I often see people quoting the seeing at their location or during a particular imaging session. My question is, how do they know? Is it from FWHM measurements on images captured or is there a more direct method? I'm really curious as to how this is monitored and measured.
Measuring the seeing
Well Written Engaging
Some weather services offer a seeing forecast, like Meteoblue: https://www.meteoblue.com/en/weather/outdoorsports/seeing/
I have found these Meteoblue forecasts to be somewhat accurate, at least better than a wild guess. If Meteoblue claims seeing to be awful, it usually is, and when it claims good seeing it probably isn't bad.
For actual measurements i just use HFR reported by NINA, and PHD2 star size statistics. Image long enough with the same kit and you have a good idea on what the HFR should be, and know how to spot the differences between stable and bad nights. With my current kit an HFR of less than 3px is excellent, 3-3.5 is good, 3.5-4 is in the average to not so great territory and more than 4 is terrible. Worst case scenario is more like 5px, in which case the jet stream would have to be above, or a local seeing effect from some kind of thermal source (like someone heating a wood fired Sauna downrange as an example). As these statistics are per pixel and so based on the kit used they do not translate to your HFR/FWHM values, but just gave these as an example.
I have found these Meteoblue forecasts to be somewhat accurate, at least better than a wild guess. If Meteoblue claims seeing to be awful, it usually is, and when it claims good seeing it probably isn't bad.
For actual measurements i just use HFR reported by NINA, and PHD2 star size statistics. Image long enough with the same kit and you have a good idea on what the HFR should be, and know how to spot the differences between stable and bad nights. With my current kit an HFR of less than 3px is excellent, 3-3.5 is good, 3.5-4 is in the average to not so great territory and more than 4 is terrible. Worst case scenario is more like 5px, in which case the jet stream would have to be above, or a local seeing effect from some kind of thermal source (like someone heating a wood fired Sauna downrange as an example). As these statistics are per pixel and so based on the kit used they do not translate to your HFR/FWHM values, but just gave these as an example.
Helpful Respectful Engaging
I already know what my Bortel is, so the one thing I look at is where the jet-stream is, how much moisture there is, and what really works best is just look at the stars. If the stars
are really steady the seeing is most likely really good, if they are twinkly, the seeing is not going to be good.
are really steady the seeing is most likely really good, if they are twinkly, the seeing is not going to be good.
Concise
Oskari Nikkinen:
Some weather services offer a seeing forecast, like Meteoblue: https://www.meteoblue.com/en/weather/outdoorsports/seeing/
I have found these Meteoblue forecasts to be somewhat accurate, at least better than a wild guess. If Meteoblue claims seeing to be awful, it usually is, and when it claims good seeing it probably isn't bad.
For actual measurements i just use HFR reported by NINA, and PHD2 star size statistics. Image long enough with the same kit and you have a good idea on what the HFR should be, and know how to spot the differences between stable and bad nights. With my current kit an HFR of less than 3px is excellent, 3-3.5 is good, 3.5-4 is in the average to not so great territory and more than 4 is terrible. Worst case scenario is more like 5px, in which case the jet stream would have to be above, or a local seeing effect from some kind of thermal source (like someone heating a wood fired Sauna downrange as an example). As these statistics are per pixel and so based on the kit used they do not translate to your HFR/FWHM values, but just gave these as an example.
Going by NINA's HFR at a pixel scale of 0.66", an HFR of 2.0 = 2.6", 3 = 3.9" and 4 = 5.3" (HFR*2*0.66). Does that sound reasonable? These numbers seem a little larger than I would expect but I suspect guiding is also playing a role in that. I've also noticed that HFR in NINA is effected by exposure time. I'm thinking that using very short exposures, on the order of a sec. or two and averaging would be more accurate.
Helpful Insightful Respectful Engaging
Bob Lockwood:
I already know what my Bortel is, so the one thing I look at is where the jet-stream is, how much moisture there is, and what really works best is just look at the stars. If the stars
are really steady the seeing is most likely really good, if they are twinkly, the seeing is not going to be good.
This is really the way to do it. You just have to learn to eyeball it. If you see a lot of twinkling at the zenith you are not going to be doing any sharp imaging.
Pulling the FWHM from, say PixInsight, is not very accurate because your optical train's flaws are baked into the measurement. However that does not make it a useless measurement, because you can use those numbers to compare night to night for your own system.
Well Written Helpful Insightful Concise Engaging
I understand that I can eyeball for a general idea and use FWHM but what I'm really looking for is a way to accurately measure the seeing. I wonder how site surveys for large telescopes are done?
Well Written Engaging
DIMM (Differential Image Motion Monitor)
I think the twinkling, or no twinkling is going to dictate what the FWHM is going to be. High FWHM numbers, bad seeing, low FWHM numbers, good seeing, simple as that.
As for the large telescope in observatories, I would think they all use AO that fixes the seeing for them.
As for the large telescope in observatories, I would think they all use AO that fixes the seeing for them.
Oskari Nikkinen:
Some weather services offer a seeing forecast, like Meteoblue: https://www.meteoblue.com/en/weather/outdoorsports/seeing/
I never really understood the number there.
The site says Index 1 and 2 indicate the seeing, with 5 as excellent and 1 as awful.
At the same time the value "Arc sec" often is larger with excellent index than with bad index, even though one would expect it the other way round.
So which value should we trust? When I use the MTF analyzer for example, it want me to enter the seeing in arc sec, which makes total sense for me. But of what use ist that value on meteoblue when the indices give you a contradictory information?
Engaging
With the provision that MeteoBlue numbers are just numerical fantasy, just look at the arcsec seeing prediction. The only reasonable good indicator of seeing quality is that of the shear term in the jet stream speed (or upper troposphere wind).
