Page 2 of 6 FirstFirst 1234 ... LastLast
Results 31 to 60 of 179

Thread: Bias effects in galaxy detection

  1. #31
    Join Date
    Jul 2006
    Posts
    9,147
    Quote Originally Posted by George View Post
    I'm more the lurker, but this topic looks like we might get some nice juice for the squeeze. Just learning all the differenct effects and pricinciples involved will be worth it for me.
    I whole heartedly agree.

    The usual Jerry interpretation, eh? "snip"
    :Picks up drink and steps from between Nereid and Jerry:

    I've been here for years and realize there's history and all, but dude!

    May I recommend decaf?
    Time wasted having fun is not time wasted - Lennon
    (John, not the other one.)

  2. #32
    Join Date
    Mar 2004
    Posts
    13,441
    Quote Originally Posted by BigDon View Post
    I whole heartedly agree.



    :Picks up drink and steps from between Nereid and Jerry:

    I've been here for years and realize there's history and all, but dude!

    May I recommend decaf?
    Put yourself in my shoes BigDon*, how would you respond?

    * well, figuratively; if I had to guess I'd say mine are rather too small for your feet

  3. #33
    Join Date
    Mar 2004
    Posts
    13,441
    Let's assume, for now, that Tal and van Dokkum (2011)'s Figure 6 is the light profile of a z = 0.34 LRG, in the five SDSS filters. Let's assume the central wavelength of each filter is (in nm):

    u 354.3
    g 477.0
    r 623.1
    i 762.5
    z 913.4

    At what redshift (approximately) would this LRG appear to be too small to be detected as a galaxy, in SDSS? Assume Tolman dimming, a 'minimum catalogue limit' of 1.4", and 'aberration' (as Disney and Lang use the word). This exercise, involving a specific (though somewhat synthetic) galaxy will - I hope! - illustrate the Disney and Lang hypothesis. It should also bring out some of the messy, real-telescopes-etc, details we may have to deal with.

    Jerry, George, and BigDon (and any other interested reader lurker): do you think you could sketch the steps you think should be taken to work out an answer?

  4. #34
    Join Date
    Sep 2006
    Posts
    1,429
    To answer this question, one would need to know many things.

    a) what is the surface brightness limit for galaxies to be detected and classified as galaxies, in each passband?
    b) what is the size limit for galaxies to be detected and classified as galaxies, in each passband?
    c) what rules are used in the SDSS software to decide than an object is a galaxy? That is, if an object is detected in the g, r, i passbands, but not the u or z passbands, is it classified as a galaxy, or discarded, or what?
    d) what is the spectrum of the sample LRG? Is it a constant, or does it vary as a function of position in the galaxy?

    With that information, one can, for any given redshift z, compute the apparent size and surface brightness in each passband (redshifting the spectrum and performing synthetic photometry as needed); determine the size and surface brightness of the galaxy in each passband; decide if the software will detect it and classify it as a galaxy in each passband; and determine how the object will appear, if at all, in the SDSS catalogs.

    Pretty basic stuff, but it will take some time.

  5. #35
    Join Date
    Mar 2004
    Posts
    13,441
    Quote Originally Posted by StupendousMan View Post
    To answer this question, one would need to know many things.
    Yes indeed.

    Jerry, George, BigDon (and other lurkers): how complete is StupendousMan's list?
    a) what is the surface brightness limit for galaxies to be detected and classified as galaxies, in each passband?
    This one is a biggie.

    For the purposes of illustrating the test, let's assume:
    u 22.12
    g 22.60
    r 22.29
    i 21.85
    z 20.32

    b) what is the size limit for galaxies to be detected and classified as galaxies, in each passband?
    1.4"
    c) what rules are used in the SDSS software to decide than an object is a galaxy? That is, if an object is detected in the g, r, i passbands, but not the u or z passbands, is it classified as a galaxy, or discarded, or what?
    For simplicity, let's assume that if it's bigger than 1.4" in any passband (filter, waveband, etc), it's a galaxy.

    d) what is the spectrum of the sample LRG? Is it a constant, or does it vary as a function of position in the galaxy?
    Another biggie.

    Tal and van Dokkum (2011) - TvD11 for short - do present results of their stacking, for the colour gradients (i.e. as a function of radius). LRGs (and ellipticals in general) become bluer away from the nucleus; however, the trend is greatest in the first few (tens of) kpc.

    Of course, colours are not spectra; for simplicity, let's assume that our LRG is bland; i.e. the spectrum taken at any position is the same, modulo level.

