Sunday, 28 April 2013

Bad Physics: ``Ballistic'' does not mean Evil!

A very quick rant about some bad physics that has been annoying me.

The tension between North Korea and the rest of the world is not a laughing matter, but has been all over the news recently. But when reporting, we often hear the that North Koreans are on the verge of producing a missile capable of carrying a nuclear warhead. But not only a missile! A ballistic missile.

Maybe I am reading too much into it, but when a reporter says ballistic, it's often said as if it is synonymous with evil! I'm sure that it's because we like to say something is "going ballistic" when something crazy happens.

But all ballistic means is that the missile is unpowered (after launch) and is moving only under the power of gravity. That's it.

There are powered missiles. Modern anti-tank missiles are usually powered and guided onto target. Here's one -

Here is a ballistic missile.
See - not so scary!

ARGOS IV: The Kinematics of the Milky Way Bulge

Another week gone, and no time for too much deep thinking (although progress is being made, so hopefully will have some interesting things to report). Luckily, the smart students out there are squirreling away, and so today I can present the latest in the study of the galactic bulge by ANU student
Melissa Ness.

I've written before about this really cool study, using the AAOmega Spectrograph on the might Anglo-Australian Telescope to measure the speeds and chemical make-up of stars in the Galactic Bulge, the centre of our Milky Way galaxy. This is hard work, as there are a lot of stars spread over a large amount of sky, and so to get lots of spectra, you need to use the multi-fibres and large field of view of AAOmega.

This has been a mammoth task over the last few years, with us taking 28,000 spectra, and in this study almost 17,5000 stellar velocities were used. But what is it we want to know?

Well, the present day shape, velocities and chemistry of the Milky Way is a consequence of its birth and growth, dependent on lots of different factors, such as what dark matter collapses to form the seed of the galaxy in the early universe, to the way that gas flows in and cool, and the way that stellar populations have evolved. So, if we can pick-apart the present day structure, then that will tell us about the formation history of the Galaxy, a topic known as galactic archaeology.

On the face of it, the bulge of the Milky Way looks quite simple, with stars basically buzzing about randomly, but in fact, the way the stars are distributed and the formation history means that the underlying dynamics can be quite complex. So, what did we find?

Firstly, here's the fields we observed
Even though the AAOmega spectrograph is mighty, you can see that the coverage is still relatively sparse, and so we are trying to work out the dynamics from a series of key-hole views through the Bulge. We, of course, would like to cover the entire Bulge, but that would take years and years of observing (and they won't let us have complete control of the telescope!).

Here is the rotation curve for the bule - yes, it rotates, and also the velocity dispersion. So, the motions of the stars are not orderly orbits, like the planets in the Solar System, but a mixture of orderly and random buzzing about.
Essentially, we can see that the velocity depends upon the location on the Bulge, and the latitude of the observations. Things get really interesting when we start to slice the stars up in terms of metallicity (which means chemical make-up).

Strongly negative numbers means the stars are chemically poor, while the values close to 0 are chemically richer (it's a logarithmic scale); now, all of these stars are almost entirely hydrogen and helium in their atmospheres (the bits we get the light from), but the higher metallicity stars are have more heavier elements than the metallicity poor stars. These extra elements come from the pollution by supernovae of gas clouds from which the stars were formed.

How do we start to understand these curves. Remember, we are looking through some key-hole views into the Bulge, but things are a little more complicated than that. The problem is that the Bulge contains structure, firstly large scale structure in the form of a Galactic Bar. Here's a model of the distribution of stars in the galaxy
We're located at 0 on the x-axis, and -8.5 kpc on the y-axis, and you can see that the Bulge (the red bit) is elongated into a bar, and it's almost along the line-of-sight. In fact, this complicated geometry has really only been known in detail for the last two decades.

The fact that we see stars with differing metallicities having different velocity signatures is telling us something interesting. It is telling us that the formation of the Bulge was a complicated thing. I could not have formed just as a single mass of stars, from a single gas cloud (if it did, we would expect the stars to be very similar metallicities) but something else happened, something more complicated.

It can be hard to unravel just what the detailed processes were, but what we can do is try and compare various models to the observed data. Here's one such comparison;
Pretty good! I don't know about you, but I am always amazed when you take ideas (and equations) formed in peoples heads, and make predictions for what you should see in a mass of stars located many light years away, and it works!

The fit is not perfect, however - there are places where the data and model differ significantly - and it is clear that we cannot have the whole story yet. But the general picture of a Bulge composed of multiple populations and a bar, which transitions into the disk populations that we are part of, is correct. Now we need to workout the complicated structure hidden away in the Bulge, which will provide us with even more clues to its hidden history is underway. Can't wait to unravel the secrets!

Well done Melissa!

ARGOS IV: The Kinematics of the Milky Way Bulge

We present the kinematic results from our ARGOS spectroscopic survey of the Galactic bulge of the Milky Way. Our aim is to understand the formation of the Galactic bulge. We examine the kinematics of about 17,400 stars in the bulge located within 3.5 kpc of the Galactic centre, identified from the 28,000 star ARGOS survey. We aim to determine if the formation of the bulge has been internally driven from disk instabilities as suggested by its boxy shape, or if mergers have played a significant role as expected from Lambda CDM simulations. From our velocity measurements across latitudes b = -5 deg, -7.5 deg and -10 deg we find the bulge to be a cylindrically rotating system that transitions smoothly out into the disk. Within the bulge, we find a kinematically distinct metal-poor population ([Fe/H] < -1.0) that is not rotating cylindrically. The 5% of our stars with [Fe/H] < -1.0 are a slowly rotating spheroidal population, which we believe are stars of the metal weak thick disk and halo which presently lie in the inner Galaxy. The kinematics of the two bulge components that we identified in ARGOS paper III (mean [Fe/H] = -0.25 and [Fe/H] = +0.15, respectively) demonstrate that they are likely to share a common formation origin and are distinct from the more metal poor populations of the thick disk and halo which are colocated inside the bulge. We do not exclude an underlying merger generated bulge component but our results favour bulge formation from instabilities in the early thin disk.

Sunday, 21 April 2013

Kinematics of Outer Halo Globular Clusters in M31

A weekend at home with the kids, with yesterday being wiped out by huge downpours from sunrise to sunset. Unfortunately, work has been full on, with a thousand different things going on, so no time to muck around. But we have another paper accepted, so here's a little summary.

This week's paper is by PhD student, Jovan Veljanoski, working at Edinburgh's Institute for Astronomy. This is another paper based upon our extensive Pan-Andromeda Archaeological Survey (PAndAS) of Andromeda. As I've noted before, this survey reveals a mass of substructure in the outskirts of Andromeda, as well as a growing population of dwarf galaxies. It also reveals lots of globular clusters, balls of a million or so stars orbiting Andromeda.

The key thing in this study is the velocities of the globular clusters. Why? Well, their velocity depends upon the mass they are orbiting, so if you get lot of velocities, then you can measure the mass. But, of course, it is never as simple as that :)

Why? Well, in the Solar System, all of the planets have nice, almost circular orbits, and we have a very well define laws of motion as the Sun is effectively is a point mass in the middle. Galaxies are trickier as the mass is spread out, and the gravitational field isn't necessarily spherical. Also, globulars have higgledy-piggledy orbits, more like the comets of the Solar System, and some are moving roughly together, as they accompanied infalling dwarf galaxy. All of this makes things a little messier.

Anyway, let's start with a picture.

The Andromeda Galaxy is lost in the mess in the middle there. The coloured circles are the globulars observed for this project; luckily, as we are collecting the light from a million stars in one go, we can get the velocities of the globulars with small (i.e. only 4m-class) telescopes. The colour-coding is the velocities that we measured. The red, yellow and blue boxes correspond to globulars that we think are associated with underlying substructure.

People have been collecting globular cluster velocities in Andromeda for a while, especially the ones closer to the centre of Andromeda. With a few hundred velocities in the bag, we can plot the velocity verses the distance of the globular (in projection) from Andromeda. What do we get? Well, this:
The velocities are "galacto-centric" meaning that we have taken the orbital motion of the Sun out of the equation, and have also subtracted the bult motion of Andromeda. Clearly, there is a slope, with positive velocities on the left and negative velocities  on the right. This tells us that there is a bulk rotation to the globular cluster population. How cool is that!

We can subtract off the rotation, and now get the velocity as a function of radius from Andromeda. Here's what we get:

So, the previous picture showed us "rotational velocity". This one tells us the "pressure velocity". I know this phrase throws some people, but it just means that the globulars are buzzing about more like comets more than the orderly motion of the planets.

With a bit of mathemagic, we can use this signature to tell us the mass of the Andromeda Galaxy. The result is that we find Andromeda to have a mass of
We have a range as there are some assumptions that we have to make about the profile of the dark matter mass distribution, but the total uncertainty is less than 20%. Cool!

I'm going to have to wrap up here, but the mass of Andromeda and the Milky Way have been a matter of a lot of debate over the years, especially the question of which one is the more massive. This puts there masses closer together again, but Andromeda still wins (just!). But more about that another day!

Well done Jovan!

Kinematics of Outer Halo Globular Clusters in M31

We present the first kinematic analysis of the far outer halo globular cluster (GC) population in the Local Group galaxy M31. Our sample contains 53 objects with projected radii of ~20-130 kpc, of which 44 have no previous spectroscopic information. GCs with projected radii >30 kpc are found to exhibit net rotation around the minor axis of M31, in the same sense as the inner GCs, albeit with a smaller amplitude of 79 +/-19 km/s. The rotation-corrected velocity dispersion of the full halo GC sample is 106 +/-12 km/s, which we observe to decrease with increasing projected radius. We find compelling evidence for kinematic-coherence amongst GCs which project on top of halo substructure, including a clear signature of infall for GCs lying along the North-West stream. Using the tracer mass estimator, we estimate the dynamical mass of M31 within 200 kpc to be M_M31 = (1.2-1.5) +/- 0.2 x 10^12 M_sun. This value is highly dependent on the chosen model and assumptions within.

Saturday, 6 April 2013

Matter Matters: Unphysical Properties of the Rh = ct Universe

An astro-ph post today. I've written previously about some papers that I've published looking at a "new" cosmological model, which is claimed to be superior to the currently favoured model, the ΛCDM  cosmology. This model, known as the Rh = ct universe, is supposedly simpler, but in truth, it is not. And what's more, it is not a good description of the Universe we observe. If you're interested, you can read about it here.

So, what's in my new paper. I take a look at a couple of claimed successes of the Rh = ct universe, namely that it explains the birth of quasars and does away with inflation. The crux of my argument is that the presence of matter in the universe actually destroys the claimed successes of this cosmology.

What I did is consider four cosmological models, the standard ΛCDM, with a present day matter density of 27% of the total, and a dark energy component making up the rest. The important thing is that the dark energy has an equation of state of -1. I also looked at a pure Rh = ct universe, which only has a dark energy component, but with an equation of state of -1/3. And then I added some matter to this pure Rh = ct cosmology, one with 27% matter at the present time, and one with 5% at the present time; this latter limit is basically the baryonic matter (atoms, stars, gas, people) we can see.

So, what do those changes do? They basically change the the look-back time verses redshift properties of the universes. In fact, they look like this

The bottom axis is in units of ~13.6 billion years. What the observational consequences of these different models.

Let's start with the quasars issue. We see very bright quasars in the distant universe, at redshifts up to ~7. These are thought to be power by supermassive black holes, with masses or order a few billion times that of the Sun. The problem is that if we assume some model for the growth of these black holes, namely that they were born a few times a mass of the Sun from supernovae explosions amongst the first stars, then there is very little time in the universe for them to grow. This picture explains the issue;
Look at the bottom left panel, which represents the ΛCDM universe. The solid black line represents the Big Bang. After the Big Bang was the dark ages, a period where gas was pooling together, but there were no stars. When the first stars ignited, the universe was reionized by the starlight (after the big dashed line).

The little red bars indicate the periods at which the first seed black holes formed. You can see the problem - in a ΛCDM universe, the seed black holes would have had to form at an epoch when there were no stars, and the most distant quasar's seed black hole would have had to be born before the universe. A real conundrum, but we don't really understand black hole growth in the early universe, i is a problem to solve.

The lower right hand panel is the same, but in the Rh = ct universe. As we've changed the look-back time verses redshift, then now all of the seed black holes are formed in the time when we have stars - success!! Well, no. Looking at the upper panels, then adding 5% matter into the mix drags the epoch of black hole birth to be at the very start of the epoch of reionization, and adding a more realistic 27% results in basically the same problem as ΛCDM.

Essentially, what is happening is that in the early universe, matter dominates, and so adding even a morsel makes a mess of the expansion needed for the Rh = ct universe.

We see the same with inflation. This is a little more complicated to explain, but here's the picture to tell you what is going on.
These are the same universes, but presented in conformal coordinates. I've mentioned these before, but if you want a reminder, have a look at this excellent article. Again, lets start with ΛCDM in the lower left; so we have distance along the bottom and time up the side, and the blue indicates the Big Bang and  the red is the epoch of recombination, where the universe had cooled enough for electrons to join with protons to make hydrogen atoms.

The triangles are light cones, with us at the apex. So, the properties of the Cosmic Microwave Background are determined by what goes on between the blue line and the red line. If you look very closely, you'll see that there are a couple of small triangles between the red and blue, and the fact that these two little triangles don't overlap is a manifestation of what is known as the Horizon problem; essentially as these little light cones do not overlap, then these patches of the sky were never in causal contact (basically, they never knew about each others properties), so why do we see different patches of the sky at (nearly) the same temperature.

The lower right presents the same situation in the Rh = ct universe. Something magical has happened! The epoch of recombination is still there, but where's the Big Bang? It's moved, and moved a long way back to a conformal time of -∞! The little triangles have become big triangles, and overlap! Horizon problem solved!!

Not so fast! Look at the top two panels, where we've added a little bit of mass to the Rh = ct universe. The Big Bang is back with a vengeance, and the triangles are little again, meaning we have the Horizon problem again. Matter in the Rh = ct universe messes things up again.

You might be asking how our standard cosmological model solves this. Essentially, we have a burst of inflation in the early universe, which extends the distance between the Big Bang and recombination, but doesn't push the Big Bang back to -∞. Here's a picture from a 1991 by Ed Harrison which explains this nicely

Now, there is a get out of jail card for the Rh = ct universe; what if we have matter in the universe but change the properties of dark energy (namely the equation of state) to make it behave as if there was only dark energy with an equation of state of -1/3. Well, what would you need.
So, the blue is no matter, and as we expect, the equation of state is always -1/3. But once we add a bit of mass and ask what the equation of state of dark energy needs to be to make it behave as Rh = ct, we get the green and red curves, and as we go back into the past, the dark energy needs to become more and more negative, eventually diverging to -∞! Very unphysical dark energy indeed.

Summary, the presence of matter messes up the Rh = ct.

Well done... well... me... :) I'm particularly proud of the opening statement of the abstract.

Matter Matters: Unphysical Properties of the Rh = ct Universe

It is generally agreed that there is matter in the universe and, in this paper, we show that the existence of matter is extremely problematic for the proposed Rh = ct universe. Considering a dark energy component with an equation of state of w=-1/3, it is shown that the presence of matter destroys the strict expansion properties that define the evolution of Rh = ct cosmologies, distorting the observational properties that are touted as its success. We further examine whether an evolving dark energy component can save this form of cosmological expansion in the presence of matter by resulting in an expansion consistent with a mean value of <w> = -1/3, finding that the presence of mass requires unphysical forms of the dark energy component in the early universe. We conclude that matter in the universe significantly limits the fundamental properties of the Rh = ct cosmology, and that novel, and unphysical, evolution of the matter component would be required to save it. Given this, Rh = ct cosmology is not simpler or more accurate description of the universe than prevailing cosmological models, and its presentation to date possesses significant flaws.