NYT: David Sobral et al "CR7" PopIII galaxy article

[marcus]

There was a NYT item today by Dennis Overbye about what I think is this:
http://arxiv.org/abs/1504.01734

CBS headline said "brightest galaxy in early universe". I don't yet understand what all the excitement is about. But I think that is the link to the professional paper that is getting the media buzz.
No time to pursue this further. Maybe somebody else can explain.

A galaxy at redshift 7 or 6.6 that has a bunch of PopIII stars or very very massive early universe stars, that last only 2 million years and then explode. something to do with "reionization era"

Haven't read the NYT piece, it might clarify. Have to go.

 

[Garth]

Thanks for the link Marcus.

It is good to confirm the existence of Pop III stars; a lot has to go on at high z, not least a lot of metallicity appears that has to be created somewhere.

With low metallicity themselves as they would have formed prior to this age, PopIII would have larger 'Jeans Mass'. As they have larger masses and much larger luminosities than Pop I or Pop II stars their lifetimes would be correspondingly shorter.

Garth

 

[marcus]

There was another thread started later today in Astrophysics about the same NYT article and PopIII observation. Websterling gave a couple of links (one of which was to the same David Sobral et al article on arXiv and one was to physicsworld, which I hadn't seen.

There was also this at physicsworld.com-
Getting the best glimpse of first-generation stars in the universe

and the paper on the arXiv-
Evidence for PopIII-like stellar populations in the most luminous Lyman-α emitters at the epoch of re-ionisation: spectroscopic confirmation

thanks to Websterling! The physicsworld article is really informative!
http://physicsworld.com/cws/article...pse-of-first-generation-stars-in-the-universe

 

[marcus]

Something really interesting going on here! There is a tricky reason why there is only a narrow window (z 6.5 to 7) where we can see PopIII.

PopIII stars don't have metal lines, obviously, and at such high redshifts we can only see their Lyman-α, which is so wavestretched that it is in the microwave.
But that is absorbed/scattered by neutral hydrogen.
So PopIII light cannot get thru from PopIII stars earlier than z = 7 because the intergalactic medium is not re-ionized enough.

for definiteness here is a table from S= 7.5 back thru 8 to 8.5
the wave stretch factor S = z+1 so that corresponds to z 6.5-7.5
$${\scriptsize\begin{array}{|c|c|c|c|c|c|}\hline R_{0} (Gly) & R_{\infty} (Gly) & S_{eq} & H_{0} & \Omega_\Lambda & \Omega_m\\ \hline 14.4&17.3&3400&67.9&0.693&0.307\\ \hline \end{array}}$$ $${\scriptsize\begin{array}{|r|r|r|r|r|r|r|r|r|r|r|r|r|r|r|r|} \hline a=1/S&S&z&T (Gy)&D_{now} (Gly)&D_{then}(Gly)&V_{now} (c)&V_{then} (c) \\ \hline 0.118&8.500&7.500&0.6962&29.297&3.447&2.03&3.30\\ \hline 0.119&8.394&7.394&0.7094&29.185&3.477&2.03&3.26\\ \hline 0.121&8.290&7.290&0.7229&29.073&3.507&2.02&3.23\\ \hline 0.122&8.187&7.187&0.7366&28.960&3.537&2.01&3.20\\ \hline 0.124&8.085&7.085&0.7505&28.846&3.568&2.00&3.17\\ \hline 0.125&7.984&6.984&0.7648&28.732&3.599&2.00&3.13\\ \hline 0.127&7.885&6.885&0.7793&28.617&3.629&1.99&3.10\\ \hline 0.128&7.787&6.787&0.7941&28.501&3.660&1.98&3.07\\ \hline 0.130&7.690&6.690&0.8091&28.384&3.691&1.97&3.04\\ \hline 0.132&7.594&6.594&0.8245&28.267&3.722&1.96&3.01\\ \hline 0.133&7.500&6.500&0.8401&28.149&3.753&1.95&2.98\\ \hline \end{array}}$$

You can see how it would be: it's only the LATE forming PopIII stars we can see, that formed between say year .75 billion and 0.8 billion. Before 0.75 billion plenty of the huge PopIII stars existed, living their short intense lives and exploding. But the reionization period hadn't run its course and the neutral hydrogen blocks their wave stretched light.