    With that information, one can, for any given redshift z, compute the apparent size and surface brightness in each passband (redshifting the spectrum and performing synthetic photometry as needed); determine the size and surface brightness of the galaxy in each passband; decide if the software will detect it and classify it as a galaxy in each passband; and determine how the object will appear, if at all, in the SDSS catalogs.
    I agree, except that ... I think there are a few other things we need to take into account (even if only to convince ourselves that they don't matter, wrt the approximate answer I'm looking for for now: this is an exercise largely in getting a feel for what sorts of things we need to consider, and determining the feasibility of doing some real research).

    For example: to what extent do we need to make explicit our cosmological assumptions (other than Tolman dimming and 'aberration')?

    Pretty basic stuff, but it will take some time.
    Yes, so far it's mostly just writing down the steps, making sure we've not missed something important, and cranking up our googling, skimming, and data-extraction skills ...

  6. #36
    Join Date
    Mar 2004
    Posts
    2,693
    You can picture an object-by-object approach, if you start by having images of a galaxy at a number of wavelengths. In a simple case, where you picked one for which you have a UV image that would redshift to a desired optical band at a redshift of interest, you would calculate the flux your galaxy would have at the new redshift along with the change in its angular size ("aberration" and all), then make a version of that image with appropriate noise. As an example, take a GALEX near-UV image of some galaxy at z~0.05, consider what it would look like at z-0.25 (HDF F30W) or z=0.87 (HDF F450W) at the angular resolution and noise per pixel of the HST HDF (or other) observations. How much of the galaxy if any) passes a desired S/N threshold? This lets you sample behavior at narrow redshift ranges but needing no assumptions about the spectral behavior, since that's all implicit in the images. The point of doing this for local galaxies which have lots of longer-wavelength data is that you then sort of understand where they fall in optical luminosity, Hubble type, concentration index, and so on, so you can ask how the mix of these entering a catalog changes with redshift.

    There were cottage industries doing this when good UV images first became available for lots of galaxies (a major program on the Astro-2 mission, for example, even before HST could do it well). Many show up doing an ADS abstract-word search on "morphogical K-correction".

    As far as I can tell, the pieces of cosmology that enter are the angular size-redshift relation, and whether time dilation follows (1+z). The latter comes in by increasing the mean time between photon arrivals. Heuristically, one can thin of the Tolman dimming as having one factor of (1+z) for photon energy, one for photon arrival time, and two for increase in angular size of distant objects compared to the Euclidean case (what Disney and co. call aberration). Additional terms may enter observationally but are purely detector-based; for example, the width of a filter band in the emitted frame decreases as 1/(1+z), but you could in principle avoid that by using a new filter which matches some standard one when used at a particular redshift. (If you know the spectrum, this can be corrected in a purely numerical way).

  7. #37
    Join Date
    Jul 2006
    Posts
    9,147
    Nereid,

    Due to time constraints, my BAUT cruising time is up and just finding out about the pop quiz, I beg permission to take up the task of rereading this thread in even greater depth, i.e. follow a link or two, tomorrow morning. I'll try to get back with some sort of answers tomorrow evening Pacific Time, even if it's "I don't know" or "I need more time to rig the mental scaffolding".

    Right now a 55gal tank full of weird and brilliant guppys demands my attention. Plus I have to figure out a way to keep possums out of my carnivorous plant collection. The possums drink the standing water they have to have in the basins and knock eveything over. Wait a sec! It just came to me as I was typing this out. I'll try setting aside a big bowl of water like a dog dish! That way they won't have to knock my plants over to drink!

    Tomorrow evening then.
    Time wasted having fun is not time wasted - Lennon
    (John, not the other one.)

  8. #38
    Join Date
    Sep 2003
    Posts
    10,399
    Quote Originally Posted by Nereid View Post
    Jerry, George, BigDon (and other lurkers): how complete is StupendousMan's list?
    All important items, no doubt. What about extinctions, since SDSS is terrestrial, or are the results already atmospherically compensated?

    I feel like a kid going to the zoo for the first time. The nuances of distant galaxy exploring is new to me.

    In earlier post, I did offer a correction (Sandage) to the Tollman inverse 4th power term, assuming I'm even on the same page of what is being addressed. Was I wrong? Won't the equation be important to resolving some of the questions?

    Other questions I have are crude:

    Wouldn't heavier intergalactic neutral hydrogen be more prevalent in earlier periods disrupting the Lyman break observations?

    Here's a wild one -- Has anyone produced a SED of a region of "empty" space to see the "color of the sky"? A great deal of scattering takes place, which might help reveal more of the light sources. [Is my heliochromology showing? ]

    What of metalicity of early galaxies as it relates to anticipated spectrums?

