Astrobiology

Chapter 3 Origins of a Habitable Planet

On a cloudy March night in 1993, Carolyn Shoemaker was peering at some photographs of the heavens that she, her husband Eugene Shoemaker (1928–1997) of the US Geological Survey, and their amateur astronomer friend David Levy had taken a couple of nights earlier from atop Mount Palomar in Southern California. The two photos, taken an hour apart, were part of a multiyear survey of comets and asteroids. The motivation behind the survey was Eugene Shoemaker’s: he was the world’s preeminent authority on impacts—that is, on the effects of meteorite and cometary strikes on the solid surfaces of the Solar System, a subject he had been studying since his days as a graduate student in the late 1950s. The survey was his attempt to quantify the number of asteroids and comets that might be expected to cross paths with the planets. But what Eugene Shoemaker really wanted was to see the effects of an impact with his own eyes. He had long dreamed of capping his career as one of the founders of astrogeology by exploring a fresh impact crater, perhaps in one of his favorite stomping grounds such as the remote deserts of Australia’s outback. No significant, crater-forming impact, however, had been witnessed during the entire span of recorded human history.*

The photos Carolyn Shoemaker was viewing that night revealed a strange, fuzzy streak in the sky. “I don’t know what I’ve got,” she said, “but it looks like a squashed comet.” The problem was that the fuzzy image on the film looked kind of like a comet, but it did not show the typical bright central core of a comet accompanied by a diffuse tail. Instead, the bright core of the object was an elongated blob, from which not one but several diffuse tails streamed off into space.

In the hour between the two photographs, the object had moved, a sure sign that it was within the Solar System (the stars are so far away that they appear as if fixed in space). Oddly, though, it seemed to be moving in the same direction across the sky—and with the same speed—as Jupiter, which was nearby in the sky and imaged on the same piece of film. Could the strange streak simply be a stray reflection of Jupiter’s bright shine? The elongated streak did not quite line up with the overexposed spot of Jupiter, suggesting that, unprecedented as it was, the streak was real. It wasn’t until the drive home that Eugene Shoemaker struck on a workable theory. What if the comet didn’t just appear to move along the same line of sight as Jupiter but was actually physically near Jupiter in the three-dimensional vastness of the Solar System? If so, the enormous gravity of this planet, by far the most massive in the Solar System, could have raised such large tides in the weak cometary material as to tear it apart. What they had seen, he surmised, was not a single comet but a train of comet pieces produced by passage a bit too close to the giant planet.

As word of the discovery spread, observations of the newly named Shoemaker-Levy 9 comet poured in. (The name stems from the fact that this was the ninth periodic comet, a comet that orbits the Sun in less than 200 years, that the team had codiscovered.) With each new observation, the orbit of this celestial oddity became more precisely defined. The first observations confirmed Eugene Shoemaker’s hypothesis; namely, that the comet had passed within 120,000 km of Jupiter on July 8, 1992, some 20 months before its discovery. By early April 1993, the data were plentiful enough to nail down an approximate orbit and, to everyone’s surprise, it turned out that the comet wasn’t in orbit around the Sun at all. It seems that this ancient resident of the outermost reaches of the Solar System had been captured into orbit around Jupiter way back in 1929, and the comet was now a moon of the giant planet. More startling still, by the end of May 1993, the data were voluminous enough to define the comet’s orbit precisely, and it was found that Shoemaker-Levy 9 was not going to survive its next trip. After circling Jupiter for 65 years, the comet’s next close pass, slated for July 25, 1994, was going to let Eugene Shoemaker see his dream come true—albeit not an impact on Earth, but an impact of the comet, which his team had discovered, onto the largest planet in the Solar System.

Researchers far and wide studied this “impact of the century” using Earth-based telescopes, the Hubble Space Telescope, and the Galileo probe, then still en route to its six-year orbital tour around Jupiter. In all, 21 impacts were observed as pieces of the comet train slammed into the deep atmosphere of Jupiter (fig. 3.1). Huge dark welts many times the size of the Earth appeared and lasted in the atmosphere of the gas giant planet for weeks. Spectroscopic studies suggest that large fractions of the estimated 1-billion-ton mass that ploughed into the planet must have consisted of substances such as water (estimated at approximately 20 million tons), ammonia, and methane, collectively known as “volatiles” in planetary science due to their relatively low boiling points. And while Jupiter wasn’t exactly in need of the extra bulk, the impact of Shoemaker-Levy 9 highlighted Jupiter’s huge influence on the movements of stray bodies in the Solar System. With a mass 0.1% that of the Sun and more than two and a half times that of all the other planets combined,* Jupiter’s gravitation is a force to be reckoned with. And as we’ll see, Jupiter’s influence on the motions of objects in orbit around the Sun is so significant that it played a critical role in the evolution of Earth as a habitable planet.

Images — **Figure 3.1** The impact of the comet Shoemaker-Levy 9 with Jupiter left enormous, Earth-dwarfing scars in the planet’s atmosphere (lower left). The impacts also delivered millions of tons of carbon, oxygen, and nitrogen compounds from the far reaches of space to the gas giant, highlighting the role that Jupiter has played in controlling the dynamics of the Solar System. (Courtesy of NASA/STScI)

The Proto-Sun

As described in chapter 2, our Sun is a middle-sized yellow star. From observations of newly born stars in relatively nearby cosmic nurseries and from detailed studies of the composition of the Sun’s planets, asteroids, and comets, astronomers have pieced together the story of the birth and evolution of our Solar System.

As the pre-solar nebula contracted, conservation of angular momentum forced the contracting dust and gas to spiral ever more rapidly around the cloud’s center of gravity. Eventually, while the large bulk of the cloud that was to form the Solar System collapsed to form the proto-Sun, a modest proportion of the nebula fell into orbit around the rapidly forming star. As the cloud slowly collapsed under its internal gravitational pull, its gravitational potential energy was converted into kinetic energy, and the inner core of the nebula became quite hot. Eventually, the center of the nebula reached 10 million kelvins, the temperature at which the fusion of hydrogen into helium ignites, and a new star was born. The ignition of fusion occurred before the pre-solar nebula had entirely collapsed, and thus the early Sun, like all very youthful stars (which are called T-Tauri stars after a cluster of well-studied examples in the constellation Taurus), was wrapped in a dense cloud of nebular materials (fig. 3.2). In those early days, the Sun was rather more active than it is today; observations of dozens of relatively nearby, T-Tauri stars in the Orion nebula demonstrate that they are much hotter, rotate dozens of times more rapidly, and expel a stream of charged particles (the stellar wind) that is thousands of times more energetic than the solar wind pumped out by our now middle-aged star.

