Sh2-105 | C27 - Crescent Nebula
The Crescent nebula is an emission nebula in the constellation Cygnus. It is also known as NGC6888, Caldwell 27 and Sharpless 105. In the center is a star that turned into a red giant about 250,000 to 400,000 years ago and started to eject its outer gas shells. The remaining star converted into a Wolf-Rayet star (WR136) and is emitting a highly energetic and fast stellar wind. This fast stellar wind is colliding with the earlier emitted gas, which is moving much slower. The result is a highly energising gas cloud with the characteristic brain-like shape of the nebula. The distance of the crescent nebula to Earth is approximately 5000 light-years.
Wolf-Rayet stars are a rare set of heterogeneous stars and typically extremely hot with surface temperatures as high as 210,000K. They have evolved from very large stars that have lost their outer hydrogen layers and are now using Helium and even heavier elements as their fuel. In their spectra, often emission lines of ionized Helium, Nitrogen and Carbon can be found. Because of their massive amount of energy being emitted, they can make for some impressive nebulae, such as the case with NGC6888.
Planning
Object
Visual Magnitude: 7.4
Apparent size: 20 x 10 arcmin
R.A.: 20h 13m 10.53s
Dec.: 38º 28’ 26.5”
Conditions
Astr. night: 20:37
Astr. dawn: 06:01
Moon: 20%
Moon set: 21.12
Humidity: 62-84%
Pressure: 1015-1050 hPa
The best time to photograph the Crescent nebula is in late summer or autumn. It reaches quite high altitudes of well in the 80º’s. The images presented here were acquired during three nights, on October 10, 13 and 15 2018. The Crescent Nebula is a typical narrow-band image with the teal-coloured veil visible in the OIII spectral line. In this case, the narrow-band images were combined with broad-band images to create proper star colours. During the observation sessions the moon was mostly absent.
Capturing
The Crescent nebula emits extensively in the H-alpha and OIII wavelengths. The almost brain-like structure is more heavy on the H-alpha side. The most challenging part for capturing and processing is to let the veil-like structure, which is very rich in OIII emission, come out extending from the main part of the nebula. There is very little signal in the SII band. This makes the Crescent nebula not a great target for a Hubble palette. Instead, only H-alpha and OIII were captured to be mapped to Red and Blue respectively. Usually a blend is then mapped to the Green channel and this is where personal taste comes in, but that will be discussed under processing.
The nebula itself is relatively small compared to the total field of view. It is embedded in a region of rich H-alpha signal, making for an overall aesthetically pleasing picture. Looking at it with a photographers eye, the nebula itself looks like it is pointing downwards. Rotating the nebula so that it would point top-right and placing it a little bit off-center would probably have enhanced the overall impression.
This was the first narrow-band image captured, so exposures were still a bit trial and error. To compensate for the reduced signal coming from a narrow-band filter, it was decided to capture 300s exposures at a gain of 300. Since that teal-coloured veil is so difficult to capture, a total of more than 11h of narrow-band exposures were captured, a true multi-session project. In this palette stars would loose quite a bit of colour, so to bring that in, additional RGB data was collected. To keep things simple, these were captured at a gain of 300 as well, but then 120s exposures. In total the exposure was almost 14h. At the time this was quite unique, but worked well.
Technical details
Telescope
Mount
Camera
Sensor Temp.
Takahashi TOA-130 + 35 flattener
10Micron GM1000HPS
ZWO ASI1600MM Pro
-20ºC
Exposures
Red
Green
Blue
H-alpha (5nm)
OIII (5nm)
Total Exposure
17 x 120s @ Gain 300/50
18 x 120s @ Gain 300/50
20 x 120s @ Gain 300/50
74 x 300s @ Gain 300/50
70 x 300s @ Gain 300/50
13.8h
Processing
This was the first narrow-band image processed and still very early in the use of PixInsight altogether. Some step-by-step guidance was required. This was found in the book The Astrophotography Manual, a practical and scientific approach to deep sky imaging, written by Chris Woodhouse. This is really an excellent book for anyone interested in astrophotography. It covers topics around equipment, sensors and how light interacts with them, setup and control, etc. A separate section on image calibration and some fundamental tutorials on PixInsight complete the theoretical part. The second half of the book describes practical examples around a dozen or so common targets. How to image them and how to process those images. Particularly useful is that throughout the book, the theory behind why things work the way they do is described, without loosing the practical application out of sight. The Crescent nebula is one of the practical examples in the book and processing of the image above was done by following the book step by step.
All images were calibrated using bias (100), dark (50) and flat (25) frames and registered and stacked using the Batch Preprocessing script. From there, three lines of processing were followed. First a colour image focusing on the nebula was created using the H-alpha and OIII stacks. Secondly, a synthetic luminance was constructed from H-alpha and OIII stacks, to be used for brightening/sharpening. Third and last, a colour image was created out of the Red, Green and Blue stacks, focusing on the star colour.
For the narrowband image, both stacks were background corrected and then the OIII stack was fitted to the H-alpha stack. Both channels were combined to a colour image, using a so-called HOO palette. This suggests that H-alpha is mapped to Red, and OIII to both Green and Blue. But sometimes both H-alpha and OIII are blended together into the Green channel. This can be done in a 50/50 blend, but Chris Woodhouse had chosen for a 30% H-alpha and 70% OIII blend. This was what was applied. A background neutralisation was applied followed by color calibration and noise reduction using MultiscaleLinearTransform. Then it was stretched to the non-linear state.
The synthetic luminance was a bit of a special case. Both the H-alpha and the OIII channels represented distinctly different areas in the image. Luminance should enhance both of these channels in equal amounts. Chris Woodhouse came up with a smart solution. He found a way to construct an image that takes its data from both the H-alpha and OIII channel by selecting for each pixel the brightest of the two. This was done using PixelMath with the equation: iif(H-alpha>OIII,H-alpha,OIII). The resulting image was then subjected to deconvolution, noise reduction and stretching. After stretching the image to its non-linear state, details were sharpened using the MultiscaleMedianTransform process.
The star colour image was created by combining the Red, Green and Blue image stacks. A background neutralisation and color calibration was applied in the linear state. Then the image was stretched to the non-linear state, some MLT noise reduction was applied and the colours were saturated.
Now it was time to add the three images together. The synthetic luminance was combined with the narrowband image using LRGB combination. This really helped to let the teal-coloured veil to come out. Further enhancing detail with MMT and noise reduction of the high frequency noise with TGVDenoise completed the nebula image. The same synthetic luminance was also added to the RGB image. This had the added benefit that luminance of the HOO image and the RGB image were the same and could be easily combined without distorting the brightness. Using a star mask on the HOO image, the RGB image was added using PixelMath. To enhance the red nebulosity in the image, the red channel was enhanced using the CurvesTransformation.
As a first narrow-band image, the results were satisfying, with the teal veil coming out nicely and the stars having good colour. However the H-alpha part of the image had become rather pale/pink and also the red colour of the background nebulosity had become a bit pale. This is one of those images one might want to re-do again after a while, to see if a better image could come out.