I have being reading through the papers that discuss the new theory which is called galactic "cosmic downsizing" and AGN "cosmic downsizing". This thread is started to summarize observations which required the creation of the "cosmic downsizing" theory. As will be shown, there are a number of new hypotheses that have been developed to explain the new observations, following the "cosmic downsizing" train of logic. It appears that the "cosmic downsizing" based theories cannot explain all of the data. (See comment.)

I would be interested in others thoughts concerning both the observations and the different hypotheses that have been developed to provide mechanisms to explain how AGN evolve and the relationship between supermassive black holes and their companion galaxies.

To start off the discussion, the first issue is what is being observed and how does it indicate that theory must be changed? Note there are other explanations of the new observations, that compete with the "cosmic downsizing" theories.

Comment:
Cosmic downsizing is a hypothesized mechanism, a fundamental theory change, to explain the observations. It is not a simple observation.

The following is a link to a Gordon Richards’ presentation which is entitled “What to do with a Million Quasars” which is a discussion of the technical methods that where used to estimate redshift and luminosity of a million quasars and related technical measurement issues. (The presentation is interesting, however, I would not recommend it if your objective was to understand the observations and their implications, as the talk is primarily a technical discussion of how to analyze the data, not what are the theoretical implications the data.) At the end of the presentation Richard summarizes what was expected based on current galactic/cosmic evolution theory and how the new observations indicate that a fundamental change to theory is required to explain the observations concerning AGN.

http://www.cfa.harvard.edu/events/co.../richards.html

http://www.cfa.harvard.edu/dvlwrap/c...007-8/0306.ram


The following is an excerpt from one of the last slides in Richard’s presentation.

"Cosmic Downsizing" is the name given to the observation that large black holes were growing the fastest in the early universe, and small black holes are growing fastest now. This is a bit counter-intuitive, since one might naïvely expect the biggest black holes to still be growing now.

Here's one recent paper that talks about it in more detail.

It turns out, though, that it's probably just a consequence of the way that AGN are fueled. Major mergers in the largest dark matter haloes happened early in the universe, and so that's when we see the growth of the largest black holes. By now, most of the big systems have merged, so the prodigious amount of fuel necessary to power quasars just isn't available. Thus, in the local universe we see weaker AGN, powered by smaller black holes, in smaller galaxies and less drastic mergers.

At least, that's the general idea. We're still working on the details.

Oh, and the last slide from the talk really is a description of what we've observed recently.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

In reply to parejkoj’s comment:

Parejkoj, the problem situation seems to be somewhat more complicated, than black holes grow at different rates depending on time period.

The following is an attempt to summarize the observations concerning AGN and super massive black holes. Is it correct? If not, what are the specific points for which we agree or disagree?

Super massive black holes have a maximum size of around 10^9 solar mass regardless of the time period. There is some physical limit and physical reason why super massive black hole can not exceed around 10^9 solar masses. There is an unresolved question as to how super massive black holes formed in the early universe. If super massive black holes formed by accretion, in the early universe, there should be a very luminous quasar period during the early universe, which is not observed.

To explain observations it is assumed the quasar luminosity period is around 10^7 years.

When comparing the current universe to z=2, the question is why is the AGN luminosity, different for different period of times. The answer to that question is dependent on what is truly causing the difference in observations.

It is assumed intergalactic gas is the same/similar, regardless of the time period. i.e. It is expected X amount of intergalactic gas in the accretion disk will create the same luminous quasar regardless of the time period. (Curiously, quasar disc metallicity is about the same comparing current period to early period, and there are some observations where it was even higher for earlier periods which is not explained.)

What the observations indicate is that quasars with same super massive black hole mass are less luminous in the current time period than in the past.

Low accretion rates at the AGN cosmic downsizing epoch, by A. Babic, L. Miller, M. Jarvis, T.Turner, D. Alexander, and S. Croom

http://arxiv.org/PS_cache/arxiv/pdf/...709.0786v1.pdf

Comment:
As I believe you have not read Bell's papers, you are not aware there is another possible explanation of the observations.

Please explain. What are you implying?

Actually, I think the maximum observed size is closer to 10^10, though what's an order of magnitude between friends? We don't yet know whether it is an actual physical limit, or (much more likely) just due to their formation mechanism: the largest dark matter haloes just aren't big enough to continuously fuel to the biggest monsters, and winds and jets blow gas away at the highest accretion rates.

The question of "how SMBHs formed in the early universe" is much better understood now than it was just a few years ago. Recent cosmological simulations do a good job of matching the observed black hole/bulge mass relationships, as well as very showing early black hole growth. Me thinks he dost protest too much!

I don't understand your point. We do see very luminous quasars early in the universe: up to redshifts of ~6.4 so far. We can't find the higher redshift ones yet, because we don't have the right equipment, and haven't conducted the right kind of surveys. The James Web Space Telescope, Pan-Starrs and the LSST should all be able to identify quasars to redshifts of >6.5. We also know that AGN are often enshrouded in dust (we've seen it with Spitzer spectroscopy) during their early stages, which could make the first quasars even harder to see.

Something like that. If by "quasar luminosity period" you mean the absolute highest accretion rates, then yes, that's probably correct. But there is some feeding going on at much lower accretion rates as well.

I'm not certain about these two...

For the first, we know that intergalactic gas changes with redshift, and it changes differently in different mass haloes. And we know that the methods for getting it down to the black hole are going to be different at different redshifts as well. But we can't make models that are sophisticated enough to properly account for this. Yet. But the current models (I linked one example above) do a pretty darn good job of reproducing what we see, even without those details.

For the second, I believe the quasar metallicity measurements required quite a few assumptions that may not be correct for those systems. This is something that I'll have to look into, but I haven't time at the moment.

That's essentially another way to phrase what I said in my previous post.

What point are you trying to make with those quotes?

I've skimmed a few. What I've read convinced me that Bell has no idea what he's talking about. If you have evidence to the contrary, feel free to post it in the ATM section, and we can talk about it there. If you do a search for posts by me and "Bell" as the keyword, I've already given some thoughts in other threads, which you are welcome to address on the ATM section of the board, if you so desire.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

In response to parejkoj's comment:

Hi parejkoy,
I do not know how to compare theory A to theory B, as when I mention theory “B”, it might invoke an emotion response and a fight. I have a problem with the innate weakness/issue which is mentioned, in the aside. I am struggling to find a process to mitigate that innate tendency. Thank you in advance for your patience.

An Aside about Process
It is very, very difficult to see the data from the perspective of a different theory, because of an innate human tendency to compete, to fight. A productive/efficient discussion is not an intellectual game, like chess, where the objective is to beat an opponent, nor is it a debate, where people pick a side and try to defend it. When I state, that you have not examined Bell’s data, observations, and theoretical way of looking at the data, I am not saying Bell is smarter than everyone else, or that I am smarter than, you. I am only trying to compare theory A to theory B.

Theory A - General
How does Theory A explain the “Step Change in Quasar Luminosity with Redshift”, comparing current universe to z=2.” and in addition how does theory A explain the luminosity variance, quasar to quasar, for quasars that are contemporary.

Theory A - Quasar Engine Efficiency & Quasar Luminosity Changes
The quasar engine should be as efficient comparing current period to z=2. Higher metallicity of the intergalactic gas should make the quasar engine more efficient as the universe evolves not less.

At z=2, theory A, explains the variance in quasar luminosity due to differences in quasar engine efficiencies. The quasar luminosity is not dependent on the total amount of gas that is available to move through the accretion disc, but at the rate, the gas is moving through the accretion disc. Based on the hypothesis, that quasar luminosity is accretion disc rate dependent, quasar luminosity should change when the quasar processes a lump of intergalactic gas.

The other theoretical method theory A could use to explain the quasar luminosity variance would be that quasar luminosity is dependent on the total mass of the super massive BH which is hypothesized to drive the quasar engine. The problem with that hypothesis is the supermassive BH mass evolution does not correlate with the evolution of quasar luminosity, either in the current universe or at z=2.

Comment:
Current quasar engine simulations have a problem, getting the accretion gas to the black hole surface. The super massive BH hole clears out the accretion gas close to the BH. Detailed long term quasar studies have found radiant structures, in regions that are closer than to the super massive BH than theoretically possible, based on accretion disc modelling.

William - if what you're really saying is "hey, I don't understand what is presently known, less certain, and not known about quasars, how they generate their luminosities, how their populations have evolved over cosmic time, how their evolution fits in with galaxy evolution, etc...could y'all answer a couple of specific questions and point me to some good references?"

...then several of us (parejkoj, ngc3314, myself, and others) would be happy to help you out.

On the other hand, if you'd rather attribute less than accurate (ahem) characteristics to quasars, which you then (with your leading questions) declare to be "mysteries" that require "Theory B" to explain, then there is another forum for that sort of thing.

This isn't a question. It's a (rather long) statement. And don't get me wrong - it does not bother me in the least that someone might question a leading scientific model. What chaps my derrière is when this is done in ignorance of the model (and the field of supporting data) being questioned.

So - would you like to put forth 1 or 2 specific questions about quasars?

As noted, the super massive black hole mass, appears to asymptotically approach 3*10^9 solar masses, at z=2.

While there are a few super massive BH which might be as massive as 10^10 solar mass, there appears to be some sort of fundamental limit to BH mass. That does not make sense from the standpoint of galaxy mergers, as it would be expected the galaxies could continue to grow. If the growth of the BH is due to accretion not mergers, viewed from the standpoint of theory A, there could be some mechanism, where there is a maximum angular velocity of the super massive BH.


http://arxiv.org/abs/astro-ph/0310267v2

“The cosmological evolution of quasar black-hole masses” by R.McLure and J.Dunlop

I just gave you a possible mechanism in an earlier post. There are hints that the slope of the black hole - stellar mass relationship changes at high mass, but there is still a strong correlation.

And what do you mean by "growth ... due to accretion not mergers?" Black holes always grow through accretion.

I think you should follow Spaceman Spiff's advice here (and mine from before): either start asking specific questions here, or post your "Theory B" in the ATM section so we can talk about it there.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

In reply to Spaceman Spiff's

Please excuse my approach. I do not want a fight or an argument.

I asked how do the observations change when comparing theory A to theory B? That is a question. Following that process, would a person learn what data supports theory A vs theory B? As you state, you have no problem with criticism of theory A.

I do not understand the last part of your comment. I asked the question in the question and answer section, because I did not want to compete with the quasar discussion, in the Astronomy section. Likewise, I did not want to interrupt that discussion. To me this seems a polite approach.

Quasar Luminosity
Gordon Richard in his presentation, noted that the quasar models must now explain quasar luminosity step changes (current universe Vs z=2) and contemporary variance of quasar to quasar at similar periods of universe evolution based on quasar efficiency changes, not evolution of BH mass. Is that correct?

What is the significance of the asymptotically super massive BH limit of 3*10^9 solar mass? (See my last comment.)

Pedantic nit (yes, I do have a license): downsizing was first used in the astronomical context to describe the phenomenon for star formation in galaxies - the characteristic mass of star-forming galaxies drops with cosmic time. An alternative description is that the more massive the galaxy, the faster it completes its star formation. Sandy Faber as presented a unified view suggesting that star formation happens as the baryonic mass of a growing galaxy passes through a particular range, with the top set by the ISM getting to hot for star formation. The data are pretty compelling about the star formation-mass-time connection, which I find vastly amusing because hierarchical models had people expecting to find just the opposite. I'm always more convinced by a piece of work (including my own) which finds the opposite of what the investigators initially expected. The AGN connection was recognized later - and indeed it is tempting (perhaps "seductive" should be the word) to wonder whether the two are directly connected.

From Night At the Museum:

__________________
I want to go back to the moon.
I don't care which rocket you use, whichever one you pick, I'll like it, I swear.
"If you think the LHC will create black holes, you might as well believe Hobbits are at the bottom of your garden."- Dr. Mike Inglis
Rovers forever! - ToSeek

I don't understand what you're asking. Observations do not change - they are what they are, regardless of the state of the current model's ability to explain them (or of the presence of competing models).

It is only recently that data from surveys such as the SDSS (and also deep surveys such as HDF, HUDF and COSMOS) that we have gained a foothold in both understanding galaxy evolution and in how quasars fit into that. The models are nowhere near mature - rather in rapid phases of development.

Cosmic downsizing - whereby the densest regions of the early universe get a head start (over less dense regions) in structure growth via gravitational collapse, but then rapidly consume most of their available supplies of cold gas (and so become mature galaxies earlier), and the related phenomenon of 'feedback' via quasar activity (accretion onto and growth of super-massive black holes) during the most active periods of galaxy growth are two of the major players in the developing model of galaxy evolution.

The title of your post was a legitimate question. But your original post did not ask it. In fact the "question" had no specific question in it. It was just a long statement, and when parejkoj tried to respond to your non-question (some misunderstandings on your part), you simply produced another long statement.

Here is a question to you: what is the significance that there are no super-massive galaxies (>> 10^12 solar masses in stars)? Models of galaxy growth and evolution predict their existence - a puzzle.

Quasars (or more broadly, the phenomenon of Active Galactic Nuclei, or AGN), initially viewed as odd, rare phenomena of minor consequence (even freaks of nature), are now recognized as having played major roles in the early evolution and growth of what resulted in massive galaxies.

There is a mechanism known as "AGN feedback", which has been shown in N-body, hydrodynamical models of galaxy evolution, and which has also left behind several observables as to its existence, that is an important player in answering questions pertaining to the properties of galaxies and the population characteristics of AGN (even the "lifetime" of individual AGN) through cosmic time. There are are variety of means by which the AGN can interact energetically with its galactic (and even galactic cluster) environment, but we do not yet understand these in detail.

In a nutshell - AGN feed back, coupled with the development of large scale structure in an evolving (expanding) universe, has placed a lid on the maximum masses of both the super-massive black holes (smbh) and their host stellar spheroid, and led to the correlation between the two (established by redshifts of ~4 or so).

A key to AGN feedback is that the quasar luminosity (itself generated by the gas accretion rate onto the smbh) is bounded by the mass of the smbh (gravity vs. radiation pressure - the Eddington Luminosity), in an environment rich in supplies of infalling cold gas necessary to rapidly form stellar populations. The galaxy makes stars as long as a supply of infalling cold gas is available to collapse. It continues to do so until the accretion rate onto the smbh is able to generate sufficient power (photon momentum and ionization, relativistic jets, etc) to blow out and unbind most of the reservoirs of cold gas. The major phases of galaxy growth via cold gas accretion comes to an end, and soon thereafter the supply of cold gas to the galactic nucleus and smbh has slowed to a trickle. The quasar has then throttled itself. This required power (to throttle both star formation and then itself) depends on the depth of the gravitational potential well that the developing galaxy sits within.

The above is a very cursory summary, and there is a lot we still do not understand, which keeps my job interesting.

While the above scenario has AGN in a starring role, that does not discount the roles of stars themselves in the evolution (especially of course, chemical evolution, but also in the heating of gas via supernovae and stellar winds) of galaxies.

Finally, observations seem to indicate that major mergers involving gas rich galaxies had largely come to a close by a redshift of ~1 (about 8 billion years ago). The general decline in star formation activity (as a function of mass) as well as activity in the most luminous of quasars are documented in the data. I'll have to look more closely at Gordon Richards' presentation, but my guess is that he was wondering if not major mergers involving gas-rich galaxies, what fed the smbh's from redshift 1 to the present. It's a precipitous decline, but by no means a step function, as William seems to imply.

Well, but the basic (cold dark matter) model of hierarchical (bottom up) merging to larger objects still holds. It's just that the deeper potential wells assembled their bits and pieces and converted the vast reservoirs of cold gas into stars more rapidly - and ended up making more massive stellar spheroids - than did the more shallow potential wells. Perhaps with a little foresight (hindsight?), we should have expected that little twist!

And the two downsizings (galaxy and AGN evolution) are, I suspect, coupled. I'll try to dig up a few references.

Spaceman Spiff, Thank-you for your comments.

The flip side of your comment/statement is what are BH, the AGN engine? I would be interested in any comments you might have on the theoretical implications of naked AGN. (See astronomy section.)

I disagree with your comment that observations do not change when viewed from the perspective of a different theory. It is quite difficult to see how the observations change, when view from a different theoretical perspective. There seems to be sufficient information, in current papers, to answer the question you pose. Perhaps the answer is related to AGN evolution.

Regards,
William

Precisely - my posts above describe that very model that answers together both your question and the one I (rhetorically) posed to you.
There is a lot of ongoing work in this area, both observationally and theoretically. The general picture has a great deal of observational support, although I'll say again, we do not yet possess a mature theory.

After doing some hunting around, it appears as though this paper, already noted above by parejkoj, is one of the leading ones in illuminating the co-evolution of galaxies and smbh. Those of you interested but without subscriptions to the ApJ should download the astro-ph arXiv eprint.

Here is an interesting take on downsizing.

Caution: Note the date of insertion and the second author's name.

I just hope this doesn't lead to more stringent policing of arxiv submissions... 'cause I don't mind occasional items like this at all. But it's odd how far into the paper you have to get to be sure of what's going on.

Well, both the author, and the last sentence of the abstract gave it away for me. As did all the footnotes after the first.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

Ha!

Douglas Scott and Ali Frollop have published 3 papers on astro-ph over the last 3 years, all on about the same date...

And all in the same journal (which was listed as "the new Journal" in their first publication)...

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report