The dregs of material that remained in orbit around the new Sun moved at first with a bewildering array of orbital inclinations and eccentricities. Nongravitational forces (primarily friction with the remaining gas), though, fairly quickly established a degree of order in the movements and confined most of the material to a thin disk in the Sun’s equatorial plane, an arrangement that still holds most of the matter in the Solar System today. How does this work? Just imagine one stray little rock orbiting the Sun on a path that is tilted relative to the thick main disk of gas and dust. During each orbit, it would pass through the disk once on its ascent above the equatorial plane and again half an orbit later on its descent. During each crossing, the stray rock would lose some of its out-of-plane velocity to friction with the gas and dust. Similar drag quickly ordered the early solar nebula, herding the remaining gas and dust into a thin disk nearly in the equatorial plane of the Sun. Although such a disk does not emit visible light, the young star warms it, and thus it does emit at infrared and millimeter wavelengths. By observing at these wavelengths, astronomers have been able to image such “protoplanetary disks” around a number of nearby young stars (fig. 3.3).

The Formation of the Planets

Protoplanetary disks are far from homogeneous. Heated from within by friction (in effect, the gravitational potential energy liberated as the gas and dust spiral inward) and from the light and solar wind emitted by their young central star, the inner portions of these disks reach several thousand kelvins. Farther out, though, the temperature of the disk falls dramatically: dust and gas in the outer reaches of the cloud lose less potential energy and, because the inner cloud is fairly opaque, receive less energy from the star. Thus, protoplanetary disks exhibit strong temperature and pressure gradients. The current theory of the formation of our Solar System, known as the equilibrium condensation model, attempts to explain the size, location, and composition of the Sun’s planets (fig. 3.4) as a function of these temperature and pressure gradients—and the chemical heterogeneity they produced—in the Sun’s protoplanetary disk. And, as we’ll see, it does a remarkably good job, albeit with two glaring, and astrobiologically relevant, exceptions.

At the low pressures found in protoplanetary disks, liquids are not stable. Thus, the only relevant phase change is the sublimation of solids directly into gases and, as a complement, the deposition of solids from the vapor. Astrophysicists refer to materials with high sublimation points, such as the iron and nickel that constitute the core of our planet, as “refractory.” Near the inner reaches of a disk, only the metals (used in the stricter, chemical sense now, i.e., shiny, electrically conductive elements) and most refractory oxides, such as alumina and a few aluminum-rich silicates, condense to form solid particles (fig. 3.5). Slightly farther out, less refractory, aluminum-free silicates, such as silica, also condense. At much larger distances from the central proto-star, molecules we consider volatile, such as water, ammonia, and methane (the cosmochemically most abundant molecular forms of oxygen, nitrogen, and carbon, respectively) condense to form ices as the temperature continues to drop with increasing distance.

These condensed materials are the fodder from which the planets in our Solar System were made. Conglomerations of them rapidly built up, first as dust, but then, as dust motes coalesced, centimeter-sized particles and meter-scale boulders. Due to gas drag, the smaller (centimeter- to meter-sized) particles of solid material that condensed from the nebula slowly spiraled inward. As they were not in purely Keplerian orbits, that is, their motions were not solely defined by gravity, but also by friction,* they suffered frequent, slow collisions, causing them to agglomerate into still larger objects. By the time this accretion had generated kilometer-sized bodies, an additional force kicked in and accelerated the process: the mutual gravitational attraction of kilometer-sized bodies is sufficiently large that it begins to dominate gas drag and greatly accelerate the rate at which protoplanets grow.

As described above, the composition of the planetesimals present in the early Solar System varied as a function of distance from the Sun, with metal-rich planetesimals dominating near the center, silicate-rich ones at the middle distances, and volatile-rich ones farther out in the cold. This equilibrium condensation model, so named because the composition of each neighborhood was determined by the equilibrium chemistry that could occur at the temperature found there, roughly predicts the composition of each of the planets (table 3.1). And even though the predictions are only rough—because there was some scattering of particles away from where they originated—the model does a reasonable job of explaining why the small, solid, terrestrial planets*— Mercury, Venus, Earth, and Mars—are in the inner Solar System; the “gas giants,” Jupiter and Saturn, come next; the “ice giants,” Uranus and Neptune, follow; and the smaller, icy outer worlds of the Kuiper belt, the scattered disk, and the Oort Cloud extend farther out still.

Table 3.1

Physical parameters of some prominent Solar System objects

Object	Distance from Sun (AU)	Composition	Mass (Earth = 1)	Density (g/cm³)
Mercury	0.39	Metals + alumina	0.06	5.43
Venus	0.72	Metals + alumina + silicates	0.82	5.24
Earth	1.00	Metals + alumina + silicates	1.00	5.51
Moon	1.00	Metals + alumina + silicates	0.012	3.34
Mars	1.52	Metals + alumina + silicates	0.11	3.93
Vesta	2.36	Metals + alumina + silicates	0.00004	3.46
Ceres	2.77	Metals + alumina + silicates + water	0.00014	2.16
Jupiter	5.21	Metals + alumina + silicates + water + hydrogen + helium	317.8	1.33
Saturn	9.58	Metals + alumina + silicates + water + hydrogen + helium	95.1	0.69
Uranus	19.28	Metals + alumina + silicates + water + ammonia	14.6	1.27
Neptune	30.14	Metals + alumina + silicates + water + ammonia	17.2	1.64
Pluto	39.88	Metals + alumina + silicates + water + ammonia + solid methane	0.003	1.86
Eris	68	Metals + alumina + silicates + water + ammonia + solid methane	0.003	2.52

Note: AU = astronomical unit.

Near the Sun, the terrestrial planets are formed of various ratios of refractory metals and silicates. As Mercury is the closest to the Sun, it should consist mostly of refractory metals, with a smaller component of silicates. Consistent with this, Mercury’s mean density of 5.43 g/cm³ (table 3.1) suggests that its bulk composition is about half silicates (mean density a bit under 3 g/cm³; table 3.2) and about half nickel and iron (mean density about 8 g/cm³). The metals, which are much denser, sank to form the core, and the silicates, floating like slag on iron in a blast furnace, formed the mantle and crust. The first direct evidence in favor of this model came during the 1973 flyby of the Mariner 10 spacecraft, humanity’s first expedition to the inner reaches of the Solar System. Mariner discovered that Mercury possesses a magnetic field, suggesting that the planet’s core is conducting, as would be expected for liquid iron. Mercury’s iron core was later confirmed by the MESSENGER spacecraft, which orbited Mercury from 2011 to 2015. Studies of how the planet altered MESSENGER’s orbit found that a core of iron-like density extends through 85% of the planet’s radius (i.e., 60% of its volume).

