Galaxy Zoo Starburst Talk

Something is not quite right about this print. . .

  • Ex103 by Ex103

    I came across a image that looks very unusual to me.

    AGS00004gp SDSS J082650.14+130813.0 RA: 126.70893, DEC: 13.136946


    Note that it is much easier to see what I'm referring to with a dark background behind the print.

    When I first started to look at this my first notion was the center must be spinning really fast. Then I though no no maybe the outer portion of the ellipse was just unlit. Then I sorta went back to my first notion and thought it might be shedding some form of heavier coalesced material.

    It has some really unusual really thin multicolor filaments that make their way outward from the brighter middle. Some seem really long as one works their way around outer brighter center of this galaxy.

    Granted it would be neat to see it in a few different spectrum sense I still suspect some aspects of this galaxy is missing from the print.

    While I wish I had a better print of it, I send a message to JeanTate who said it reminded him of NGC1275. (BTW Jean, Thanks again for the info.)

    Maybe someone on the science team can take a peek at it?


  • JeanTate by JeanTate

    Here's what it looks like in DR10:

    enter image description here

    And DR7:

    enter image description here

    At least some of the faint, filamentary features apparently radiating from the central bulge seem to be 'visible' in at least the DR10 image.


  • Ex103 by Ex103

    Jean - was there any time lapse between the other DR images? Or are they reprocessed from an original capture? And the print I posted is the from the DR9 set right?

    The first thing I noticed when I looked at them and compared them to the DR9 I posted, was the star in the upper right hand corner . . . Go right a bit more in both the DR7 and DR10 and the edge of another star appears. I put them in a slideshow viewer and started flipping and everything in the image I posted suddenly got larger and thus I think a tighter cropping was made of that print.

    Other things. . . DR7 looks heavily processed and appears 'lossy' for lack of a better way to describe the print. And it looks like the filament's I spoke of were filtered out with the rest of the 'noise' (at least that is what appears to have happened when I compare them.) The shapes around them still look like they may have existed but were removed like badly cutout puzzle pieces (for lack of a better description.) Still the two small clumps do appear.

    Of the 3 I would say the images DR9 is a tad over saturated and that makes things in it look wider. Still the varied colors seem consistent though out each of the prints.

    Need more time to think about them.

    And thanks for these pictures.


  • JeanTate by JeanTate in response to Ex103's comment.

    Ex103 - you're very welcome.

    The apparent difference in scale is just that, apparent. For the DR7 and DR10 images, I copied the image URL from their respective Explore tool pages, and edited it (to make it 424 x 424, the size of the Quench image; to remove the grid; to change the scale to 0.16" per pixel). I don't know what scale the Quench images - which are DR8 images - are; however, as you discovered, it's not 0.16!

    In general, there's only one set of images for each SDSS object; the camera 'took photos of' each part of the sky just once.

    However, each field overlaps, round the edges, somewhat. This can produce two (sometimes even three) sets of images, sometimes with strange results (I wrote about this in the GZ forum: NGC 5020 nucleus in SDSS: why the apparent resolution differences?).

    There's one part of the sky which was imaged by the SDSS camera lots - and I mean LOTS - of times. It's called Stripe 82, and we zooites classified lots of galaxies there, back in Galaxy Zoo 2. For some examples of how different (Stripe 82) images can appear when they are 'co-added' (I think the term used among amateur astrophotographers is 'stacked'), check out this GZ thread (and the GZ blog posts it links to): Speckly and bright enhanced images: What, why & how.

    When the Multi-wavelength viewer tool becomes available - in our own Galaxy Zoo Examine - we can look at the raw data, in each filter, for this object. In my experience, you can sometimes see structure in one of those images that is not apparent in the JPG (DR8 in this case).

    If you're really interested, you could obtain the FITS which contains the object, and process the image yourself, using FITS Liberator (if you have Photoshop), or DS9, or another image processing package.


  • Ex103 by Ex103

    Hum interesting, I suppose it will remain a bit of a mystery. The oddness of the resizing has me a bit baffled in respect to the print but like I said if there is a bit of residual heating or what not left in the CCD or I suppose I should expand that and say it is possible that one of the filters, lenses, mirror, or some other object within the telescope could have held that residual image heating as well.

    I suspect some of this is due to saturation limits on the CCD too, but one can never be too sure unless their involved in the process of capture process. And an good feel of the limits of the telescope, angle of capture, weather, and maintenance being done at the time of image capture all helps. While it also helps to move the telescope off of fixed position points too and have a few sets of images taken a few months if not a year apart (given the distances where talking about,) I can generally understand why this wasn't done. However if one is looking for signs of motion, possible reflected objects, and the like, somethings can be ruled out a bit easier with a extra set of prints. I do admit the whole system produced a enormous volume of data and I'm sure as those 'false returns' are found out, someone is working diligently to be rid of them and thus improve the images.

    I suppose I should stop taking such a detailed look at these prints. I have noticed inconsistencies but wasn't sure as to the source, generally I suspected most of these things were due to atmospheric heat waves and the like.

    Don't get me wrong, I still think SDSS is a wonderful telescope and source for information. The automation of the telescope is also very interesting and I'm sure that as time wares on, some out there will come up with tons of ideas on how to expand or contract certain aspects of how these types of telescopes process and capture the data. So I tend to hope that while people might sense some disappointment in what I'm writing, I don't want them to get discouraged when it comes to trying things or making refinements along the way that improves the ability to interpret the data.

