by Katie Mack
Andrew Pontzen, a colleague and coauthor of Peiris’s (and my former officemate at Cambridge) works on theoretical aspects of dark matter, and has done some of the pioneering work to explain why dark matter takes the shape it does in galaxies. He contends that we have a very good understanding of cosmology, in the sense that our data line up extremely well with a picture that includes dark matter and dark energy, and that it seems unlikely anything will suddenly appear to change that picture. We know how much stuff is out there and how it behaves. On the other hand, we don’t know how to connect either dark matter or dark energy, which together constitute 95 percent of the universe, to fundamental physics. “So in that sense we don’t understand at all,” he says.
Meanwhile, the view from particle physics is frustratingly similar. Back in the 1970s, physicists developed the Standard Model of particle physics to describe all the known particles of nature: the quarks that make up protons and neutrons, the leptons like the neutrinos and the electron and its cousins, and the so-called gauge bosons that act as go-betweens carrying the fundamental forces between particles (electromagnetism and the strong and weak nuclear forces). Notwithstanding some minor tweaks, like taking neutrinos from being strictly massless to being very, very light, the Standard Model has been fantastically successful, passing every experimental test thrown at it. It even predicted the existence of the Higgs boson—the final piece of the Standard Model puzzle. In the years since, nothing has been discovered in particle experiments that the Standard Model didn’t tell us we would find.
You’d think this would be hailed as a triumph. The theory works! Everything is as we predicted!
Why aren’t we sitting back and basking in our brilliance and success?
Because this is, in some ways, a worst-case scenario. As great as the Standard Model is at matching experimental results, we know that it, like the Concordance Model in cosmology, has to be missing some very important pieces. In addition to having nothing at all to say about dark matter or dark energy, it has some major “tuning problems”—places in the model where a parameter has to be set juuuuust right or else everything falls apart. Ideally, we should have some theoretical framework that tells us why a parameter is what it is. It’s disconcerting when we find that the only reasons we have to set the parameter to that value are “otherwise bad things would happen to us” or, worse yet, “that’s just what the measurement says.”
For decades, there was hope on the horizon that we might be able to step seamlessly from confirming the important aspects of the Standard Model to finding the edges of its validity and making new discoveries with whatever model we found to replace it. In the 1970s, a model known as supersymmetry (SUSY, for short) was proposed to fix some of the theoretical niggles of the Standard Model by hypothesizing new mathematical connections between different kinds of particles and explaining the confusing structure of the Standard Model and its parameters. It came with a tantalizing promise too: a whole slew of new particles (“supersymmetric partners” of the Standard Model set) that might be produced in particle collisions just a little more powerful than what could be achieved by colliders at that time. SUSY has also been widely held up as a stepping-stone toward string theory, the leading idea in the quest to bring gravity and quantum mechanics together into a unified whole.
Unfortunately, despite working for decades to improve and upgrade the LHC, we’ve seen no sign of supersymmetry’s promised particles. Some physicists still hold out hope for SUSY by proposing tweaks that would make the new particles harder to find, but at some point the tweaks become so extreme that SUSY has just as many theoretical problems as the Standard Model. And the signal just isn’t there. Now and then, some quirk of the data will produce a whirlwind of excitement as physicists rush to explain why there are a few more events in a particular detector channel than expected. But so far, none of these blips has turned out to be more than a statistical fluke destined to fade away in the next data release.
I spoke to Freya Blekman, an experimental physicist who searches for beyond–Standard Model signatures in LHC data, about the current conundrum. “I’ve been in the field twenty years now and I’ve seen my share of excesses come and go, and I’ve also seen my share of popular models come and go,” she said. “Depending on who you speak to, there are people who are disillusioned… people have been telling them for a very long time that they should be seeing something. And what the experiments see is only the Standard Model.” From her perspective, though, the disillusionment is misplaced. Not because people are missing hints that are really there after all, but just because there was never any guarantee that anything new would be found with these experiments.
Still, the lack of direction from experiments can be troubling—enough to push some researchers out of particle physics entirely, and into cosmology. One of those is Pedro Ferreira, a cosmologist at Oxford University who switched from quantum gravity to cosmology during his PhD and who now studies the cosmic microwave background and general relativity in astrophysics in the hope that they might provide some better insights. “There hasn’t been anything revolutionary that particle theory has done which has led to observational results since 1973,” he says. There have been lots of new theoretical ideas, and some of them are very appealing, but without clear experimental evidence for something beyond the Standard Model, it’s hard to know where to go next, or which of the various proposals are likely to be right. “There’s all this beautiful stuff that’s come out. But have we solved the problem of quantum gravity? I don’t think so. And the problem is, how would we know if we’d solved it?”
Fortunately, no one is giving up hope. I spoke with dozens of cosmologists and particle physicists about where this whole thing is going (where by “whole thing,” I mean both theoretical physics/cosmology and the actual universe), and while there was no agreement about the optimal approach, there were a few common themes. One was diversification: whatever big multinational experiments or observational programs we decide to invest in, it’s important to diversify our approaches and come up with ideas that will give us new perspectives on these old problems (which goes for the theory side as well as the data-taking side). The other was the importance of continuing to get as much new data as we can, and to analyze it in every way possible.
