Sh2-184 - Pacman Nebula
NGC281 - Pacman Nebula. Click here for full resolution image.
The Pacman Nebula is a bright emission nebula in the constellation of Cassiopeia. The nebula is registered in several catalogues. In the New General Catalogue it is classified as NGC281, in the Sharpless catalogue it is known as Sh2-184 and in the Index Catalogue of Nebulae, the code is IC-11. But the resemblance with the videogame character from the 1980’s makes the name Pacman Nebula the one remembered by most.
The nebula is part of a large HII region, with in the center the open star cluster IC1590. A multiple star, B1 and several dense dark clouds known as Bok globules are some of the other highlights in this region. The region is located in the Perseus Spiral arm of the Milky Way, about 4,100 lightyears away from earth.
Planning
Object
Visual Magnitude: 7.4
Apparent size: 35 x 30 arcmin
R.A.: 00h 53m 59.65s
Dec.: 56º 42’ 34.3”
Conditions
Astr. night: 18:56
Astr. dawn: 05:37
Moon: 20%
Moon set: 15:44
Humidity: 84-96%
Pressure: 1010-1032 hPa
The Pacman nebula is a circumpolar object, with highest altitudes during autumn. The images were taken during many different sessions, the first one on October 10, 2018. Additional images were taken on October 13, November 03, 15, 16 and 19 and January 19. 2019. The conditions mentioned above are from November 03, 2018.
Capturing
The Pacman nebula makes a very good narrowband target, with strong emissions in the H-alpha spectrum. It was decided to plan for a Hubble palette processing, mapping SII to Red, H-alpha to Green and OIII to Blue. With narrowband imaging using only the narrow-band spectral lines, often the star colours of the surrounding stars are difficult to get right. Therefore a separate small set of RGB-data was captured, purely to give the stars a bit of extra colour.
Quite some setbacks were experienced during capture in October and November 2018. On one night the temperature had dropped from 14 ºC to 4 ºC overnight. No temperature-dependent focusing routine was setup, so images of the second part of the night were noticeable softer than from early in the evening. Also on several nights there was high humidity in the air, leading to parts of the night being unsuitable for imaging. And the third major setback was dew settling in on the lens which turned out very difficult to remove. All of this was clearly a learning experience! To compensate for the many rejected frames, an extra 3h of exposures was taken in January 2019, bringing the total exposure time to 16h.
With a diameter of just over half a degree, the nebula nicely fills the frame of the TOA-130 + ASI1600 combination. At the time not a whole lot attention was spent on rotating the frame properly, other than making sure the rotation was the same across nights. It would probably have been nicer if the framing would have rotated by 90 degrees anti-clockwise. It would require too much of a crop to change this rotation in post-processing.
All images were taken at gain 139 (unity gain) and exposures were 300s for the narrowband images and 60s for the broadband images. The exposure of the broadband images was quite short on purpose. Because the main focus was on the colour of the stars, most star colour would be retained if the stars would not be too bright.
Technical details
Telescope
Mount
Camera
Sensor Temp.
Takahashi TOA-130 + 35 flattener
10Micron GM1000HPS
ZWO ASI1600MM Pro
-25ºC
Exposure
H-alpha (5nm)
SII (5nm)
OIII (5nm)
Red
Green
Blue
Total Exposure
62 x 300s @ Gain 139/21
59 x 300s @ Gain 139/21
53 x 300s @ Gain 139/21
30 x 60s @ Gain 139/21
30 x 60s @ Gain 139/21
30 x 60s @ Gain 139/21
16h
Processing
Before the final set of images could be processed, a thorough assessment of the frames had to be done. Based on visual judgement using the Blink process in PixInsight, the ones that were either not sharp, or had too much problems with fog or dew were discarded. In total 45 frames of 300s were eliminated this way, which is almost 4h of exposure. In reality even more frames had been lost. They were so bad that they did not even make it to the evaluation stage.
The final set of images was calibrated with Bias (100), Dark (50) and Flat (25) frames, registered and stacked using the BatchPreprocessing script. Processing started with the narrow-band images.
To bring some balance in the exposure of the three narrow-band filters, a LinearFit was applied to OIII and SII with H-alpha as the reference. Especially the OIII channel with a mean ADU value of 126 was a lot brighter than H-alpha with 80. SII was very similar to H-alpha but was still fitted to be consistent. Then each of the three channels was subjected to noise reduction with the MureDenoise script. This really had a significant effect and resulted in very clean images. The three images were then combined to an RGB image using PixelMath, mapping SII to Red, H-alpha to Green and OIII to Blue. The resulting image had a strong magenta halo around the nebula, which could be largely reduced using DynamicBackgroundExtraction. The image was then stretched into its non-linear stage. Next up was the colouring within the Hubble palette. Cyan, Magenta, Yellow and Green masks were created using the ColorMask script and a blur of 3. Using a process described in detail for the Rosette Nebula, the colours were adjusted with alternating use of the various masks. The resulting image was then subjected to some further noise reduction, using MultiscaleLinearTransform on layer 1 (3,0.33,3), layer 2 (3,0.33,3), layer 3 (1,0.33,3) and layer 4 (0.5,0.33,3). The DarkStructureEnhance brought out the details of the Bok globules a little bit more. Stars were shrunken with MorphologicalTransformation and a sharpening step was applied using the Detail Layer section of MultiscaleLinearTransform on layer 1 (0.05), layer 2 (0.05) and layer 3 (0.025). This completed the processing of the SHO-image.
Next was processing of the RGB image. The goal here was to only add some colour to the stars, so not a whole lot of processing had to be done. No noise reduction was applied. Only a DynamicBackgroundExtraction was applied on the individual channels before they were combined. Then a Background Neutralization and Color Calibration step was applied before the image was stretched. Since the purpose was to add some colour to the stars, colours were boosted using the ColorSaturation process. To make sure that the colours would eventually blend in well with the SHO image, a synthetic luminance was created from the SHO image and this was added to the RGB image using LRGB combination.
The actual replacement of the colours of the stars was quite simply done by selecting only the stars in the SHO image using a star mask, and then apply the RGB image onto the SHO image using PixelMath. The resulting effect was subtle, but definitely noticeable, as can be seen in the above images.
After processing in PixInsight, some final touches were applied in Affinity Photo. These include lifting the black-point to enhance the dark edges of the image, an overall unsharp mask to sharpen the image, and a vibrance increase to enhance the colours and contrasts.
This image has been published on Astrobin.