    [I'm reluctant to ask questions that might slow the pace.]


    For example: to what extent do we need to make explicit our cosmological assumptions (other than Tolman dimming and 'aberration')?
    Yes, what are the more solid givens?
    We know time flies, we just can't see its wings.

  9. #39
    Join Date
    Sep 2006
    Posts
    1,429
    Quote Originally Posted by George View Post
    All important items, no doubt. What about extinctions, since SDSS is terrestrial, or are the results already atmospherically compensated?
    The SDSS results have been corrected to remove (most of) the effects of the Earth's atmosphere.

    In earlier post, I did offer a correction (Sandage) to the Tollman inverse 4th power term, assuming I'm even on the same page of what is being addressed. Was I wrong? Won't the equation be important to resolving some of the questions?
    Your earlier posting mentioned "Lobin and Sandage 2001." I believe you were referring to a series of a 4 papers by Lubin and Sandage, the last of which is

    http://adsabs.harvard.edu/cgi-bin/np...cead52d9b01416

    In that series of papers, the authors look at the empirical relationship between surface brightness in one particular passband and redshift. They fit the results to models which look like (surface brightness) = K (z)^n, and find values of the exponent "n" which are less than 4 for both R and I passbands. This is not inconsistent with the Tolman hypothesis, since the properties of galaxies observed in a fixed passband will change with redshift. Tolman's formula is an ideal one, which would apply if one could move a single, constant source throughout the universe at will, and measure its bolometric energy output (measure all energy emitted at all wavelengths). As Lubin and Sandage conclude in their abstract,

    We conclude that the Tolman surface brightness test is consistent with the reality of the expansion to within the combined errors of the observed <SB> depression and the theoretical correction for luminosity evolution.



    Wouldn't heavier intergalactic neutral hydrogen be more prevalent in earlier periods disrupting the Lyman break observations?
    No.

    Here's a wild one -- Has anyone produced a SED of a region of "empty" space to see the "color of the sky"? A great deal of scattering takes place, which might help reveal more of the light sources. [Is my heliochromology showing? ]
    Yes, there have been several papers describing the spectrum of "blank" regions of the sky. I recall seeing one such paper, which described scattered light within the Milky Way ... ah, here it is:

    http://arxiv.org/abs/1109.4175


    What of metalicity of early galaxies as it relates to anticipated spectrums?
    There will be small changes to the spectra of galaxies at high redshift due to the lower metallicity, but those are minor compared to the large changes in the spectra due to the younger stellar populations.

  10. #40
    Join Date
    Mar 2004
    Posts
    13,441
    Quote Originally Posted by ngc3314 View Post
    You can picture an object-by-object approach, if you start by having images of a galaxy at a number of wavelengths. In a simple case, where you picked one for which you have a UV image that would redshift to a desired optical band at a redshift of interest, you would calculate the flux your galaxy would have at the new redshift along with the change in its angular size ("aberration" and all), then make a version of that image with appropriate noise. As an example, take a GALEX near-UV image of some galaxy at z~0.05, consider what it would look like at z-0.25 (HDF F30W) or z=0.87 (HDF F450W) at the angular resolution and noise per pixel of the HST HDF (or other) observations. How much of the galaxy if any) passes a desired S/N threshold? This lets you sample behavior at narrow redshift ranges but needing no assumptions about the spectral behavior, since that's all implicit in the images. The point of doing this for local galaxies which have lots of longer-wavelength data is that you then sort of understand where they fall in optical luminosity, Hubble type, concentration index, and so on, so you can ask how the mix of these entering a catalog changes with redshift.


    Would it be accurate to say that Disney&Lang might claim that the Tolman dimming was not (adequately) accounted for in this approach?

    There were cottage industries doing this when good UV images first became available for lots of galaxies (a major program on the Astro-2 mission, for example, even before HST could do it well). Many show up doing an ADS abstract-word search on "morphogical K-correction".
    I see many enjoyable hours ahead of me, spent reading some of these papers! Thanks.

    As far as I can tell, the pieces of cosmology that enter are the angular size-redshift relation, and whether time dilation follows (1+z). The latter comes in by increasing the mean time between photon arrivals. Heuristically, one can thin of the Tolman dimming as having one factor of (1+z) for photon energy, one for photon arrival time, and two for increase in angular size of distant objects compared to the Euclidean case (what Disney and co. call aberration). Additional terms may enter observationally but are purely detector-based; for example, the width of a filter band in the emitted frame decreases as 1/(1+z), but you could in principle avoid that by using a new filter which matches some standard one when used at a particular redshift. (If you know the spectrum, this can be corrected in a purely numerical way).
    There's one other that I had in mind, if we were to use TvD11: reversing their calculation of (LRG-frame) radial distances ... the x-axis in their Figure 6, for example, is in kpc (not arcsecs).

