Thursday, January 15, 2009

Press releases from last week's AAS meeting in Long Beach, CA.


A variety of astronomy highlights and press releases from last week's AAS meeting in Long Beach, CA made it into the popular science media, in particular Space.com, but also the BBC and the NYT (unless otherwise noted, the links are to a Space.com article). I chosen to cover four stories from the AAS that made it into the popular science media.

Unfortunately the meeting website and the AAS don't seem to link to or archive the these press releases. In some cases all that is available is the pop science write-up, which may or may not accurately reflect what the scientists involved wanted to present. It used to be the case that press releases only happened on the publication of a full peer reviewed Journal article, but this is the AAS where you typically announce work-in-progress and so full papers may still be in the works.

(1.) "Mystery Roar from Faraway Space" by Andrea Thompson discusses an excess component in the Extragalactic Radio background (in the 3 to 90 GHz band) discovered in data from the ARCADE 2 balloon experiment. Emission in this band is a combination of a signal due to the Cosmic Microwave Background (the remainder radiation from the Big Bang), a foreground signal from our own Galaxy, and the integrated emission from extra-galactic sources (in particular galaxies). The claim is that after accounting for known sources and the expected contribution from galaxies there is a residual component to the radio background with a spectral index of -2.56 +/- 0.04 (the actual submitted paper is Seiffert et al, arXiv:0901.0559).

The actual paper spends quite a bit of time trying to demonstrate that they have good "evidence for isotropic radio emission... beyond what can be explained Galactic emission and the unresolved emission from the known population of discrete sources" [i.e. galaxies], but the paper doesn't really speculate as to what the source of this component is. The argument that this excess can not be due to galaxies is a pretty good one. Astronomers have a good idea of how much radio emission to expect from all the galaxies in the Universe because galaxies obey the relatively tight radio/FIR correlation (the massive stars whose explosions ultimate create the cosmic ray electrons responsible for galactic radio emission also heat dust to IR-emitting temperatures), and if galaxies were responsible for the excess radio emission then we'd expect a higher IR background than we see.

Thompson's write up (nicely) doesn't go beyond the submitted paper: 'For now, the origin of the signal remains a mystery. "We really don't know what it is," said team member Michael Seiffert of NASA's Jet Propulsion Laboratory in Pasadena, Calif.'

This story is also covered by Dennis Overbye at the NYT on January 7th ("Theory Ties Radio Signal to Universe's First Stars"). This article goes rather further than the actual submitted paper, spending quite a bit of time discussing Alan Kogut's (NASA/GSFC) admittedly "wildly speculative" suggestion that the excess emission originates from black holes formed by the collapse of the first stars to form in the Universe (Population III stars).

(2.) "Explosions Starve Black Holes" by Andrea Thompson discusses results presented by Daniel Wang (Hi Daniel), but I'm confused as what new result its reporting, or whether its theory or observational data despite the subject matter (galaxies, X-rays, supernova-heated hot gas) being very close to stuff I do. I could not find an actual press release on the UMass or JPL web sites, nor a paper on arXiv. The the idea that supernova heating in prevents/explains the lack of accumulation of large masses of cool/warm interstellar medium (by cool or warm I mean gas with temperatures less a few million degrees) from recycled gas from evolved stars in old stellar populations (ellipticals, bulges) is not new, dating back to the 1960's. That AGN will accrete less in a hot low density ISM is also
self-evident. Nevertheless, despite my lack of understanding regarding the news item I've included it in this list because I know Daniel and hot gas in galaxies also deserves more media attention.

(3.) "Surprise! Milky Way Much More Massive" by Andrea Thompson discusses new Very Long Baseline Array (VLBA) radio observations of Masers that indicate that the rotational velocity (and hence mass) of our Milky Way galaxy is higher than previously thought (by approximately 15% above the old standard of rotation speed 220 km/s, and ~50% higher in mass). This is also covered by Kenneth Chang at the NYT ("Data Uncover Bigger Galaxy in Cosmos, and It’s Ours" ), at the BBC ("Milky Way 'bigger than thought'" ), and NRAO's full press release ("Milky Way a Swifter Spinner, More Massive, New Measurements Show") is here. The image shown at the start of this post is taken from the NRAO press release, and shows an artist's impression of the Milky Way (as viewed from above) along with our location (red circle) and the location of the maser sites (green, blue circle?).

The basic method is to exploit the amazing spatial resolution given by very long-baseline radio interferometry to measure trigonometric parallaxes (and hence the distance) to methanol masers (simpler wikipedia article on masers here) in massive star forming regions in our galaxy. This is a direct distance measurement, and hence doesn't rely on multiple error-prone steps in the Cosmological Distance Ladder. From the doppler shifts of the radio emission one also has the line-of-sight velocity of the source with respect to us. Thus you have size and velocity measurements, although how you go beyond that to get a rotation model for the Galaxy isn't immediately obvious to me (remedial reading of Galactic Dynamics or Galactic Astronomy is hereby self-assigned).

Unfortunately this is another result that doesn't appear (AFAIK) to be associated with a publically available recent paper (no obvious arXiv preprint, no published paper that I can find, no paper mentioned in the NRAO press release). Papers on distances to individual star forming by members of this team are on arXiv, but don't mention changing the previously accepted value of the Milky Way's rotational velocity (Brunthaler et al, arXiv:0811.0713; Rygl et al, arXiv:0812.0905 ).