    Still those things said I'm just trying to understand the limits of the data and I tend to get curious when I see oddities I'm not used to seeing. . . So questions and possibilities pop up and while I might be a bit more verbose about what I'm thinking, oddly enough I'm ruling out a lot of it too usually 5 minutes later.

    When it comes to Photoshop I had at one time the CS2 version but somewhere along the line I've lost the disks for it. And given the raw expense of the program I can't justify the upgrade to the latest version, considering I'm not editing photographs anymore nor transferring print film over to digital format. Folks reading these things need to realize that when I was doing some photo editing I was always on the lookout for dust, hair, and other things. But most of the time dust and hair are only a few pixels across in digital prints unless your really blowing up or resizing a portion of the captive image. It's also the reason why some of these things within the print stood out to me so much.

    I generally thought that the images produced came from one date on a run though the galaxy sense I didn't see or sense any motion in relation to the stars. Still I wasn't sure. I never really did dig into the background on the telescope or how it ran though the sky.

    Sense were looking at what is essentially a layered mosaic I do tend to believe that one will see oddities crop up. I had noticed what I thought were edge lines traveling though that particular cropped print as well. So that could be a contributing factor.

    *My apologies for not editing this a few hours ago, things sometimes get rather busy about the house. . . And distractions break up my line of thought sometimes.


  • JeanTate by JeanTate

    It's great to be as enthusiastic and excited as you obviously are, Ex103! 😃

    I would encourage you to pursue your passion, and use your fascination with the galaxy to drive you to learn the details of how the SDSS camera, photometric pipeline, and Explore tool work. A good place to start might be the SDSS SkyServer, especially the Education section.

    If you have questions as you learn, don't be shy about asking them, either here or in the Galaxy Zoo forum; lots of zooites have traveled down similar roads before you, and are only too happy to help you.


  • Ex103 by Ex103

    Hey there Jean,

    Again thanks for the replies, I've been looking at the SkyServer stuff as suggested. I'll have to do some more digging to see what other information is there. Some of what I read justifies some of things I've been thinking while trying to digest these images. It is still going to be difficult to for me to pinpoint where the source of these filaments lie. But I think there from the telescope refocusing and moving for each of the various bands as each print is taken for the mosaic, (provided some lens, coating, or mirror with in the telescope held some of the residual heating. Or even the CCD.) The only troubling aspect of this is that it is not repeated with the star.

    I figured I'd post another print for the hell of it. . . Sense I saw a anomaly or basically something different about it that stood out to me. Maybe someone is collecting these things anyway out there. So I tend to hope this helps.

    AGS00003ef or SDSS J095205.63+115855.1

    SDSS J095205.63+115855.1

    What spiked my interest was the dullness of the lower 5 o'clock edge of the ellipse. Granted the image is poor and it would be tough for me to make anything further than a guess as to what it is. - - - And I began to wonder if this is a candidate for a very large Hanny's Voorwerp or just a odd merger in progress. I also thought it could be some process being linked to a jet of a black hole stripping off gas as the much larger galaxy consumes up what appears to be a very small one. However, something about it had the feel of a merger still in progress. Sense even that half of the ellipse seemed to be off color slightly and gave the appearance that it was distorted. Now those things said, it very well could be the relative speed, spin, and gravity tidal forces simply stripped off any excess gas out of the darker colored mass, assuming the darker portion has heavier coagulated elements (thus, reacting somewhat slower to a change in direction, momentum, . . . well you catch the drift.)

    But alas it's just another one of those prints I wish was a bit sharper. And some of these color differences could be miss leading me. Especially if someone took a tear dropper tool and removed certain color ranges from the print like what was done with the DR-7 print way above this posting. (That is why I said the filaments looked like they were cut out like bad puzzle pieces.) Yes it stands out to me. . so dose the Gaussian blur, (what appears to be some) sharping, and an cross hatch filter.

    BTW. . .
    How are the calculations going? With respect to what information the group has gathered thus far?
    I tinkered with it a bit, but I don't always have the necessary time to really pull up my sleeves and get into it. In fact I stopped classifying for a bit again as things got sorta busy for me again.

    Take care!


  • JeanTate by JeanTate in response to Ex103's comment.

    Good to see where your interest is leading you! Hope you keep digging.

    Have you found an app which can help you visualize the data in a FITS file (have you found out how to download a FITS file? worked out what one is)? When I started doing this, I found it very sobering to realize just how huge the dynamic range in the SDSS photometric data is! The JPG images we see, in Galaxy Zoo, have just 256 (? 8-bit) levels of intensity, while the data in FITS files are derived from 16-bit (65536 levels) data coming from the A/D converter (more on the camera here, and even more here). Having said that, it's also interesting that the faint outer parts of a galaxy are often just ~1 level1 brighter than the sky (in a given band, and a given pixel).

    As for the data analysis, well ... I didn't download the classifications data quickly enough, and now the download part of the tool doesn't seem to be working. And I see several other zooites have said they're having trouble with Tools too. A couple of zooites who seem to be quite a bit more than just ordinary zooites (I won't name names) have done some wonderful analyses, but they used tools few of us have (or could easily get), and I doubt any of us could replicate their analyses using just a spreadsheet and database (such as you find in MS Office or OpenOffice). Certainly none of those analyses would be possible using Tools.

    But I did have a lot of fun writing today's Object of the Day, based on some of the analyses I've been doing, in the GZ forum: White.

    1 I think these levels are called "ADU"s (Analog-to-Digital Unit), or "DN"s (Data Number); this PDF explains some of the nitty-gritty: Signal-to-Noise in Optical Astronomy (from a University of California Observatories website)