#329. Every Story Tells a Picture : Cloudy Orion (OTUS-style)

By pascaljappy | How-To

Feb 26

While I recently proclaimed my love of clouds in photographs, astronomers see things very differently. Clouds are bad for business, not just for ice-cream vans. They mask tiny details and wash away faint light sources.

For us photographers, it’s a different story. And clouds can actually add something interesting to photographs of stars …

The constellation of Orion half-covered in clouds (Sony A7r & Zeiss OTUS 85/1.4)

The constellation of Orion half-covered in clouds (Sony A7r & Zeiss OTUS 85/1.4)

Above is the lower half of the constellation of Orion. And a rapidly approaching layer of stratus that covered the sky in 10 minutes and didn’t take much longer to clear away after that.

As you can see, it adds two important qualities to the photograph:

  • A yellow – blue colour and luminosity contrast between the left and right of the frame
  • A smearing of the main stars that makes them glow and draw the constellation more clearly.

For the astro-minded viewer, the fuzzy patch below the 3 aligned stars (Orion’s belt) is the M42/M43 nebula known as the Orion Nebula. Long exposures with much longer focal lengths reveal a very beautiful and complex system of filaments and stars. See here for a stunning 12 hours exposure with a 1140mm f/7.5 APO telescope !

Smaller and fainter, but equally interesting is the Flame Nebula (aka NGC 2024 or Sh2-277) that can be seen immediately to the top left of Alnitak (ζ Ori), the left-most star in the belt. Click the image to enlarge, for a better view, and see below in a cloud-free photograph. Nearby Alnitak rips the gas of its electrons by beaming powerful ultraviolet onto it and the recombination of these into atoms is what makes the nebula glow. It’s a giant neon light. Longer exposures (stay tuned) reveal the famous but elusive Horsehead Nebula (NGC 2023) immediately South of (below) Alnitak.

Both nebulae are part of a huge star-forming complex called the Orion Molecular Cloud Complex, which I hope to talk about in more detail soon.

 

Lens, Position & Framing

The best lenses for astrophotography are those that can be used at wide apertures with no aberrations.

The focal length is chosen depending whether you want to take in a vast expanse of the sky (such as the Milky Way) or focus more tightly on constellations, or even on individual objects, using long-focus telescopes. With the wider lenses, it’s a good idea to incorporate foreground as the pattern-less sky might not hold visual interest for long in the final image.

Here, I am using the fab Zeiss OTUS 85/1.4 wide open. The field of view of a  50mm lens  would be better to photograph the whole constellation, but 85mm is fine for the bottom part where the clouds are gathered.

Framing depends on what’s there to be seen. On a cloudless day, I would use a vertical frame to encompass to complete constellation. In my future experiments with longer exposures, I will try to capture what is called the Barnard loop, requiring a shorter focal length or some stitching with the OTUS 85.

DSC01251

10 second exposure, without clouds. Click to enlarge.

Exposure time is particularly important for good results. Exposure is a compromise between several factors.

The longer the exposure, the fainter the stars and celestial objects you will be able to detect, the higher the noise and the more elongated stars will appear (leading to star trails in really long exposures).

A famous rule, called the 600 rule, recommends that you divide 600 by your focal length to obtain the longest admissible exposure. It’s no use. It was devised in the film era when 30-micron circles of confusion were the norm and is totally overoptimistic for today’s sensors. Besides, it doesn’t apply equally to all cameras given their varying pixel pitch.

The photo above is a 10 second exposures, which is not that far from the 600/85 (7 seconds) but way too long (click to enlarge and see for yourself).

For the mathematically minded, here’s how to calculate the proper times for your setup. The Earth rotates at roughly 15 arsec/s. Multiply by your focal length and you exposure time, then divide by 206 (more or less the tangent of one arc-sec in microns). 15 x 85 x 10 / 206 = 62. That’s 62 microns, 12 or pixels. At least 3 times what I would recommend for sharp images.

For my setup I’d consider 2.5 s the best time, but the first image on this page is pushed to 4 seconds (I think). As a rule of thumb I’d use a 300/focal length time for modern sensors with 5-6 micron pixels and even 200/fl for Micro 4/3.