However, I did remember this issue was mentioned in Shattow and Loeb's paper on the LMC 's orbit (2008, MNRAS, 392, L21 ), and sure enough they say: "Also, within the past five years, the circular velocity of the MW and the distance between the Sun and the Galactic Centre have been updated by Reid & Brunthaler (2004; hereafter RB04) and Gillensen, Genzel & Eisenhouer (private communication; hereafter GGE08), respectively. These increased the likely circular velocity of the MW from the IAU standard of Vcirc = 220 to 251 ± 15 km s−1. A value of 220 km s−1 now corresponds to a reduction of the best-fitting value by two standard deviations (and equivalent to moving the Sun a total of 1 kpc closer to the centre of the Galaxy). Uemura et al. (2000) used parallax measurements from Hipparcos and SKYMAP to obtain a similar value of Vcirc = 255 ± 8 km s−1."

Here are the links to the Reid and Brunthaler (2004) and Uemura et al (2000) papers:
1. "The Proper Motion of Sagittarius A*. II. The Mass of Sagittarius A*," Reid and Brunthaler, 2004, ApJ, 616, 872. (Note that this paper gives theta_0 = 236 +/- 15 km/s, with the error dominated by the 0.5 kpc uncertainty in R_0.)
2. "Galactic Rotation Derived from OB Stars Using Hipparcos Proper Motions and Radial Velocities," Uemura et al, 2000, PASJ, 52, 143. They obtained v_theta = 255 +/- 8 km/s.

So although this result might actually be a few years old and not "hot of the presses" new, it certainly was news to me and is a nice example of how we continue to reassess the nature of our own Galaxy.

(4.) "Black Holes Preceded Galaxies, Discovery Suggests" by Andrea Thompson. This is a story that got a lot of press and blog attention, much of which typically fails to include the important qualifier "suggests" that Thompson includes in her title (good for you Andrea!). The BBC article is another offender ("Black holes 'preceded galaxies'" ). Interestingly this story was not touched by the NYT. The core scientific result is an impressive technical achievement and is scientifically interesting and important. Unfortunately the press releases go beyond what the data is saying.

One of the most striking differences between our current view of the supermassive black holes (SMBHs) that power AGN and the galaxies within which they reside, and what we knew say 15 years ago when I was in grad school is that we now that galaxy growth and black hole growth are linked some how. In the local z=0 (here and now) Universe, 13.7 billion years after the Big Bang, those galaxies that have SMBHs (not all do, many lower mass galaxies appear not to have SMBHs at all) have SMBHs that weigh typically about 0.1% of the galaxy mass (there are of course arguments over whether the fundamental relationship is BH mass to galaxy stellar (baryonic) mass or to just the mass of the galaxies Bulge stellar population, or the mass of the entire halo including the Dark Matter, but you get the picture).

This strongly suggests that galaxy growth (star formation from gas) and black hole growth (accretion of gas) are somehow linked and possibly regulate one another. This was a shocking result. Some astronomers now speculate that AGN control or limit star formation on galactic scales, other speculate that star formation regulates/limits accretion onto SMBHs, and other suggest that the size of the galaxy's dark matter halo ultimately controls both star formation and the accretion onto the SMBH. All very different and much more complex and almost ecological compared to the view back in the early 1990's where theories of galaxy formation and AGN growth were essentially separate, their evolution disconnected from one another.

The Carilli et al press release is basically about measuring the ratio of the mass of the SMBH to the galaxy in four high redshift AGN. Do these much younger galaxies, only a billion or two years old, have the same ratio as the older galaxies now do? That is, were the BH and stellar growth rates constrained by whatever process in the same way over the entire history of the Universe? This is a fascinating question, and the answer appears to be no. In Carilli et al's four high-z AGN the SMBH appears to be proportionally much more massive than in local galaxies. Or alternatively the galaxies (mainly stars) are less massive.

Whichever way you put it, the black holes in these particular galaxies in early Universe appear to have grown faster relative to the galaxy than the long term average we now see imprinted in galaxies at z=0 .
Qualitatively this result is not entirely unexpected (read the nice introduction to Haiman, 2006, New Astronomy Reviews, 50, 672 ) but actually managing to get quantitative measurements is a great piece of science nonetheless. Good job (so far) guys!

Its at this point things start to go wrong, IMHO. These galaxies they're observing are young, but not primeval - and they still have most of their mass in stars. The period is still ~1.5 billion years after the Big Bang and about a billion years after the first stars and galaxies began to form.

Yet the result is extrapolated to much earlier times and to discuss what came first: Black Holes or stars? “Black holes came first and somehow—we don’t know how—grew the galaxy around them,” Carilli says [sciam.com ]. "The implication [emphasis mine] is that the black holes started growing first." Fabian Walter is quoted as saying [BBC article ]. But the data itself doesn't tell you what happed. Its at least equally plausible that the first stars came before the black holes, but then the black hole growth rate briefly surged.

The data says that the SMBHs are proportionally more massive with respect to the galaxy mass at high redshift than at low redshift. A reasonable extrapolation of this is to say that the black holes grew proportionally faster than the galaxies, averaged over the entire 1.5 billion year period after the Big Bang. But its a big leap to say the Black Holes formed first. If the stellar/galactic and Black Hole growth rates are no longer inextricably coupled then we have less, not more, of a handle on the age!

Yet from reading the popular press Joe Public would get the incorrect impression that science has solved the problem and worked out that Black Holes came before galaxies. I only highlight this particular case because this seems me to be a typical problem with popularizing science and with press releases: The qualifying statements get stripped out, and the possible-but-speculative implications of a result end up getting higher billing than the real science itself.

As a side note, the SciAm article on this has an update mentioning that even the basic science result may not be robust but instead be a statistical artifact. No doubt this work will spur other groups to perform similar studies, so time will tell whether the results of relatively larger SMBH to galaxy mass in the early Universe holds true. Nevertheless, very interesting work despite the annoying PR spin.

Post a Comment