Table 3.2

Properties of common materials in the protoplanetary disk

Material	Density (g/cm³)
Primordial gases (at 25°C)
Hydrogen (1 atm)	9 × 10⁻⁵
Hydrogen (10⁶ atm)	0.3
Helium (1 atm)	1.2 × 10⁻⁴
Helium (10⁶ atm)	0.5
Common ices (at their melting points)
Nitrogen	0.81
Methane	0.49
Ammonia	0.82
Carbon dioxide	1.50
Water	0.92
Common silicates (at 25°C)
Quartz	2.65
Granite	2.69
Basalt	3.01
Common metals (at 25°C)
Iron	7.87
Nickel	8.91

Stepping out from Mercury, we find the other terrestrial planets Venus, Earth, and Mars. The densities of Venus and Earth are slightly less and slightly greater than that of Mercury, respectively (table 3.1). Venus and Earth, however, are some 14 and 17 times more massive than Mercury, and the increased gravity associated with these greater masses leads to greater compression of the material in their cores. Were it not for this compression, it is estimated that the bulk densities of Venus and Earth would be about 4.2 g/cm³, which is equivalent to about a 3:1 mixture of rock and metal. Correspondingly, the Earth’s dense metal core extends through only a bit more than half of its radius, rendering it, relative to the size of the planet, significantly smaller than Mercury’s. This trend of increasing rock-to-metal ratios continues with Mars, which has an observed density of 3.93 g/cm³ and an estimated uncompressed density of just 3.3 g/cm³, indicating that it is composed of an even larger fraction of rock (the exact size of its core awaits confirmation from the ongoing InSight Mars lander mission). In short, these planets formed at distances far enough from the early Sun that silicates readily condensed, and thus all three consist of a thick rocky mantle surrounding a relatively small metallic core.

Further corroboration of the equilibrium condensation model is seen in the ratios of potassium to uranium found on the rocky planets. Both elements are radioactive, rendering them easy to detect. Both also tend to remain in the crust, where they can be observed, rather than partitioning into the iron-nickel core. But uranium-containing minerals are far more refractory than those that contain potassium, and thus the ratio of these two elements reveals the temperature at which a planet’s constituents condensed. The Russian Venera 8 and Vega 1 and 2 landers, which each spent a few tens of minutes sampling the surface before being overwhelmed by the heat, found the potassium-to-uranium ratio on our nearest neighbor to be 7,000:1. This ratio rises to about 12,000:1 for the Earth, 18,000:1 for Mars, and to 70,000:1 by the time we reach the outer asteroid belt (fig. 3.6). As the theory goes, as we move farther from the Sun and the temperature of the protoplanetary disk drops, more of the volatile potassium compounds solidify to be swept up into planets. This said, there is an oddity: at 2,500:1, the Moon’s ratio is less than a fifth that of the Earth’s, suggesting the stuff from which it was made is far more refractory than the material that formed our planet. More on this in a moment.

Beyond Mars lies the asteroid belt and beyond that, the outer Solar System. This area of space is dominated by Jupiter, more than 300 times as massive as the Earth, and the other gas giant planets each more than a dozen times as massive as the Earth (itself the largest of the rocky, terrestrial planets of the inner Solar System). A key step toward acceptance of the equilibrium condensation model was its ability to explain this enormous inequity. Even though we think of Jupiter as a gas giant, it isn’t made entirely of gas, and the key to understanding its size and location lies in one of its other components: water.

As oxygen is one of the most cosmically abundant of the elements heavier than helium, water (a combination of oxygen with the Universe’s most abundant element) was a major component of the solar nebula. Water, however, is volatile. In the low pressure of the planetary disk, water vapor could not form liquid water at all, but it would condense to form ice when the temperature of the disk hit approximately 150 K at the so-called “snow line.” Studies of meteorites and the asteroids they arise from indicate that asteroids beyond about 2.7 AU (astronomical units; 1 AU is equal to the mean Earth-to-Sun distance of about 150 million kilometers) contain a significant fraction of water. For example, Ceres, the largest asteroid (now classified as the only “dwarf planet” in the asteroid belt), which orbits just outside the snow line at 2.8 AU, has a mean density of just 2.16 g/cm³ and is alone estimated to contain as much water as all the Earth’s oceans despite weighing in at only about 1/7,000 the mass of our planet. Indeed, the Dawn spacecraft, which dropped into orbit around Ceres in March 2015, even found signs of relatively recent (last few tens of millions of years) water-driven “volcanism” (fig. 3.7). In contrast, the 3.46 g/cm³ density of the second most massive asteroid, Vesta, which orbits the sun at a distance of 2.4 AU, suggests that it is comprised predominantly of basalt surrounding a small iron-nickel core.

Because water must have been abundant in the protoplanetary disk—the density of which decreased with increasing distance from the Sun—Jupiter, which resides immediately beyond the asteroid belt, was in a prime location. At this distance from the early Sun, water ice condensed in larger amounts than anywhere else and rapidly accreted with metals and silicates to form a “proto-Jupiter” with a mass 10 to 15 times that of the present-day Earth. (A major goal of the Juno spacecraft, which entered a polar orbit around Jupiter in July 2016, is to better understand the nature of the Jovian core.) At this mass, gravity became strong enough to pull in and hold onto the gases hydrogen and helium. The terrestrial planets never reached this step; they could only accrete by collisions of solid objects as their gravity was too weak to hold onto the much more abundant hydrogen and helium that were by far the dominant components of the nebula. In contrast, young Jupiter acted like a giant cosmic vacuum cleaner, rapidly sweeping up all the material in or near its orbit until it mostly consisted of hydrogen and helium. As a result, Jupiter grew rapidly, ultimately becoming, by a factor of more than three, the most massive planet in the Solar System despite having a density of just 1.33 g/cm³. Indeed, it is thought that Jupiter grew so rapidly that, even before it had finished growing, its gravity disrupted the accretion process nearby and prevented the formation of a planet where we now see the asteroid belt. It may also have starved the growing proto-Mars, leaving the Red Planet significantly smaller than it would have been without its giant neighbor.

The story for Saturn starts similarly, with the rapid formation a rock and water core that began to attract hydrogen and helium. But as one moves outward from the center of the protoplanetary disk, not only does the density fall, but also the orbital speed. These two effects slowed accretion, leaving Saturn unable to accrete as much hydrogen and helium as Jupiter before these gases were driven off by the intense solar winds of the T-Tauri–stage Sun. Saturn is thus less than a third the mass of Jupiter and, because its weaker gravity does not compress its gases as much, only reached a density of 0.69 g/cm³.

