..and further proof that scientists are not dogmatic but actually explore the evidence and make hypotheses: the most convincing evidence I've seen for the existence of dark matter (rather than MOND) is from HongSheng himself in comparing the two body relaxation time of globular clusters compared to dwarf galaxies of the same mass (see astro-ph/0511713).
The assumption is that at the dark matter can in principle interact with other things in the universe through non-gravitational means (most likely the weak force), so at some point in the early universe when reaction rates were much much much higher, it would have been in equilibrium with everything else (including the photons that would eventually become the cosmic microwave background (CMB)).
At some point the reaction rates would become low enough that the average dark matter particle would no longer interact with anything else, and so the dark matter would decouple from everything else and cool adiabatically. Photons decoupled at redshift 1100 and cooled adiabatically since then, and are now at 3K. If the dark matter decoupled earlier (a safe assumption), then assuming it's relativistic (ie. it would cool down at the same rate as the photons), it would be even cooler now. Non-relativistic things cool somewhat slower, but it couldn't be so hot today if it didn't start out relativistically hot.
There are a few assumptions in there, but it's surprising.
...and yes, black holes indeed are one hypothesized form of dark matter. Dark matter is matter that has gravitational effect but no electromagnetic effect (ie. it's dark), and solo black holes certainly fit that description to a T.
However, as the other responses have mentioned, MACHOs (MAssive Compact Halo Objects, ie. solo black holes, neutron stars and white dwarfs) cannot explain a number of observations. Their gravity will actually lens background stars, which we would be able to detect. The observed number of lensing events is much too small for MACHOs to make up a significant fraction of the dark matter.
Another severe problem is that we don't know of any way to make MACHOs except using baryons, and observations of the ratios of light elements (H, D, He-3, He-4, Li-6, Li-7, Be-8) combined with our knowledge of nuclear reaction rates tells us that the density of baryons is too small for the dark matter to be made significantly of baryons.
But the main point I wanted to make is that you shouldn't say "maybe there's no dark matter, but instead there are lots of black holes that we can't see" - postulating black holes that we can't see is exactly postulating a form of dark matter.
Probably the orbital period is so long that the Roche lobe is quite large - too large for a dwarf star to overflow its lobe and fuel the accretion onto the compact object... haven't run the numbers, so don't quote me on that, but that would be my naive guess.
On the last point, I can't remember the exact numbers but it's not as simple as "light particle = hot dark matter" / "heavy particle = cold dark matter". The heavier the particle, the earlier it decoupled, ie. the hotter the universe was when it decoupled. Whether the distribution function is relativistic ("hot") or not ("cold") depends on the average particle velocity at decoupling, ie. on both the temperature at decoupling and on the particle mass. So higher mass = lower velocity = less likely to be relativistic, but on the other hand higher mass = decouples at a higher temperature = higher velocity = more likely to be relativistic.
There end up being several mass ranges that produce cold dark matter, and they depend on what other species the dark matter particles interact with. I don't remember the numbers off-hand, but the paper points out that the axion, which has been proposed as a cold dark matter candidate, is expected to have a mass somewhere around the number they calculate.
the universe is made up of: 25% non-relativistic matter that only interacts gravitationally and has particle mass much less than one million solar masses, 5% normal baryonic matter, and 70% cosmological constant
after inflation, the overall density was almost completely smooth except for very small random perturbations with a power spectrum of -1 and amplitude such that if all of the gravitational growth was linear, the current rms amplitude in spheres of radius 8/h Mpc would be 0.9
after inflation, the universe was expanding at a rate such that the current rate of expansion is 70 km/s/Mpc
the universe as a whole and all of the matter which makes up the universe obeys General Relativity
That's the model. The simulation is what we use to turn that model into a set of predictions about the structure in the universe that we can test (and it's not the only tool we have for doing that - for the sake of predicting the properties of galaxies, so-called "semi-analytic models" are just as popular). But the simulation can be wrong without the model being wrong... in the trivial case, if there was a bug in the code or in the initial conditions, but more seriously if there are numerical problems that we don't understand whereby the process of turning a fluid into a set of discrete much-more-massive particles means that the output of the simulation is not the same as the actual behaviour of the model!
Yes, you missed the bit about the rotation curve. From the hydrogen measurements, they can measure the rotation curve and therefore calculate the mass. The amount of mass necessary to cause that amount of rotation is about 100 times as much mass as is detected in hydrogen. Since there are no stars either, either it's dominated by dark matter or molecular hydrogen... and I don't think anyone has a good way of making that much molecular hydrogen without dust, which comes from stars. Ergo, it's a galaxy made up mostly of dark matter.
Not really. A gravity wave is one where gravity is the restoring force. There is no restoring force in a gravitational wave... it's a wave of the entire gravitational field.
And yes, I know it's annoying that two very different things have similar names.;-) Gravity waves got named long before the concept of a gravitational wave was dreamed of.
The blurb correctly says that they are looking for gravitational waves. The title incorrectly calls these gravity waves.
Gravity waves are waves where displacement from equilibrium in a medium is counteracted by the force of gravity. For example, the waves on the surface of a pond are due to regions that are higher getting pulled down by gravity.
Gravitational waves are a phenomenon in general relativity where accelerating dense masses cause waves in the space-time metric that propogate at the speed of light.
The depth of mercury used in LMTs is usually 0.5-1mm (reference), so it's more like 50-100 tonnes.
As others have pointed out below, mercury doesn't work anyway, and the liquids they're looking at are much lighter. Of course, if you're not using mercury, you need to find a way of aluminizing the surface of the liquid, so you still need to get some metal up there. But aluminum is light, and you only need a thickness of about 0.1 microns.
(aside: why is it that the majority of comments on any astrophysics story are really really lame off-topic jokes? I mean, even worse than in your average/. story, which is saying something...)
Since it looks like most people here haven't read the journal articles (the sciencenews.org article is pretty light on the details), here's the basic idea:
In the early universe, the universe is mostly smooth except for small density fluctuations. The universe is made up of 3 basic fluids: photons, dark matter, and baryons. Density waves ("acoustic waves") pass through these fluids as far as they can at a given time - in the early universe, the horizon scale is quite small and the waves can't get all that far. Therefore you get horizon-sized acoustic oscillations. Dark matter is pressureless while the photon-baryon fluid isn't, so they react differently to compression... the end result of that is that there are some oscillations on quite large scales that are there because of the baryons.
What these groups have done using two different surveys (the 2 degree Field Galaxy Redshift Survey (2dFGRS) and the Sloan Digital Sky Survey (SDSS - they're using the Luminous Red Galaxy (LRG) subsample)) is look at galaxies at moderate redshift (medians of about 0.1 for 2dFGRS and 0.35 for the SDSS LRG sample) and compute the correlation function (in Dan Eisenstein's paper) or power spectrum (in Shaun Cole's paper). These tell you how clustered galaxies are as a function of how far away they are from each other.
What they both find is that there's a peak around 150 Mpc, exactly as you'd expect for a universe with about 75% vacuum energy and 25% matter, of which about 15% (ie. 15% of 25% = 4% total) is baryonic. The test is pretty sensitive to all of those numbers, and thus provides further evidence that the universe is dominated by a vacuum energy that drives acceleration.
I want the highest bandwidth solution of all. I'm also in Australia, and I don't care so much about seeing the numbers (or rather, I do, but I have full confidence in my ability to find them on my own) as I do about hearing the commentary. Information on the House and Senate races, where the surprises are, what interesting patterns there are that don't come up in lists of "x votes for A and y votes for B in state C".
So... does anyone know if there will be online TV (preferably) or radio broadcasts?
Heheh... I purposely avoided using calculus to make sure I didn't lose anyone.:-)=
Yes, it's quite possible for two objects to have relative speeds that are greater than the speed of light. What I think you're getting at is that photons that are currently emitted from an object that is currently moving away from us at greater than the speed of light will never reach us. But we can see photons that were emitted from these objects in the past.
I guess my problem is that I more or less got the idea that the the BB involved an object of extremely small dimensions - again that doesn't make much sense if both space and time are really products of the BB.