  11. #41
    Join Date
    Mar 2004
    Posts
    13,441
    StupendousMan has already commented on much of your post George; some additional comments ...
    Quote Originally Posted by George View Post
    I feel like a kid going to the zoo for the first time. The nuances of distant galaxy exploring is new to me.
    Well I think this is very interesting and exciting stuff, but was expecting to be going solo on the exploration. So I was glad StupendousMan said "An open discussion on this thread might be interesting and educational for many readers"

    In earlier post, I did offer a correction (Sandage) to the Tollman inverse 4th power term, assuming I'm even on the same page of what is being addressed. Was I wrong? Won't the equation be important to resolving some of the questions?
    Did StupendousMan identify the paper you had in mind?

    Other questions I have are crude:
    I don't know about crude; as long as you're learning (and enjoying yourself), all questions are good, aren't they?

    Wouldn't heavier intergalactic neutral hydrogen be more prevalent in earlier periods disrupting the Lyman break observations?
    Evolutionary changes - of any kind - are certainly interesting.

    However, Disney&Lang's main point is that the galaxies we see, at redshifts of ~0.5+, cannot possibly be 'like' the ones which are so prominent locally (M31, M101, M87, M82, ...), in a universe with "aberration" and Tolman dimming. What I'm interested in doing is exploring their hypothesis (or hypotheses, I suspect there's actually more than one), and testing it (them). Starting with 'assume aberration and Tolman dimming, but no galaxy evolution'.

    [I'm reluctant to ask questions that might slow the pace.]
    Please don't be reluctant!

    Yes, what are the more solid givens?
    How well did you understand ngc3314's most recent post?

  12. #42
    Join Date
    Mar 2004
    Posts
    13,441
    Quote Originally Posted by StupendousMan View Post
    Quote Originally Posted by George
    What of metalicity of early galaxies as it relates to anticipated spectrums?
    There will be small changes to the spectra of galaxies at high redshift due to the lower metallicity, but those are minor compared to the large changes in the spectra due to the younger stellar populations.
    This ties back to several things ngc3314 said, starting from the OP:

    "gains from the very blue colors of young stellar populations, detection of galaxies from their UV-bright compact regions at substantial redshift, "

    "This is what would make Milky Way analogs "drop out" at higher redshifts, although the details depend on how we observe them - higher resolution lets us see the bright cores and star-forming regions, while redshifting ultraviolet into the visible band would exacerbate light loss due to dust while presenting us brighter and more compact star-forming regions when they are not heavily dust-reddened."

    "the maximum surface brightness of starburst regions in galaxies is pretty constant with z when taking Tolman dimming into account, although the linear scale and thus luminosity changes a lot."

  13. #43
    Join Date
    Mar 2004
    Posts
    13,441
    This thread's been a bit quiet; loss of interest?

    It might be helpful, for lurkers and others, to get a bit of a feel for the SDSS filters/bandpasses and redshift.

    For simplicity, consider this: what redshift would an object have for light emitted in one bandpass to be seen in the next (longer wavelength) one? Of course, to answer this in detail is quite complicated, not least because the central wavelengths of each bandpass are not related to each other in a simple way, nor are the detector response plus filter transparency a square function! But at what redshift is light emitted at 354.3 nm (centre of the u-band) observed at 477 nm (centre of the g-band)? 0.346. Here are redshifts for g->r, r->i, and i->z, respectively: 0.306, 0.224, 0.198.

    Crudely, then, for the 'average' LRG in TvD11 - with a redshift of 0.34 - the g-band light profile (in Figure 6) is the u-band one, in the rest frame (i.e. as if it were observed at zero redshift), and the r the g*.

    The centre wavelengths of the two GALEX filters are 227.5 nm (NUV) and 155.0 nm (FUV); the TvD11 average LRG u-band profile is not really the same as a zero redshift NUV one (264 nm becomes 354 nm, when redshifted 0.34), but it's close.

    Now for a research question: given the light profiles in Figure 6, how to convert them into 'zero-z' ones? For simplicity, leave the passbands alone (i.e. derive curves for an SDSS u-band (g-band, r-band, ...) blueshifted by 0.34); just 'reverse out' the (1+z) "aberration"** and the (1+z)^-4 Tolman dimming.


    * except for "aberration" and Tolman dimming, of course!
    ** I don't like using this term; does anyone have a better, equally pithy, alternative?

