Structure of the Universe

Galaxies are gravitationally bound assemblages of millions to tens of trillions of stars. Galaxies are not scattered randomly across the universe but instead are grouped together in galaxy clusters, consisting of several tens to many thousands of individual galaxies. There is growing evidence of connective patterns between galaxy clusters stretching across regions at least as large as many hundreds of millions of light-years. Explaining the observed features of this large-scale structure is a challenge to cosmologists that puts limits on acceptable models of the origin and evolution of the universe.

Overview

The contents of the universe are arranged in a hierarchy of structures. Stars, of which the Sun is a familiar example, are hot balls of gas that generate energy through nuclear fusion reactions. There is growing evidence that many stars are orbited by families of planets, analogous to the solar system. Stars (and their families of planets, if they have them) are gravitationally bound together into galaxies, vast collections of millions to tens of trillions of stars, spanning thousands to hundreds of thousands of light-years. The Sun and solar system are part of the Milky Way galaxy, a moderately large spiral galaxy consisting of several hundred billion stars along with gas and dust.

Galaxies, in turn, are gravitationally bound together into galaxy clusters, containing from several tens to many thousands of galaxies. The galaxy is part of the Local Group, a small galaxy cluster containing about forty members. Galaxy clusters group together to form superclusters. The Local Group is part of the Virgo supercluster, containing several tens of thousands of individual galaxies and spanning more than 100 million light-years. On a larger scale, galaxies, clusters, and superclusters appear to be arranged in a network of filaments and walls surrounding large, nearly empty voids or bubbles hundreds of millions of light-years across. This overall large-scale structure has been described as “frothy,” or analogous to Swiss cheese with all its holes.

Understanding the structure and evolution of the universe requires evidence about the distribution of matter on the largest scales. Most models of the universe satisfy the cosmological principle. This is the assumption that, at any moment, the large-scale structure of the universe looks the same from all places (homogeneity) and in all directions (isotropy).

When Albert Einstein published his general theory of relativity in 1916, in which gravity is treated as a warping of space-time, almost immediately, it was recognized that it would have a profound impact on cosmological models. Einstein himself tried applying it to a static universe and found that he needed to add an arbitrary term called the cosmological term or cosmological constant to obtain a static solution.

In 1922, the Russian mathematical physicist Alexander Alexandrovich Friedmann, and in 1927, the Belgian priest and cosmologist Georges Lemaître independently derived two classes of homogeneous and isotropic solutions (without the cosmological constant) in which the universe expands. In one of these classes, space has a uniform positive curvature and finite extent though no boundary; the universe is said to be closed because it expands to some maximum size and then contracts. The two-dimensional surface (not including the interior volume) of a three-dimensional sphere is such a space; motion on the surface is never blocked by a barrier (perimeter line), but the area of the surface is finite. In the other class, space is negatively curved and of an infinite extent; the universe is said to be open because it expands forever. In two dimensions, such a surface is termed “hyperboloid;” a saddle is an example of a finite part of a surface that mathematically extends to infinity if it lacks a boundary curve. In 1932, Einstein and the Dutch astronomer Willem de Sitter proposed a third expanding model without the cosmological constant. In it, space is flat and infinite, geometry is Euclidean, and the universe barely expands forever—it is the boundary between the two Friedmann-Lemaître solutions. In two dimensions, the surface of a plane is flat and infinite. These are the only three possible spatial geometries and types of models that can be solutions to Einstein’s gravitational field equations under the assumptions of homogeneity and isotropy.

The common feature of these models is that their scale factors (the distance between representative points, such as clusters of galaxies) change with time; the universe must be either expanding or contracting. In 1929, Edwin Powell Hubble published the first observational evidence that the universe is indeed expanding. Using the 100-inch on Mount Wilson (then the largest telescope in the world), he found that the more distant a galaxy is, the greater its spectrum is redshifted. This means all features in the spectrum are shifted to longer than normal wavelengths, which can be due to motion away from the Earth.

