« Reading List: The B-58 Blunder | Main | Reading List: Humans to Mars »

Saturday, May 28, 2016

Reading List: The Cosmic Web

Gott, J. Richard. The Cosmic Web. Princeton: Princeton University Press, 2016. ISBN 978-0-691-15726-9.
Some works of popular science, trying to impress the reader with the scale of the universe and the insignificance of humans on the cosmic scale, argue that there's nothing special about our place in the universe: “an ordinary planet orbiting an ordinary star, in a typical orbit within an ordinary galaxy”, or something like that. But this is wrong! Surfaces of planets make up a vanishingly small fraction of the volume of the universe, and habitable planets, where beings like ourselves are neither frozen nor fried by extremes of temperature, nor suffocated or poisoned by a toxic atmosphere, are rarer still. The Sun is far from an ordinary star: it is brighter than 85% of the stars in the galaxy, and only 7.8% of stars in the Milky Way share its spectral class. Fully 76% of stars are dim red dwarves, the heavens' own 25 watt bulbs.

What does a typical place in the universe look like? What would you see if you were there? Well, first of all, you'd need a space suit and air supply, since the universe is mostly empty. And you'd see nothing. Most of the volume of the universe consists of great voids with few galaxies. If you were at a typical place in the universe, you'd be in one of these voids, probably far enough from the nearest galaxy that it wouldn't be visible to the unaided eye. There would be no stars in the sky, since stars are only formed within galaxies. There would only be darkness. Now look out the window: you are in a pretty special place after all.

One of the great intellectual adventures of the last century is learning our place in the universe and coming to understand its large scale structure. This book, by an astrophysicist who has played an important role in discovering that structure, explains how we pieced together the evidence and came to learn the details of the universe we inhabit. It provides an insider's look at how astronomers tease insight out of the messy and often confusing data obtained from observation.

It's remarkable not just how much we've learned, but how recently we've come to know it. At the start of the 20th century, most astronomers believed the solar system was part of a disc of stars which we see as the Milky Way. In 1610, Galileo's telescope revealed that the Milky Way was made up of a multitude of faint stars, and since the galaxy makes a band all around the sky, that the Sun must be within it. In 1918, by observing variable stars in globular clusters which orbit the Milky Way, Harlow Shapley was able to measure the size of the galaxy, which proved much larger than previously estimated, and determine that the Sun was about half way from the centre of the galaxy to its edge. Still, the universe was the galaxy.

There remained the mystery of the “spiral nebulæ”. These faint smudges of light had been revealed by photographic time exposures through large telescopes to be discs, some with prominent spiral arms, viewed from different angles. Some astronomers believed them to be gas clouds within the galaxy, perhaps other solar systems in the process of formation, while others argued they were galaxies like the Milky Way, far distant in the universe. In 1920 a great debate pitted the two views against one another, concluding that insufficient evidence existed to decide the matter.

That evidence would not be long in coming. Shortly thereafter, using the new 100 inch telescope on Mount Wilson in California, Edwin Hubble was able to photograph the Andromeda Nebula and resolve it into individual stars. Just as Galileo had done three centuries earlier for the Milky Way, Hubble's photographs proved Andromeda was not a gas cloud, but a galaxy composed of a multitude of stars. Further, Hubble was able to identify variable stars which allowed him to estimate its distance: due to details about the stars which were not understood at the time, he underestimated the distance by about a factor of two, but it was clear the galaxy was far beyond the Milky Way. The distances to other nearby galaxies were soon measured.

In one leap, the scale of the universe had become breathtakingly larger. Instead of one galaxy comprising the universe, the Milky Way was just one of a multitude of galaxies scattered around an enormous void. When astronomers observed the spectra of these galaxies, they noticed something odd: spectral lines from stars in most galaxies were shifted toward the red end of the spectrum compared to those observed on Earth. This was interpreted as a Doppler shift due to the galaxy's moving away from the Milky Way. Between 1929 and 1931, Edwin Hubble measured the distances and redshifts of a number of galaxies and discovered there was a linear relationship between the two. A galaxy twice as distant as another would be receding at twice the velocity. The universe was expanding, and every galaxy (except those sufficiently close to be gravitationally bound) was receding from every other galaxy.

The discovery of the redshift-distance relationship provided astronomers a way to chart the cosmos in three dimensions. Plotting the position of a galaxy on the sky and measuring its distance via redshift allowed building up a model of how galaxies were distributed in the universe. Were they randomly scattered, or would patterns emerge, suggesting larger-scale structure?

Galaxies had been observed to cluster: the nearest cluster, in the constellation Virgo, is made up of at least 1300 galaxies, and is now known to be part of a larger supercluster of which the Milky Way is an outlying member. It was not until the 1970s and 1980s that large-scale redshift surveys allowed plotting the positions of galaxies in the universe, initially in thin slices, and eventually in three dimensions. What was seen was striking. Galaxies were not sprinkled at random through the universe, but seemed to form filaments and walls, with great voids containing little or no galaxies. How did this come to be?