Clifford V. Johnson, a theoretical physicist at the University of Southern California, works on string theory, black holes, extra dimensions of space, and the subtleties of entropy. He is about as deep into pure theory as anyone I know, and he is very excited about data right now. “My feeling is that we are maybe lacking a good sort of single idea, but we are not lacking in huge sources of data,” he said. “And that reminds me of the immediate pre-quantum days, right?” In those days, theory was booming, with lots of half-formed ideas about the structure of atoms and nuclei, though none were all that compelling. “But we just then got all this wonderful data that began to eventually take shape. I don’t see why that can’t happen again. Looking at the history of science, that’s how this works.”
So let’s talk about data. What we’re looking at, and how, in both cosmology and particle physics. What it might tell us about both the physics of the universe today and how it’s all going to come to an end in the future. And then we’ll check back in with the theorists. Because some of the ideas they’re talking about right now are absolutely wild.
TOUCHING THE VOID
If we want to learn anything about the far future of the cosmos, we’d better address the giant invisible ever-expanding killer elephant in the room: dark energy. When the accelerated expansion of the universe was discovered in 1998, the new paradigm placed us squarely in the path of a dark-energy-dominated future: one in which the cosmos gets progressively emptier, colder, and darker until all structure decays and we reach the ultimate Heat Death. But this is just an extrapolation, one that’s predicated on dark energy being an unchanging cosmological constant. As we’ve seen, if whatever is responsible for cosmic acceleration falls into the category of phan
tom dark energy, or if it somehow changes over time, the implications for the cosmos are drastically different.
Unfortunately, as far as observations are concerned, dark energy doesn’t give us a lot to hold on to. It is, as far as we can tell, invisible, undetectable in laboratory experiments, completely uniformly distributed through space, and only really noticeable at all by its indirect effects over scales much larger than our galaxy.
Generally speaking, there are two things we can measure. The first is the expansion history of the universe, which at the moment we study primarily by looking at very distant supernovae and figuring out how fast they’re receding. The other is the history of the formation of structure, where by “structure” we are generally referring to galaxies and clusters of galaxies, because all the little things like stars and planets are just annoying details if you’re a cosmologist. Measuring this is a bit less straightforward, but also allows for a lot of creative uses of massive piles of data. The trick is to get images and spectra of as many galaxies as possible, over a giant volume of space (and a large patch of cosmic history), and use statistical methods to infer how all that matter came together over time. Together, these two kinds of measurements can tell us how the space-stretching properties of dark energy have affected the universe as a whole and how much it’s impeded the efforts of matter to clump together and form things like galaxies and clusters and us.
When you only have two things you can measure to determine the WHOLE FATE OF THE UNIVERSE, it makes sense to invest a lot in measuring them very, very well. There’s been a surge of interest in the last couple decades in new telescopes and surveys with “dark energy” prominently featured in their science cases. Some are designed around the promise of how well they can use expansion and structure growth measurements to determine the dark energy equation of state parameter w (discussed in Chapter 5). If w = -1 exactly, now and in the past, we have a cosmological constant, and if it’s measurably different by any amount at all, we have a lot of Nobel Prizes. But even if you don’t care about dark energy, or if you subscribe to the pessimistic view that we’re fated to forever just narrow in on a garden-variety cosmological constant, dark energy surveys tend to be popular among astronomers of all stripes by doubling as all-purpose galaxy-gathering missions.
The upcoming LSST (Large Synoptic Survey Telescope), recently renamed the Vera C. Rubin Observatory (VRO), is a fantastic example. An 8.4-meter telescope on a high-desert mountain in Chile, VRO will take images of a few million supernovae and 10 BILLION galaxies, piecing together new images of the whole southern sky every few days. That kind of repeated coverage is great for supernova studies because it’ll let us see the rise and fall of the brightness of each supernova over the several days during which the explosion is visible. But it’s also great for studying galaxies, because it means you can stack up images night by night and see fainter and more distant galaxies than any other survey of its kind.
(As an aside, I recently attended a conference session about Planetary Defense in which the speakers were discussing the kinds of observations you need to spot potentially hazardous asteroids that might be on a collision course with our fragile little planet. The VRO will, at least for the southern sky, revolutionize our ability to pick these things up early, which might make it easier to find ways to stop them. I get a kick out of the idea that by attempting to understand the dark energy that will eventually destroy the universe, we might have a better chance of, on a much shorter timescale, saving the world.)
Whatever else its uses, the cosmological value of the VRO can’t be overstated, if only because having massive piles of exquisite data gives us a very good chance of finding something new and surprising. According to Peiris, the VRO will be a game changer. “We are looking at the universe in a different way than what’s been done before,” she says. “Any time we’ve looked at the universe in a way that we haven’t before, we learn new stuff.”