If the universe is expanding, the scale factor was smaller in the past, and the local densities of both matter and radiation were greater. However, the mass density of matter varies inversely as the scale factor is cubed, while the energy density of radiation varies inversely as the scale factor to the fourth power. At sufficiently early times when the scale factor was smaller, the radiation density was greater than the density of matter, and the universe was dominated by radiation. With time, as the scale factor increased and the universe expanded, radiation density decreased faster than the density of matter. Eventually, the universe became dominated by matter.

The modern universe is “lumpy” on a variety of scales: stars, galaxies, galaxy clusters, superclusters, and “walls” and “voids.” There has been growing interest in trying to understand how an initially homogeneous Big Bang could produce the lumpy inhomogeneities observed. This is called the homogeneity problem.

Applications

Galaxy surveys carried out to great distances show that the density of galaxies in voids is typically a factor of ten less than average, and the density in the narrow but long walls and filaments is typically a factor of ten thousand greater than average. Explaining the origin of such variations in density from an early universe that was amazingly homogeneous is a major challenge in modern cosmology.

The cosmic microwave background radiation (called the CMB or CBR for short), coming from a few hundred thousand years after the Big Bang when the matter in the expanding universe became transparent, is remarkably uniform and isotropic. However, detailed observations of it made by the Cosmic Background Explorer (COBE) satellite, the Wilkinson Microwave Anisotropy Probe (WMAP) satellite, the Balloon Observations of Millimetric Extragalactic Radiation and Geomagnetics (BOOMERANG) project in Antarctica, and the Planck spacecraft do reveal small variations. Completed in 2018, the Planck mission provided accurate measurements of CMB and provided new data about the universe's age and cosmic makeup. The data perfectly fit a blackbody Planck curve for a temperature of 2.735 kelvins. Removing a slight asymmetry in temperature (about 0.007 kelvins) in opposite directions, presumably due to the motion of the solar system through the background radiation, leaves very small temperature fluctuations on the order of ten micro-kelvins, about one degree in angular size. The hotter regions had slightly higher densities, and they were about the right size to develop into clusters and superclusters of galaxies.

The initial formation of the hotter, denser regions and their evolution into galaxy clusters and superclusters may involve dark matter. This is the modern name used to refer to what formerly was called “missing mass.” In many situations in astronomy, the amount of matter that can be detected through the electromagnetic radiation that it emits (whether radio waves, visible light, or other wavelengths) is much less—typically by a factor of about five to fifty—than what is needed gravitationally to hold galaxies, clusters of galaxies, and superclusters together and to account for gravitational lensing of distant objects. Most astrophysicists consider the gravitational estimates reasonably well established and, thus, believe the mass is not “missing” but is not emitting electromagnetic radiation. The challenge is to explain the nature of dark matter, which is much more abundant than ordinary luminous matter.

Various observational tests suggest that nonluminous ordinary matter (perhaps in the form of boulders, planet-sized objects, black dwarfs, black holes, and other known non-emitting entities) can account for no more than about 10 to 12 percent of dark matter. Thus, most dark matter must be in some more exotic form. One suggestion involves weakly interacting massive particles (WIMPs), subatomic particles that, like neutrinos, rarely interact with ordinary matter (other than gravitationally) but are much more massive than neutrinos. The Large Synoptic Survey Telescope (LSST) and experiments like LUX-ZEPLIN (LZ) have made advances in observational techniques for the purpose of detecting WIMPs or other exotic dark matter.

Whatever dark matter is, it is necessary for the modern universe. Temperature (and hence density) fluctuations revealed in the background radiation are about the right size to develop into galaxy clusters and superclusters. Still, they do not contain enough ordinary matter to contract gravitationally into protogalaxies as quickly as galaxies seem to have formed after the Big Bang. Adding a lot of dark matter to these density fluctuations stimulates the rapid growth of galaxies because of the increased gravitational attraction.

Temperature and density fluctuations recorded in the background radiation probably are due to very small quantum fluctuations that occurred spontaneously during the first 10^-35 seconds after the Big Bang and then were magnified by many orders of magnitude during the era of cosmic inflation, when the universe expanded exponentially in the next 10^-30 seconds. Since the ordinary matter was opaque to electromagnetic radiation until a few hundred thousand years after the Big Bang, it would have been buffeted and kept smoothed out by the strong radiation field until finally, the ordinary matter became transparent and decoupled from the radiation. However, since dark matter does not seem to interact with electromagnetic radiation, it would have been able to collect in the magnified quantum fluctuations as soon as inflation ended and, thus, built up density concentrations to attract ordinary matter later.

Another idea that may be relevant to the formation of large-scale structures in the universe is cosmic string theory. According to string theory, space-time has ten or eleven dimensions, but most are “rolled up” or compacted so that only the familiar four space-time dimensions (length, height, width, and time) are noticeable. Cosmic strings are hypothetical long, thin, line-like concentrations of unbroken symmetry left over from the spontaneous symmetry breaking that occurred when the electromagnetic, weak nuclear, and strong nuclear forces separated to close the grand unification era. Specific multidimensional modes of vibration of cosmic strings are thought to be manifested as all the particles and forces in the universe. In the early universe, cosmic strings may have served as “seeds” for forming long concentrations of matter, like droplets condensing on a wire, which later evolved into filamentary chains of galaxies. However, into the mid-2020s, direct evidence of cosmic strings remained elusive.

Context

The history of the quest to understand the universe's structure has been a progression toward recognition of ever more subtle organization at ever-larger scales. Nearly all cultures and societies have divided naked-eye stars of the night sky into patterns called constellations. In early Greco-Roman cosmology (and in many others), stars were attached to the inside of a hollow sphere—the celestial sphere—that enclosed the Earth fixed at the sphere’s center. In the early 1600s, Galileo, in the first recorded use of telescopes to systematically study the sky, discovered many stars too faint to be seen with the unaided eye. Gradually, during the 1600s, the idea developed that stars were similar to the Sun, and, therefore, their different apparent brightnesses meant that they were at different distances from the Earth. Consequently, the stars in a given pattern or constellation might not be a real grouping in space but could be at very different distances from the Earth.

However, this idea could only be confirmed once the distances of stars could be measured directly. The first successful measurements of stellar distances were made in 1838 and 1839, independently by Friedrich Wilhelm Bessel in Germany, Thomas Henderson in South Africa, and Friedrich Georg Wilhelm von Struve in Russia. They all used the method of trigonometric parallax. That involved measuring small changes in the apparent positions of nearby stars relative to more distant stars as seen from Earth at various points in its orbit around the Sun. Trigonometric Parallax is still the most assumption-free method of measuring distances, but it is limited by the ability to measure small angular shifts accurately. Hipparcos, the High-Precision Parallax Collecting Satellite operated by the European Space Agency, has accurately measured parallaxes to distances of about 1,600 light-years. Beyond that distance, other methods must be used, but their calibration is ultimately tied to the distances obtained by trigonometric parallaxes. Thus, constellations came to be seen as merely convenient direction indicators from a person's vantage point on Earth, not physical associations of stars.

In the early 1600s, Galileo discovered telescopically that the hazy white band of light known as the Milky Way consisted of lots of faint stars. Because the Milky Way forms an excellent circle band of light around the sky, by the 1700s, its overall shape was described using terms like sheet, disk, millwheel, and grindstone. In 1784, William Herschel tried to determine its size and shape by counting the stars seen in various directions. These early models all placed the Sun near the Milky Way’s center. It was not until 1918 that Harlow Shapley found that the Milky Way galaxy was much larger than previous estimates, and the Sun was far from the center.

At this time, there was still uncertainty about whether the Milky Way galaxy comprised the entire universe. As far back as the 1700s, there was speculation by Immanuel Kant and others that the spiral nebulae might be other large collections of stars like the Milky Way. During the 1920s, Edwin Hubble, using the Mount Wilson 100-inch reflecting telescope, could resolve individual stars in some spiral nebulae; moreover, some were cepheid variable stars, which may help determine distances. The spiral nebulae were shown to be located far beyond the Milky Way galaxy and hence were galaxies comparable to the Milky Way. A few years later, Hubble found that most galaxy spectra were redshifted, thus confirming the cosmological models of an expanding universe derived from Einstein’s general relativity.

