New from the Zoo – gas clouds around active galactic nuclei
With the excitement about new observations of Hanny’s Voorwerp, this might be a good time to point out that Zoo participants are finding new examples of a related (albeit less spectacular) class of objects. (That was evidently my roundabout way of saying “hey, everyone, you can look for these too!”). What do these four galaxies have in common?
(If you want a closer look at the SDSS data, from top to bottom they are 587739845390761994 = NGC 5972, 588017991773520114, 587736809916399664>, and 587732483292266549 = Markarian 266.)
First off, they all have active galactic nuclei (AGN) as revealed by the spectra. The SDSS images all show oddly colored filaments or loops around the nucleus. (Kevin named that third one the Teacup because of the loop). Colors this pure don’t come from starlight – and indeed, wherever we have the additional information to say, we find that these are ionized gas illuminated by the nucleus. (This should sound familiar to member of the Voorwerp Van Club). Seeing these giant emission-line clouds tells us about the environment of the AGN – how much gas is out there, what its density and ionization state are… It also tells us about the AGN, since the gas views it from various directions that we don’t and over time spans of tens of thousands of years. Some AGN are partially surrounded by clouds of absorbing material so that our view isn’t necessarily representative.
What else do we know about these galaxies? NGC 5972 attracted some interest as possibly being a spiral with a giant double-lobed radio source, which would be quite rare. In a study published in 1995, Phillippe Veron and Mira Veron-Cetty showed that the “spiral arms” are filaments of gas ionized by the central AGN, whose motions are mostly smooth rotation about the core. Within the last couple of months, colleagues at Georgia State University kindly measured a spectrum for part of the handle of Kevin’s Teacup, using the 1.8m Perkins telescope at Lowell Observatory. Same story – the ratios of various emission lines tell us that the gas is ionized by the copious UV and X-radiation of an AGN. The loop shape is interesting, spanning something like 15,000 light-years. Could this be a bubble driven by some kind of hot outflow from the core? Scribble, consult Chandra instrument guide…
Mkn 266 (AKA NGC 5256) is a picturesque merger with (at least one) embedded AGN. Its emission-line features have received attentional the way back to 1988, with interpretations including gas shocked by collisions during the merger. I’d be willing to bet that there are many more of these hiding in the SDSS data. In the usual color mapping, strong [O III] emission shows up as blue and H-alpha as green at low redshifts. H-alpha shifts to the red (i) band at greater redshifts. The Teacup shows both [O III] and H-alpha, and at its redshift these add to purple. Find another one? Feel free to chime in and share it here! Sharing common features but being much smaller, maybe we could call them voorwerpjes.
This request was first posted as a thread on the forum, and Zooites have come up with some more excellent candidates. Based on the ones we know most about, the most likely cases have the spectral signature of an active galactic nucleus when we have a nuclear spectrum, and the anomalous gas doesn’t necessarily follow any kind of spiral or even symmetric pattern. I will also point out that finding these kinds of patterns is something that is not particularly easy to have a computer do, since it would have to start by winnowing through a quarter million or so fairly bright galaxies and do point-by-point analysis of multiple images for each one. Picking up color differences is something our visual systems are very good at (even through a computer monitor).
How to tell which colorful blobs are likely to be these ionized gas clouds, rather than star-forming regions? When there is an SDSS spectrum, one good sign is when the spectrum has the hallmarks of an AGN rather than a star-forming galaxy. This is easy when it’s a so-called Type 1 Seyfert nucleus (or even a type 1.5, for those who are really picky). In these nuclei, the hydrogen emission lines (espcially H-alpha and H-beta) are enormously broad, betraying gas motions of many thousand of km/s in the core. As an example, here is the spectrum of 587742773491531836, posted in the forum by stellar190.
We see spectra like this when we have a clear view down close to the central black hole. When we don’t, perhaps because of material around the accretion disk, we see a type 2 Seyfert nucleus. The emission lines are still broad compared to the cores of normal galaxies but we no longer see the very wide, smeared contributions to the hydrogen lines (and we also don’t see much of the blue continuum from the nucleus, so the type 2 galaxies have a continuum spectrum that comes mostly from the surrounding stars in the galaxy). What distinguishes these from star-forming galaxies is more subtle – ratios of the emission lines, which tell us how much radiation of what energies the gas is absorbing. The easiest ones to check are the strong [O III] line (lab wavelength 5007 Angstroms), which will be at least three times as strong as H-beta, and the ratio of the [N II] 6584 line to H-alpha. This latter ratio should be 0.5 or more, which is never seen in gas ionized by hot stars. Here is a good example, 588010358543351857, posted on the forum by 1000GGG and ElisabethB.
Star-forming galaxies have much narrower emission lines, even when you look at the whole SDSS spectrum, and when they have strong [O III], the [N II] line next to H-alpha is still much weaker. This is an example spectrum of such a star-forming galaxy, part of the wildly irregular galaxy NGC 3991.
So, as a first cut, oddly colored spots are more likely to be AGN clouds if the nucleus has a spectrum like one of the first two samples above. But what if there isn’t an SDSS spectrum? So far, the AGN clouds are likely to be fairly isolated, not part of rings or complete spiral patterns (unlike blue star-forming regions in many ordinary galaxies). Sometimes they do come in symmetric pairs, and some are slightly curved. And some make loops or twists that jump put right away from the images. The final test has to await spectrum measurements for the clouds, which we hope to be getting as our sample grows (thanks to all your keen eyes out there). We already have more candidates than have been previously discussed in research papers. Does that last part sound familiar to longtime Zoo participants?
Great stuff Bill, thanks for the info.
I never knew there was so much info in spectrums. I guess it’s because I understand little of it. After this though I’ll watch more carefully.
Thanks for info Bill.
I made the extraction like overlap if you needed.
Super. Where do we log our suspects?
We’re collecting candidates on the forum in the thread starting at http://www.galaxyzooforum.org/index.php?topic=275014.0
Hello, my name is Cristina and i’m a spanish girl, and I have one problem: I don’t understand the page and the job, and I want help. Is possible make a spanish page for the spanish people? Thank you very much, kisses!!!!! (Sorry for my orthography :P)
Thank you so much, Bill for the interesting article.
I posted this on the ‘Could this be a Hanny’s Voorwerp’ thread a while back but haven’t seen any comments about it. Given your comment about objects that haven’t got spectra yet, I thought it might be of interest – here’s a link to the post:
Maybe it’s just a merger or something and not of interest, but I’d be interested to know what you think!
Bill. Great stuff, but does Galaxy Zoo have a primer on reading spectra?
Well, I realize this is an old post but the most interesting thing here is that 2012 sees an incoming gas cloud with a nucleus (that black hole thing again) and the projection is a merger in 2012. One site IOP has a paper proposing the resolution will be the cloud will precip into drops. I say 4. The nucleus of the cloud will be absorbed by the larger electromag body.
The 4 resulting gas clouds? They will dissapate outward to the ring of stars. We’ll see.