I will start this small article with my premise:
“If you over-process a night sky image to the point of destruction it will have better results for the general public”
To make my point more clear I will show what I mean with a couple of images. I will start with this photo I took of the Winter Milky Way from the South Hemisphere:
This is a photo I like. It was taken from a very dark location away from light pollution and shows the beauty of the Milky Way from Carina to Scorpius. Dark nebulas as the pipe nebula in Scorpius or the Coalsack nebula in Crux are clearly visible. Other nebulas as the Lagoon (M8) and Triffid can also be identified. The image shows the Milky Way brighter than what can be seen with our eyes because the camera is more sensitive to light but there’s not a huge difference. For an observer that was in that place this photo will be a good representation of what he saw and felt at that moment.
Now let’s apply a lot of contrast, saturation and sharpening to that very same image. The result is this:
In astronomical and photographical terms the photo is now destroyed. The sky is never pitch black as the photo shows it, the Milky way is never that bright, the fine details are gone and everything is now reduced to a bright blob of light against a very dark sky. It sounds terrible and it is terrible but believe me that the general public will prefer this overdone image to the original. And I also think I didn’t overdo the image enough, more damage can be done and more “spectacular” the photo will be.
I’m not going to do the experiment of uploading both photos to photo-sharing communities because I don’t like to use my viewers as Guinea Pigs but I’m totally convinced that the overprocessed image would win the battle by a huge margin.
So why is this happening? I think it is because the public, without a knowledge of astronomy, is likely to believe in almost any image of the night sky you present as something real. They have not enough knowledge of the sky or astronomy to say the photo was totally overdone.
If you present a photo of a green cow the general public will reject it saying things as “I like the photo but not the processing”, “This is not real”, “overdone”, “photoshopped” and if they have a bad day you can get something as “this is not photography”. Been there, done that.
This happens because everybody knows cows are not green, so when they see a green cow they know the image has been manipulated and they feel the photographer tried to fool them, the result is a rejection towards the photo. If you do the same with a night sky image presenting a bright green Milky Way arching above the hills of a landscape the public will love it. They just don’t know the Milky way can’t be that bright, they just don’t know it is not green and they just don’t know what astronomical features were destroyed in the processing. So without a reason to think the photo is overdone they will just admire what they see and love the photo. The comments will be “stunning”, “I never thought the sky could be so beautiful”, “your location has some amazing skies” and so on.
Even photographers will think the photo is great because they can’t tell the degree of processing applied if they don’t know hoq the real thing is. When photographers without any familiarity with the night sky start their journey in astrophotography or night landscapes they tend to overprocess the images too. This is easy to explain as they try to produce with the photos the result they will like as vieweres.
So what happens if you are a photographer with a knowledge of astronomy? Do you try to keep your photos honest and real but with your artistic touch or do you just overprocess the photo to the point of destruction to impress the public? To be honest I have no idea of the answer to this question.
As an example Iwas asked to present some photos for an exhibit recently and I had to decide between honest photos with a low impact to the public or destroyed photos to generate some “wows” I went with the first option because I need to like my photos too but from a sales, marketing or visibility point of view that’s certainly the wrong decision.
This is a view of the Milky Way above a lake in Patagonia. I took the artistic license to make the sky a little more blue than what it really was but there’s not a huge difference from the real thing. The Magellan clouds are visible on the left and they have the brightness that matches what you would see from such a dark location. There are even some traces of airglow near the horizon, that’s the brightness of earth’s atmosphere and it can only be seen in very dark places without light pollution. You can see them as bands or streaks in a greenish color. I was there and the photo represents what I saw, and what I liked in a good way.
It’s interesting in astronomical terms and I hope it’s also a beautiful view of the night sky, but can I do it better? worst?
When I show this overdone version people say “wow” they point how bright the Milky Way and the Magellan Clouds are, they ask about the location, and viewers with good eyes signal there’s a hint of “aurora” at the horizon. I can either be happy with that or just embarrassed because nothing they say is real and the photo has been destroyed. The big Magellan cloud looks like a light tube up there, I feel terrible to even show this as an exercise but print this photo big in metal paper and you have a winner. You will see people gathered around the photo, you will see photographers that want to take a workshop with you and there’s a chance you can even win some contests with such a photo, it’s novelty, it’s unique, it’s bright, it’s destroyed.
