Early Galaxy Pinpoints Reionization Era

by Nancy Atkinson on November 6, 2009

Want to stay on top of all the space news? Follow @universetoday on Twitter

This is a composite of false color images of the galaxies found at the early epoch around 800 million years after the Big Bang. The upper left panel presents the galaxy confirmed in the 787 million year old universe. These galaxies are in the Subaru Deep Field. Credit:  M. Ouchi et al.

This is a composite of false color images of the galaxies found at the early epoch around 800 million years after the Big Bang. The upper left panel presents the galaxy confirmed in the 787 million year old universe. These galaxies are in the Subaru Deep Field. Credit: M. Ouchi et al.


Astronomers looking to pinpoint when the reionozation of the Universe took place have found some of the earliest galaxies about 800 million years after the Big Bang. 22 early galaxies were found using a method that looks for far-away redshifting sources that disappear or “drop-out” at a specific wavelength. The age of one galaxy was confirmed by a characteristic neutral hydrogen signature at 787 million years after the Big Bang. The finding is the first age-confirmation of a so-called dropout galaxy at that distant time and pinpoints when the reionization epoch likely began.

The reionization period is about the farthest back in time that astronomers can observe. The Big Bang, 13.7 billion years ago, created a hot, murky universe. Some 400,000 years later, temperatures cooled, electrons and protons joined to form neutral hydrogen, and the murk cleared. Some time before 1 billion years after the Big Bang, neutral hydrogen began to form stars in the first galaxies, which radiated energy and changed the hydrogen back to being ionized. Although not the thick plasma soup of the earlier period just after the Big Bang, this star formation started the reionization epoch.

Astronomers know that this era ended about 1 billion years after the Big Bang, but when it began has eluded them.

We look for ‘dropout’ galaxies,” said Masami Ouchi, who led a US and Japanese team of astronomers looking back at the reionization epoch. “We use progressively redder filters that reveal increasing wavelengths of light and watch which galaxies disappear from or ‘dropout’ of images made using those filters. Older, more distant galaxies ‘dropout’ of progressively redder filters and the specific wavelengths can tell us the galaxies’ distance and age. What makes this study different is that we surveyed an area that is over 100 times larger than previous ones and, as a result, had a larger sample of early galaxies (22) than past surveys. Plus, we were able to confirm one galaxy’s age,” he continued. “Since all the galaxies were found using the same dropout technique, they are likely to be the same age.”

Ouchi’s team was able to conduct such a large survey because they used a custom-made, super-red filter and other unique technological advancements in red sensitivity on the wide-field camera of the 8.3-meter Subaru Telescope. They made their observations from 2006 to 2009 in the Subaru Deep Field and Great Observatories Origins Deep Survey North field. They then compared their observations with data gathered in other studies.

Astronomers have wondered whether the universe underwent reionization instantaneously or gradually over time, but more importantly, they have tried to isolate when the universe began reionization. Galaxy density and brightness measurements are key to calculating star-formation rates, which tell a lot about what happened when. The astronomers looked at star-formation rates and the rate at which hydrogen was ionized.

Using data from their study and others, they determined that the star-formation rates were dramatically lower from 800 million years to about one billion years after the Big Bang, then thereafter. Accordingly, they calculated that the rate of ionization would be very slow during this early time, because of this low star-formation rate.

“We were really surprised that the rate of ionization seems so low, which would constitute a contradiction with the claim of NASA’s WMAP satellite. It concluded that reionization started no later than 600 million years after the Big Bang,” remarked Ouchi. “We think this riddle might be explained by more efficient ionizing photon production rates in early galaxies. The formation of massive stars may have been much more vigorous then than in today’s galaxies. Fewer, massive stars produce more ionizing photons than many smaller stars,” he explained.

The research will be published in a December issue of the Astrophysical Journal.

Source: EurekAlert

About 

Nancy Atkinson is Universe Today's Senior Editor. She also is the host of the NASA Lunar Science Institute podcast and works with Astronomy Cast. Nancy is also a NASA/JPL Solar System Ambassador.

Don Alexander November 6, 2009 at 3:24 PM

Typical press release… Gives “age after the Big Bang” but not the REDSHIFT… :|

Well, after reverse-engeneering it, I think it’s actually the z = 6.96 galaxy reported years ago in Nature already.