 

[marcus]

http://arxiv.org/abs/1504.01734
==David Sobral et al page 1==
... (the need for deep near-infrared exposures). At these redshifts (z > 6.5), the Lyα line is virtually the only line available to confirm sources with current instruments. However, Lyα is easily attenuated by dust and neutral hydrogen in the inter-stellar and inter-galactic medium. Indeed, spectroscopic follow-up of UV-selected galaxies indicate that Lyα is suppressed at z > 7 (e.g. Caruana et al. 2014; Tilvi et al. 2014) and not a single z > 8 Lyα emitter candidate has been confirmed yet (e.g. Sobral et al. 2009; Faisst et al. 2014; Matthee et al. 2014). If the suppression of Lyα is mostly caused by the increase of neutral hydrogen fraction towards higher redshifts, it is clear that z ∼ 6.5 (just over 0.8 Gyrs after the Big Bang) is a crucial period, because re-ionisation should be close to complete at that redshift (e.g. Fan et al. 2006)
==endquote==

==Sobral et al, from the abstract==
Our findings are consistent with theoretical predictions of a PopIII wave, with PopIII star formation migrating away from the original sites of star formation.
==endquote==
My take on this is that these PopIII stars can be a CHAIN REACTION, because they all quickly explode (lifetime ~ 2 million years) compressing the next bunch of gas, so it can start to collapse and form a bunch more of these giant stars, which then soon explode and send a shock wave of compression thru the gas, triggering more collapsing into stars. So there is a kind of ripple effect.

So this wave of PopIII stars could have gotten started much earlier sometime after year 0.5 billion and progressed along until around year 0.8 billion and only then was the interstellar medium hot enough and ionized enough to be transparent to their light and let it through this narrow window, around z ~ 6.5

 

[marcus]

Helpful explanation in physicsworld piece by Susanna Kohler
http://physicsworld.com/cws/article...pse-of-first-generation-stars-in-the-universe

==excerpts==
...
...
...A population that has remained conspicuously absent, however, are the very first stars, which should have formed before the recycled materials existed. These hypothetical, extremely metal-poor stars, termed "population III" stars, should have started forming between 10⁶ and 10⁷ years after the Big Bang. Not only would these early stars have been extremely hot and enormous – several hundred or even a thousand times more massive than the Sun – they would have exploded as supernovae after only about two million years, seeding the universe with the metals to form population II stars. But while they have been theoretically predicted, they are yet to be directly detected, as spotting these stars is very difficult. They were extremely short-lived and would have shone at a time when the universe was largely opaque to their light, towards the end of the "dark ages".

A team led by David Sobral at the University of Lisbon, and Leiden Observatory in the Netherlands, may have changed this paradigm with the recent detection of an extremely bright galaxy in the early universe. Indeed, as the team's survey looks at exceedingly distant galaxies, it let's us look back in time, revealing the universe as it was a mere 800 million years after the Big Bang. The survey uncovered several unusually bright galaxies, including the brightest galaxy ever seen at this distance – an important discovery in itself.

But further scrutiny of this galaxy, named CR7, produced an even more exciting find – a bright pocket of the galaxy contained no sign of any metals, and further observations with other telescopes confirmed this initial detection. "By unveiling the nature of CR7 piece by piece, we understood that not only had we found by far the most luminous distant galaxy, but also started to realize that it had every single characteristic expected of population III stars," says Sobral.

Formation waves
Sobral and his team postulate that we are observing this galaxy at just the right time to have caught a cluster of population III stars – the bright, metal-free region of the galaxy – at the end of a wave of early star formation. The observations of CR7 also suggest the presence of regular stars in clumps around the metal-free pocket. These older, surrounding clusters may have formed stars first, helping to ionise a local bubble in the galaxy and allowing us to now observe the light from CR7.

