Centaurus A Extreme Deep Field - 120 Hours

[Ross G]

A unique and beautiful photo Rolf.

What an amazing project.

You are an inspiration.

Ross.

[marc4darkskies (Marcus)]

To say that I'm impressed is an understatement Rolf! What a tremendous effort for a spectacular result. Without equal IMO!

Cheers, Marcus

[pixelsaurus (Mike)]

Well done that man!

[SkyViking (Rolf)]

Thank you everyone for your kind comments and overwhelming response, it is all very engouraging (and I'd need some serious encouragement to top this exposure time! )

To answer a couple of the questions that were raised:

Originally Posted by icytailmark

wow 120 hours for just 1 image thats incredible work. How do you find the patience and time?Just out of curosity how many Gigs of data did you capture in total? This deserves to be image of the year.

Thank you Mark, it wasn't really too bad because I have a permanent setup and can just flick the switch so to speak. I let the scope collect all it could on every clear night while the galaxy was visible from my location. I started out when it was rising around midnight, and over the months the imaging obviously shifted earlier and earlier. On these 43 nights I was able to take many other images on the side, either before Centaurus A was in a favorable altitude, or after it had passed out of my view. For example I did all my NIR work while grinding away on this project, so I didn't grow too impatient. I still have a lot of data for a few other targets captured during this time that I just haven't had time to process yet.
In terms of raw FITS files I got around 45GB of them. And all up my Centaurus A folder is currently siting at 75GB, but that includes everything such as batch stacked files (more on that below).

Originally Posted by rustigsmed

Wow wow wow!
Awesome stuff, congratulations on the result Rolf, that is a lot of work and its truly paid off.

that's a lot of hours, how long were the subs? I would imagine your computer would battle digesting the stacking!!

Cheers

Rusty

Thank you Rusty - I use 300s sub frames for luminance, and 380s, 280s, 280s for R,G,B respectively. I have experimented and found that this gives me the optimal S/N ratio for my combination of average sky conditions and telescope/camera.
I had to do the stacking in batches for two reasons: Firstly, I found that in Maxim the plain old Average combine method actually produce the smoothest result, quite a lot better than any of the the other fancy stacking options. But that meant I had to weed out the bad frames by hand... Secondly, since I had around 1500 frames there was just no way it would stack them in one go without throwing errors (and it would take hours probably). So I manually stacked them all in batches of 10, and then stacked those together for the final FITS. Fortunately I could do the batching little by little as I gathered more frames each night.

Originally Posted by Peter Ward

OK..the dark lanes look like someone put a parking sticker on the galaxy...one does need to "respect the light"

but....... Jaysuswepped !!!

Rolf! Bltized it!

Sublime outer halo detail.

The benchmark deep-sky image for 2013 IMHO.

Thanks Peter, I just knew you would comment on the dust lane
I actually did consider this quite a bit but decided that the dust lane is the defining feature of this galaxy and that it should therefore feature prominently in the image. My goal was to create an image that showed all the main features at their best, as such an image has never been made before of this target. So while I may not have 'respected the light' (am I one of those now) I did so contiously based on my subjective preferences.

Originally Posted by strongmanmike

Well my friend, I knew the day would come for me to hand over the Centaurus A mantle, I knew some day someone would take the bull by the horns and surpass my effort from May 2008 at Wiruna, and pow you have certainly done that .... frankly I am surprised it took so long...aaaalthough in my last vestiges of relevance before being relegated to mediocrity...my field is wider so does show the "full" outer polar extentions of the galaxy

I have had a good look at all your presentations of the data and this is a spectacular result Rolf (not to mention out right interesting) and something to be very proud of

Something I related to was the pleasure you mention when working with that volume of data, I remember with my 20hrs taken from a true dark sky it was indeed easy to play with and gradient and noise reduction were almost non existent...as for having 120hrs at your fingertips...I can only imagine...

120hrs..... The Germans invented this didn't they blitzkrieg ? and 43 nights .....I'll just go and rock back and forth in the corner dribbling now...

Mike

Thank you very much Mike, I'm glad to see you take it like a gentleman .
Out of respect for your efforts I actually had my reservations about going this deep on Centaurus A in particular, but given the season and visibility from my location this is actually one of the targets that I get the most observing time on over the year. So for that reason, in order to collect as much exposure as possible, it was an easy choice.
I also simply find it an extremely interesting galaxy with so much stuff to show both in the bright and faint regions, so a good target all up for this kind of image.
Actually I do link to your great image from my site (if you don't mind?) for the very reason you point out. Let's face it, I didn't quite fancy doing a 3 x 120 hour mosaic to include the outer polar extentions of the halo...
Thanks for your comments Mike, your image has certainly been an inspiration for me.

Originally Posted by troypiggo

Massive! Such dedication and commitment. Most of us impatiently wrestle with maybe a few hours of data on one target. 43 nights on one target! Your mount must almost know the way by itself after that!

Thanks yeah yes the mount almost did know its way!
I still home in on my targets the old fashioned way, by star hopping. I must admit that in the end it took me only literally a few seconds to get Centaurus A lined up and centered on the chip - and that often done during twilight when all I had to go by in the finderscope were nearby Omega Centauri and a few lonely stars.

Originally Posted by jase

A splendid image Rolf! I admire your dedication in sticking to the plan and the write up on your approach is just as good as the image providing even further insight into your methodology. Its not easy to put a familiar target in a different light, but you've done it with exquisite results.

I do however wonder...90hrs of luminance? Really? Is that how much you acquired and used or did you acquire far more than that and throw away the subs that didn't meet FWHM requirements? I suspect the signal to noise ceiling was well below this, perhaps offering diminishing returns beyond 50hrs. Did you measure this? What however such a large volume of data provides is greater choice. Choice to include only the exceptionally low FWHM subs.

Thoroughly enjoyed the jet and along with the shell extensions as it fades naturally into the galactic cirrus. A minor technicality in the name. This is not Centaurus A, but NGC 5128. Centaurus A is the radio source within NGC 5128, but a rose by any other name would smell as sweet. ...and talking names, 'Aotearoa' or the land of the long white cloud has become a myth. Clearly you do get endless nights of good astronomical weather! Probably rival many places in Oz.

Once again, excellent work. If you produce a solid image each quarter, you should be happy. In this game, its quality over quantity. Well done!

Thanks very much Jase, yes I did collect more data than the 'mere' 120 hours included here. Somewhere around 127 hours all up if I remember correctly. So I certainly lost some frames to poor guiding, reflections, bright satellite trails etc. It was a pitty having to throw away 7 hours, but overall that is still a pretty good success rate. I usually get to use almost all my data.
I thought about doing a separate stack of only the best frames to achieve maximum detail just in the dust lane. After around 40 hours of processing etc. I didn't get around to it. But maybe some day, perhaps a Sidonio is coming up...

Regarding the weather, we certainly do get plenty of clear nights over here. It is not uncommon that the clouds will disappear at night so the conditions are quite favorable. But still probably nothing compared to the interior of Australia I imagine!

Originally Posted by Dennis

Hello, Rolf

Well, you’ve really smashed this one right out of the park…

No doubt you got bored with Milky Way globulars, so you’ve spread your wings and flown beyond our galaxy to pick up just a few hundred or so elsewhere, in the vast reaches of our cosmos.

A most wonderful and inspirational piece of work Rolf, it really lifts the “what’s possible” genre to a whole new level and injects amateur astronomy with inspiration for new horizons.

Cheers

Dennis

Thank you Dennis, yeah it's pretty crazy with the extragalactic globulars. My tally is now on 709, plus 135 in the Sombrero and a handful in Atoms for Piece Galaxy, so all up ~850 in total... Shall we say first to a thousand?

Originally Posted by Joshua Bunn

Fantastic Rolf, really admire your dedication to this target.

Was it hard to decide from all the objects out there, which one to spend this much time on?

Josh

Thank you Josh, see my reply to Mike above. This is one of the galaxies that I get the most observing time of, so it was an obvious contender.


Thanks again everyone, it is always great to share ones results with such a great bunch of astronomy aficionados!

Regards,
Rolf

[SkyViking (Rolf)]

Originally Posted by clive milne

Rolf, kudos for the work that went in to this image and the result...
It is truly gratifying to see that you did it with a humble 10" newtonian as well!

One minor point;
The magnitude limit of Christian's deep field image is realistically 27.5 or there abouts, but.... that was done with a deep depletion sensor and IR filters if memory serves me correctly..

However you slice it, you have set a new standard that I hope others will adopt.

respect
~c

Thank you for your comments Clive
I have been studying Christian's image quite a bit and I just want to add some notes about faint magnitudes:

If we want to know how faint we can see we must be careful with what is actually measured. One obvious problem is that to obtain the deepest views possible we use the luminosity frame, where all the signal has been picked up, as opposed to the RGB frames where only a portion of the spectrum is recorded. However, official catalogues give magnitudes as seen though specific filters B, V, R etc.
So what I did in my own image was to calibrate the FITS against a white star (within 0.1 magnitude difference in the three bands B/V/R). After having done this it is then possible to measure other stars and read a magnitude for them. But all that this tells us is really just the white light magnitude of the given star. It may have a much fainter magnitude in a particular colour band, which is often the case, but I don't think that can be fairly reported as the faintest magnitude in an image.

While Christian's quasar image is certainly a magnificent effort, and indeed does show the most distant object ever captured by amateurs, I cannot quite get the supposed record magnitudes to add up with what the image is showing.
And I think this point is essential in order to establish how faint we really are able to see in these long exposures.
This is not to say that the reported magnitudes in the qasar image are incorrect, but merely that they may not exactly represent what we are looking for when talking about the faintest stuff that we can see.