Yeah, that's probably the problem. The BB is not an object, it's a time.
I assumed that once we have space and time, also those oldest photons would have gone bucketing on past any matter in the vicinity, have to have. But perhaps it was more or less zooming straight around some intensely curved geodesics, straight ahead all the way around space-time?
We think the universe is globally close to flat, so the photons move on nearly straight lines. The photons have gone zooming past anything near where they started... they're coming from objects that are currently very very far away. They've had to travel a long way to reach us, which has taken a long time. - most of the age of the universe. So they come from a time not long after the BB.
If I understood what you wrote though, it seemed that you are suggesting that the relative rate of separation between two remote bodies can be greater than SOL (e.g. the most separated of Gamow's "raisins"), but the article is implying that we can see almost all the way back to the BB.
Ah, but remember that the distance between the two objects is getting larger. Which means that the red shift has been getting larger with time. If we see a photon from an object that's currently moving away from us at the speed of light, the recession velocity was less than the speed of light for the entire time the photon was travelling!
For example, let's imagine a photon leaving an object that is currently moving away from us at about 25% faster than the speed of light. At the time the photon is emitted, the distance between us and the object was much smaller, and so the relative velocity was much smaller than the speed of light. So the photon easily traverses a large fraction of the distance. Now it's 3/4 of the way to us, and the original object is expanding away from us at the speed of light. But the distance the photon has left to travel is only 1/4 of that, so the recession velocity is only 1/4 of the speed of light, and it has no problems travelling the rest of the way.
Photons from close to tht horizon are red-shifted, they lose energy. remeber the hot glowing soup from the big bang? It was really really hot, giving off X-rays and gamma rays and everything. However the light we now see from the big bang is from really really far away, almost at the horizon. X-rays with a 1 nanometer wavelength have been getting constantly stretched by space itself over the umpteen billion year journey. They stretched out by a factor of tens of thousands, stretched out into microwaves with wavelengths a few centimeters long.
The cosmic microwave background actually comes from the time of decoupling, at redshift ~1100... it was mainly optical light at the time, not X-rays (the temperature was about 3000K, cooler than the Sun, which emits mainly in the optical), and has redshifted by a factor of 1100 since then (not tens of thousands) to become microwaves now.
For it to be topologically open, you'd have to throw out the big bang theory that the universe originated at a single point.
According to Big Bang theory, if you extrapolate back toward t=0, all objects get closer and closer together. That's not really the same as "originating at a point", which seems to imply that it started at a specific physical location. The origin is a singularity, in the sense that the density rises to infinity and distances between any two objects gets asymptotically smaller, but it can still be infinite.
I think the problem is that you're imagining an explosion in an empty universe, whereas you should be thinking about a universe that's infinitely big, but insanely dense and insanely hot that all expands and cools together.
Presumably the universe and the matter in it could not expand more rapidly than light.
Those two statements are different... matter cannot move faster than the speed of light, but the expansion of the universe can.
In fact, the expansion of the universe doesn't have a physical velocity associated with it - it's a fractional rate of change. So if the universe expands at "0.1 Gyr^-1", then proper distances increase by 10% per gigayear (*). If the distance you're interested in is larger than 10 billion light years, then it increases at faster than the speed of light. But that same rate of expansion corresponds to a much smaller velocity if you're dealing with a much smaller distance.
A photon's important length is its wavelength, lambda. This wavelength increases because of universal expansions at a rate of lambda * H_0... or about 10^-24 m/s for an optical photon (wavelength of 500nm). But this isn't even a real velocity, it's just the rate of change of the wavelength - even if it were greater than the speed of light, it has nothing to do with causality.
(*) This rate of expansion (also known as the Hubble constant, H_0) is, for historical reasons, usually expressed in units of km/s/Mpc... but you'll notice that the km and Mpc cancel out, giving a fractional rate. If the Hubble constant is 70 km/s/Mpc (consistent with current measurements), that is equal to 0.072 Gyr^-1.
Yes (well, technically it doesn't in a curved space-time, but since the universe is globally flat, any deviations are extremely small on average).
In which case, shouldn't any radiation created by the big bang be at least 13 billion light years away from it's point of origin by now?
Yes.
So, unless they are reflecting off something or the universe wraps around at the edges, why can we still detect them?
Uh... here's where you've lost me. They're going in straight lines - that's why we can see them. When you look straight up and detect CMB photons, they were emitted from a point 13 billion light years away in that direction. When you look straight down, those photons were emitted from a point 13 billion light years away in that direction.
The big news is that they've measured the polarized power spectrum, and it agrees extremely well with the theoretical predictions. Which means that not only do the density fluctuations match what's expected, but the matter is moving in the gravitational field of those density fluctuations exactly as expected.
There's a big difference between those two statements! The first one is quite correct (well, except for issues relating to the opacity of the early universe - if you did it with a neutrino telescope or gravitational radiation telescope, you could theoretically see back to almost the Big Bang). The second isn't - if you travel, you're going forward in time, whereas when you look at light from far away, you are seeing back in time.
Slashdot Mirror
I'm sorry... "evidence" would have been a more accurate word. Point taken.
[TMB]
..and further proof that scientists are not dogmatic but actually explore the evidence and make hypotheses: the most convincing evidence I've seen for the existence of dark matter (rather than MOND) is from HongSheng himself in comparing the two body relaxation time of globular clusters compared to dwarf galaxies of the same mass (see astro-ph/0511713).
[TMB]
The equivalent rest mass of the light in our Galaxy is about a thousand solar masses, compared to 10^12 solar masses in matter.
So it's a cute idea, but it doesn't work in practice...
[TMB]
What a good question!
The assumption is that at the dark matter can in principle interact with other things in the universe through non-gravitational means (most likely the weak force), so at some point in the early universe when reaction rates were much much much higher, it would have been in equilibrium with everything else (including the photons that would eventually become the cosmic microwave background (CMB)).
At some point the reaction rates would become low enough that the average dark matter particle would no longer interact with anything else, and so the dark matter would decouple from everything else and cool adiabatically. Photons decoupled at redshift 1100 and cooled adiabatically since then, and are now at 3K. If the dark matter decoupled earlier (a safe assumption), then assuming it's relativistic (ie. it would cool down at the same rate as the photons), it would be even cooler now. Non-relativistic things cool somewhat slower, but it couldn't be so hot today if it didn't start out relativistically hot.
There are a few assumptions in there, but it's surprising.
[TMB]
Looking at the Hubble diagram, it's mainly predicated on the 4 highest-z GRBs. In fact, it smacks of Malmquist bias.
[TMB]
...and yes, black holes indeed are one hypothesized form of dark matter. Dark matter is matter that has gravitational effect but no electromagnetic effect (ie. it's dark), and solo black holes certainly fit that description to a T.
However, as the other responses have mentioned, MACHOs (MAssive Compact Halo Objects, ie. solo black holes, neutron stars and white dwarfs) cannot explain a number of observations. Their gravity will actually lens background stars, which we would be able to detect. The observed number of lensing events is much too small for MACHOs to make up a significant fraction of the dark matter.
Another severe problem is that we don't know of any way to make MACHOs except using baryons, and observations of the ratios of light elements (H, D, He-3, He-4, Li-6, Li-7, Be-8) combined with our knowledge of nuclear reaction rates tells us that the density of baryons is too small for the dark matter to be made significantly of baryons.
But the main point I wanted to make is that you shouldn't say "maybe there's no dark matter, but instead there are lots of black holes that we can't see" - postulating black holes that we can't see is exactly postulating a form of dark matter.
[TMB]
Probably the orbital period is so long that the Roche lobe is quite large - too large for a dwarf star to overflow its lobe and fuel the accretion onto the compact object... haven't run the numbers, so don't quote me on that, but that would be my naive guess.
[TMB]
On the last point, I can't remember the exact numbers but it's not as simple as "light particle = hot dark matter" / "heavy particle = cold dark matter". The heavier the particle, the earlier it decoupled, ie. the hotter the universe was when it decoupled. Whether the distribution function is relativistic ("hot") or not ("cold") depends on the average particle velocity at decoupling, ie. on both the temperature at decoupling and on the particle mass. So higher mass = lower velocity = less likely to be relativistic, but on the other hand higher mass = decouples at a higher temperature = higher velocity = more likely to be relativistic.
There end up being several mass ranges that produce cold dark matter, and they depend on what other species the dark matter particles interact with. I don't remember the numbers off-hand, but the paper points out that the axion, which has been proposed as a cold dark matter candidate, is expected to have a mass somewhere around the number they calculate.
[TMB]
That's the model. The simulation is what we use to turn that model into a set of predictions about the structure in the universe that we can test (and it's not the only tool we have for doing that - for the sake of predicting the properties of galaxies, so-called "semi-analytic models" are just as popular). But the simulation can be wrong without the model being wrong... in the trivial case, if there was a bug in the code or in the initial conditions, but more seriously if there are numerical problems that we don't understand whereby the process of turning a fluid into a set of discrete much-more-massive particles means that the output of the simulation is not the same as the actual behaviour of the model!
[TMB]
Yes, you missed the bit about the rotation curve. From the hydrogen measurements, they can measure the rotation curve and therefore calculate the mass. The amount of mass necessary to cause that amount of rotation is about 100 times as much mass as is detected in hydrogen. Since there are no stars either, either it's dominated by dark matter or molecular hydrogen... and I don't think anyone has a good way of making that much molecular hydrogen without dust, which comes from stars. Ergo, it's a galaxy made up mostly of dark matter.
[TMB]
Not really. A gravity wave is one where gravity is the restoring force. There is no restoring force in a gravitational wave... it's a wave of the entire gravitational field.
;-) Gravity waves got named long before the concept of a gravitational wave was dreamed of.
And yes, I know it's annoying that two very different things have similar names.
[TMB]
The blurb correctly says that they are looking for gravitational waves. The title incorrectly calls these gravity waves.
Gravity waves are waves where displacement from equilibrium in a medium is counteracted by the force of gravity. For example, the waves on the surface of a pond are due to regions that are higher getting pulled down by gravity.
Gravitational waves are a phenomenon in general relativity where accelerating dense masses cause waves in the space-time metric that propogate at the speed of light.
[TMB]
The depth of mercury used in LMTs is usually 0.5-1mm (reference), so it's more like 50-100 tonnes.
As others have pointed out below, mercury doesn't work anyway, and the liquids they're looking at are much lighter. Of course, if you're not using mercury, you need to find a way of aluminizing the surface of the liquid, so you still need to get some metal up there. But aluminum is light, and you only need a thickness of about 0.1 microns.
[TMB]
Since it looks like most people here haven't read the journal articles (the sciencenews.org article is pretty light on the details), here's the basic idea:
In the early universe, the universe is mostly smooth except for small density fluctuations. The universe is made up of 3 basic fluids: photons, dark matter, and baryons. Density waves ("acoustic waves") pass through these fluids as far as they can at a given time - in the early universe, the horizon scale is quite small and the waves can't get all that far. Therefore you get horizon-sized acoustic oscillations. Dark matter is pressureless while the photon-baryon fluid isn't, so they react differently to compression... the end result of that is that there are some oscillations on quite large scales that are there because of the baryons.
What these groups have done using two different surveys (the 2 degree Field Galaxy Redshift Survey (2dFGRS) and the Sloan Digital Sky Survey (SDSS - they're using the Luminous Red Galaxy (LRG) subsample)) is look at galaxies at moderate redshift (medians of about 0.1 for 2dFGRS and 0.35 for the SDSS LRG sample) and compute the correlation function (in Dan Eisenstein's paper) or power spectrum (in Shaun Cole's paper). These tell you how clustered galaxies are as a function of how far away they are from each other.
What they both find is that there's a peak around 150 Mpc, exactly as you'd expect for a universe with about 75% vacuum energy and 25% matter, of which about 15% (ie. 15% of 25% = 4% total) is baryonic. The test is pretty sensitive to all of those numbers, and thus provides further evidence that the universe is dominated by a vacuum energy that drives acceleration.
