I have started a bit of astrophotography, an important part of which is the editing of the image to bring out stars etc not even visible after a ~40 second exposure. To do this the image must retain as much data as possible - what are the best kinds of setting/procedures to achieve this?
Light pollution is the biggest challenge when it comes to Astrophotography, optimum exposure and retaining as much detail and information possible. Therefore, my first advise, that is only if you are not already doing this, is to seek out a dark sky. There are apps and websites that will help you locate areas closest to you.
This alone will help increase detail quite considerably.
The next step, taking into account that the Camera is set to RAW and turned to manual mode, I would suggest, is to use a exposure method called ETTR or Exposure To The Right. This method uses overexposure to get the highest dynamic range by capturing the maximum amount of light. When you view this data via the Histogram, the graph will show the data pushed to the right. in other words, looking from the left, it will be fairly flat for the first 50/60% of the Histogram and there will be a sudden elevation which will reach the top of the chart with 10/20% of the histogram left and continue off the chart to the right.
It is here, in the latter half of the histogram where the data resides that you need to bring to fruition in Lightroom.
An Example of an ETTR Exposure will be 30 Seconds shutter with an aperture of f/2.8 and ISO of 3200/6400 -
The high ISO is a key to the extra detail and will make a significant difference to the final image which will have great detail, show less luminance noise, less colour noise and appear cleaner and smoother than if taken on a slower ISO of say 1600 or 800 with a faster shutter speed of say 15 seconds and appears to be more pleasing straight from the camera.
The reason for this is due to it having a higher signal to noise ratio, and even though the initial RAW file looks completely overexposed and unusable, it has in fact captured more usable Data that can now be used in Lightroom to bring out the details in the stars and the gases of the Milky way.
Such an image when first viewed in Lightroom will look over exposed and very washed out, even if has been taken from a very dark sky and away from light pollution.
The next stage is to work in Lightroom to bring out the details without blowing out any highlights and keeping the sky as dark as possible reducing the original Orange tint from the ETTR exposure.
My personal work flow is to bring down the exposure first whilst increasing contrast, but ensuring that there is a balance between the stars, Planets, gases, Milky way and the dark sky. The next stage is to alter the White Balance to get rid of the orange cast.
If there is any light pollution from the ground, say a nearby city, then I use a Gradient filter to correct the light spillage.
To summaries, ETTR will capture greater dynamic range, detail and image data with a higher signal to noise ratio than would a normal exposure. You can then make optimal use of this data in Lightroom to bring down the exposure, increasing contrast and reducing any orange cast to bring out the desired detailed effect
The first and most important thing is to shoot in RAW.
Next, you need to make sure that you are not losing anything in the RAW conversion process. This could happen in two ways. First, low values could be clipped at 0 by a "black level" setting. The second and much more subtle is that faint detail could be lost to noise reduction. I'm not talking about the NR applied by an explicit setting in the RAW converter, but that applied by the demosaicing algorithm itself. Yes, a RAW conversion algorithm can apply noise reduction. For this reason I would favor an open source RAW converter such as RawTherapee, or at least a RAW converter that uses an open source demosaicing algorithm.
There are tons of sites that cover astrophotography out there. One that I would recommend (while not solely about astrophotography, it has tons of other interesting articles) is www.clarkvision.com .