It was previously thought that population III stars might only be found in small, dim galaxies, making them impossible for us to detect. But CR7 provides an interesting alternative: this galaxy is bright, and the candidate population III stars are surrounded by clusters of normal stars. This suggests that these first-generation stars might in fact be easier to detect than was originally thought.The work is to be published in the Astrophysical Journal. A preprint is available on arXiv.
==endquote==

So earlier PopIII stars at the first part of a "wave" of giant star formation could have enriched the star forming regions AROUND the pocket of pure metal free gas. And that allowed ORDINARY stars to form and they helped (with their UV) to ionize the medium around that pocket and make it transparent to the later PopIII end of the wave stars.

Sobral et al are presenting a scenario with a DIFFERENT TIMING in which the formation periods overlap. There is still some PopIII forming in metal-free gas going on even after the surrounding regions have lots of normal stars (metals facilitate dumping waste heat of condensation and allow smaller mass, longer lifetime, stars to form). These normal stars help to make the medium transparent.

 

[marcus]

This is interesting stuff, and NYT Dennis Overbye is a good science journalist so want to thank PF member DiogenesNY for spotting the NY Times article and quoting. I was too busy earlier this morning to quote
==Diogenes NY==
I thought this might be of interest: NY Times article on evidence of pop III stars.

diogenesNY

--------------------------

New York Times - June 17, 2015

Astronomers Report Finding Earliest Stars That Enriched Cosmos
By DENNIS OVERBYEJUNE 17, 2015

Astronomers said on Wednesday that they had discovered a lost generation of monster stars that ushered light into the universe after the Big Bang and that jump-started the creation of the elements needed for planets and life before disappearing forever.

Modern-day stars like our sun have a healthy mix of heavy elements [...]

[article continues] http://www.nytimes.com/2015/06/18/s...-region&region=bottom-well&WT.nav=bottom-well
==endquote==
I'll quote some more excerpts from Dennis Overbye's excellent article:
==Overbye excerpts==
...
...
Modern-day stars like our sun have a healthy mix of heavy elements, known as metals, but in the aftermath of the Big Bang only hydrogen, helium and small traces of lithium were available to make the first stars.

Such stars could have been hundreds or thousands of times as massive as the sun, according to calculations, burning brightly and dying quickly, only 200 million years after the universe began. Their explosions would have spewed into space the elements that started the chain of thermonuclear reactions by which subsequent generations of stars have gradually enriched the cosmos with elements like oxygen, carbon and iron.
...
...
Now, in a paper to be published in The Astrophysical Journal, an international crew of astronomers led by David Sobral of the University of Lisbon, in Portugal, and the Leiden Observatory, in the Netherlands, said they had spotted the signature of these first-generation stars in a recently discovered galaxy that existed when the universe was only about 800 million years old. Its light has been traveling to us for 12.9 billion years, while succeeding generations of stars have worked their magic to make the universe interesting.

The galaxy, known as CR7, is three times as luminous as any previously found from that time, the authors said. Within it is a bright blue cloud that seems to contain only hydrogen and helium.

In an email, Dr. Sobral called this the first direct evidence of the stars “that ultimately allowed us all to be here by fabricating heavy elements and changing the composition of the universe.”
...
...
Garth Illingworth, an astronomer at the University of California, Santa Cruz, and a veteran of the search for early galaxies, pointed out, however, that these stars were appearing far later in cosmic history than theory had predicted.
...
...

As in much of astronomy, the nomenclature of these star generations is awkwardly rooted in history and Earth-centered. Modern stars like the sun, with healthy abundances of so-called metals (anything heavier than helium), are now called Population I, mainly because they were the first known. They mostly inhabit the spiral arms and younger parts of galaxies like the Milky Way. In the middle of the 20th century, however, the astronomer http://www.phys-astro.sonoma.edu/brucemedalists/baade/noticed that the stars in older parts of the galaxy, like its core or globular clusters, are older and have fewer metals. He called them Population II.

The advent of the Big Bang theory of the origin of the universe forced astronomers to realize that the first stars must have had no metals at all; those are known as Population III.

Stars of both Population II and Population III are probably present in CR7, Dr. Sobral and his team report.

While the blue cloud is metal-free, according to spectral measurements, the color of the rest of the galaxy is consistent with more evolved stars making up most of its mass. This suggests, they write, that the Population III stars there are late bloomers of a sort, forming from leftover clouds of pristine material as the galaxy was sending out its light 12.9 billion years ago.
...
...
==endquote==

 

[Garth]

Thank you Marcus, diogenesNY and Websterling for your threads.