If an astronomer sees the photo, professional or amateur he will be disgusted. But how many astronomers do you see around you now? As I say “you can’t argue with success”.
If you browse online you will find plenty of images of the Milky Way and other night sky features described as “stunning” when they are actually overprocessed shots to the point of destruction. The question is how many viewers notice that and if that is or not important to the photographer. In most cases the photographer is honest with his own processing, he just doesn’t know he is destroying the sky in the photo, he processes until he likes it. Honest photo, honest viewers, but nothing is real.
This is something that I have been thinking in the last weeks and I think it can create an interesting debate about what is the right way to go. It’s a terrible Dr Jekill and Mr Hyde feeling, I know I can make my photos more succesful if I just make them more horrible to me.
Maybe this is in some way similar to what happens with HDR. The general public loves HDRs, they are bold, bright and they look very real but many photographers don’t because they know the image is overdone to a point they don’t like it any more. So what do you do? Do you process to your likes or do you process to be succesful? Believe me you don’t want to feel that way.
If you ask me I prefer to avoid the wow factor and I hope the viewer can get interested in the night sky and learn how many beautiful things can be seen out there, the importance to fight against light pollution and that if the photo is honest there’s probably a lot to learn from it and that it can be beautiful too. If I get a “wow” from a photo that I know is not overdone then I will feel really good, the only problem is for that to happen I need to go thru many many low impact photos when I could just do a little overprocessing. The debate is now open.
To state before hand, I am also a nightscape (astro-landscape, whatever you call it) photographer. I have also put on a couple exhibitions, presentations, etc and this is my opinion on everything you just posted:
The public does like the over-processed* images. They like to see the Milky Way highlight in the image, the bright colors of the night sky, and stars to ‘twinkle’ within a still image – but not necessarily in ways that are destructive to the photo. A good majority of people have not really seen the night sky from a really dark location. They don’t know what it should and should not look like. From doing public observation nights (with telescopes) for years, when they look at a galaxy in a telescope they expect to see a hubble-quality view, not a fuzzy grey blob. I think they apply the same thought process to nightscape images. They know the Milky Way is out there and maybe they have seen one of the giga-pixel panos of the entire Milky Way that went viral on the internet months ago. Or if not, at the very least, have the thought pattern that if all these nebulae, star clusters, and galaxies have so much color in them, with an amazing resolution that allows you to see these faint details, then nightscape images should as well!
Of course, we know better. We know that images without the Milky Way can have at most 4 distinct colors – and those are colors of the stars. We know that the sky is not completely black, but rather has a small bit of color to it. We know that unless you are imaging under a full moon (and lose many stars/the Milky Way) the landscape is likely to be black. And all of this is not what our naked eye sees. When we are out in the field, even under a new moon, we still detail (faint) in the landscape. We see color all around us. We see a muted Milky Way that looks like a large fuzzy cloud in the sky and we see stars twinkling along the horizon. But our camera does not see any of this. We get a pitch black foreground, a super bright Milky Way, and thousands of stars with a simple 30 second shot. So the reality of situation is that we are trying to appease 3 separate views of the night sky: what we see, what the camera sees, and the expectation by the public.
Thankfully, especially with the advent of digital imaging and more digital image programs/algorithms out there than a single person could know about, we can find a good balance between what we see and the expectations of the public by abusing what the camera sees. First, we must accept (and I think that most/all of us do) that what the camera records is not an accurate representation of the physical world. When we talk about nightscape shots this is especially true. Example: Take an image of the full moon so the moon is properly exposed, or rather to be as close as what the naked eye sees. Right there, you will not see any stars in the field of view, where you do with your eye. So if you were to take an exposure that captures the same number of stars as you see with your eye, you would overexpose the moon and you could easily spend hours in photoshop trying to combine the two images to try an approximate what the eye saw – and more than likely you would not achieve the same results. But now, what if you were to take an image of the full moon, properly exposed, then come back and image the same part of the sky 14 days later during the new moon. Using a few tricks in photoshop and you could merge the 2 and create a much more realistic image of what you saw that night during the full moon.
So if this moon/sky combo was part of a nightscape shot, the public would (more than likely) like the 2nd version of the image much better than the first – mostly because when they look up on a night of a full moon, even from inside a medium-city, this is the view they see – a hand full of stars next to a full moon with all the crater and maria visible. But the image was made by ‘cheating’. In this hypothetical case the image was not over-processed, nothing would be over/under-exposed.
The same can be applied to your Milky Way image/argument. The public ‘knows’ the dark sky to be black. In your first Milky Way image, there is a gradual transition between the red sky to a medium-dark black. In the bottom image there is an ‘aurora’ of brilliant reds that overlap the pitch-black night sky.
Likewise, the top Milky Way has a bunch of different shades of reds, browns, and yellows in them which are produced by undefinable shapes of gases in space. But take a look at those hubble images – galaxies all have well-defined structures, nebulas have cut off points, you can see the details of where different gasses are. In these super-wide field shots like your, we these undefinable shapes blend in to each other. I look at the dust lanes leading away from Anteres and see a handful of stars on the edges and even within the dust – clearly that is not a well-defined region of space. In the bottom image though, I see the dust lanes and they are black – they have shape – just like hubble. I see plenty of structure within other dust lanes so clearly that must be what it is really like. One part of space is completely empty (or blocking the light) while right next to it is a part of space that is exciting, filled with colors and life. So the public might say the top image must just be out of focus or blurred, or something else along those lines is wrong with it.
So far, I have agree with a lot of what you said – the public would view the bottom Milky Way image as better, even though it is not what the naked eye would see, and it is not even what the camera would see. However, I don’t agree with your premise. I do not think that we need to be destructive with our images to produce a more ‘spectacular’ image. Again, we just need to abuse the camera.
If I may critique the Milky Way image, here is what I think could have been done to make the public like it more, while still maintaining some semblance of what it is “really” like: Start by shooting from a darker location (assuming the bottom is light pollution) or wait until there are not as many clouds near the horizon (assuming the bottom is thin clouds). Shoot with a longer focal length and mosaic the images together. By doing this you will increase the resolution of the image and can see more details/structure within the Milky Way. Maybe apply a small amount of HDR – not enough to be destructive or noticeable, but still enough to provide a clear separation between the background sky color and the main band of the Milky Way. Also by doing this, you can enrich the colors in the Milky Way, again, without over-saturating them. Lastly, try and control the number of stars that we see. In shots with the Milky Way it is very easy to confuse stars for noise – especially right below (in this image) the Anteres region (the northern hemisphere folks have this problem as well just on the other side of the galactic center).
Now your second set of images posted, with the bottom half of the Milky Way, Magellanic Cloud(s?) and the blue sky. I think that the blue-er image is accepted more readily by the public because that is the other, traditional, representation of the sky – a bright blue color. Think about all the illustrations/cartoons/even children’s books that feature the night sky. What is the color of the sky? Either black or deep blue. But our cameras are not as sensitive to blue light as our eyes, and again our eyes do not show us what the public expects.
In particular with your images, I find that I like the idea of the bottom (hyper-blue) image a bit better. The sky in the top image just seems too muted. However by just cranking up the saturation, you have introduced a number of artifacts that, as an photographer, I really do not like. Examples in this image of that include highlight (or creating) a harsh color gradient, defining an unnatural boundary around the Magellan Cloud(s), casting blue halos around ALL the stars, highlighting a region of the Milky Way that is uninteresting and distracting from the image, and creating an “aurora” from the light pollution in the image.
If I may once again take the liberty of describing some of the things that I think could be done to make the image more public-friendly while maintaining the astronomy/photography good-ness(?) of the image: Take 2 images, one of the landscape and one of the night sky – so you can process them separately; take a vast under-exposed image of the night sky to get the majority of the stars but not enough to get the Milky Way then take a much longer image to get more data within the Milky Way and combine these shots; take an overexposed image to bring color to the night sky, or even better, take a shot at the beginning of astronomical twilight and use that image as the basis of color for the sky; take an image of the Magellan Cloud(s) with a longer focal length and overlay that in the correct position to get more detail within the cloud – also this will allow you to tweak the color/contrast of the object separate from the sky.
Now all of these changes (in general) may require a lot of time in Photoshop (or some other image software program) but by carefully controlling the dynamic range you can get the added detailed, the enhanced color, the sharper image that the public would expect out of most other types of photography without destroying your image (while maintain most/all of the scientific accuracy).
All of this said, let me just say that I do love your work. You have done many very creative nightscape shots and have been the inspiration for several of my images!