If we look at the magnitudes reported in the image here:
http://www.itelescope.net/display/Sh...=1336724493881
To the left of the quasar is a normal star, labelled magnitude 23.1 and the quasar itself is m23.3 in IR. Notice how there is an enormous difference in signal even though these two supposedly have almost the same magnitude. This is because the quasar light is redshifted out of the spectrum so that only a small portion reaches the sensor, where on the other hand the star emits light across the entire spectrum that the chip is sensitive to. As Christian points out: "Its redshift is so large that it has no 'visible' magnitude as such!" So a lot more signal is picked up from the star than from the quasar, although they may both have roughly the same magnitude in the IR band.

Similarly, there is a supposed magnitude 26.9 source immediately to the right of the quasar, but judging from the image this one seems to be of nearly equal signal strength to the quasar itself, although there are supposedly over three magnitudes of difference between them. Again this can be explained easily if these magnitudes are in fact given for particular bands.
I don't know if the 26.9 magnitude comes from a catalogue (and if so, which band is it?), but if not then another potential issue is that this source is so close to surrounding sources that it cannot be reliably measured by aperture photometry. This method would in fact give an inaccurate reading that would be much too faint. This is because aperture photometry relies on comparing the signal strength on a point source with that of the surrounding background (there has to be enough un-obscured background around it). But the background in this case is heavily influenced by the bright sources that nearly touches the supposed magnitude 26.9 source, and as a result the S/N ratio between the source and the background would be measured much lower than it really is - and thus leading to a much too faint magnitude reading.
Aperture photometry can only reliably be used for point sources that lie away from others. In my image I have only measured those stars for which Maxim aperture photometry was not impacted by surrounding sources. There were actually plenty of even fainter readings when I analysed the FITS file, but I did not report these as actual magnitudes because I think they were not reliable (S/N very low and magnitudes fluctuating wildly just by moving the aperture from pixel to pixel, which indicate the measure is unreliable)

As outlined, there is just not enough information to establish how Christian arrived at these numbers so the above are just my observations and comments based on my own experience in determining limiting magnitudes.

As a result of this, I do genuinely believe this 120 hour image of Centaurus A is the deepest image ever recorded with amateur equipment, but I'm certainly open for all scrutiny and more than happy to stand corrected if this is not the case.

[Osirisra (Ken)]

Congrats on The Universe Today article, I expect to see this awesome image in magazines and website all over the place, APOD next

[alpal]

Originally Posted by Osirisra

Congrats on The Universe Today article, I expect to see this awesome image in magazines and website all over the place, APOD next

Yes it's here:

http://www.universetoday.com/102473/...a/#more-102473

And yes - it just has to be an APOD.

[DavidNg (David)]

Wow..

[John Hothersall]

I am out of words, a great thread to read. 120hours would take me 2 years. Great achievement which encourages everyone as the potential for moderate sized scopes has just increased dramatically.

Admiration, John.

[Greg Bock (Greg Bock)]

whoa,...fall off your chair stuff that Rolf, that halo is so extensive. Congrats on the APOD too, well deserved.

[alpal]

Originally Posted by Greg Bock

whoa,...fall off your chair stuff that Rolf, that halo is so extensive. Congrats on the APOD too, well deserved.

No APOD yet!

[Jeffkop (Jeff)]

I cant imagine trying to process all that .. its a head ache with 10 hrs of data trying to align subs of the same type then align them all .. painful hours of trying to eliminate hot pixels cold pixels column defects .. as someone else aptly put it ... shaysus .. WOW .. you must have a camera that has little or no defects and low noise .. An award winning result mate .. up there with the best astro photos Ive seen .. You would have to be ... well .. over the moon

[gregbradley]

This image should spark a thread about signal to noise ratio and how much gain extra exposure time will provide.

As I understand it the signal to noise ratio improves to the square root of the increased exposure time. So it must take a LOT more hours to get that last little bit.

Did you find that Rolf with this image? Or did you wait until you had the full 120 hours to process it. I wonder what it looked like at say 40 hours for example.

Greg.

[naskies (Dave)]

Whoa! Beyond amazing... you've outdone yourself again, Rolf! I really enjoyed your write-up too. Many thanks for sharing!

[seeker372011 (Narayan)]

This image has a momentous air about it..like climbing Mt Everest ..as a country man of yours did some 60 years ago...or breaking the four minute mile or something of that ilk

But beside all that just simply enjoyed the image

Thank you

I do hope it is adequately recognised by being selected for an APOD or whatever. I have no doubt we'll all be reading about it in a future issue of Sky and Tel or Astronomy and seeing the actual image in a double spread in one of those

[Stevec35 (Steve)]

Another incredible result Rolf! Congratulations!

Steve

[alexch (Alex)]

Rolf,
I only saw your image now and WOW! What an effort and the result has left me speechless for a while. I'll be coming back to look at it again and again.

Thanks for the view!

Alex

[LucasB (Lucas)]

That is incredible. Super deep and fantastic processing!
Lucas

[Forgey (Paula)]

Fantastic work Rolf, absolutely stunning!

[AG Hybrid (Adrian)]

Very impressive work.