Markarian’s Chain

*M84, M86 and 6 other galaxies forming Markarian’s Chain - Click image for full resolution*

Markarian’s Chain is a string of 8 galaxies in the constellation Virgo. Together with 1300 more galaxies they belong to the Virgo Cluster. The Virgo Cluster, Local Group (Milky Way, Andromeda Galaxy and others) and another 100 galaxy groups form together the Virgo Super Cluster. But let’s not drift too far away in space, as the Markarian’s Chain itself is already impressive enough.

Two of its largest member galaxies are Messier objects, M84 and M86. both elliptical galaxies. Observations at radio wavelengths and by the Hubble Space Telescope have revealed that M84 has a massive black hole in its center. M86 has the highest blue-shift of any Messier object, as it approaches the Milky Way with a speed of 244 km/s. The remaining 6 galaxies are smaller and dimmer. They are discovered by William Herschel and are usually referred to by their NGC-catalogue numbers: NGC 4477, NGC 4473, NGC 4461, NGC 4458, NGC 4438 and NGC 4435.

These are distant objects, with M84 and M86 being 55 and 52 million lightyears away from Earth. The apparent size ranges from 1.5 arcmin for the smallest galaxy to 7.5 arcmin for the largest galaxy. The term Markarian’s chain is usually referring to these eight galaxies, but that is somewhat arbitrary. Surrounding the chain are dozens more galaxies, visible with moderate telescopes. The Virgo Cluster is really packed with galaxies. If you’re in for some galaxy hunting, this is really the place to be. No surprise then that this is an often photographed part of the night-sky.

Planning

Markarian’s Chain is a typical object to photograph during Galaxy Season, spring time. Then it is high in the sky in the early parts of the evening. This image was shot during February. At that time of the year the object appears somewhat later in the evening, with a transit time of 03:29h. So when photographing this object at this time of year, an automated setup is highly recommended.

To make maximal use of the available time, an ‘early’ target (M35) was photographed during the early parts of the night and Markarian’s chain was photographed during the second part of the night.

*Visibility charts showing altitude at 22:00h throughout the year (left) and throughout the session on February 12, 2021 (right).*

Conditions

All images were taken from the backyard in Groningen, The Netherlands (53.18, 6.54), divided over two nights: February 12 and 13, 2021. The first night had good observing conditions, but the second night was a different story. There was more haze in the air and throughout the night clouds kept passing by. The SubframeSelector tool in PixInsight was used to separate the bad from the good frames, based on SNR and FWHM criteria. In total 120 luminance frames were shot, but only 80 of those were usable, with 51 of those from the first night. RGB channels also had poor keeper rates.

Capturing

The image was captured using the Takahashi TOA-130 in combination with the ASI6200MM-Pro camera. The horizontal Field of View of this combination (2.1 degrees) fits very well the total stretch of the target (approximately 1.5 degrees). Many photographs of Markarian’s Chain are framed diagonally. Diagonal compositions usually give a balanced perspective. In this case however a horizontal framing was chosen. One might see in it a dragon with the larger galaxies to the right forming its head and the smaller ones to the left representing its tail.

Technical Details

Telescope
Mount
Camera
Filters
Accessoires
Software

Takahashi TOA-130 + FL67 flattener, Sesto Senso 2
10Micron GM1000HPS, Berlebach Planet
ZWO ASI6200MM Pro, cooled to -15 ºC
Chroma 2” LRGB unmounted, ZWO EFW 7-position
Fitlet2 (Linux Mint 20.04), Pegasus Ultimate Powerbox v2, Flip Flat
KStars/Ekos 3.4.3, INDI Library 1.8.8, Mountwizzard4 1.04, SkySafari 6.7.2, openweathermap.org

Frames

The image was captured using LRGB broadband filters. Most of the individual galaxies have magnitude 9 or 10, so are pretty faint. Under these circumstances, Luminance frames of 3 mins and RGB frames of 5 minutes usually give good results. There was minimal clipping on the brighter cores of the galaxies and enough exposure time to capture faint details. Below are the frames listed taken in each session that made it to the final image.