The story for Uranus and Neptune is a bit different. Although born in still less-dense portions of the protoplanetary disk, these planets had the advantage of being so far down the temperature gradient that ammonia and methane (which, as the hydrides of nitrogen and carbon, respectively, are also cosmologically abundant) also condensed, allowing the planets to build up rock and ice cores several times the mass of the Earth. However, neither planet grew bulky enough to pull in and hold significant amounts of molecular hydrogen or helium before these gases were blown out of the Solar System, so their bulk densities come in at 1.27 and 1.64 g/cm³, respectively. Given this, astronomers now categorize these near twins as “ice giants” to distinguish them from the true gas giants Jupiter and Saturn.

Still farther out from the Sun, low densities and centuries-long orbital periods greatly slowed accretion, ensuring that only small, icy bodies could form. Some of these bodies form a disk analogous to the asteroid belt, named the Kuiper belt after the Dutch-American astronomer Gerard Kuiper (1905–1973), stretching from 30 AU (approximately the orbit of Neptune) to 50 AU from the Sun. Still others form a much more irregular disk, the “scattered disk” that extends to greater than 100 AU and is comprised of icy bodies tossed into the outer Solar System by Jupiter and the other giant planets (more on this in a moment). The largest known members of the Kuiper belt and the scattered disk are Pluto and Eris, respectively, the latter of which was discovered in 2005 by Mike Brown of Caltech and named by him after the Greek goddess of strife and discord and which led to the former losing its planetary status. These two “dwarf planets” weigh in at only 0.22% and 0.27% of the Earth’s mass, or about one-fifth the mass of the Earth’s Moon. At 1.86 and 2.52 g/cm³, the bulk densities of these frozen orbs suggest that Pluto and Eris are comprised of slightly more than and slightly less than half ice, respectively, with the rest being rock.

How rapidly did these condensation and accretion processes occur? Chondritic meteorites have given us an indication of the timescale. These rocky space travelers are the most abundant type of meteorite; they consist of the oldest, most primitive and unaltered material in the Solar System and thus offer a window into the first condensation events. The oldest chondrite precisely dated using radioisotope dating (see sidebar 3.1) is 4.567 billion years old, providing a date for the first condensation events. Most other chondrites date to within 20 million years of this age, suggesting that the condensation process was very rapid indeed relative to the age of the Solar System. Computer simulations of the condensation of the pre-solar nebula that take into account the physics and chemistry of the problem suggest that, once condensation into meter-sized bodies was complete, the accretion of these planetesimals into the planets we see today required only another few million to few tens of millions of years. Given the magnitude of the construction process (after all, we are talking about the creation of an entire planetary system here), this is an astonishingly short time. But as described above, studies of T-Tauri stars indicate that, within a few million years of the start of fusion, the fierce solar wind of the early Sun would blow most of the remaining dust and gas out into interstellar space, limiting any further accretion to that involving larger planetesimals. Specifically, surveys with infrared telescopes (infrared, again, being particularly sensitive to warm dust) indicate that, while more than 80% of stars exhibit evidence of a protoplanetary disk when they’re a million years old, by the time their 10 millionth birthday comes around, fewer than 1% do. In short, if it were to happen at all, the accretion of the Solar System had to happen rapidly.

Sidebar 3.1

Radioisotope Clocks

Many nuclei are unstable and decay through a range of processes collectively known as radioactivity. Because the rate of decay of any one type of nucleus (an isotope) is well known and is completely independent of environmental conditions, the decay process can be used as a clock with which to date geological events.

In principle, the concept of radioisotope dating is straightforward. We simply compare the amount of a given isotope present today in a sample with the original concentration of the isotope. Using the known rate of decay (as given by the isotope’s half-life), it is a simple matter to calculate how long the decay has been taking place and thus the age of the sample. But radioisotope dating is straightforward only in principle. The rub is figuring out how much of the relevant isotope was there in the first place.

For some isotopes and some decay routes, the task at hand isn’t so hard. For example, the isotope potassium-40 (⁴⁰K) decays with a half-life of 1.26 billion years, either by the emission of an electron, which converts a neutron into a proton to form calcium-40 (⁴⁰Ca), or by capture of an electron by a proton to form a neutron, to produce argon-40 (⁴⁰Ar). Unfortunately, ⁴⁰Ca is the most abundant isotope of calcium and is extremely common in minerals, and thus it is not possible to distinguish the ⁴⁰Ca produced from the decay of ⁴⁰K from the ⁴⁰Ca initially present when the rock was formed. The ⁴⁰Ar produced by the second decay process, in contrast, is well suited for radioisotope dating: argon being an extremely inert gas, if any was originally present it would escape when the rock was last molten. The ⁴⁰Ar in a rock today, then, must have come from the decay of ⁴⁰K since the rock crystallized. Thus, using the known half-life of ⁴⁰K and knowledge of the amounts of ⁴⁰K and ⁴⁰Ar in a rock, we can estimate the time that has passed since the rock solidified to its present form.

Other isotopes also make suitable targets for radioisotope dating, but the determination of their original concentrations requires more effort. For example, uranium-238 (²³⁸U) transmutes through several steps into lead-206 (²⁰⁶Pb) with a half-life of 4.47 billion years. Some minerals, such as zircon, preferentially exclude lead ions from their crystal structure while including the better-fitting uranium ions. Thus, when a zircon crystallizes, its initial lead concentration is zero. Over time, any ²³⁸U in the zircon decays into ²⁰⁶Pb. Knowledge of the current ²³⁸U content plus the current ²⁰⁶Pb content therefore indicates the original ²³⁸U content, which in turn provides a means of dating the crystallization of the material.

Another commonly employed isotope for dating is rubidium-87 (⁸⁷Rb), which decays into strontium-87 (⁸⁷Sr) with a 48.8-billion-year half-life. Here the availability of an isotope of strontium that is not the product of radioactive decay, strontium-86, provides a means of estimating the original strontium concentration. With this, and with knowledge of the current amounts of ⁸⁷Rb and ⁸⁷Sr, we can calculate the original and current ⁸⁷Rb concentrations and thus the time that has elapsed since the rock crystallized.

All three of these radioisotope clocks—and others—have been used to date the formation of more than 70 different meteorites and so provide a lower limit on the age of the Solar System. And while the observed dates vary over the range 4.53 to 4.58 billion years ago, the best estimate is 4.56 billion years. Given that many of these meteorites are very primitive (and thus unlikely to have had their radioisotope clocks “reset” since they condensed from the pre-solar nebula), this “lower limit” is probably within 10 million years of the actual date on which the pre-solar nebula started to condense to form the Solar System.