  14. #44
    Join Date
    Mar 2004
    Posts
    2,693
    Quick (airport) post on something George brought up. At the highest redshifts we can reach (z>6), the amount of neutral H in the intergalactic medium does matter. As we finally see the hydrogen Gunn-Peterson effect completely block light shortward of redshifted Lyman alpha, the effective location of the Lyman break shifts from 912 to 1216 A (plus a bit for line broadening) in the emitted frame, which can now be modeled realistically since we have QSOs and GRBs to show us how fast this happens as a function of redshift..
    Last edited by ngc3314; 2011-Sep-29 at 09:41 PM. Reason: typos

  15. #45
    Join Date
    Sep 2003
    Posts
    10,399
    Quote Originally Posted by StupendousMan View Post
    The SDSS results have been corrected to remove (most of) the effects of the Earth's atmosphere.
    Thanks, and thank goodness.

    Your earlier posting mentioned "Lobin and Sandage 2001." I believe you were referring to a series of a 4 papers by Lubin and Sandage, the last of which is

    http://adsabs.harvard.edu/cgi-bin/np...cead52d9b01416

    In that series of papers, the authors look at the empirical relationship between surface brightness in one particular passband and redshift. They fit the results to models which look like (surface brightness) = K (z)^n, and find values of the exponent "n" which are less than 4 for both R and I passbands. This is not inconsistent with the Tolman hypothesis, since the properties of galaxies observed in a fixed passband will change with redshift. Tolman's formula is an ideal one, which would apply if one could move a single, constant source throughout the universe at will, and measure its bolometric energy output (measure all energy emitted at all wavelengths). As Lubin and Sandage conclude in their abstract,...
    I assumed that at the time of the formula involving z that the age of the universe, Hubble constant, and other factors have been better refined such that the 4th power got tweaked by Sandage et. al., but apparently only for the R & I band. [Free time has become in short supply, but this weekend might grant me time to read-up on this and the tweaking, assuming tweaking is the right word.]

    Yes, there have been several papers describing the spectrum of "blank" regions of the sky.
    Wow, that's great.

    I recall seeing one such paper, which described scattered light within the Milky Way ... ah, here it is:

    http://arxiv.org/abs/1109.4175
    Another something I must look at.

    The thought is that scattering might prove to be a helpful indirect tool of the deeper regions, if we can tickle out the foregound. [I know an astronomer that added the known atmospheric spectral extinctions to the sp. irr. of the Sun and matched it to a star -- I found which one it was, too: Procyon (F5) -- then observed its color as seen from Kitt Peak. By observing Procyon terrestrially (after extinctions) he was, in effect, seeing the Sun's true color. Such spectral manipulations I know are common, but I thought I could combine the spectral scattering idea with adding a grin to this same astronomer who is reading this.

    Perhaps scattering information might be another source of evidence to help discern all those ancestral species.
    We know time flies, we just can't see its wings.

  16. #46
    Join Date
    Jul 2006
    Posts
    9,147
    Still reading. Trouble is every read through I come away with something different.

    But I will say...

    Have you tried running this issue passed people skilled in electronic warfare? Especially "active" electronic warfare? I knew several back in the day who would have been fascinated with this issue. Hiding entire galaxies and dynasties of galaxies through signal manipulation! Which is "all" you are dealing with. Instead of ionispheric bounces, reflective chaff and intentionally designed false return signals you have the gravity of galactic clusters, the expansion of the universe and local noise bending and altering the light to the point we don't receive it.

    It's the same issue just 10^23 orders of magnitude larger.

    I think a meeting of the minds between people working on this issue and the folks at NAESU (NAY-su) would benefit both of you.
    Time wasted having fun is not time wasted - Lennon
    (John, not the other one.)

  17. #47
    Join Date
    Mar 2004
    Posts
    13,441
    Thread's gone quiet again ...

    Continuing with TvD11; Figure 2 has a convenient scale bar, in which 50" is equated with 215 kpc (presumably at z = 0.34). That'll be helpful, at least at the level of a first analysis.

    Looking at Figure 6, at z = 0.34, the surface brightness (mag/arcsec^2) at ~2.3" from the centre of our 'typical' LRG, is 26, 24, ~22-23, for the u-, g-, r-, i-, and z-bands, respectively.

    Reversing out the "aberration", this means that surface brightness (in the same units) will be these same values at 1.7", in the blueshifted u-, ... z-bands (1.7 = 2.3/(1.34)).

    Yes? No? Maybe? Don't know??

    Next: the (1+z)^-4 Tolman dimming, applied to the same data ...