Actually, given the rare number of nights where we dont have clouds here, probably the best solution is to go out and hope for the best anyway.
Which is why I'd like to be able to actually measure the value. A DIMM is not practical because of cost and even the Santa Barbara Instruments SM-3 seeing monitor is over $1500.00
I suspect that the programming inside PHD2 is capable of doing something similar to the SIMM technique that Santa Barbara uses. Obviously, you wouldn't do this while guiding or even with tracking running but I could imagine having the system look at Polaris or a dimmer star near the NCP for measurement. I know it's not something we absolutely need to know but it would make for an interesting project.
I suspect that the programming inside PHD2 is capable of doing something similar to the SIMM technique that Santa Barbara uses. Obviously, you wouldn't do this while guiding or even with tracking running but I could imagine having the system look at Polaris or a dimmer star near the NCP for measurement. I know it's not something we absolutely need to know but it would make for an interesting project.
Oskari Nikkinen:
Some weather services offer a seeing forecast, like Meteoblue: https://www.meteoblue.com/en/weather/outdoorsports/seeing/
I never really understood the number there.
The site says Index 1 and 2 indicate the seeing, with 5 as excellent and 1 as awful.
At the same time the value "Arc sec" often is larger with excellent index than with bad index, even though one would expect it the other way round.
So which value should we trust? When I use the MTF analyzer for example, it want me to enter the seeing in arc sec, which makes total sense for me. But of what use ist that value on meteoblue when the indices give you a contradictory information?
* The index and arc sec value are different things, index is for lucky imaging and arc sec is for long exposure. The index is the high frequency "wobble" part of the seeing and the arc sec value is the width of that wobble, or so i have understood it. I have had nights with a terrible index of 1 and 2, but with a fwhm of less than 2 which is average. On those nights deep sky imaging is ok, but planetary is a waste of time.
Those estimates are really just a guideline. If Meteoblue claims 2.7 arc sec fwhm seeing, its going to be bad, but no accurate info on just how bad. Likewise if the value approaches 1 its going to be better than average, but doesnt actually guarantee excellent seeing. The jet stream portion of that site is the most important and accurate data and gives decent info on that night.
Helpful Insightful Respectful Engaging
Oskari Nikkinen:
Now that you say it, this kind of makes sense. I did not try lucky imaging so far, but will definetely keep this in mind, once I do. Thank you!
Tony Gondola:Oskari Nikkinen:
Some weather services offer a seeing forecast, like Meteoblue: https://www.meteoblue.com/en/weather/outdoorsports/seeing/
I have found these Meteoblue forecasts to be somewhat accurate, at least better than a wild guess. If Meteoblue claims seeing to be awful, it usually is, and when it claims good seeing it probably isn't bad.
For actual measurements i just use HFR reported by NINA, and PHD2 star size statistics. Image long enough with the same kit and you have a good idea on what the HFR should be, and know how to spot the differences between stable and bad nights. With my current kit an HFR of less than 3px is excellent, 3-3.5 is good, 3.5-4 is in the average to not so great territory and more than 4 is terrible. Worst case scenario is more like 5px, in which case the jet stream would have to be above, or a local seeing effect from some kind of thermal source (like someone heating a wood fired Sauna downrange as an example). As these statistics are per pixel and so based on the kit used they do not translate to your HFR/FWHM values, but just gave these as an example.
Going by NINA's HFR at a pixel scale of 0.66", an HFR of 2.0 = 2.6", 3 = 3.9" and 4 = 5.3" (HFR*2*0.66). Does that sound reasonable? These numbers seem a little larger than I would expect but I suspect guiding is also playing a role in that. I've also noticed that HFR in NINA is effected by exposure time. I'm thinking that using very short exposures, on the order of a sec. or two and averaging would be more accurate.
* HFR is close to fwhm i believe (just a different form of measurement), so 3px at 0.66 is around 2 arc seconds. And youre right, short exposures get a more accurate readout, which i do since i focus manually. I use 3s focus exposures and get the seeing info from those. Any wind or guiding issue will add to the HFR during a longer exposure so less useful for seeing estimation.
I should say that the HFR measurement in NINA is often much worse than the actual stack fwhm because it is measured from a single sub which NINA debayers so the measurement is rather noisy and favours bright stars close to or at saturation (i shoot OSC). For data that during capture was measured as 3px HFR i get a stacked fwhm of somewhere around 1.4->2".
So its all relative to your kit. But having used my kit in all sorts of conditions those numbers do give a decent estimate on what to expect when its all stacked.
Helpful
This is what I use for jet-stream. At the moment, for me, aside from moon being full, at my location it would be pointless to image with the upper
winds hitting around 100mph. Not sure if this site is available outside of the US.
Jet Stream Winds at 300 mb (~30,000 ft) (GFS 10-day Forecast) (weatherstreet.com)
winds hitting around 100mph. Not sure if this site is available outside of the US.
Jet Stream Winds at 300 mb (~30,000 ft) (GFS 10-day Forecast) (weatherstreet.com)
Helpful
New to or Considering Astrophotography/astronomy?
I had posted this with a new image I’d uploaded as a change from the “typical” review of my processing steps/acquisition. A friend/commentor suggested I post in a forum, so this is the result. This is VERY subjective and based entirely on my experien...
Super Sample Anti Aliasing aware Registration and Stacking
Here to document a technique I'm working on for improving the quality of image stacking especially from CFA/Bayer camera images. In images, the overlapping of pixels on other pixels causes a pattern called Moiré interference. This can be seen opt...