The loss of protoplanetary disk material during the T-Tauri stage is quite dramatic. The total amount of, say, iron in all the Solar System’s planets is equivalent to about 3% of the mass of all the iron in the Sun, implying that, if the protoplanetary disk started out with the same composition as the Sun (a good bet), then the disk’s mass was at least 3% that of the mass of the Sun. Moreover, if, in addition to losing volatiles, the disk lost iron-containing dust (also a good bet), it would have been even more massive. Consistent with this, the protoplanetary disks circling young, Sun-like stars typically weigh in at about one-tenth the mass of the Sun. In contrast, the total mass of all the planets, dwarf planets, and asteroids in our Solar System today amounts to 0.14% of the mass of the Sun, meaning that 95% to >98% of the protoplanetary disk was blasted into interstellar space during the Sun’s T-Tauri stage.

The Mysterious Moon

The accretion model makes for a planetary system that changes gradually and logically from the inner planets toward the outer planets. However, one particular Solar System body falls outside this logic in a number of aspects: the Moon. As we’ll see, the Moon’s gravitational effects here on Earth have exerted a significant influence on life on the planet, so we should have a closer look at why the Moon is there.

Until the 1990s, there was no clear consensus regarding the origins of the Earth’s natural satellite. Indeed, before the 1970s, three theories had been put forth as to the Moon’s origins. The first, the “wife hypothesis,” suggested that the Moon originated elsewhere in the Solar System and only later was captured into orbit around the Earth. The “sister hypothesis” was based on the premise that the Moon originated in the same neighborhood as the Earth but failed to fuse with it. The third, not surprisingly called the “daughter hypothesis,” postulated that the Moon, perhaps due to the extremely rapid rotation of the early Earth, was torn from the Earth. One imaginative version of the latter theory postulated that the large hole left behind after the split became the Pacific Ocean—this was well before the acceptance of plate tectonics, a paradigm that implies the Pacific is far younger than the Moon. By the late 1960s, the scientific community had largely discounted the daughter hypothesis based on the improbably high rotation rates (equivalent to about a 2-hour-long day) required before centripetal forces would overwhelm the gravitational forces that held the proto-Earth together. But there was no clear advantage to either of the two remaining hypotheses. Everybody expected the question would be settled by the early 1970s once the Apollo missions brought back Lunar samples with which to test the two competing models. It was not.

The Apollo missions and the minerals they brought back only uncovered new contradictions. The isotopic distributions found in the Lunar samples (e.g., the ratios of oxygen-16 to oxygen-17 and oxygen-18 in various minerals) are similar to those found on Earth. This ruled out the wife hypothesis because objects accreted in different parts of the solar nebula exhibit different isotopic patterns. Likewise, however, the sister hypothesis is ruled out by the observation that the elemental makeup of the Moon is vastly different from that of the Earth. The Moon’s density, for example, is only 3.34 g/cm³, meaning that, unlike the Earth, it lacks a substantial iron-nickel core and consists almost entirely of silicate rocks. Likewise, when compared with Terrestrial rocks, the minerals collected from the Lunar surface are strangely depleted in the lower boiling point metals, such as sodium and potassium. The ratio of potassium to uranium in the Lunar crust, for example, is only 2,500:1, indicating that the Moon is greatly depleted in the former, rather volatile element relative to the level found in the Earth’s crust (see fig. 3.6).

After a couple of decades of hand-wringing about these conflicting bits of evidence, the planetary science community has now achieved some degree of consensus regarding the origins of the Moon. According to the currently in-favor theory, the Moon formed by a “daughter-like” mechanism. However, the force that disrupted the planet was not rapid rotation but, instead, the impact of a Mars-sized object late in the accretion process that very nearly ripped the Earth apart. The force of the impact would have melted both the proto-Earth and the impactor and splashed an enormous amount of liquid rock into space. Given that the Earth had already differentiated into a dense, iron-nickel core and a rocky mantle, the splashed material would have been almost entirely silicates (the impactor was presumably differentiated too, but its metal core would have sunk and merged with the Earth’s). With the heat of the impact, any volatiles and even sodium- and potassium-containing minerals would have evaporated, leaving the Moon as it is today: a piece of Earth rock (explaining the Earth-like isotopic abundances) depleted of both volatile elements and metals (accounting for its differing bulk composition). And when did this cosmic collision occur? The radioactive decay of hafnium-182 into its daughter isotope tungsten-182 gives us a clue. Tungsten is a “siderophile,” an element that dissolves readily in iron-nickel alloys, so any tungsten present in the proto-Earth would sink into the core. Hafnium, in contrast, remains in the crust. Measurement of the tungsten-182 currently in Lunar rocks thus provides an indirect measure of the amount of hafnium-182 that was present in the crust when the Moon formed. Comparing this with the amount of tungsten-182 in the oldest meteorites (the parent-body sources of which did not differentiate to form a core, thus leaving their primordial tungsten levels unchanged), combined with knowledge of the 9-million-year half-life of hafnium-182 (see sidebar 3.1), suggests that the Moon formed some 30 million years after the beginning of accretion. Thus, the Moon—which, as we’ll see, is thought to play a vital role in the evolution of life on Earth—was formed in a fluke accident near the tail end of the accretion period.

Cleaning Up the Mess and Delivering the Goods

Next to the odd composition of the Moon, the abundance of water and other volatiles on Earth is the other striking deviation from what the equilibrium condensation model of planet formation would predict. According to this theory, the temperature of the gas and dust that condensed to form the proto-Earth was far too high to allow water to condense, and thus the Earth should not contain significant amounts of this or any of the other volatile hydrogen, nitrogen, and carbon compounds critical for life. And yet the bulk composition of our planet includes, for example, about 350 parts per million water, about 300 parts per million carbon, and about 50 parts per million nitrogen (equivalent to 0.035%, 0.03%, and 0.005%, respectively). This, obviously, is not a lot, but it is enough to support all life on our planet and cause some problems for the equilibrium condensation model. Where did these volatiles come from? Most “planetologists” believe that they came from out beyond the snow line.

The presence of Jupiter, Saturn, Uranus, and Neptune had far-reaching consequences for life on Earth. In particular, the intense gravity of these massive planets affects orbital dynamics throughout the outer Solar System, which ultimately led to the “cleanup” of most of the planetesimals remaining after the end of accretion. The idea is that, through orbital resonances or close encounters, the outer giant planets would perturb the orbits of any planetesimals in their neighborhoods. Whenever they’d toss a planetesimal in toward the Sun, this would cause the orbit of the planet doing the tossing to migrate slightly outward. And whenever they’d toss anything out away from the Sun, this would cause the orbit of the tossing planet to migrate a bit inward. Jupiter, though, threw a bit of a wrench into the works: because it is so much larger than all the rest of the planets, it alone has enough gravity to toss planetesimals entirely out of the Solar System. The fact that Jupiter tossed many planetesimals entirely “out of the ballpark” created a net asymmetry that, over the first few hundred million years of the Solar System, caused the orbits of Neptune and, to a lesser extent, Saturn and Uranus to migrate outward and the orbit of Jupiter to migrate inward. As these planets migrated, almost all the planetesimals between Jupiter and Neptune were either tossed inward, where they would eventually collide with the rocky, inner planets, or were lifted into the scattered disk, into the more distant Oort Cloud, or even entirely out of the Solar System.* This “cleanup” produced two effects that were so vital to the formation of life on Earth that, in a nutshell, without Jupiter we would not exist.