VRO isn’t the only new observational program to get excited about. There are a slew of other new telescopes and surveys coming up, each of which is poised to show us the cosmos in ways we’ve never seen before. Some of the most hotly anticipated are a class of new space telescopes like the James Webb Space Telescope (JWST), Euclid, and the Wide Field Infrared Survey Telescope (WFIRST), which will take deep images and spectra with infrared light, helping us to see galaxies so far away that their light has been stretched out of the visible part of the spectrum altogether.
Even cosmic microwave background observatories are getting in on the dark energy game. We saw in Chapter 2 how studying the CMB can tell us about the early universe and the origins of cosmic structure. At the time the CMB light was emitted, dark energy was completely unimportant in the universe, its effects totally swamped by the extreme densities of matter and radiation. So it may be surprising that CMB observations give us any insights into how dark energy is acting today. The trick is that all the cosmic structure we want to study—every galaxy and cluster of galaxies—is between us and the CMB, and each one of those objects distorts the space it’s in just a little bit with its gravity.
Imagine you have a snapshot looking down into a clear-water pond at the pebbles below. Even if you don’t know exactly where every pebble should have been placed, or all their exact shapes, you could probably tell the difference between very still water and water that has some ripples in it by noticing distortions in how the pebbles look, because you have a sense of what pebbles should look like in general. In a similar way, we understand the cosmic microwave background so well that we can see, at least in a statistical sense, the tiny distortions in its light due to all the stuff between here and there. This is called CMB lensing, and it’s a fantastic tool for studying the growth of cosmic structure. New CMB observatories will help us refine the method, but we’ve already used CMB lensing to make a map of ALL OF THE DARK MATTER IN THE OBSERVABLE UNIVERSE. Granted, the map is an extremely low-resolution, blurry kind of map, like a map of the world reproduced from memory with fingerpaints, but still, it’s pretty impressive that this is a thing we can do at all.
Renée Hložek, a cosmologist at the University of Toronto, uses the CMB and galaxy surveys to better understand our cosmological model, with a particular interest in dark energy and the universe’s ultimate fate. She points out that combining data between things like VRO and new CMB observatories will become especially powerful as each data set improves. Using a technique called cross-correlation, we can take what we know about the positions of individual objects from galaxy catalogs and compare that with what we know about the largest-scale distribution of matter from CMB lensing. This can give us more precise results that make it harder to miss any deviations from the Concordance Model. Alternative theories that use changes in gravity to mimic the effects of dark energy will look very different in the combined data, Hložek says. “Basically, I think we’ll run out of places to hide.”
What other cool things can you see if you have images of billions of galaxies? A big one is strong gravitational lensing, in which a galaxy or cluster of galaxies is distorting the space it’s in so much that the light from an object directly behind it gets split into multiple images, or spread out as an arc of light encircling it. Think of looking at a candle through the base of an empty wineglass—the curved glass spreads the light out in broad arcs or a circle instead of showing it to you as a single flame. When a gravitational lens does this, the individual images follow different paths through the distorted space. That means that if, for example, a supernova goes off in the lensed galaxy, it might be seen in one of the images before it shows up in another, because the light making up the second image took a longer path to get to us.
Aside from being a fabulous party trick,III time delay measurements like this give us a new way of measuring the expansion rate of the universe, since the distances involved are so large that the expansion becomes an important factor in the calculation. And we desperately need new ways to measure the expansion rate, because our current methods are givi
ng us weirdly different answers.
As you’ll recall from Chapter 5, measuring the expansion rate (also known as the Hubble Constant) using supernovae gives us one number, and measuring it via the CMB gives us another. A slew of other measurements have failed to resolve this contradiction, generally falling on one side or another. (A very recent result found something in between, but in a way that, unhelpfully, didn’t agree with either side.) Gravitational lensing time delay measurements might be a way to resolve the problem, because with the VRO, the number of these systems we can use will go from a few to hundreds. Gravitational wave measurements from instruments like LIGO (discussed in Chapter 7) could give us insight here too, and in the next decade or so might reach the precision necessary to finally settle the question.
THE VIEW FROM LEFT FIELD
One of the things I love about cosmology is how much it requires thinking creatively, trying to approach the physics of the universe from a totally new direction. This doesn’t mean fully unconstrained flights of fancy. You can’t just randomly make stuff up. But what you can (and must) do is constantly find new ways to look at problems to wring a little more insight out of whatever data the universe has to offer.
This kind of creative thinking becomes especially important when we’re faced with a conundrum like “How do we improve on Concordance Cosmology or the Standard Model?” Everything we’ve tried so far has been frustratingly consistent with predictions; where are we supposed to find clues leading us to new models if we can’t get something in the current model to break?
Clifford Johnson is optimistic, and points out that this lack of clear direction might be good for us. “I don’t have a thing I can point to and go, ‘This is the future!’ ” he told me. “I just feel the diversity of things that we’ve been driven to do… is probably somewhat healthy.”