It has been a mystery why we couldn't see Pop III, and now we know they are there but their Lyman $\alpha$ emission is absorbed by dust and HI.

It is also interesting that Garth Illingworth commented they "were appearing far later in cosmic history than theory had predicted". So these must be stragglers of some sort, presumably their formation process is not simultaneous but spread out over a period of time.

One interesting question is what happened next? If, as super-sized stars they go supernova then presumably these would be extremely bright hyper-novae that should be easy to spot.

Furthermore such hyper-novae would be expected to leave behind a population of black holes, perhaps the progenitors of the SMBHs that have been detected even at high red shift.

Garth

 

[Ken G]

What I am not understanding is why there needs to be normal stars to make the surroundings of the pristine region transparent. Hot sars emit light copiously in the ultraviolet continuum, which would not care about hydrogen lines in neutral hydrogen. And even if these pop III stars emit huge Lyman alpha lines (which I gather is the claim being made), would they not be emitted in a wind, so broadened by the wind speed? Wind speeds for massive stars are usually pretty fast, so those lines should be way broader than any absorbing lines in the interstellar medium around the stars. So why do we need to ionize hydrogen to be able to see stars? And if we do need to ionize hydrogen, we only need stars, not normal stars. In fact, normal stars put most of their mass into low-mass stars that do not ionize hydrogen, so we'd be much better off putting that mass into pop III stars that pump out ionizing photons far more efficiently. Finally, I don't see why this new information is being touted as an answer to a puzzle-- it seems to me it creates a puzzle that did not exist before: why did these pop III stars wait so long to form, if they are in huge galaxies? Is there some source of pristine gas that is infalling into that galaxy that is not mixing with what is already there? It seems to me the question is not why can we see these stars (though I don't understand the explanation given for that), it's how did they get there. So there are just some elements of this story that are not hitting home for me, though I confess I have not read all the material cited.

 

[marcus]

Ken G, you are asking a bunch of very substantive questions it seems to me. I'd guess you probably know more about it than i do and have thought more carefully about PopIII stars and the end of the reionization. I'll add one point in case it might help resolve some of these puzzles.

Isn't it true that we can't see individual stars in a galaxy like CR7? So what does it mean to detect or observe PopIII stars?

It must mean finding a region in a larger galaxy (or a small dim galaxy) whose light is comparatively free of metal lines and has a strong Lyman α signature.

If all the light is broadened out by whatever (the wind effect you mentioned or something else) then it doesn't count. PopIII stars might be in the region but we haven't detected them. To detect PopIII we have to be lucky. there has to be a strong dominant Lyα and negligible metal.

The Lyα line, just by good fortune, has to be coming to us from that region w/o being blurred and broadened and scattered so that we don't know it is coming from that region.

We used to think (I imagine) that we would only have that good fortune in the case of small dim galaxies. But these people have evidence that another favorable circumstance is possible. there can be a pristine cloud of gas that somehow persisted as part of a bright larger z~6.5 galaxy. Somehow, by good fortune, that cloud did not get contaminated by heavier elements and normal stars couldn't form in it, and all it contains is some "late bloomer" PopIII stars.

They are not guaranteed to find other galaxies where there was that favorable accident, but they can go looking. Maybe they will find other bright z~6.5 galaxies that by coincidence have "late bloom" PopIII regions.

 

[Ken G]

I guess what must be happening is that the starlight is getting reprocessed in surrounding neutral hydrogen, which converts the continuum coming from the star into Lyman alpha when the ionized hydrogen recombines. So you need some neutral hydrogen to convert the continuum into Lyman alpha, but not so much that the Lyman alpha cannot escape once created. If you didn't ionize the surrounding gas sufficiently, the Lyman alpha optical depth would be too large, and it would not escape to be seen. In that case, you might create other hydrogen lines that could escape, perhaps Balmer lines or something, so I would think you could look for those as well. Since the galaxy is redshifted, you can get the Lyman alpha through the neutral hydrogen in our own galaxy, but you could get Balmer lines through also. So I don't see why it's so important to have the other stars around.