The first effect of all this planetesimal tossing was to deliver volatiles to the otherwise “dry” proto-Earth. Such delivery is still occurring today. A dramatic example of the probable extraterrestrial origins of Earth’s volatile organic inventory occurred on January 18, 2000, when an approximately 60 ton carbonaceous chondrite meteorite was seen to explode over Tagish Lake in the far north of Canada’s British Columbia province. As the fall occurred in the middle of winter, it was easy to snowshoe out onto the lake and collect pristine samples of fresh meteoritic material. Fortunately for those of us interested in this sort of thing, one of the few inhabitants of this remote area was an outdoorsman by the name of Jim Brook, who understood the importance of such pristine materials. Venturing out on the ice a week after the fall, he carefully collected uncontaminated samples in clean plastic bags and stored them in his freezer.**

The couple of hundred meteoritic samples that Brook and later expeditions recovered are some of the freshest extraterrestrial material we have on hand to study. Spectroscopic studies of the Tagish meteorite indicate that it is a good match for the asteroid 773-Irmintraud, which orbits in the outer reaches of the asteroid belt, well beyond the snow line. Sometime in the past 100 million years, a meteor impact broke off a chunk of this asteroid, and the gravitational perturbations of Jupiter sent the chunk earthward. Of note, the Tagish meteorite is composed of 5% total carbon and about 3% organic material, mostly as aromatic hydrocarbons. The water content of Tagish is somewhat harder to determine because the meteorite fell on snow. But typical carbonaceous chondrites are 5% to 20% water and, with comets, could have been major suppliers of the Earth’s oceans. Even today, giant outer planets in our Solar System continue to push volatiles inward toward the rocky inner planets.

In addition to moving volatiles inward, the era of gas giant migration and its resultant planetesimal tossing had a second profound impact on the origins of life on Earth: it put a hard stop to the process of accretion. The history of the end phase of the cleanup is readily visible on any clear, moonlit night: the impact craters left behind by these outer Solar System planetesimals have been preserved on the Moon’s face due to the absence of any remodeling of the Lunar surface by erosion or plate tectonics. With the Lunar rocks they brought back, the Apollo astronauts obtained more quantitative evidence regarding this cleanup stage. The 382 kg of Lunar samples allowed geologists to isotopically date various surfaces on the Moon, then calculate the rate of crater formation during various epochs of the Moon’s history. In doing so, they found that the cleanup phase ended with what is now called the “late heavy bombardment,” some 3.8 billion years ago (fig. 3.8).

One of the last major craters formed in the late heavy bombardment was the Imbrium Basin, which, at 1,160 km in diameter, is easily visible from Earth as the right eye of the Man in the Moon. The size of this crater amply demonstrates that these impacts delivered not only volatiles but also enormous kinetic energy. The planetesimal that formed the Imbrium Basin is estimated to have been about 400 km in diameter. An impactor of this size striking the Earth would provide enough kinetic energy not only to boil all the water in the Earth’s oceans but also to vaporize hundreds of meters of the crust over the entire globe. Because impacts even significantly smaller than the one that formed Imbrium pack sufficient energy to sterilize the entire planet, it is safe to assume that we owe our existence today to the era, billions of years ago, when the outer planets largely swept the Solar System free of such impactors. Indeed, studies of the effect suggest that Saturn, Uranus, and Neptune are too small to have cleaned up the Solar System in any reasonable time frame, and so, were it not for Jupiter’s vast bulk, the Earth would have been subjected to sterilizing impacts for billions, rather than hundreds of millions, of years after its formation.

The Volatile Inventories of the Other Inner Planets

The other inner, rocky bodies, of course, were also subjected to the late heavy bombardment. But even if you have a rocky planet to start with and get the volatiles delivered to your doorstep, things can still go spectacularly wrong. On innermost Mercury, for example, intense heat and relatively weak gravity allowed most of the volatiles to escape into space. Most, but not all. Mercury’s spin axis is almost perfectly perpendicular to its orbital plane. Because of this, there are no seasons on Mercury, and the bottoms of craters at the poles, forever in shadow, are chilled to a temperature of a few kelvins. These craters act as cold traps and thus, as shown by both radar studies from the Earth and neutron spectrometer measurements of the hydrogen content of the surface by the MESSENGER Mercury orbiter, are filled with ice (fig. 3.9). The Moon yields a similar picture; the mean temperature of the Moon at its distance from the Sun is much lower than that of Mercury, but the Moon’s gravity is still lower, and thus it is able to hold only the tiniest whiff of atmosphere. Indeed, the Moon’s atmosphere is so thin that, between rocket exhaust gasses and the astronauts venting their cabin air, each Apollo landing temporarily doubled its mass. The excess gasses, though, were lost to space over the course of days, returning the Lunar atmosphere to its normal state. Like Mercury, though, the Moon does harbor some deep, ice-filled craters at its poles.

But what about Mars and Venus? Mars we’ll discuss in detail in chapter 9 as a potential abode for life. But Venus is nowhere near as hospitable. This is ironic because, in many respects, Earth and Venus are near twins. Venus is our nearest planetary neighbor, orbiting only 28% closer to the Sun than Earth. And, being only 5% smaller in diameter and 18% less massive than the Earth, Venus is also quite similar in size, suggesting that the two should have been on the receiving end of similar amounts of volatiles delivered from the outer Solar System. Consistent with this, their estimated carbon inventories are nearly identical, and their estimated nitrogen inventories are the same to within a factor of three (fig. 3.10). But here the similarities end. Whereas more than 99% of the Earth’s carbon is stored in the crust (primarily as carbonate rocks, such as limestone), almost all of Venus’s carbon is in its 96.5% carbon dioxide atmosphere, which blankets the planet with a pressure some 93 times greater than that found on Earth. Carbon dioxide, of course, is a greenhouse gas. So much so that the average temperature on the surface of Venus is more than 460°C,* hot enough to melt lead, tin, and zinc. From whence does this dramatic difference arise? The answer lies in Venus’s water, or lack thereof: Venus is 200,000 times drier than the Earth.