In parallel with this patient observational work, theorists were working out the history of the early universe based upon increasingly precise observations of the cosmic microwave background radiation, which provides a glimpse of the universe just 380,000 years after the Big Bang. This ushered in the era of precision cosmology, where the age and scale of the universe were determined with great accuracy, and the tiny fluctuations in temperature of the early universe were mapped in detail. This led to a picture of the universe very different from that imagined by astronomers over the centuries. Ordinary matter: stars, planets, gas clouds, and you and me—everything we observe in the heavens and the Earth—makes up less than 5% of the mass-energy of the universe. Dark matter, which interacts with ordinary matter only through gravitation, makes up 26.8% of the universe. It can be detected through its gravitational effects on the motion of stars and galaxies, but at present we don't have any idea what it's composed of. (It would be more accurate to call it “transparent matter” since it does not interact with light, but “dark matter” is the name we're stuck with.) The balance of the universe, 68.3%, is dark energy, a form of energy filling empty space and causing the expansion of the universe to accelerate. We have no idea at all about the nature of dark energy. These three components: ordinary matter, dark matter, and dark energy add up to give the universe a flat topology. It is humbling to contemplate the fact that everything we've learned in all of the sciences is about matter which makes up less than 5% of the universe: the other 95% is invisible and we don't know anything about it (although there are abundant guesses or, if you prefer, hypotheses).

This may seem like a flight of fancy, or a case of theorists making up invisible things to explain away observations they can't otherwise interpret. But in fact, dark matter and dark energy, originally inferred from astronomical observations, make predictions about the properties of the cosmic background radiation, and these predictions have been confirmed with increasingly high precision by successive space-based observations of the microwave sky. These observations are consistent with a period of cosmological inflation in which a tiny portion of the universe expanded to encompass the entire visible universe today. Inflation magnified tiny quantum fluctuations of the density of the universe to a scale where they could serve as seeds for the formation of structures in the present-day universe. Regions with greater than average density would begin to collapse inward due to the gravitational attraction of their contents, while those with less than average density would become voids as material within them fell into adjacent regions of higher density.

Dark matter, being more than five times as abundant as ordinary matter, would take the lead in this process of gravitational collapse, and ordinary matter would follow, concentrating in denser regions and eventually forming stars and galaxies. The galaxies formed would associate into gravitationally bound clusters and eventually superclusters, forming structure at larger scales. But what does the universe look like at the largest scale? Are galaxies distributed at random; do they clump together like meatballs in a soup; or do voids occur within a sea of galaxies like the holes in Swiss cheese? The answer is, surprisingly, none of the above, and the author explains the research, in which he has been a key participant, that discovered the large scale structure of the universe.

As increasingly more comprehensive redshift surveys of galaxies were made, what appeared was a network of filaments which connected to one another, forming extended structures. Between filaments were voids containing few galaxies. Some of these structures, such as the Sloan Great Wall, at 1.38 billion light years in length, are 1/10 the radius of the observable universe. Galaxies are found along filaments, and where filaments meet, rich clusters and superclusters of galaxies are observed. At this large scale, where galaxies are represented by single dots, the universe resembles a neural network like the human brain.

As ever more extensive observations mapped the three-dimensional structure of the universe we inhabit, progress in computing allowed running increasingly detailed simulations of the evolution of structure in models of the universe. Although the implementation of these simulations is difficult and complicated, they are conceptually simple. You start with a region of space, populate it with particles representing ordinary and dark matter in a sea of dark energy with random positions and density variations corresponding to those observed in the cosmic background radiation, then let the simulation run, computing the gravitational attraction of each particle on the others and tracking their motion under the influence of gravity. In 2005, Volker Springel and the Virgo Consortium ran the Millennium Simulation, which started from the best estimate of the initial conditions of the universe known at the time and tracked the motion of ten billion particles of ordinary and dark matter in a cube two billion light years on a side. As the simulation clock ran, the matter contracted into filaments surrounding voids, with the filaments joined at nodes rich in galaxies. The images produced by the simulation and the statistics calculated were strikingly similar to those observed in the real universe. The behaviour of this and other simulations increases confidence in the existence of dark matter and dark energy; if you leave them out of the simulation, you get results which don't look anything like the universe we inhabit.

At the largest scale, the universe isn't made of galaxies sprinkled at random, nor meatballs of galaxy clusters in a sea of voids, nor a sea of galaxies with Swiss cheese like voids. Instead, it resembles a sponge of denser filaments and knots interpenetrated by less dense voids. Both the denser and less dense regions percolate: it is possible to travel from one edge of the universe to another staying entirely within more or less dense regions. (If the universe were arranged like a honeycomb, for example, with voids surrounded by denser walls, this would not be possible.) Nobody imagined this before the observational results started coming in, and now we've discovered that given the initial conditions of the universe after the Big Bang, the emergence of such a structure is inevitable.

All of the structure we observe in the universe has evolved from a remarkably uniform starting point in the 13.8 billion years since the Big Bang. What will the future hold? The final chapter explores various scenarios for the far future. Because these depend upon the properties of dark matter and dark energy, which we don't understand, they are necessarily speculative.

The book is written for the general reader, but at a level substantially more difficult than many works of science popularisation. The author, a scientist involved in this research for decades, does not shy away from using equations when they illustrate an argument better than words. Readers are assumed to be comfortable with scientific notation, units like light years and parsecs, and logarithmically scaled charts. For some reason, in the Kindle edition dozens of hyphenated phrases are run together without any punctuation.

Posted at May 28, 2016 20:10