That’s if your camera is not guided to track the Earth’s rotation, of course. More on that in a few weeks.

 

Post processing “Cloudy Orion”

Post processing this image had two goals:

  • Correct the horrible colour cast and liven up the “print”
  • Find a pleasing combination of contrast and saturation. This is not about scientific rigueur, but aesthetics.

White balance is a notoriously tricky area for astrophotography. Serious amateurs and pros use monochromatic sensors and a set of calibrated RGB or MCYK or narrowband filters with corresponding exposure ratios to achieve repeatable balance. Hubble colours are obtained through one such process called the Hubble Palette.

We, lowly togs, have to make do with white balance + temp & tint correction in post-processing (unless you own a Leica M Monochrom or Phase IQ260 Achromatic, in which case I’m too jealous to talk to you, go away).

Unfortunately, we cannot rely on the Daytime WB setting (5600K) because, at night, the light pollution in the atmosphere has a much stronger impact on colour than during the daytime. The more light-pollution in your area, the colder you will have to set the white balance (around 2800K). In my suburban conditions, 3300K is a good starting point. And in excellent observing sites, 4000K is possible (or more).

After this initial setting, here’s an efficient solution for fine tuning.

(1°) Go wild with saturation. In LightRoom, set Saturation and Vibrance to +100. Do the same with individual colour channels. This leads to this very subtle image 🙂

An over-saturated view of the constellation of Orion in post processing. Sony A7r &OTUS 85

Warhol was here

You’ll notice the very weird square boxes around individual stars, in Sony’s Bionzy world.

(2°) Then zoom to 100% in a dark area in the frame. Adjust the Temperature slider in your post-processing software up to a point where all stars show colour. Too cold and all will be blue. To warm and all will be orange. You want individual colours to show clearly. Repeat the procedure with the Red/Green Tint slider. You may have to repeat this a few times but it shouldn’t take more than 10 seconds.

(3°) Set all saturation and vibrance sliders back to their original neutral point.

The photograph now seems dull and flat, but it is well-balanced for the conditions. Quite a let down after the previous pop-art jewel, right ? 😉

Note that if you don’t have clouds handy to smear bright stars into colourful fuzzy balls, you can use older lenses with spherical aberration to the same effect, you a take second out-of-focus shot and stack or you can use dedicated plugins for LightRoom and Photoshop.

DSC06618-3

Ugly Betty Orion

From here on, it’s just a matter of seasoning to taste. There are many possible interpretations from this stage. The one chose at the top of this page is just one of them. Here are a few more.

DSC06618-4 DSC06618-Modifier DSC06618-Modifier-2 DSC06618-Modifier-3

Real astronomers would probably disapprove strongly of these methods. As you can see, some salvage info, others bury it. But this is not science and when handed lemons, make lemonade. I’m just saying clouds shouldn’t stop you making interesting pictures of the heavens.

Oh, that and – again – that the Zeiss OTUS 85/1.4 is an absolute star ! Astronomy clubs should definitely consider buying one for wide-field work ! I’m sure imaging the Flame Nebula in 4 seconds is quite the selling proposition …

Have fun 🙂

 


Email: subscribed: 4
  • Jens says:

    Hey Pascal, you’ve given us exposure times, aperture and WB settings – how about a clue as to the ISO setting – or am I missing something here? Thanks, Jens

    • pascaljappy says:

      Ahem, so sorry Jens 😉 😉 😉 I think it was 12800. The really great thing about the A7r is that it gets a bit noisy at those ISO settings, but only in luminance. There are very little chrominance issues at all. So the files are ever so easy to clean up (at the expense of very fine detail, I suppose). ISO 12800 @f/1.4 is really cool 🙂

  • Jens K says:

    Thanks, Pascal. I’ve migrated to the A7S (because I also do some video), so high ISOs shouldn’t be too much of a problem. I’ll give it a try asap.

    • pascaljappy says:

      Wow ! With the A7s, you should be able to double that and get even better pictures (shorter times, shorter trails). Please send some our way, if you like your pictures !

  • >