Ever since, central goals of observational and theoretical cosmology have been to produce accurate maps of the distribution of these galaxies and explanations of the origin of the features in this distribution. Pioneering work by George Abell and others, beginning in the 1950s, showed that galaxies were grouped into clusters, some rich with thousands of member galaxies and others poor with fewer than one hundred members. Some clusters are regular, with a spherical shape and many galaxies concentrated near the center; others are irregular, with galaxies scattered across an extended region of space. In the 1980s, Margaret Geller and John Huchra at Harvard-Smithsonian Center for Astrophysics began mapping the large-scale distribution of galaxies out to great distances. They, and teams at other institutions, have shown that out to distances of several billion light-years, superclusters of galaxy clusters are arranged in filaments and walls surrounding large, nearly empty voids. The Sloan Digital Sky Survey (SDSS), a multi-spectral imaging and spectroscopic redshift survey that began in 2000, and the Dark Energy Spectroscopic Instrument (DESI), an instrument used to study the expansion of the universe and dark matter which began its survey in 2021, expanded understandings of the large-scale structure of the universe. Together, they have located the positions of millions of galaxies.

Understanding the origin of structure in the universe on such large scales challenges the creativity of cosmologists. Although the exact details are not yet clear, it seems that all the patterns of organization seen in the universe were determined by events and processes in its earliest moments after the Big Bang.

Bibliography

Belusević, Radoje. Relativity, Astrophysics, and Cosmology. Wiley-VCH, 2008.

Bennett, Jeffrey, et al. The Cosmic Perspective. 10th ed., Pearson/Addison-Wesley, 2024.

Brüning, Jochen, and Matthias Staudacher. Space-Time-Matter: Analytic and Geometric Structures. De Gruyter, 2018.

Chaisson, Eric, and Steve McMillan. Astronomy Today. 9th ed., Addison-Wesley, 2018.

Cohen, Nathan. Gravity’s Lens: Views of the New Cosmology. John Wiley & Sons, 1988.

Drexler, Jerome. Discovering Postmodern Cosmology: Discoveries in Dark Matter, Cosmic Web, Big Bang, Inflation, Cosmic Rays, Dark Energy, Accelerating Cosmos. Universal, 2008.

Duncan, Todd, and Craig Tyler. Your Cosmic Context: An Introduction to Modern Cosmology. Pearson Addison Wesley, 2009.

Ferreira, Pedro G. The State of the Universe: A Primer in Modern Cosmology. Phoenix, 2007.

"Large Scale Structures." National Aeronautics and Space Administration, universe.nasa.gov/galaxies/large-scale-structures. Accessed 14 Jan. 2025.

Lemoine, M., J. Martin, and P. Peter, editors. Inflationary Cosmology. Springer, 2008.

Liddle, Andrew, and Jon Loveday. The Oxford Companion to Cosmology. Oxford University Press, 2008.

North, John. Cosmos: An Illustrated History of Astronomy and Cosmology. University of Chicago Press, 2008.

“Planck.” European Space Agency, www.esa.int/Enabling‗Support/Operations/Planck. Accessed 14 Jan. 2025.

Schneider, Peter. Extragalactic Astronomy and Cosmology: An Introduction. Springer, 2016.

Schneider, Stephen E. ISE Pathways to Astronomy. 6th ed., McGraw-Hill, 2020.

"Sloan Digital Sky Survey." Alfred P. Sloan Foundation, sloan.org/programs/research/sloan-digital-sky-survey. Accessed 14 Jan. 2025.

“Successful Start of Dark Energy Spectroscopic Instrument (DESI) Follows Record-Setting Trial Run.” Berkeley Lab, 17 May 2021, newscenter.lbl.gov/2021/05/17/start-of-dark-energy-survey. Accessed 14 Jan. 2025.

Weinberg, Steven. Cosmology. Oxford University Press, 2008.