Jon Hanford November 6, 2009 at 5:54 PM

@ Don, would that z=6.96 galaxy be IOK-1 ( http://arxiv.org/ftp/astro-ph/papers/0609/0609393.pdf )? If so, why would this photometrically measured redshift be especially relevant compared to the 2006 spectroscopically derived redshift (possibly as means to confirm the validity of the Lyman dropout value derived with their new apparatus?).

Jon Hanford November 7, 2009 at 5:40 AM

After parsing the press release and re-reading the paper I linked above, it appears this is indeed an extension of the work published in the 2006 paper. The current PI, Masami Ouchi, appears as a coauthor on the 2006 paper, which mentions in its summary that new more sensitive observations of the Subaru Deep Field would be conducted using newer red sensitive CCDs in the SuprimeCam on the 8.2m Subaru (not 8.3m as noted in Carnegie and Eurekalert press releases). In the 2006 paper, only two ‘Lyman-alpha Emitters’ were positively detected using the ‘dropout’ technique (IOK-1 & IOK-2), but only IOK-1 had a detectable redshift using Subaru’s spectrograph. The PR states that 22 LAEs have been found now in the SDF. Interestingly, page 7 of the 2006 paper has 6 band images of IOK-1 & 2 using their older detectors.

Lawrence B. Crowell November 7, 2009 at 6:15 AM

There is of course a 1-1 relationship between z and the radial factor in the FLRW spacetime metric. So

z = a(t)/a(t0) = [1 + (t - t0)H]^{-1} .

This does suggest that gravitational clumping of matter and early star formation (popIII stars I presume) happened rather rapidly. This might suggest more local inhomogeneous distribution of matter than thought.

In the WMAP data on the distribution of anisotropy that fit the Legendre polynomial distribution I always thought it funny that the lowest term fell short of the model curve.

LC

Jon Hanford November 8, 2009 at 7:21 AM

LBC, have you seen any work concerning “Dark Stars” as possible candidates for the first stars? Just read a 2008 paper ( http://arxiv.org/PS_cache/arxiv/pdf/0812/0812.4844v1.pdf ) that explores whether the first stars to form may have been powered by DM annihilation instead of fusion.

The beginning of the paper on these WIMP powered stars states “The first stars to form in the universe, at redshift z~10-50, may be powered by dark matter annihilation for a significant period of time.”.

The paper notes these would be very massive (100-800 solar mass, similar to Pop III stars) and may point to a mechanism for formation supermassive black holes seen in the very early universe (SMBH 10^9 solar mass at redshift z=6).

Makes for an interesting read, but how plausible are these objects (the paper points to discriminating factors between Pop III stars and Dark Stars)?

Lawrence B. Crowell November 8, 2009 at 8:01 AM

Thanks for that paper. This idea has been kicked around for a while. Of course you have to remember I am not an astronomer or astrophysicist, but a general relativist and quantum field theory wog. Yet these ideas or varieties of this idea have been kicked around for a the last couple of years.

I mostly log on here to catch some of the astrophysics and astronomy stuff I might not hear about otherwise.

The one question I have about the idea of dark stars is how they actually form. While it is popular to think of black holes as vacuum cleaners that suck everything up, it is actually hard to get something into a black hole. This is counter intuitive, but assume from a safe distance you throw something at a black hole and just miss by say about 10 times the Schwarzschild radius. The object will go into an orbit, which of course a periapsis precession, but in a pure two-body problem it is in an “eternal orbit. What causes things to go into the black hole is when you have a collection of particles or a gas that exhibits friction or attenuating physics that degrades the collective orbit of the stuff. Accretion disks are the perfect example of that. This is the same with any gravitational clumping. There has to be some attenuating process which reduces the total energy

E = p^2/2m – GMm/r

(summation over p and M and m assumed, and gravity by itself does not do that exactly.

Dark matter does not strongly interact with anything, including itself. It is “cold” as a result. So from a model position it is hard to know what mechanism would cause dark matter to accumulate rapidly. Maybe some virial process where some dark matter was removed further out by gaining energy, say maybe making a galactic halo, while the stuff which loses energy to the outward moving DM implodes inwards. Then maybe neutralino annihilations take place.

I am suggesting the above as one possible model. I am not sure if others have worked this sort of thing up. The equation 2 for the rate per volume of annihilations will only result in significant channel processes

nt + nt-bar –> gamma + Zinos and photinos

if the volume these particles sit in is made small enough. Maybe there is some astrophysical work to be done here.