Planets tend to lose light atoms, such as hydrogen and helium, in a process called Jeans escape, named after the same Jeans whose “Jeans mass” defines the minimum sizes of stars. Jeans escape occurs when the velocity of the most rapidly moving atoms exceeds the planet’s escape velocity and they are lost to space. At a given temperature, molecular (or atomic) velocity is inversely proportional to the square root of mass, and thus the light hydrogen isotope ¹H, which moves on average 41% more rapidly than the heavy isotope ²H (deuterium), is more likely to escape, leaving the remaining water on the planet enriched in the heavier isotope. In 1979, the Pioneer Venus mission sent four entry probes into the atmosphere of Venus and measured, for the first time, the planet’s hydrogen to deuterium ratio. At 62.5:1, the ratio Pioneer found is 100 times greater than the ratio seen on Earth. Venus, in fact, is so deuterium rich that it is believed to have lost enough hydrogen to fill several oceans’ worth of water. It is this lost water that is the ultimate cause of the vast climatic differences between the Earth and its twin.

Water plays a vital role in the regulation of carbon dioxide levels on Earth and so is critical in regulating the Earth’s temperature. When atmospheric carbon dioxide levels rise (due to volcanism, for example), the greenhouse effect increases temperature, which in turn increases rainfall and the fluvial (i.e., by water) and chemical (reactions speed up at higher temperatures) weathering of rocks. This weathering releases calcium and magnesium ions from the rock, which flow into the ocean where they react with carbon dioxide (as carbonate ion) to form calcium carbonate (limestone) and calcium-magnesium carbonate (dolomite) in a process that, on Earth, removes carbon dioxide from the atmosphere on a timescale that is geologically rapid (fig. 3.11). This sequestration of carbon dioxide into rocks reduces atmospheric carbon dioxide levels, thus lowering the temperature and reducing weathering. Enough so that, were it to continue unabated, it would remove all the carbon dioxide from our atmosphere, reducing temperatures enough to drop the Earth’s mean temperature below freezing.

What halts the cooling is plate tectonics. That is, the slow, inexorable creation of new oceanic crust at “spreading centers,” such as the mid-Atlantic ridge, a 16,000 km volcanic cleft in the middle of the Atlantic from which new crust spreads out at a rate of about 2.5 cm a year. The island nation of Iceland sits astride the ridge, accounting for both its exuberant volcanism and its many tourist-friendly hot springs. Given that the total surface area of the planet is fixed, these spreading centers must be matched by “subduction zones,” in which oceanic crust is driven down into the mantle. The Mariana Trench, the deepest spot in the Earth’s oceans, is an example. It’s where the Pacific plate is diving under the Mariana plate. As the Pacific plate descends and its volatile-rich sedimentary rocks heat up, they melt, rise to the surface, and produce volcanoes, the tops of which form the Mariana Islands. This volcanism releases the carbon dioxide stored in the sediments, starting the process anew. This “carbonate cycle” or “carbonate-silicate cycle” forms a negative feedback loop that maintains a constant temperature on the Earth’s surface. Without liquid water, the cycle fails and carbon dioxide builds up in the atmosphere, leading to runaway greenhouse warming, as apparently happened on Venus.

The carbon dioxide cycle plays an absolutely critical role in, if not the origins of life, at least the maintenance of Earth’s habitability. The issue in question is “the faint early Sun paradox,” first pointed out in the early 1970s by Carl Sagan (1934–1996). Stars are relatively simple and predictable systems, and our understanding of their physics is quite mature. It leads us to the well-established extrapolation that early in its life the Sun must have been around 20% dimmer than it is today. (As fusion causes helium to build up in the Sun’s core, the zone of fusion moves outward, increasing the surface temperature and therefore the brightness.) Thus a planet situated at a habitable distance from the early Sun (i.e., a distance at which liquid water could form) would, as the Sun grew brighter, heat up enough to lose its water by upper-atmosphere photolysis (a process that breaks water into its constituent hydrogen and oxygen) and turn into a Venus. This problem is exacerbated by the fact that water vapor is a potent greenhouse gas, thus accelerating the heating as the oceans begin to evaporate in earnest. This slowly increasing brightness is a universal feature of stars, and thus astrobiologists need to distinguish between habitable zones and “continuously” habitable zones. The former is the type of zone in which liquid water can form at a given period in a star’s life, and the latter is the much, much narrower region in which water stays liquid over billions of years. The continuously habitable zone, although still fairly narrow, is broadened significantly by the carbon dioxide cycle; it is estimated that, for an Earth-type planet in orbit around a Sun-like star, the continuously habitable zone ranges from 0.95 to 1.15 AU. If the Earth had formed outside this tight band, life would not have flourished here for the billions of years it has. This said, continuously is a relative term; the Sun’s steadily increasing brightness will push the inner edge of the supposedly continuously habitable zone past us in just 2 billion years. Don’t get too comfortable!

The importance of the carbon dioxide cycle suggests that plate tectonics—the geological process in which crustal rocks are recycled into the mantle—plays a critical role in maintaining a habitable planetary environment. The precipitation of carbonate rocks is very efficient and over geological time would lead to the removal of effectively all atmospheric carbon dioxide. The resultant reduction in the greenhouse effect would be greatly exacerbated by the fact that snow and ice are white and thus reflective. This, in turn, would lead to further cooling, plunging the planet into a global ice age (termed the “snowball Earth”). The existing evidence suggests that this may have happened several times in the previous billion or so years of our planet’s history. When the Earth’s surface was covered by ice, however, the atmosphere was isolated from the oceans and atmospheric carbon dioxide levels were free to rise. All that was needed was plate tectonics, with which the carbon dioxide in carbonate rocks is recycled by volcanoes back into the atmosphere. In this way, ice-covered oceans lead to increasing atmospheric carbon dioxide levels, warmer temperatures, and melting of the snowball Earth, and the global carbonate cycle starts anew.

Liquid water, then, is required to prevent runaway greenhouse warming (à la Venus), and plate tectonics is required to prevent runaway sequestration of carbon dioxide leading to a snowball planet. It is already clear that liquid water is rare in the Solar System. Plate tectonics may also be rare; Mercury, Venus, and Mars do not exhibit any compelling evidence of the effect. We are not sure why the Earth exhibits plate tectonics while its near twin, Venus, does not, but, ironically, it has been hypothesized that water may be the missing ingredient on our nearest neighbor. Without water as a lubricant, the theory suggests, Venus’s crustal plates are too rigid to sink into the mantle the way that Earth’s crustal plates do.

