M35 - Open Cluster
The Open Cluster M35 and nearby NGC2158 - Click here for full resolution image
M35, also known as NGC2168, is an Open Cluster, located in the constellation Gemini. It is a loose and irregular cluster of several hundred stars, only 120 brighter than magnitude 13. Thanks to its close proximity to earth of only 2800 lightyears, its apparent size is one of the largest star clusters visible in our skies. The bright blue stars are quite young at 175 million years.
A close neighbour of M35 is NGC2158, a much smaller open cluster. The reason it is much smaller is because it is much further away than M35. Approximately 16,500 lightyears are between this 2 billion years old cluster and our sun. The old red stars make a beautiful contrast with the bright blue stars of M35, creating some depth in the universe.
This is the first broadband image that was captured using the newly acquired ASI6200 camera. For a more detailed technical description of the first light experience with this camera, see this blog post.
Planning
Object
Visual Magnitude: 5/09
Apparent size: 40 x 40 arcmin
R.A.: 6h 10m 10.58s
Dec: +24º 19’ 45.1”
Conditions
Astr. night: 19:37
Astr. dawn: 05:58
Moon: 1%
Moon set: 18:12
Temperature: -8 ºC
Humidity: 84%
Pressure: 1041 hPa
M35 is a typical winter object, with best observation period from December through March. At the backyard observatory, during time of observation, there was clear visibility of M35 from the start of the night until just after midnight. Then it disappeared behind a tree. Images were taken on two consecutive days on 12 and 13 February 2021, with the equipment remaining in place between both nights. The new moon created perfect conditions for broadband imaging. Temperatures were freezing at -8 ºC, but resulted in a fairly clear sky, somewhat better on the first night than on the second night.
The two objects are 18th on the list of the ‘Two in the View’ observing program of the Astronomical League. This is a list of well known objects that have a second, or even third object easily in the same field of view of regular telescopes. The local astronomical society has adopted this program as well and many of the objects have been observed and/or photographed already.
Capturing
The image is captured using RGB broadband filters as well as a luminance filter. The challenge with star clusters is to keep clear separation between individual stars. Therefore the exposure should be short enough to not over-saturate individual stars, while long enough to record even the faintest of stars in the cluster.
The ASI6200 has a dual-gain sensor, meaning that at gain 100, the dynamic range is almost the same as at gain 0. However, the full well depth at gain 0 is significantly larger. Therefore, for this image, the camera was used at gain 0, to allow bright stars to fill the wells with photons without over-saturating the signal. This comes at the cost of some extra noise, but for this image that is a good trade-off to make. A star cluster can typically be photographed with shorter total exposures than faint nebulas. In this case a total of 4.2h of exposures were acquired.
The object(s) were significantly smaller than the field of view of the telescope/camera combination. Exact framing was not critical. NGC2158 was placed in a diagonal line relative to M35, to make for an aesthetically pleasing image.
Technical details
Telescope
Mount
Camera
Sensor Temp.
Takahashi TOA-130 + FL67 flattener
10Micron GM1000HPS
ZWO ASI6200MM Pro
-15 ºC
Exposures
Luminance
Red
Green
Blue
Total Exposure
63 x 120s @ Gain 0/21
14 x 180s @ Gain 0/21
14 x 180s @ Gain 0/21
14 x 180s @ Gain 0/21
4.2h
Processing
All frames were calibrated with Bias (100), Dark (50) and Flat (25) frames, registered and stacked using the WeighedBatchPreprocessing script. As a start, all four image channels were subjected to noise reduction using the MMT-method, which cleared up the images significantly.
From here, the luminance channel was treated separate from the colour channels and combined all the way at the end again. The stacked luminance image looked quite clean and did not require any background extraction. A careful deconvolution was applied. The background was protected using a range-mask, a PSF image was created using the PSFImage script, and the larger stars got separate protection using a star-mask. A total of 20 iterations with local deringing of 0.027 gave good results, clearly contracting the stars to their peak center-values. The image was stretched carefully not to over-saturate the stars. Rather than rely on an auto-stretch, this was achieved by manually stretching in the ScreenTransferTool while zoomed in on some of the bigger stars. Using the ScreenTransferTool in manual mode is a nice alternative to using HistogramTransformation in preview mode. Preview modes in most tools have a low resolution and can give insufficient feedback for precise manipulation. The last bit of stretching was achieved by applying contrast with the CurvesTransformation tool using an S-shaped curve. M35 is located in an extremely densely populated part of the sky, so there are many, many stars in the image. To keep the focus on the cluster, an overall shrinking of stars was applied using MorphologicalTransformation (Erosion, amount 0.36, size 7). In an attempt to further enhance the image, an UnsharpMask was tried. But this undid the star shrinking to an extent, and gave a too contrasty image. So this was undone. The luminance was now ready.
The colour images had quite a dis-balance in intensity, with the green channel having much higher mean brightness (220ADU) than either Red (151 ADU) or Blue (152 ADU). A linear fit was applied to both Green and Blue with Red as the reference. The three channels could now be combined to an RGB image. A diagonal background-gradient was evident, from red/magenta top-left to green bottom-right. A gradient this straightforward can easily be removed using a simple AutomatedBackgroundExtraction, with function degree of 2. BackgroundNeutralization was tried but abandoned. It did not improve the image much and introduced some background colour-blotches. Colours were calibrated using the PhotometricColorCalibration tool. This gave very good results. The image was now ready to be stretched in the same careful was as was done with the Luminance image. Enhancements were made using the CurvesTransformation tool, this time including the addition of some colour saturation and a small increase of the c-component. This c-component usually warms up the image and has a pleasant, but strong effect, so should be applied with care. There was a little bit chromatic noise still present in the image. This was reduced by a small convolution (StDev 1.2).
The Luminance channel was now added to the RGB using LRGBCombination and the result was pretty much ready immediately. Only a slight further enhancement of the contrast applied using CurvesTransformation completed the final image. The field of view of the ASI6200 was too big for this target, so a final crop was applied, resulting in a final image of 6000 x 4000 pixels, or 24 MP.
This was the second image using the ASI6200, this time using broadband imaging. The issues with calibration frames from the first image did not repeat themselves, so must have ben related to some wrong settings at the time. All files calibrated perfectly. Also in general the images showed a lot of latitude for processing with only minimal risk of over-saturation and a lot of detail to work with. Also the images have a low noise-floor, and remaining noise can be effectively eliminated using the MMT method. So far the experience with this camera is very good. It is most likely going to be the major workhorse camera in the observatory for quite some time to come.
This image has been published on Astrobin.