Cheers LC

Jon Hanford November 8, 2009 at 12:33 PM

LBC: thanks for your thoughtful, thought-provoking and timely response to my query.

“Then maybe neutralino annihilations take place.” Of course you may follow the many celestial and terrestrial observatories searching for signs of DM both in the heavens but also down here on earth (and maybe from the sun :) ). These are fertile years in the quest for understanding DM, and the quest has barely begun.

Jon Hanford November 8, 2009 at 12:42 PM

The Dark Star hypothesis was elaborated on in a 2007 article from PhysOrg: http://www.physorg.com/news115880789.html .

Jon Hanford November 8, 2009 at 12:45 PM

Of course, the Wiki page has many interesting links to the Dark Star hypothesis: http://en.wikipedia.org/wiki/Dark_star_%28dark_matter%29 . This is where I found the link to the 2008 paper on Dark Stars.

Lawrence B. Crowell November 8, 2009 at 8:02 PM

I am not sure how much dark matter there is in say the solar system. If I were to assume there is some density of dark matter in the solar system, rho, then we can easily use Gauss’ law

int nabla^2V dv = int F-da = int dv*rho,

where a is an increment of area which encloses the volume V and dv are increments in volume. For the Gaussian surface inside the mass it is easy to get

F = (4pi/3)G*m*rho*r

which is a harmonic oscillator force on the mass m. So any planet or asteroid in the solar system would have this perturbation in their orbit.

The orbits of the planets are very well established and I suspect the error bars for this DM in the solar system pretty small. I doubt that DM accumulates inside ordinary materail bodies. The reason is that if they are noninteracting they just pass right through the sun or the Earth and continue on what ever orbit they were on. There is nothing which would attentuate the energy of these particles.

The idea of the dark star seems to only make sense as something which results from a virial theorem. We might assume clumps of DM in the early universe which interact gravitationally to cool down some of the DM into a slowly collapsing bunch and conversely heat up some of it into the galaxy halo. The collapsing dark star might then be a seed for galactic black holes.

This is somewhat reminiscent of hyperstars that Feynman and Wheeler tried to work up back in the 1950-60s.

Cheers LC

Jon Hanford November 9, 2009 at 12:12 AM

The “Dark Stars” paper mentions that these stars would appear quite luminous (L=10^6 solar) but their surfaces would be rather cool (6000-10000K) as opposed to Pop III stars (>30000K). This would allow some discrimination for whatever detectors will search this early epoch.

Lawrence B. Crowell November 9, 2009 at 10:09 AM

The authors of this paper seem to consider the dark star as something which might transition from DM annihilation powered to fusion powered.

I am not an astrophysicist, and certainly not one intimately familiar with the massive computer modeling done. Yet I could imagine a sort of model which might go as follows. The universe around 500my after the big bang has anisotropic and inhomogeneous distributions which are amplified due to some Virial process where accumulated matter has a kinetic energy equal to half the potential and due to “friction.” The Virial process ejects lots of DM into higher kinetic energy and begins for form the halo of galaxies. This friction is largely with ordinary matter, which interacts with itself strongly, but this triggers an amplification of gravitational clumping. Now we consider the primary species of dark matter as the neutralino, which might be a Majorana spinor field and act as its own anti-matter. This depends upon what the SUSY partner of the Higgs field is. The accumulated DM begins to generate radiation and if there is enough that might initiate fusion as well. The process begins to “run away” where luminous matter flows in. Eventually this object implodes into a black hole, forming the early galactic black hole.

A DM fired star which initiates the PopIII star might work, for one problem with these stars is that pure hydrogen has a low opacity and the energy produced rapidly escapes rather than being trapped to further continued nuclear fusion. Yet if there is a DM source of energy that might be sufficient to keep the system going. For large enough clumps of radiating matter they might then over time implode into a large black hole.

Well, thought for a Monday morning.

Cheers LC

Jon Hanford November 10, 2009 at 10:14 AM

What I like in this theory is a plausible and rather straightforward way to generate IMBHs and SMBHs early on in the universe. This is one possible answer to this largely unknown process.

Lawrence B. Crowell November 10, 2009 at 1:24 PM

I might be something interesting to numerically model. If one were to set up a lot of particles in motion, maybe according to a hydro-code, then the evolution might clump some of this matter into the center with increased radiation production from neutralino annihilations and the eventual production of compact bodies towards the center.

Cheers LC

Comments on this entry are closed.

Previous post:

Next post: