What caused my astrophotography images to come out hazy?

Question

I attempted a bit of astrophotography tonight and I was surprised with the outcome. I didn't expect to see so many stars in the images! I loved it!

However, when I went to look at the pictures, they appear hazy. They don't have that crisp clarity that comes with really beautiful night shots.

I do all of my photography in full manual mode. I shoot with a Nikon D3300 and the lens I used was the Nikkor AF-S 18-55mm DX VR with no hood (probably need a hood for it).

This image was taken at f/3.5 18mm at ISO 6400 for 20 seconds.

This image was taken at f/5.6 55mm at ISO 6400 for 20 seconds.

This image was taken at f/5.6 55mm at ISO 6400 for 20 seconds.

The images are saved in JPG format (I didn't think about it at the time but I won't make that mistake when I go back and do it again--I'll shoot RAW).

What caused them to appear hazy? These are out in the country in rural Alabama. I'm sure I can go further out but I'm not near any metropolitan areas, so, I don't think it was ambient light. The moon wasn't out, either. There was a street lamp but I was in a shadow and pointed completely away from it.

From looking online, I saw a ton of tips on how to shoot at night but nothing really addressing haziness. So, what did I do wrong? What caused this and how do I avoid it happening again?

---

I do want to add that I don't process my images. I don't edit them. I don't particularly fancy tools like that because I feel that using them as I am learning how to improve my skills as a photographer (hobby, not career) may hold me back or possibly act as a crutch. I refuse to use anything like that unless I absolutely have to.

Answer 1

These look like perfectly normal shots to me. You have some star streaking going on, and some light pollution, but nothing out of the ordinary.

Most, if not all, astro shots require heavy processing. Here's one of mine showing a wide field around Andromeda. This was produced by taking 35 shots of about 15 seconds exposure each (plus in-camera noise reduction), doing basic corrections for things like white balance in Lightroom, then aligning and performing LRGB processing in Photoshop to reduce the noise and pull out all the detail. That's basic stuff in astro.

Now, you don't have to use that kind of thing if you don't want to, but don't expect to get results like you see online. This doesn't just apply to astrophotography but all genres. Post processing has been part of the the art form since its earliest days. Are you into photography purely to learn the camera in as much detail as possible? Or to make great images? If it's the latter, then you need to be using post-processing. Everyone else is.

Answer 2

What caused my astrophotography images to come out hazy?

• A hazy sky. Light pollution combined with the air, and everything else blowing around in it (water vapor, dust, other particulate matter), between you and the stars.
• A single exposure. Most "wow" astrophotography photos are the result of combining multiple exposures of the same field of view.
• A stationary camera. The stars moved during your exposure. The camera/lens didn't (at least not in the right direction). When you changed the focal length from 18mm to 55mm you tripled how far across the frame each star moves in the same amount of time. To counteract that you must reduce the exposure time by the same ratio.
• Chrominance noise. The result of low signal-to noise ratio. You can overcome this by using multiple frames to cancel out the random noise. You use long exposures to reduce the influence of read noise by increasing the signal. More signal with same read noise means a higher signal-to-noise ratio (SNR). To use long exposures you need to be tracking the sky. To use multiple frames without loosing the edges of photos where part of the sky has moved out on one side and another part has moved in on the other side of successive frames you need to be tracking the sky.
• The limits of your lens. If you were tracking the sky this would be more in play. As it is the movement of the sky caused most of the blur in two of the three photos.
• The limits of your ability to accurately focus your lens on very small points of light.
• Using the default in-camera JPEG engine to process the raw data from the sensor into a viewable image. The in-camera defaults are designed to work best with typical well lit daylight scenes of things that reflect well diffused light and are very much larger than single points emitting light. They are not designed to process very small and dim points of light in a hazy sky.

Other than the sun, the largest stars in the sky are only about 50 milliarcseconds (mas) wide as seen from Earth. Typical stars visible to the naked eye are less than 2 mas wide. A milliarcsecond is 1/1000 of an arcsecond, which is 1/60 of an arcminute, which is 1/60 of an angular degree. A milliarcsecond is 1/3600000 of a degree! So most stars are less than one two-millionth of a degree wide! The sun is roughly one-half degree, or 1.8 million mas in width. This makes the sun roughly one million times as wide as a typical star visible to the naked eye.

I do want to add that I don't process my images. I don't edit them. I don't particularly fancy tools like that because I feel that using them as I am learning how to improve my skills as a photographer (hobby, not career) may hold me back or possibly act as a crutch. I refuse to use anything like that unless I absolutely have to.

That's about like if Ansel Adams had refused to do anything other than straight contact prints in the darkroom! None of his masterpieces would exist in anything resembling their current form had he took such a position. In the film era there was plenty of "processing" going on in the darkroom. Much of what we do on our computers in the digital age has an analog counterpart in the wet darkroom. This is especially the case when we are working with images that stretch the technical capabilities of our cameras and display devices to the absolute limit.

Adams' adjustments of exposure times and development times was to allow him to capture a scene with a wider dynamic range than the limits of the film emulsions available to him as they were conventionally used. The dodging and burning that Adams raised to an art form in his darkroom was to allow him to create a print of a scene that had a wider dynamic range than the limits of the photographic papers available to him.

It is true that it is always best to get as close as we can in camera to what we want our final image to look like. There are technical reason why this is so. But it is equally as true that the best photographers often use that well shot image as their starting point rather than as their final product.

Not everything done in post-processing is to correct a mistake made when the image is captured. The modern digital environment does give us much more leeway as far as what errors can be corrected and the degree to which we can miss and still salvage a usable image. But that isn't the primary purpose of shooting raw and individually customizing for each image the parameters with which that raw data is transformed into a viewable image. Rather, much of digital postprocessing is a way to enhance and expand the capabilities of our cameras by adjusting the subtle differences between one pixel and the next detected by the camera in a way that emphasizes some of those differences between adjoining pixels and makes those differences more explicit while at the same time minimizing other differences between one pixel and the next. Nowhere is this more true than in astrophotography.

Answer 3

Atmospheric phenomena can enhance light pollution. Even if you are not right next to a city, you can still suffer from their light pollution if you point in the right direction. I believe your first image is definitely suffering from some LP, probably exacerbated by an inversion layer or other haze phenomena.

I often image at a dark site that is 35 minutes out of town, deep in a rural area with very dark skies at 21.3mag/sq". However, to the west is the greater Denver metropolitan area and all the surrounding cities. The LP bubble from this highly populated area stretches nearly 40 degrees into the sky, and affects up to 60 degrees, above the western horizon. As such, I only image to the east, north, and sightly south. To the south there are additional LP bubbles, albeit lower towards the horizon (most not topping 30 degrees).

As for your other two images. Astrophotography is an extremely low-signal form of photography. The vast majority of the pixels in any given frame will be barely above the read noise floor, and would be considered "black" pixels in most normal photos. Only stars will reach much above the noise floor, and even then, not many will be much more than 50% of the signal range of the sensor.

For this reason, in order to get decent astro photos, it is necessary to track unless one is going solely for ultra wide field milky way images. Even then, some level of tracking is usually necessary to get quality results. For most astro photos, total integrated exposure time in the realm of a couple hours to many tens of hours is usually necessary, depending on how much noise the camera has, how dark your skies are, how transparent the sky is, and exactly how faint the details are.

You can take a look at my astrophotography on AstroBin for some examples of how much total integration time it takes to get various kinds of objects:

http://www.astrobin.com/users/jrista/

Most of my work is with a Canon 5D III with a large f/4 scope from a dark site. However I also have some images with a new mono CMOS camera purchased this year, using narrow band filters, as well as some older work from my light polluted back yard with DSLRs. You can get an idea of how much total exposure time is required...and it is often in the 8-10 hour range for DSLR RGB, 2-4 hours for bright nebula w/ narrow band filters, and 15-30 hours for faint nebula with narrow band filters.

Answer 4

Your pictures can be improved with the method I explained here:

What can be used to "fight" light pollution in astrophotography?