Image

Unfortunately due to weather conditions, a lot of frames from the second session had to be rejected. In total 40 Luminance, 4 Red, 2 Green and 4 Blue frames were of insufficient quality to make it to the final set. Still with all the remaining frames, the final image covered a total exposure of 8.1h, roughly half in luminance and half in the Red, Green and Blue channels.

After a little crop to straighten the edges, the final image has a resolution of 9120 x 5898 pixels, or 53.8 Megapixels. It covers a field of view of almost 2 degrees horizontally. What is officially known as Markarian’s chain contains ‘only’ 8 galaxies, but just around it, there are many more galaxies to be seen and in that only gets more when looking at the total Virgo.

Processing

All frames were calibrated with Bias (100), Dark (50) and Flat (25) frames, registered and stacked using the WeighedBatchPreprocessing script. This time a slightly different approach was used towards noise reduction. In recent processing noise reduction was applied to each individual channel at the start. To make things simpler, noise reduction for the colour channels was now applied towards the end of the linear processing of the combined RGB image.

Both green and blue had a higher signal than red, so they were linearly fit to the red channel. After combining all three channels into one RGB image the background was addressed. This turned out to be quite difficult and required several rounds of DBE. Not sure what caused the issues, perhaps the poor imaging conditions and therefore the poorer quality of the images of the second night played a role. But as much as was tried, the background kept having a bit of a magenta colour cast in certain parts and green and blue colour cast in other parts of the image. It was only at the very end of processing where the last bit of background colour casts were removed (see below). PhotometricColorCorrection went smoothly and noise reduction using he MMT-method, worked well. From there a careful manual stretch led to the non-linear RGB image. Not being satisfied with the background colours another attempt was made to clear that up, but with only limited success. The colour image was tweaked by several curves adjustments and saturation increases to give it a decent contrast and colour. As final step on the RGB image a mild convolution made for smooth colour tonal transitions.

The luminance image had a linear background gradient, which was much easier removed using DBE. Noise reduction using the MMT-method was applied and deconvolution used to sharpen up the image. There were no big stars in the image, so no local de-ringing support mask was used. In recent deconvolution processes a range mask was used. But that gave a problem with this image. Range masks become binary, which is good to protect the background. But they also expose fainter galaxy areas fully to the deconvolution process. This often results in artefacts and a kind of craquelure effect. Therefore a clone-mask was used that was stretched to clip the blacks and stretch the highlights, but not too much. A convolution then ensured smooth transitions between affected and non-affected areas. Deconvolution is a tricky and labour intensive process and each time you do it, you learn a bit more. Next was a manual stretch to convert into a non-linear image and regular contrast enhancing and star shrinking was applied to get to the final luminance image.

In combing the luminance with the RGB image something weird appeared as the images were apparently not aligned. It turned out that the pre-processing of luminance and RGB images had been done separately from each other and each had used its own reference image. A quick re-aligment of the Luminance image to the RGB image was able to correct this. Another clear lesson to always be very precise and follow closely what happens at each step of the way. Zooming in, blinking, before/after all are very important to identify issues early on and solve them.

As mentioned above, subtle colour gradients in the background were difficult to remove. Recently Affinity Photo 1.9 was released and one of the improvements is a specific astrophotography persona. It allows for calibrating, stacking and editing of images, all from within the same photo-editing software. While it does not allow for the detailed fine-tuning of processes like PixInsight, it gives access to some very powerful editing techniques without the steep learning curve of a package like PixInsight. One dedicated astro-filter is the background removal tool. By clicking on several places in the background, the software tries to create an even background without colourcast. The filter was used on this image and resulted in a marked improvement. A full assessment of Affinity Photo 1.9 for astrophotography is left for some other time, but in this case, it came in pretty handy.

A few smaller finishing touches were applied in AP as well. This included increasing vibrance and saturation, reducing saturation in the overall background, increasing contrast as well as cooling down the white balance a little bit.

This image has been published on Astrobin.