Highest magnitude to image?

[Calzune]

I'm new to Astrophotography and I'm just wondering what is the "top" magnitude you image? 

I know that the lower the magnitude is the more visible it is, so what is your limit when imaging and how do you know if an dso has to high magnitude? 

Im using stellarium, skysafari etc and in the settings you can adjust the magnitude to a limit so higher magnitude won't show up, so that's why I'm asking so I don't waste any time on very faint dsos. 

 

I hope you understand my question.. 

[vlaiv]

Things are not as simple as that.

There are couple of magnitudes reported for DSOs - and in general, they don't really tell you much.

First there is total magnitude. That is total integrated brightness of DSO:

It is number marked by arrow in Stellarium.

Then there is surface brightness number - sometimes expressed in mag/arcmin^2 and sometimes in mag/arcsec^2. Luckily there is easy conversion between the two - just add 8.89. In above image surface brightness is 19.89 in mags per arc second squared - or 11 mags per arc minute squared.

Problem is that these numbers generally don't tell you much. If we take 7.5 magnitude of this object - it would seem that it is fairly bright - you can see 7.5mag star with even small telescope, but with DSOs you need to account for their size as that light is spread over some surface. Larger the dso - dimmer it is for same total magnitude.

This is why we have surface brightness - that tries to avoid above problem and gives us idea how bright object is per unit surface. This is rather good measure for objects that are uniform in their illumination - but sadly there is only one such object (and even it has gradients) - background sky. All other objects show some variation in brightness.

For example - let's take a look at M51:

It says mag21.45 as surface brightness. But is it really? Have a look at this diagram:

It is rough surface brightness diagram of M51 that I made from one of my data sets. It is luminance brightness diagram. In it brightness ranges from mag17 to mag26.

If you want to capture tidal tail of galaxy you need to achieve good SNR at mag27. You see now that mag21.45 is in fact something like average value of brightness values of the galaxy.

Now the final question - how "deep" can you go? That really only depends on how much time you have. More time you spend on target - deeper you will go. In principle you want to get SNR>5 for interesting parts of the target to render it nicely. That means that you need to get signal to noise ratio of 5 or more.

You can calculate the rest - take magnitude that you are interested in and see how much photons it is producing per second and per exposure, ~880000 photons per second per cm^2 on top of the atmosphere from mag0 target, account for atmospheric extinction, optical losses in telescope, clear aperture of telescope, quantum efficiency of sensor, pixel size, focal length - and hence sampling resolution, level of light pollution, read and dark noise and sub duration - times number of subs and you will roughly get what sort of SNR you can expect in what amount of time given your setup and conditions.

Most galaxies that we image go down to mag29 - 30 in faintest parts. Not sure about nebulae.

As you see - this question is rather complex, but there is very simple way to go about it. Interested in imaging particular target? Look for results that people got with similar setups in similar conditions - pay attention to total imaging time. If you can replicate it or spend more time with better gear - it's worth having a go. With time you will get the sense of how bright / faint things are and you will be able to decide without looking for reference.

[Calzune]

7 minutes ago, vlaiv said:

Things are not as simple as that.

There are couple of magnitudes reported for DSOs - and in general, they don't really tell you much.

First there is total magnitude. That is total integrated brightness of DSO:

It is number marked by arrow in Stellarium.

Then there is surface brightness number - sometimes expressed in mag/arcmin^2 and sometimes in mag/arcsec^2. Luckily there is easy conversion between the two - just add 8.89. In above image surface brightness is 19.89 in mags per arc second squared - or 11 mags per arc minute squared.

Problem is that these numbers generally don't tell you much. If we take 7.5 magnitude of this object - it would seem that it is fairly bright - you can see 7.5mag star with even small telescope, but with DSOs you need to account for their size as that light is spread over some surface. Larger the dso - dimmer it is for same total magnitude.

This is why we have surface brightness - that tries to avoid above problem and gives us idea how bright object is per unit surface. This is rather good measure for objects that are uniform in their illumination - but sadly there is only one such object (and even it has gradients) - background sky. All other objects show some variation in brightness.

For example - let's take a look at M51:

It says mag21.45 as surface brightness. But is it really? Have a look at this diagram:

It is rough surface brightness diagram of M51 that I made from one of my data sets. It is luminance brightness diagram. In it brightness ranges from mag17 to mag26.

If you want to capture tidal tail of galaxy you need to achieve good SNR at mag27. You see now that mag21.45 is in fact something like average value of brightness values of the galaxy.

Now the final question - how "deep" can you go? That really only depends on how much time you have. More time you spend on target - deeper you will go. In principle you want to get SNR>5 for interesting parts of the target to render it nicely. That means that you need to get signal to noise ratio of 5 or more.

You can calculate the rest - take magnitude that you are interested in and see how much photons it is producing per second and per exposure, ~880000 photons per second per cm^2 on top of the atmosphere from mag0 target, account for atmospheric extinction, optical losses in telescope, clear aperture of telescope, quantum efficiency of sensor, pixel size, focal length - and hence sampling resolution, level of light pollution, read and dark noise and sub duration - times number of subs and you will roughly get what sort of SNR you can expect in what amount of time given your setup and conditions.

Most galaxies that we image go down to mag29 - 30 in faintest parts. Not sure about nebulae.

As you see - this question is rather complex, but there is very simple way to go about it. Interested in imaging particular target? Look for results that people got with similar setups in similar conditions - pay attention to total imaging time. If you can replicate it or spend more time with better gear - it's worth having a go. With time you will get the sense of how bright / faint things are and you will be able to decide without looking for reference.

Wow thank you for a detailed answer! 

Now I got even more confused.. But that's astrophotography right? 😂

 

Big thanks! 

[wimvb]

Wow! Excellent write up, @vlaiv

@Calzune: välkommen till forumet. I think this part is the most important for you, atm

9 hours ago, vlaiv said:

Now the final question - how "deep" can you go? That really only depends on how much time you have. More time you spend on target - deeper you will go.

Furthermore, the darker skies you have, the shorter time it takes to go to a certain "depth". The more light pollution you have, the longer it takes to get to a certain "depth". Light pollution adds noise, and you need more data to average that out.

Edited January 15, 2020 by wimvb

[jjosefsen]

Apologies if this is hijacking, but I find it is sort of related.

Does anyone know of a program that can platesolve and anotate stars up to and including mag19?

I can't seem to find reference to more than mag 17.

[vlaiv]

54 minutes ago, jjosefsen said:

Apologies if this is hijacking, but I find it is sort of related.

Does anyone know of a program that can platesolve and anotate stars up to and including mag19?

I can't seem to find reference to more than mag 17.

Have you tried Astrometry.net? It is also available for local install and you can even build your own indices for plate solving. You can download catalogs that contain stars down to mag19 and use them to build indices for plate solving (maybe annotation as well - not sure about that).

Have a look here:

https://archive.stsci.edu/prepds/atlas-refcat2/

 

[laser_jock99]

Modern CCD cameras are very sensitive to incoming individual photons, couple that with a large aperture scope and it's remarkable what faint magnitudes can be reached.

I attempted to image the same 'blank' area of sky used for the famous Hubble Deep Field Image. Using a 12" F4 fast Newtonian and Lodestar X2 camera, I was able to pick up the four main magnitude 21-22 galaxies in just 180 seconds.