  18. #48
    Join Date
    Mar 2004
    Posts
    2,693
    Quote Originally Posted by Nereid View Post
    Thread's gone quiet again ...

    Continuing with TvD11; Figure 2 has a convenient scale bar, in which 50" is equated with 215 kpc (presumably at z = 0.34). That'll be helpful, at least at the level of a first analysis.

    Looking at Figure 6, at z = 0.34, the surface brightness (mag/arcsec^2) at ~2.3" from the centre of our 'typical' LRG, is 26, 24, ~22-23, for the u-, g-, r-, i-, and z-bands, respectively.

    Reversing out the "aberration", this means that surface brightness (in the same units) will be these same values at 1.7", in the blueshifted u-, ... z-bands (1.7 = 2.3/(1.34)).

    Yes? No? Maybe? Don't know??

    Next: the (1+z)^-4 Tolman dimming, applied to the same data ...
    The Tolman signal includes the effects of "aberration", so it's evaluated at the same radii for like objects. (BTW, I also wish there were a better term than "aberration" for "departures from Euclidean angular size-distance relation").

    The way I would view this might be that the surface brightnesses in shifted bands matching emitted wavelength and projected radius would all be brighter by [2.5 log (1.34^4)] = 1.27 magnitudes at z=0. Using magnitudes is tricky, since various magnitude systems have essentially arbitrary zero points tied to particular spectral responses (so working in F-lambda, or using a well-established set f K-corrections to account for the spectral shape, is needed).

  19. #49
    Join Date
    Mar 2004
    Posts
    13,441
    For the benefit of lurkers, I think a quick refresher on some key terms might be helpful.

    The various 'magnitudes' used in SDSS (DR8) are described here: Measures Of Flux And Magnitude.

    Now several terms are sprinkled liberally throughout that webpage (and in many others on observational astronomy), e.g. 'flux', 'luminous flux', 'intensity', 'surface brightness', 'luminosity'.

    There's also reference to 'profiles' (some of the magnitudes the photometric SDSS pipeline produces involve 'fits', using one profile or another); here is a quick primer on the general topic of the distribution of observed 'brightness', within the galaxy, of various (classes of) galaxies, and various widely used models thereof: Structural Components of Galaxies (here is another, focussing on ellipticals: Properties of Elliptical Galaxies).

    So, how are all these things related? Here is one quick backgrounder: Intensity, Flux Density and Luminosity*, and the next page, Magnitudes.

    So, in terms of the physics, (observational) astronomers are measuring (and discussing) energy of something, per something (or some things): of detected electromagnetic radiation, per unit time (so it's more power than energy), per unit wavelength (or per unit frequency), per unit solid angle, ... and they often do so in terms of a unit ('magnitude') that's sometimes difficult to tie back to energy cleanly. IOW, it's not always straight-forward to convert an observational astronomer's 'intensity' to Janskys (Jy, 1 Jy = 10^-26 W m^-2 Hz^-1).

    * One fly in the ointment: "The geometry of the situation results in the interesting fact that the observed surface brightness is independent of the distance of the observer from the extended source" (at the bottom of the first of those pages); one of the key parts of the Disney&Lang paper, and of any research into is, concerns the fact that, in a GR-dominated universe, there are (in general) "departures from Euclidean angular size-distance relation" (to quote NGC3314).

  20. #50
    Join Date
    Mar 2004
    Posts
    13,441
    Here is a DR7 spectrum of a galaxy (from SDSS, obviously).

    Note the label on the y-axis: "Fλ [10-17 erg cm-2 s-1-1 ]". Clearly this is flux, with dimensions of ergs per second (so that's power, not energy), per square cm, per Angstrom (which is 0.1 nm) ... and a constant.

    Would any reader (other than StupendousMan, NGC3314, or parejkoj) like to have a go at expressing "10" of these flux-per-Angstrom as a (Pogson) magnitude?

    ETA: Here is the DR8 spectrum of the same galaxy.
    Last edited by Nereid; 2011-Oct-07 at 05:50 AM. Reason: Added DR8 spectrum

  21. #51
    Join Date
    Sep 2003
    Posts
    10,399
    Quote Originally Posted by Nereid View Post
    Here is a DR7 spectrum of a galaxy (from SDSS, obviously).
    Having read your prior post, I'm now bedazzled on what those magnitudes really represent for each band.

    Note the label on the y-axis: "Fλ [10-17 erg cm-2 s-1-1 ]". Clearly this is flux, with dimensions of ergs per second (so that's power, not energy), per square cm, per Angstrom (which is 0.1 nm) ... and a constant.
    Right, I get this. Integration would give you the actual power amount we receive.