This, of course, simply pushes the question one step farther: why did Venus lose its water when Earth so clearly did not? The answer to this question lies in the precise locations of the two planets. Although, because of greenhouse warming, the mean temperature on the Earth’s surface is 15°C, given its albedo (reflectivity) and its distance from the Sun, its mean temperature would otherwise be well below freezing. Because of this, by 10 km above sea level, which is above most of the greenhouse gas, atmospheric temperatures reach −50°C.* Any water vapor that diffuses up toward the upper atmosphere thus eventually condenses and falls back as rain or snow (fig. 3.12). The mean temperature at Venus’s orbit, in contrast, is above the freezing point of water, so water can diffuse far higher in the Venusian atmosphere, ultimately being subjected to the full intensity of the Sun’s UV light, which tears the molecule into its constituent atoms. The oxygen, being extremely reactive, drifts back down and oxidizes Venus’s surface rocks. (Radar images of the surface of Venus indicate that the planet is covered with relatively young lava flows, and thus there are plenty of fresh, reduced mantle rocks to oxidize.) In contrast, because it is very light, a small but significant fraction of the hydrogen is moving faster than the Venusian escape velocity and is lost into space. Over the course of the past 4.5 billion years, this mechanism has removed all but a tiny trace of Venus’s original inventory of water. Indeed, as it turns out, Earth might also have lost a great deal of water by this mechanism.* Specifically, at 6,000:1, the ¹H to ²H ratio of Earth is one-seventh that of Jupiter, which, because Jupiter is so massive, is not thought to have lost any hydrogen via Jeans escape, suggesting that its ratio reflects the primordial ratio of the protoplanetary disk.

Conclusions

Our planet formed in the inner reaches of the solar nebula, and thus like all terrestrial planets it is composed predominantly of refractory metals and silicates. These, however, are not the materials of which life is made. Those we owe to Jupiter, whose mighty bulk tossed icy, volatile-rich material from the outer Solar System inward to the nascent Earth. This cosmic cleanup also saved us from later catastrophic, planet-sterilizing impacts, and thus Jupiter played a significant role not only in life’s origins but also in providing the billions of years of stability required for complex life to evolve. But even with these favorable conditions helped along by Jupiter, the development of an Earth-like planet can easily go off toward a state that does not support a diverse biosphere, as the example of Venus shows (see also sidebar 3.2).

The impact of Shoemaker-Levy 9 was an example of the final vestiges of this cleanup. It also marked the pinnacle of the long and successful career of Eugene Shoemaker, who, sadly, was killed a few years later in an auto accident while exploring impact craters in Australia’s outback. In a fitting tribute to this well-liked and extremely influential planetary scientist, a small sample of his ashes was placed on the Lunar Prospector orbiter that was launched not long after his death and that later confirmed suspicions that the deeply shadowed craters on the Lunar poles contain hydrogen, no doubt as water ice. On July 31, 1999, Lunar Prospector was intentionally crashed into a crater at the Moon’s north pole in an attempt to toss up a plume of dust and steam. (Although telescopic observations of this event from the Earth failed to spot the water, it was eventually confirmed during a similarly intentional crash by the LACROSS mission in 2009.) With this impact, Eugene Shoemaker, one of the most powerful and eloquent advocates of the benefits of astrogeological research, became the first human interred on another world.

Sidebar 3.2

Weighing the Probabilities

So what are the chances of everything coming together just right to form a habitable planet? The question of the frequency with which suitable stars host planetary systems is covered in chapter 9. But this is a reasonable place to ask: of all the potential solar systems out there, what fraction includes planets capable of supporting the formation and evolution of life? We simply do not know.

One of the issues is how frequently rocky, terrestrial planets form. Again, we discuss this in detail in chapter 9, but for now let’s just point out that the first successful planet-finding techniques were limited to the detection of gas giant planets. Only since 2012 have we begun to get a handle on how often Earth-sized planets orbit in the habitable zones of suitable stars. Moreover, we need to know something even more specific: what fraction of all planetary systems include both Earths and Jupiters? The problem is that, as noted in this chapter, without Jupiter to clean out the newly formed Solar System, the late heavy bombardment would have continued effectively unabated, with dinosaur-killing-sized impacts happening once a century or so, and larger, planet-sterilizing impacts happening a few times each billion years. How frequently Sun-like stars host both Earth-like and Jupiter-like planets remains an open question.

And if we’re interested in the evolution of higher life, a second element comes into play: the Moon. Due to our planet’s rotation, centripetal force deforms the Earth from a perfect sphere—Earth has a distinct bulge at the equator. The Sun’s gravity pulls on this bulge and, over the course of millions of years, should be able to very significantly shift the Earth’s rotational axis. If this force ever succeeded in pushing our planet’s axial tilt much below its current value of 23.5°, the seasons would run amok. The Moon, however, weighing in at a quite reasonable 1.25% of the mass of the Earth and residing a mere 384,000 km away, exerts a similarly large force on the Earth’s bulge. Because the Moon’s pushing and shoving happens over a different time frame from the Sun’s pushing and shoving, the Moon prevents the Sun’s perturbations from building up and keeps our axial inclination at a nice, mild level. Were it not for this effect, the Earth’s climate would fluctuate wildly over the course of a few tens of millions of years. And while this would not necessarily have prevented the formation of life, it could well have made the origins of life that much less likely.

Evidence for these putative fluctuations can be seen on our neighboring planet Mars, where thickly layered terrains near the poles are thought to represent massive climatic fluctuations arising from the wandering of the planet’s axial tilt. Although its current axial tilt is an Earth-like 25.2°, Mars lacks large moons (its moons, Phobos and Deimos, are but a few tens of kilometers across), and therefore the Sun’s persistent, small torques build up over time, causing the Martian poles to make massive excursions. Jack Wisdom of the Massachusetts Institute of Technology and Jacques Laskar of the Institut de Mécanique Céleste et de Calcul des Ephémérides and the Paris Observatory have independently shown that, under the influence of these torques, the axial tilt of Mars migrates randomly over tens of millions of years from 10° to a devastating (for climate and the evolution of advanced life) 50°.

So, without our massive Moon, our climate would be unstable and life would be that much less likely to have formed and thrived here. And how improbable was the formation of our Moon? Again, we can only speculate, but the answer may well be that it was very improbable. The Moon is thought to have formed through an off-center collision between the accreting proto-Earth and a Mars-sized object. The frequency of such collisions is probably low; after all, none of the other three rocky planets have large moons. On the other hand, Charon, one of the three moons of icy Pluto, is a whopping 15% as massive as the dwarf planet (or planet—your choice!) it orbits, suggesting that the probability of such impacts, while low, may not be astronomically low.

Touma, Jihad, and Jack Wisdom. “The Chaotic Obliquity of Mars.” Science 259, no. 5099 (1993): 1294–97.

Chapter 3