07/21/2013:
Robert raises a host of very interesting points with this new thread. His post will have us ransacking the databases for weeks. I also thank Patrick again for making the original observations in enough detail that we all await the end of the bright moon cycle to chase after these goodies ourselves.
Robert is certainly right that we don’t have enough data to say anything very certain about the IC 4329 distance—especially his point about the Tully-Fischer’s uncertainties in instances where a galaxy’s disc rotation curve can’t reliably be determined. Robert, could I ask how much is certain about the disc rotation rates of S0s in general and whether line-of-sight (LOS) position angle affects rotational power-law distributions the way it does ellipticals? I notice that we don’t have any recent bolometric or spectral energy distributions for either of the pair. We also don’t have enough absolute motion data—we know the pair is receding from us at redshift z=0.01689, but that is line-of-sight or 3-D reference frame? The data sets are too imprecise to determine what angular velocity components exist, and along which axes. Said another way, we can’t parse Doppler reddening from recessional reddening. For that matter, we don’t have intergalactic medium reddening, either.
There are a couple of other worthy lines of analysis than can help us determine whether Robert’s basic point—eyepiece angular size and brightness evidence as we observe it gives one to believe this pair is closer than its literature data say. I think all of us could take a visual look at these two and compare their apparent sizes with other Cen/Norma galaxies of the same visual magnitudes. I’d love to know what Patrick’s second look will tell us after reading all this stuff. IC 4329 is vm 11.54 and 3.2 x 1.6 arcmin; and IC 4329A’s is vm 13.1 and size 1.3 x 0.4 arcmin. Sad for me, by the end of this bright moon cycle these galaxies will be getting rather low in the sky. They won’t give me much to work with due to my aperture limit of 200mm. Patrick, could you have another look at the estimated size of the 4329 core region? Does your 30 arcsec estimate still hold?
IC 4329A’s AGN core radiates at such high UV and xray luminosity that its absolute bolometrics must be devlish to measure. The AGN is certainly hot—
this 2009 A-J paper states that the AGN has an active 130-million solar-mass black hole which is the source of the AGN’s high energy yield in the 10-30 KeV xray band. The 10-30 KeV strength/distance ratio point to a Hubble redshift of z = 0.01605, which correlates to a comoving radius of 71 Kpc.
Robert’s querying the real distance of the Ab3574 cluster raises the issue of how quickly data becomes dated as new and better measuring equipment comes on line. Much of the Ab3574 data dates from the late-1990s back to the 1970s—15 to 40 years ago.
There are a couple of workarounds to this dilemma. One is the absolute magnitudes and metallicity functions of globulars associated with either of the pair. The GC metallicities of IC 4329 have been studied as recently as
Harris et al 2009. In general, GCs fall into three age bins: two-thirds are >10*Gyr, one-third are between 5 and 10*Gyr, and a small fraction have an age <5*Gyr. Most GCs are bimodal—they consist of two generations—a primordial very low metallicity one (metal-poor), and a second population roughly 200 Myr younger. The 1st generation is called a ‘red’ GC because it is slightly older and hydrogen-rich/metals-poor. The 2nd generation is higher in iron and the alpha elements because its stars were formed out of the SN 2b core-collapse and SN 1a white dwarf supernova of the first generation. The 2nd gen is called a ‘blue’ GC because its stars tend to be lower-mass and less evolved on the main sequence; this would mean a high abundance of stars under 0.8 solar masses. A globular regeneration cycle typically takes 200 Myr. There is usually no 3rd or later generation because the supernova cycle and hot AGB stellar winds of the 2nd generation eject nearly all remaining starforming gas. (The GC NGC 2808 has a third & bluer main sequence, which has GC folks happily scratching their heads and plotting scope-time strategies.)
Earth-bound and orbital spectroscopy have really taken off since around 2006. Grating resolutions have gone from 25,000 to 46,000 in that time, and signal-to-noise ratios are now 120-to-1 instead of the 40-to-1 of a decade ago. We have ROSAT and GALEX these days. With a cluster as remote as Ab3574, astronomers need ever photon they can get because metallicity bimodality depends on the accuracy of colour bimodality. Colour bimodality data depend in turn on the magnitude differences between the red giant branch and the red clump (the red end of the horizontal branch), and the magnitude difference between the blue horizontal branch and the main sequence turnoff point. A GC bimodality difference of Δ(B – I)≈0.3 would correspond to age differences of 4.5 Gyr and 7.5 Gyr for metallicities of [Fe/H] = –2.25 and –1.5, respectively. If we know the above red/blue magnitude data, we know the GC absolute magnitudes from which to calculate their distances.
