture just jumps out and is good.
The comparison is very good to show what seeing does to a picture.
01-03-2012, 10:19 PM
great info for me. Good to see the side by side comparison. As a not quite there beginner yet, i try and soak in all this info i read.
And what a great image that is of Centaurus A Ultra Deep Field - Multi Award Winner. :eyepop: Images like this inspire me to sell everything thats not bolted down just to get a modest little setup under way. :thumbsup:
01-03-2012, 10:42 PM
Geez some of you people are hard to bloody please.
Mike makes a simple thread to show how poor seeing can make imaging a bit harder than it already is and he is torn a new one because he claims to have discovered Fire, has nothing compared to the Planetary Imagers . . . sheesh!
Thanks Mike for giving us a visual example of the difference between good and bad seeing. Us oldies may know it but I'm sure the newbies would like to know and see it.
We read about it, so now it's good to see a side by side example.
02-03-2012, 08:50 AM
:lol: All's good Kenny, I think the spirit and intention of the post was interpreted and responded to appropriately by most ;)
02-03-2012, 09:54 AM
It's an academic discussion Ken, not a personal attack. People are allowed to discuss this and not agree, or have a difference of opinion, or even suggest that there is nothing really to moan about when imaging at native focal lengths. I am not ripping Mike a new one, I am discussing the topic and suggesting he has nothing really to moan about. That is a valid discuss point. I hear this business all the time from DSO imagers and it always makes me laugh.
Just for the record, seeing defects are multiplied many times over when imaging at 6000+mm. At 12,000 mm it is like watching something in a washing machine. Imaging at 800-1000mm has little effect and can only be seen via guide record generally. My point is to show that everything is relative. From my perspective talking about seeing at low focal lengths is quite humorous.
Next time I just won't bother.
02-03-2012, 11:27 AM
Yeh I guess so but even at the puny 1140mm FL, if you look at my comparison again, I think I know which of the two results you would prefer to have as the Lum in your LRGB ;)
Please do, no issues here mate, your second comparison was just as revealing as my Cen A's :thumbsup:
02-03-2012, 04:04 PM
Some of the degraded seeing could be at the scope itself, a scope bought out of a warm building causes air currents, I always set up my rig as soon as the sun is no longer on it to give as much cooldown time as possible, and only use just enough anti dew heating to keep dew at bay.
A friend of mine into viewing with a 16 inch truss dob is installing a special low virbration fan and a special baffle that will draw air across the front of the primary mirror to hopefully eliminate any boundry layer turbulence. He says he doesnt want to capture starlight thats been travelling for hundreds or thousands of lightyears only to be degraded in the last few inches of its travel.
06-03-2012, 05:02 AM
I might be asking the obvious but would it then be preferable to shoot binned rgb images on nights of poor seeing?
Just thinking that not only will the image scale be more forgiiving but blurring less noticeable when combined with a good luminance frame?
06-03-2012, 09:05 AM
RGB is very forgiving and can be taken in poor conditions and/or binned. It is the Lum, or what ever you use for Lum, that needs to be at it's best if you want the best resolution possible...but sometimes beggers can't be choosers :sadeyes:. Of coourse if you take your RGB in good seeing and bin 1X1 you can always then use it as a synthetic Lum to augment your actual lum :)
14-03-2012, 11:32 PM
Thanks Mike and Paul for posting these, it makes us beginners feel a lot better knowing it isn't *always* our fault :D
I find for planetary bad seeing is amazingly awful, it makes it impossible to focus as the planet dances around like a cork in a washing machine. What do you do when it is like that, or do you just pack up and try again on a better night?
15-03-2012, 12:16 AM
Go straight to bed and get that much needed sleep.:)
15-03-2012, 12:39 AM
I have only once packed it in due to seeing and that was the night I took that Centaurus A image... I just lump the softer images :)
15-03-2012, 08:43 AM
After many futile attempts on the planets, I've been doing deep sky whenever the seeing is poor, as the planets literally look like a blurry earthquake, and I have no idea how you would set a focus point.
On a good seeing night, planetary is a joy, but in bad seeing I am baffled.
vBulletin® v3.8.7, Copyright ©2000-2015, vBulletin Solutions, Inc.