    The example spectrum, if it were a star, would be somewhat close to a Planck temp. of roughly 4,300K -- compensating for redshift. I have no idea, of course, what value this might have in reprsenting galaxies, but it's something....maybe.

    Would any reader (other than StupendousMan, NGC3314, or parejkoj) like to have a go at expressing "10" of these flux-per-Angstrom as a (Pogson) magnitude?
    Right! We don't need no stinkin' astronomer. What good's a vulcanist to a dive-bombin' moth?
    We know time flies, we just can't see its wings.

  22. #52
    Join Date
    Sep 2003
    Posts
    10,399
    Tolman Diming mentioned from the start is intriguing.

    I went to New Wright's calculator web site and plotted this. [I don't know if he included abberation in his equations.]

    Ang Size for Expansion.jpg
    Attached Images Attached Images
    We know time flies, we just can't see its wings.

  23. #53
    Join Date
    Mar 2004
    Posts
    13,441
    Quote Originally Posted by Nereid View Post
    Here is a DR7 spectrum of a galaxy (from SDSS, obviously).

    Note the label on the y-axis: "Fλ [10-17 erg cm-2 s-1-1 ]". Clearly this is flux, with dimensions of ergs per second (so that's power, not energy), per square cm, per Angstrom (which is 0.1 nm) ... and a constant.

    Would any reader (other than StupendousMan, NGC3314, or parejkoj) like to have a go at expressing "10" of these flux-per-Angstrom as a (Pogson) magnitude?

    ETA: Here is the DR8 spectrum of the same galaxy.
    By chance, I was browsing astro-ph, and came across The Cosmic Origins Spectrograph.

    The abstract includes this: "For faint targets, with flux F_lambda ~ 1.0E10-14 ergs/s/cm2/Angstrom, COS can achieve comparable signal to noise (when compared to STIS echelle modes) in 1-2% of the observing time." And I thought, Huh?!? the galaxy whose SDSS spectrum I linked to surely isn't 'faint', yet from ~6000 to 9000 ┼, F_lambda is ~1.3E10-16 ergs/s/cm2/Angstrom!

    Reading the paper, I think I can see why SDSS galaxy spectra can be easily two orders of magnitude brighter than COS* ones (in F_lambda):
    By eliminating windows in the detector systems, the throughput is enhanced and the noise level is greatly reduced, resulting in large gains in signal to noise, particularly for the faintest targets with Fλ < 1.0 Î 10−14 erg cm−2 s−1−1 at 1200 ┼
    BTW, the galaxy is SDSS J104155.65+074513.9, and CAS gives its photometric pipeline outputs as:
    fiberMag_r 18.62 mag
    petroMag_r 17.22 mag
    devMag_r 17.07 mag
    expMag_r 17.41 mag
    psfMag_r 18.60 mag
    modelMag_r 17.07 mag
    petroRad_r 5.810 arcsec

    * interestingly, the COS 'aperture' is 2.5", close to that of the SDSS spectrograph (which is 3")

  24. #54
    Join Date
    Sep 2003
    Posts
    10,399
    Quote Originally Posted by Nereid View Post
    The abstract includes this: "For faint targets, with flux F_lambda ~ 1.0E10-14 ergs/s/cm2/Angstrom, COS can achieve comparable signal to noise (when compared to STIS echelle modes) in 1-2% of the observing time." And I thought, Huh?!? the galaxy whose SDSS spectrum I linked to surely isn't 'faint', yet from ~6000 to 9000 ┼, F_lambda is ~1.3E10-16 ergs/s/cm2/Angstrom!
    Approximating the integration, that's a little less than about 3,000 photons per sec. [Sunlight at 1 AU is about thousand trillion times greater in flux (~ 1e18 photons per sec).]
    We know time flies, we just can't see its wings.

  25. #55
    Join Date
    Mar 2004
    Posts
    2,693
    COS works in the UV (where detectors still don't approach the 90% of a good CCD in the optical), and is designed specifically to use higher dispersion than in the SDSS - they mention 20 km/s in Doppler width which maps to roughly 0.05 A pixels, assuming Nyquist 2-pixel sampling, rather than the ~2 A/pixel of SDSS spectra.

  26. #56
    Join Date
    Mar 2004
    Location
    Ocean Shores, Wa
    Posts
    5,255
    Quote Originally Posted by Nereid View Post
    Jerry, George, and BigDon (and any other interested reader lurker): do you think you could sketch the steps you think should be taken to work out an answer?
    You a proposing a freshman astronomy approach to a problem that is much more complex than calculating the sky drop-out in different bandwidths.