A typical GC has a half-light radius of 3 pc (a bit under 10 light years). A GC at IC 4329’s estimated distance of 58 Mpc would subtend an angle of 0.02 arc sec, which is smaller than the earth-based scopes’ pixel resolution even on nights of superior seeing at 0.5 arcsec or better. The Hubble Wide-Field/Planetary Camera resolves to 0.01 arcsec, but all that means is a 2-pixel dot instead of a one-pixel dot. In our best equipment IC 4329’s GCs still appear star-like. All we can do is measure their blue-to-red ratios. Hence we have poor spectral distribution numbers for the IC 4329 pair. Here’s the problem in graph form, based on 2006 data from the Gemini-South GMOS camera (
Cockcroft, Harris et al, Astronomical Journal 138 (2009) 758):
The dotted boxes in Image 1 below [Figure 2 in the linked Cockcroft-Harris article above] show us the sampling limits of IC 4329’s globulars. The sample consisted of 3558 de-contaminated and then measured objects. Instrument error sharply limits the data set to the small sample within the dotted boxes. The vertical scale is the important one in this graph: it basically puts an error limit on data reliability at one-tenth of a magnitude.
In Image 2 attached [Figure 4 in the Cockcroft-Harris article] is where the IC 4329 globulars lie on a normalised line radiating from the core to the instrumental limits of observation. The dotted vertical lines demark the inner zone of reliable data, de-noised for proximity to the galaxy core, and the outer zone where field objects contaminate accuracy:
Even constrained like this, some C-M diagram data can be extracted from this data. Image 3 below [Figure 3 in the article] shows three CMDs of galaxies with similar brightness and distance properties, one of them IC 4329. The objects in the dotted boxes are the 3 galaxies’ globular clusters binned by colour and brightness.
And finally (drum-roll, please), Image 4 below [Figure 6 in the article] is the resulting CM diagram, plotted as bins of the g (green-visual 551 nm) minus i (infrared 806 nm) colour bands. The vertical measure is luminosity and the horizontal measure is colour band.
In this plot, the upper set of large squares are the measured blue (lower metallicity) population and the bottom squares are the red globular population. (The lines are means lines for various interpretive calculations for the globulars not important to us here.)
All of this sheds light on the interpretation of IC 4329 as a S0 galaxy with a spiral-like GC distribution. In general, spiral GCs are more alpha-enhanced in the halo (more SN 2b contribution) and iron-group enhanced (more 1a contribution) towards the core. This points to a spiral ancestry for the S0 IC 4329, but it doesn’t help us determine how far away IC 4329 is, or its combined vector motions within the Hubble Flow.
IC 4329’s globulars have a mean recessional velocity of 4808 ± 21 km/sec in the CMB reference frame. That would locate Abell 3574 as a member of the Hydra-Centaurus Supercluster. But we knew this already, so all we’ve done here is go to considerably lengths to eliminate IC 4329’s GCs as concluding anything either way with regard to Robert’s point. (It was fun, though.)
A second line of reasoning is related to the black hole in the core of IC 4329’s recently interacted neighbour IC 4329a. The two galaxies had a grazing brush with each other, the xray remnant of which is a
hot spot in a bridge between the two:
Note that in
Robert’s image of IC 4329A, there appears to be slight deformation of the outermost part of the visible disc, bending at the tip CCW in the direction towards 4329 itself; there is a somewhat less noticeable deformation of the upper left part of the disc as well. (The whole thing looks like a skinnier version of the Integral Sign Galaxy.) An ancient galactic graze would not boost our knowledge of the galaxies’ present distance, but it adds to our evidence whether IC 4329 is an S0 or E. A shallow graze would add no net globular accretion to either galaxy (though a few might change hands), since the mass-metallicity relationship of ellipticals is sharply weighted to the blue due to gas impoverishment during and after an elliptical’s formation. The Milky Way’s MMR is red-rich and blue-poor; it still has plenty of gas for new, metals-enhanced star formation. One argument in favor of NGC 104, the Sombrero, being an S0 with a disc is that it has a blue MMR (
Rhode & Zepf 2004). OTOH, M49 has no blue MMR and astronomers have no explanation (
Strader et al. 2006). Go figure.
I wonder if a higher-rez image would reveal IC 4329A to be a warped remnant of the long-ago interaction. The comparative modesty of the warp with respect to the mass of IC 4329 points to a barely grazing or non-touching disc-disc interaction. The elevated-luminance bridge evident in the ROSAT image traces the path hot gas and railing stellar streams would take. Add that info to IC 4329’s blue-weighted GC population and I wonder whether we have in the IC 4329 pair an example of interaction disturbance turning a spiral turning into an S0.
The upshot of all this is that we need better, more modern data. It would be helpful to know (a) Na/O anticorrelation, CN correlation, [Mg-Ti/Fe] ratios from SN 1a contributions; and (b) Cr/Ni/Ba abundance and r-process neutron-capture abundances from Type 2b supernovae.
Unless Robert knows someone with access to Hubble or ESO proposal-vetting committees, we’ll probably be waiting awhile.
Wish I had a bigger scope. Two meters would be nice. Couldn’t afford the upkeep. Pity.
Thanks again to Robert and Patrick.
=Dana