Here's links to preprints of the papers:
[TMB]
I want the highest bandwidth solution of all. I'm also in Australia, and I don't care so much about seeing the numbers (or rather, I do, but I have full confidence in my ability to find them on my own) as I do about hearing the commentary. Information on the House and Senate races, where the surprises are, what interesting patterns there are that don't come up in lists of "x votes for A and y votes for B in state C".
So... does anyone know if there will be online TV (preferably) or radio broadcasts?
[TMB]
Yes, it's quite possible for two objects to have relative speeds that are greater than the speed of light. What I think you're getting at is that photons that are currently emitted from an object that is currently moving away from us at greater than the speed of light will never reach us. But we can see photons that were emitted from these objects in the past.
Yeah, that's probably the problem. The BB is not an object, it's a time.
We think the universe is globally close to flat, so the photons move on nearly straight lines. The photons have gone zooming past anything near where they started... they're coming from objects that are currently very very far away. They've had to travel a long way to reach us, which has taken a long time. - most of the age of the universe. So they come from a time not long after the BB.
The Big Bang is a time, not a place.
[TMB]
For example, let's imagine a photon leaving an object that is currently moving away from us at about 25% faster than the speed of light. At the time the photon is emitted, the distance between us and the object was much smaller, and so the relative velocity was much smaller than the speed of light. So the photon easily traverses a large fraction of the distance. Now it's 3/4 of the way to us, and the original object is expanding away from us at the speed of light. But the distance the photon has left to travel is only 1/4 of that, so the recession velocity is only 1/4 of the speed of light, and it has no problems travelling the rest of the way.
[TMB]
The cosmic microwave background actually comes from the time of decoupling, at redshift ~1100... it was mainly optical light at the time, not X-rays (the temperature was about 3000K, cooler than the Sun, which emits mainly in the optical), and has redshifted by a factor of 1100 since then (not tens of thousands) to become microwaves now.
[TMB]
I think the problem is that you're imagining an explosion in an empty universe, whereas you should be thinking about a universe that's infinitely big, but insanely dense and insanely hot that all expands and cools together.
[TMB]
In fact, the expansion of the universe doesn't have a physical velocity associated with it - it's a fractional rate of change. So if the universe expands at "0.1 Gyr^-1", then proper distances increase by 10% per gigayear (*). If the distance you're interested in is larger than 10 billion light years, then it increases at faster than the speed of light. But that same rate of expansion corresponds to a much smaller velocity if you're dealing with a much smaller distance.
A photon's important length is its wavelength, lambda. This wavelength increases because of universal expansions at a rate of lambda * H_0... or about 10^-24 m/s for an optical photon (wavelength of 500nm). But this isn't even a real velocity, it's just the rate of change of the wavelength - even if it were greater than the speed of light, it has nothing to do with causality.
(*) This rate of expansion (also known as the Hubble constant, H_0) is, for historical reasons, usually expressed in units of km/s/Mpc... but you'll notice that the km and Mpc cancel out, giving a fractional rate. If the Hubble constant is 70 km/s/Mpc (consistent with current measurements), that is equal to 0.072 Gyr^-1.
[TMB]
Yes.
Uh... here's where you've lost me. They're going in straight lines - that's why we can see them. When you look straight up and detect CMB photons, they were emitted from a point 13 billion light years away in that direction. When you look straight down, those photons were emitted from a point 13 billion light years away in that direction.
[TMB]
The press release at the CBI website is much more informative.
The big news is that they've measured the polarized power spectrum, and it agrees extremely well with the theoretical predictions. Which means that not only do the density fluctuations match what's expected, but the matter is moving in the gravitational field of those density fluctuations exactly as expected.
[TMB]
There's a big difference between those two statements! The first one is quite correct (well, except for issues relating to the opacity of the early universe - if you did it with a neutrino telescope or gravitational radiation telescope, you could theoretically see back to almost the Big Bang). The second isn't - if you travel, you're going forward in time, whereas when you look at light from far away, you are seeing back in time.
[TMB]
[TMB]
The internet went to hell just 5 years ago?? It's been September since 1994...
[TMB]