Where Is The Missing Dark Matter?
April 7, 2010 1 Comment
It’s a very exciting time to be a cosmologist. There are great big fundamental questions that are unanswered yet: What is the universe made of? What bigger question could you ask? What’s even more exciting is the realization that it’s within our grasp to answer these questions. We’ve narrowed down the possibilities and we’ve got all the machinery in place to find the answer. And that answer is likely to be one that is unrivaled in the physical sciences in terms of beauty because of the prospect of being able to understand the formation of galaxies and the movements of cosmic structures—essentially the behavior of the cosmos as a whole—in terms of the properties of subatomic particles. We’ll be able to explain the universe as the result of the properties of its most basic constituents.
Over the last 15 years or so, computer simulations have become the primary tool that theoreticians have at their disposal to understand the formation of galaxies and of structure in the universe. The aim of the simulations is to take what we call initial conditions, which is an early state of the universe, and then see how that primeval, amorphous state evolves into an approximation of the universe we can compare with current observational surveys. Through these simulations we can arrive at an understanding of what the universe is made of, how it is structured, and how it came to be.
Computer-simulated universes are a very powerful tool because they allow you to produce material evidence for what various assumptions about the universe translate into, and then you can take this material evidence and compare it against reality. Because the universe is so complex, most mathematical treatments require many approximations and simplifications, so they are of limited applicability. Yet with a computer simulation you don’t need to make any of those approximations. You solve the equations in the full generality, so it’s a very appealing activity for theoreticians to do.
In the classic Einsteinian view of the universe, everything is smooth at the beginning and stays smooth forever. That clearly is not what our universe is doing because today our universe is very inhomogeneous—it is broken up into islands that we call galaxies and galaxy clusters. If the universe had been entirely smooth, we wouldn’t be here to talk about it.
Instead, there must have been a small departure initially from this simplest assumption of a perfectly uniform universe. So the universe was not perfectly homogeneous either when it began or shortly after it began but, rather, it was slightly inhomogeneous. It had small regions where the density of matter was slightly higher than average and other regions where it was slightly lower than average. They were really tiny, these inhomogeneities, so tiny that for practical purposes it is hardly much of a departure from the simplest version of the theory. Yet tiny as they are to begin with, these inhomogeneities are very important because they are the seeds from which star clusters, galaxies and, eventually, human beings, will grow.
In April 1992, there was a very important discovery in cosmology that made the headline news all over the world—the discovery of ripples in the structure of the microwave background radiation. These ripples are nothing other than these little inhomogeneities we are talking about.The COBE satellite that discovered these ripples was short-sighted—it had a very blurry vision of the early universe. The ripples that COBE saw were much larger than the scales of the initial galaxies, so we haven’t yet detected directly the progenitors of the galaxies in the large-scale structure in the microwave background, but we have discovered or we’ve directly imaged very closely related entities that correspond to larger structures today.
What goes into the computer simulation is the nature of the lumps that we’ve studied using the COBE satellite. And then the simulation follows the dynamic evolution of those small inhomegeneities as the universe expands and as it cools, taking these very tiny little lumps and making them grow bigger. As the process unfolds the lumps move around fairly quickly, and, as they do, some of them bump into each other and coalesce, and the computer follows these coalescences beautifully. Eventually one sees the mock universe grow from an almost, but not quite, homogeneous initial state to one which is really complex, irregular in structure and corresponding to the universe we see at the present day.
In the real universe, the whole evolutionary process is driven by gravity and gravity is produced by mass, so in order to create a simulated universe, we need to know what sort of mass our universe has. One of the critical discoveries of astronomers in the last 25 or 30 years is the realization that there must be more mass in the universe than is accounted for by what we can see.
That means most of the mass in the universe is made up of what we call dark matter, which simply describes matter that doesn’t shine. To perform a successful computer simulation one needs to specify: what is the dark matter? What is it made of and how much is there?
The amazing thing is that if you make different assumptions you end up with different universes. So what many of us have been working on for the last 20 years is exploring various possibilities, evolving them in the computer to the present, and picking out those that look more like the real universe than others. Each mock universe that’s made up in the computer can be compared with the real universe in a variety of ways. You could look at different properties of the real universe and you could ask, “How many lumps are there?” or “How big are the lumps?” or “How are the lumps distributed?” You then can ask corresponding questions in the real universe and compare the two.
There are various candidates for the dark matter, but today one of the most popular is a very exotic elementary particle we call a WIMP, or weakly interacting massive particle.The WIMPs are just elementary, subatomic particles—fundamental constituents of matter. Tiny little individual things, they themselves come basically in two types: the so-called hot dark matter and cold dark matter. Hot dark matter consists of quickly moving small particles such as neutrinos, a particle which may or may not have a mass and therefore may or may not contribute to the shape of the universe. Cold dark matter is made up of particles that are sluggish—they move more slowly and are therefore cold. Predicted in a certain class of theories of fundamental interactions called supersymmetric theories, they have yet to be discovered experimentally.
The reason many people believe the dark matter is a cold-dark-matter WIMP is precisely because the cold dark matter simulation that we can create in the computer looks a lot like the real universe, whereas every other possibility we’ve tried, including hot dark matter, has turned out to look nothing like the universe. When we started cold dark matter simulations over 15 years ago, our intention was to rule them out as a candidate.
We were following a methodology where you put forward a candidate with the goal of ruling it out in order to narrow down the possibilities. With cold dark matter we failed miserably in that sense. We haven’t been able to rule it out. In all the calculations that we did and the many follow-ups people have done that have extended our work, they all come back to the same thing: cold dark matter universes look a lot like the real thing.
The fact that cold dark matter looks so good in a computer simulation doesn’t prove, of course, that it is the force shaping the universe. Today it is the front runner candidate, but until we actually see a WIMP, we can’t be sure. There are other possibilities that need to be explored and those can be explored within the context of computer simulations.
The key point of these theories is they require the existence of these hypothetical elementary particles. The proof of the pudding is in the eating; you have to capture one of these particles. So the ultimate test of this cold dark matter theory is to find the cold dark matter directly. Physics is, after all, an empirical, experimentally based human activity. You can’t prove that something is correct by theory. The Greeks thought that the truth could be established by pure thought, but we now know better: the universe is not made that way. We cannot prove the reality of anything just by thinking about it.
It’s hard to prove these particles exist because they’re very weakly interactive; that’s why they are dark: they don’t interact with anything. Cold dark matter doesn’t experience electromagnetic or nuclear interactions like protons and electrons do. They don’t interact with your apparatus, so trying to detect them in the laboratory is like trying to catch water using a bucket full of holes; it just goes through it.
Still, there are experiments to detect even these very weakly interacting particles by side effects. If you have a semiconductor, occasionally one of the WIMPs could have a head-on collision with a silicon atom and cause the atom to recoil . Now these hits are very, very rare so you have to have several kilograms of semiconductor and you are trying to find one atom moving just because it gets hit by a WIMP. Until these experimental searches succeed we cannot be certain that the theories are correct. But the exciting part is that the experiments are in place and the particles are detectable. If they exist, we will know about them in a few years.