    Look at the galaxies that we can see in the ultra deep Hubble fields. they are warped, torn and distorted. The lensing elements are both gravimetric and (likely) distorted by many pockets of intervening dust and gas. The Lyman forest tells us that the light is reaching us in tatters.

    Freshman calculations aside, the best we could hope for, is that the most distant events are directly mirrored by fairly local events and we can then make statistical assumptions based upon this premise. But current theory also dictates an evolution element; which is 'confirmed' by the fact that all of our freshman calculations indicate galaxies were brighter in the past than they are today. We can accept this at face value; but as our depth of knowledge about the spectral features of these most distant observations become clearer; if their root structure, metallicity and other evidences of evolution are found wanting; we should wonder how this new evidence fits with our best prior explanations.

    Years ago, someone on this board discribed a scientist as someone who immediately explains the reason behind something that they told you was impossible in last year's lecture. This year, we are giving a Nobel prize this to three scientists who 'found' what Einstien called his biggest blunder. Something is missing in this equation, throwing in another ad hoc parameter is not the best solution.

    What do I propose? I have already stated several times: Start with a shotgun of initial of possible theories and run the light through all the different wringers: Which one provides the best match with all of the modern evidence? Don't forget to remove your deeply ingrained preconceptions - which have been proven impossible - from your analysis.

  27. #57
    Join Date
    Sep 2006
    Posts
    1,429
    Quote Originally Posted by Jerry View Post
    What do I propose? I have already stated several times: Start with a shotgun of initial of possible theories and run the light through all the different wringers: Which one provides the best match with all of the modern evidence? Don't forget to remove your deeply ingrained preconceptions - which have been proven impossible - from your analysis.
    So, go ahead and do what you propose. No one is stopping you.

  28. #58
    Join Date
    Mar 2004
    Posts
    2,693
    Quote Originally Posted by Jerry View Post
    What do I propose? I have already stated several times: Start with a shotgun of initial of possible theories and run the light through all the different wringers: Which one provides the best match with all of the modern evidence?
    That is effectively an infinite universe of starting theoretical possibilities. Your low opinion of the intellect and intellectual honesty of my entire profession is noted, but surely you could propose a realistically attainable alternative?

  29. #59
    Join Date
    Mar 2004
    Posts
    13,441
    Quote Originally Posted by George View Post
    Approximating the integration, that's a little less than about 3,000 photons per sec. [Sunlight at 1 AU is about thousand trillion times greater in flux (~ 1e18 photons per sec).]
    Can you expand a little please?

    I'm not 100% sure what the "that" refers to (the COS number, or the SDSS example?); and in any case, I can't see how you arrived at the 3,000 photons/sec conclusion (the detectors in the SDSS spectrograph, and COS, certainly aren't 100%, especially after the losses before photons hit them, but 'faint = 3,000 photons/sec' seems wrong).

    One heuristic I've heard: 0 mag is 10,000 photons/sec at the top of the atmosphere, at 5500 ┼ (per square cm, per ┼). I've no idea how accurate it is (can we check it?), nor where it comes from, nor how widespread it is (among observational astronomers, in the 'optical').

  30. #60
    Join Date
    Mar 2004
    Posts
    2,693
    Quote Originally Posted by Nereid View Post
    C
    One heuristic I've heard: 0 mag is 10,000 photons/sec at the top of the atmosphere, at 5500 ┼ (per square cm, per ┼). I've no idea how accurate it is (can we check it?), nor where it comes from, nor how widespread it is (among observational astronomers, in the 'optical').
    The rule I heard from Joe Wampler in grad school was V=0 corresponds to 1000 photons/(cm2 second Angstrom) in the middle of the V band. I once did a more careful calculation and got something like 1090, but adding significant figures starts to need so many details (spectral shape, filter passband) that it ceases to have very general use.

Similar Threads

  1. Arp 291 A "wind effects" galaxy
    By RickJ in forum Astrophotography
    Replies: 1
    Last Post: 2010-Nov-28, 03:23 AM
  2. observational bias
    By toothdust in forum Against the Mainstream
    Replies: 23
    Last Post: 2008-Jul-18, 10:29 PM
  3. Replies: 2
    Last Post: 2007-May-09, 06:47 PM
  4. Bias in Papers
    By RussT in forum Against the Mainstream
    Replies: 13
    Last Post: 2006-Oct-24, 08:31 AM
  5. The artful bias of the BBC
    By Glom in forum Off-Topic Babbling
    Replies: 33
    Last Post: 2006-Feb-07, 11:08 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
here
The forum is sponsored in-part by: