CHAPTER 2: THE COSMIC SHOOTING GALLERY

Some ideas are important, even revolutionary. One such idea is that the Earth is a cosmic target, vulnerable to global catastrophe. In the previous chapter we outlined the case for impacts in Earth history, but we now step back to the situation about 1980 for a more detailed view. To many scientists at that time, the concept of Earth's vulnerability to impacts appeared to contradict common sense and the lessons we were taught about global history. In many ways their uniformitarian philosophy makes much sense; it is difficult to imagine that processes or events far beyond our experience might have dominated the history of our planet. Yet, as we recount in this chapter, the discovery of ubiquitous impact craters on many planetary bodies reintroduced the ideas of catastrophism into geology. We use the record of the Moon to derive the surprisingly violent cratering history of our own planet. Next we describe the comets and asteroids, which are the culprits in this cosmic bombardment, including the results of recent radar studies and space missions. Geologist Gene Shoemaker enters the story as one of our chief characters, and we describe how Gene and his colleagues (including his wife Carolyn) search for asteroids and comets. The chapter concludes with the primary product of these searches, a description of how often the Earth is struck by comets and asteroids of various energies, from less than a kiloton up to more than a million megatons of TNT.

====================
CHAPTER 3: TARGET EARTH
"While there are no definite data to reason from, it is believed that an encounter with the nucleus of one of the larger comets is not to be desired." -- Nineteenth-century Astronomer Herbert Howe in his 1897 textbook
Is the Earth really being struck by comets and asteroids? The Moon, Mercury, Mars, and even Venus are heavily cratered, but there are few impact features readily found on our planet. Why is this? Once it was thought that the atmosphere protected us from impacts, but now we know enough about meteors to recognize that only smaller impacts -- those by asteroids or comets less than about the size of a football field -- are filtered out by the atmosphere. If there were any doubt as to this conclusion, the mapping of Venus by the Magellan radar satellite in the early 1990s dispelled it.

The atmosphere of Venus is about 100 times denser than that of the Earth, yet Venus has many impact craters. None of the craters on Venus is smaller than about 10 km across, however. Scientists have studied the process of meteoric breakup in an atmosphere to interpret the cratering record on Venus, concluding that the ability of an atmosphere to fragment an incoming projectile is roughly proportional to the amount of gas in the atmosphere. The atmosphere of Venus filters out rocky asteroids with a diameter of less than about 1 km (and hence a cross-section of about 1 square kilometer); these are the objects that produce the 10-km craters. The Earth's atmosphere, being a hundred times less massive, shields us from impacts of objects only if their cross-section is less than 1/100 square kilometer, corresponding to diameters of about 100 m, or the size of a football field. The atmosphere protects us from most small objects, but not large ones; thus the paucity of large craters on Earth cannot be an atmospheric shielding effect.

What most distinguishes our planet from the Moon, Mercury, Mars and Venus is the Earth's higher levels of geologic activity, including volcanism and plate tectonics. These effects, together with water and ice erosion from the continents and sedimentation in the oceans, effectively fill in, erode, and erase impact craters almost as quickly as they are made. Depending on where a crater is formed, it may remain visible for only a few million years, a mere blink of the eye compared with the 4.5-billion-year age of the planet. We should expect to see only the most recently formed terrestrial craters, and even for them it is more productive to look in areas of geological stability and low rates of erosion, such as mid-continental deserts. Only in the past few decades have geologists begun to study these recent craters and learned to use more subtle clues to locate the faint ghosts of older, more heavily eroded impact scars.

Tourists driving east across Arizona on Interstate 40 pass through some magnificent scenery. First there are the dark twisted mountains south of Lake Meade, then the high plateau cut by the Grand Canyon of the Colorado River. Near Flagstaff the road skirts the snowcapped peaks of the San Francisco volcanoes, fringed with recent cinder cones and fresh lava flows near Sunset Crater. Proceeding eastward, a driver descends across a broad treeless plain with distant views of the Painted Desert and, eventually, the Petrified Forest National Park. About midway between Flagstaff and Petrified Forest, however, is another natural wonder, Meteor Crater. Many tourists zip past, perhaps failing to distinguish the signs announcing the privately-owned Meteor Crater from the commercial rock shops and dinosaur exhibits that are also advertized along this lonely stretch of highway. But if they do pause, they can visit the best preserved and most accessible impact feature on our entire planet.

Seen from the air, Meteor Crater is a striking circular bowl, looking very much like one of the smaller lunar craters. From the ground, it is barely visible as a low (50 m) mesa (originally called Coon Butte) until you reach the rim and look down into its cavernous interior. The circular crater is almost one mile across (1.2 km) and 570 feet (175 m) deep, with a steep trail descending to the crater floor. Surrounding it for many miles are scattered small fragments of nickel-iron alloy, some of it melted and mixed with the local limestone rock.

Meteor Crater was first noted by local sheepherders in the 1880s, and they were followed by prospectors who collected the nickel-iron fragments from the surface. By the 1890s some geologists were already convinced that this was an impact structure from the apparent meteoritic nature of the metal samples. However, the uniqueness of the feature and its proximity to the San Francisco volcanic features confused others. Even Grove K. Gilbert, the Chief Scientist of the U.S. Geological Survey and one of the first geologists to advocate an impact origin for lunar craters, concluded that this terrestrial crater had probably been formed by a volcanic steam explosion.

One man who believed passionately in the meteoritic origin of the Coon Butte feature was mining engineer Daniel M. Barringer, who began studies of the crater early in the 20th century and eventually purchased the land that contains the feature. Barringer was motivated in part by his conviction that the bulk of the meteorite was buried within the crater, and that this would prove to be an economically attractive mining investment. He brought in drilling rigs, and by 1908 he had drilled 28 bore holes in the crater floor, all to no avail. Later Barringer drilled holes around the crater thinking that perhaps the body had impacted at an oblique angle. By the time of his death in 1929, Barringer had spent nearly a million dollars, a huge fortune by the standards of the time, on his fruitless search for the iron-nickel mass. Eventually Barringer's heirs gave up on the search and began to develop the property as a tourist attraction.

Shortly before his death, Barringer hired as a consultant the British astronomer and mathematician Forrest Ray Moulton. Moulton was one of the first scientists to calculate the energy of a body impacting at more than 15 km/s speed, and to estimate the amount of rock that would be melted or vaporized by the release of this energy. He recognized that an impact crater would be much larger in diameter than the size of the incoming projectile, and he calculated that the projectile itself would be largely destroyed in the impact. Writing in 1931, Moulton noted that "the energy given up in a tenth of a second would be sufficient to vaporize both the meteorite and the material it encountered -- there would be in effect a giant explosion that would produce a circular crater, regardless of the direction of the impact." Although bitterly disappointing to those who hoped to find a large body of iron ore in Meteor Crater, these conclusions pointed the way toward a proper understanding of the impact process on both the Earth and the Moon.

It was not until the 1950s that another major scientific study was carried out of Meteor Crater. Gene Shoemaker, then a young field geologist, had been studying the craters formed by nuclear explosions. At that time the United States was developing its arsenal of fission bombs, and these devices were tested in the deserts of Nevada. Shoemaker had the bright idea of comparing Meteor Crater with these man-made features. He was the first geologist to map the crater systematically and to compare its structure with that of lunar craters as well as bomb craters, and his studies of Meteor Crater eventually formed his doctoral thesis. Even today he leads groups of fellow scientists and students on fascinating (and energetic) field trips into the crater, where he appears to know every rock like an old friend. Shoemaker's work clearly established that the crater is of impact origin. It also set the stage for detailed studies of lunar craters carried out by the Apollo astronauts, many of whom Shoemaker had trained in Meteor Crater. Shoemaker estimated the energy of the impact at 15 megatons, and later one of his students determined what is now accepted as the age for Meteor Crater, 50,000 years.

No humans lived in North America when the half-million-ton mass of iron and nickel (probably in form of an irregular lump 40-50 m across) came screaming down into the Arizona desert; the ancestors of the Native Americans had not yet crossed the land bridge from Asia. Because iron is so dense and strong, the atmosphere had little effect on the incoming projectile, which probably struck the ground with a speed of more than 15 km/s. The resulting explosion must have raised huge clouds of dust and spread a blanket of debris for hundreds of miles. Originally, the crater was nearly twice as deep as it is today, but subsequent erosion has partially filled the interior and obliterated most evidence of the ejected material, except near the crater rim. Even in its current state, however, Meteor Crater remains one of the best examples on our planet of a relatively fresh impact feature. The tourists who do stop to visit on their drive across Arizona witness the most characteristic of planetary landforms, rare on our planet but actually common elsewhere in the solar system.

Meteor Crater is a spectacular a hole-in-the-ground, but it is extremely young geologically speaking. Arizona is a place of considerable geological upheaval, as can be judged from the recent volcanic peaks that are visible from Meteor Crater itself. An impact crater will not survive for many millions of years in such an environment, but there are other places on our planet where the weather is less stormy, subterranean forces are mild, and change is slow. Such a place is the Outback of central and western Australia. For hundreds of millions of years, there has been no faulting, no mountain-building, and no volcanism. Erosion works slowly in a level landscape where anything over 50 m high is called a "mountain", making central Australia one of the Earth's best scorecards for recording extraterrestrial impacts.

Many years ago Gene Shoemaker was attracted to this bleak, unchanging landscape to hunt for impact craters alongside Australian geologists. As his retirement approached, Shoemaker and his comet-hunting wife, Carolyn, began to spend two months every year camping in the Outback, away from the hubbub of America, researching the craters of Australia.

In September 1990, springtime in Australia, the Shoemakers decided to show off nine of their favorite craters to 50-odd fellow geologists and astronomers who were coming to Australia for a scientific conference. One of the largest tour groups to venture into the Outback, their caravan (using an especially constructed 6x6 army-green outback bus) traversed 6,000 km in just 3 weeks. Besides a few emus and patches of thorny spinafex grass, there was little to see but craters, and most of them were difficult to see as well. One crater was barely larger than a camel wallow, but some of the biggest ones were even more difficult, requiring a trained geologist to pick out.

Camping in the desert night after night, hundreds of miles from the nearest town, we could visibly connect the sky with the land. With the Southern Cross off to the side and newly-discovered Comet Levy shining overhead, accompanied by meteors that streaked across the dark sky, it was easy to imagine that celestial objects can plummet to the ground to wreak devastation on the Earth.

There are many lessons to be leaned from this, the most densely cratered terrain on Earth. The smaller craters, a mile across and less, were (like Arizona's Meteor Crater) all formed by metallic meteorites, remnants of which could still be found in the soil near the craters. This fact proves that the much more common stony boulders from the sky that would otherwise make craters of such sizes (boulders less than about the size of a football field) are torn apart during their atmospheric plunge. Only the stronger iron meteorites make it through to form mile-wide (or smaller) craters. Australia even offers one example -- the Henbury crater field near Alice Springs -- where an incoming stony object began to break up in the atmosphere and formed a tightly-packed cluster of craters.

A two-part side trip to Lake Acraman and the faraway Flinders National Park was particularly dramatic. Acraman is a vast waist-deep lake where a huge crater formed millions of years ago. Several hundred kilometers away, in the low but colorful Flinders Range of hills, Gene Shoemaker showed the travellers a layer in the rocks where ejecta from Acraman Crater was preserved in the geological strata. The composition of this ejecta layer is identical to the target rock in the Acraman area -- with a little bit of extraterrestrial iridium mixed in.

Returning from Flinders, the Shoemaker party stopped at a winery, reached civilization at Adelaide, then flew to Perth in time for the annual meeting of the international Meteoritical Society. There Shoemaker and his colleagues addressed one of the most important scientific issues that inspires the study of Australian craters: what is the rate at which asteroids and comets strike the Earth? From dating the time of formation of the Australian craters, Shoemaker concludes that craters larger than 10 km in diameter (produced by an asteroid just under 1 km across) are produced on the land area of the Earth about once every million years -- a value in good agreement with the impact rate deduced from counting asteroids or evaluating the numbers of craters on the Moon (as we discussed at the end of the last chapter). So we conclude that there is nothing special about the Earth or about our particular time -- we are currently pelted by cosmic debris just like any other planet, and at about the average rate.

Only a small fraction of the impact structures on Earth are as readily visible as those in Australia. Usually, the surface expression of a crater is obliterated by erosion and sedimentation within a few million years of its formation. The underlying structures may still be detected, however, using the modern techniques of geophysical exploration. One byproduct of the global search for oil and ore deposits has been the discovery of buried impact structures.

Paradoxically, as a large crater erodes it tends to evolve from a depression into a hill. An example is Gosses Bluff, west of Alice Springs -- the crater is gone, but its central peak remains. The inner part of the crater rebounds to fill in the original cavity formed by the explosion. Many of the larger craters on the Moon have similar central peaks. These anomalous upthrusts or hills have in the past been called crypto (meaning hidden or secret) volcanic structures. One example is in the flat prairies of Iowa, where the so-called Manson cryptovolcanic structure has recently been confirmed as the remnant of the central peak of a crater 35 km in diameter that formed 70 million years ago. Other cryptovolcanic structures in the U.S. Great Plains include Kentland in Indiana and Serpent Mound in Ohio. This outdated terminology is another example of the confusion of impact structures with volcanoes that has characterized terrestrial geology until recently.

When a geological feature is badly eroded, arguments about its origin are inevitable. What geologists need are objective criteria for distinguishing an impact structure from one that has been formed by volcanism or other geological forces. Two such geological clues have been discovered, both of them signatures of the explosive shocks associated with impact, producing pressures that far exceed anything that can be generated in a volcano.

Silica (silicon dioxide) is a compound found in many rocks. It has several mineral forms, the most common of which is quartz. One very rare form of silica, originally produced in laboratory experiments in 1953, has a very dense and highly stable structure that can be formed only at very high pressures. The chemist who discovered this mineral was named Loring Coes, and the mineral is called coesite. Gene Shoemaker discovered natural coesite in 1960 in samples from Meteor Crater, and subsequently in association with other known impact craters. Another shocked mineral was found in 1961, called stishovite after one of its Russian discoverers, S.M. Stishov. In addition, ordinary quartz grains subjected to shock can develop characteristic internal striations visible under a microscope. Today, shocked silica is generally accepted as the most convincing fingerprint of an impact, a fingerprint that survives long after the crater itself has faded into invisibility.

A second indication of the high pressures of impact is found in structures called shatter cones. As a shock wave passes through the target rock, it fractures the rock in a characteristic cone-shaped pattern that is visible to the trained eye of the geologist and requires no detailed laboratory confirmation. Shatter cones have been identified in the underlying rock of all of the younger terrestrial impact craters.

Armed with these tools and using space-age remote sensing from aircraft and orbiting satellites, geologists have discovered more than 150 impact scars on the Earth. In Chapter 6 we will discuss the discovery of the 200-km Chicxulub Crater in Mexico, but here let us mention two other giant scars, the Bushveldt-Vredefort complex in South Africa and the Sudbury basin in Canada.

The Transvaal region of South Africa contains some of the Earth's most ancient rocks, more than two billion years old. The features that interest us here are the Vredefort Ring, a half-circle of rock about 75 km in diameter, and a nearby region about 300 km long called the Bushveldt Igneous Complex. This area interests geologists not only for its age and complexity, but also for its concentration of valuable ore deposits. In the 1960s investigators found numerous shatter cones in the Vredefort ring as well as abundant shocked quartz throughout the region. By the 1970s, some geologists were interpreting the entire Bushveldt-Vredefort complex to be the result of four nearly simultaneous impacts producing overlapping craters, the largest of which was about 250 km in diameter. Unfortunately, the leading geologist studying this problem was killed in an accident in 1975, and to date there remain many unanswered questions about his interpretations of the observations.

The Sudbury basin, on the north shore of Lake Huron in Canada, contains the word's largest deposit of nickel and one of the most productive areas for mining iron as well. Like Bushveldt-Vredefort, Sudbury is an extremely complex region containing violently shattered rocks. There are also huge deposits of rapidly cooled lava, in spite of the absence of supporting evidence of extensive volcanism elsewhere in this part of the Canadian Shield of ancient rocks. In 1964 Arizona geologist Robert Dietz, who was the leading proponent of impact geology at the time, proposed that the Sudbury basin was an impact structure 1.7 billion years old. Presumably the meteorite was of iron-nickel composition, and it is the source of the rich nickel deposit in the area. A single metallic asteroid 4 km in diameter could provide all the iron-nickel ores present. Shatter cones were found in 1962, followed by the identification of impact minerals. Although controversy remains, today it seems likely that this feature, which has been so important to the economy of Canada, is one more example of our planet's long history of cosmic impacts.

Impacts were especially important for the Earth during the first half-billion years after it formed. Like the Sun and the rest of the solar system, our planet was born approximately 4.5 billion years ago in the collapse of a cloud of gas and dust. This collapse, triggered perhaps by the explosion of a nearby star, produced a spinning disk called the solar nebula. The central part of this solar nebula became the Sun, while the building blocks of the planets formed from the small solid bodies -- called planetesimals -- that formed out of the dust in the spinning disk.

The first solid matter in the solar nebula probably consisted of tiny grains of dust that condensed from the cooling gas like raindrops forming in a rising cloud of water vapor. Bits of dust clung together to form larger rocks and boulders, and these collided to build up still larger objects. The Earth is thought to have formed in this cauldron of swirling debris through a process of accretion of these planetesimals. As the planet grew, its gravity attracted more and more material, and the accreting material struck with higher and higher speed. Energetic impacts generated so much heat that the upper layers of the planet melted to form a global ocean of liquid rock. No craters date from that period, of course, since the Earth had no solid surface to preserve them.

The Earth was formed without a Moon. At some point during the period of accretion, however, it is likely that our planet was struck by another coalescing world about the size of Mars today -- that is, with a mass about 10% that of the Earth. The result of such an impact between worlds was catastrophic on a scale that, fortunately, is no longer conceivable within the solar system. The smaller, Mars-sized planet was completely destroyed, and even the larger Earth was shattered to its core. The explosion formed a massive atmosphere of hot rock vapor and ejected gargantuan quantities of molten and vaporized rock into space. Some of that ejected material continued to orbit the Earth as a giant ring, which cooled and collapsed to form our Moon. Scientists have deduced this scenario from a detailed comparison of the composition and structure of the Earth and Moon, which testifies to just such a catastrophic birth for our satellite. Understanding the role of impacts in the formation of the Moon is one of the many products of the Apollo expeditions and the priceless moonrocks returned for laboratory analysis.

If the Moon-forming impact had been just a little larger, the Earth itself would have been disrupted beyond repair. Several planetary collisions of comparable magnitude probably occurred during the early days of solar system history, but any direct evidence is long gone. We do see planetary peculiarities, however, that are best understood as the product of random collisions among planetary-scale bodies. Venus spins in the opposite direction from its orbital motion about the Sun, probably as the result of a late collision that struck it a glancing blow and reversed its direction of rotation. The small planet Mercury appears to be just the metal-rich remnant of a larger parent, stripped of most of its rocky mantle in another giant collision. It is largely a matter of luck that the final product of this chaos was the four inner planets we have today: Mercury, Venus, Earth, and Mars, plus the Moon.

Calculations predict that all four of the inner planets should have been made from the same building blocks of rock and metal. Because these building blocks were formed in the inner solar system, they lacked the liquid or solid forms of water and other volatile materials. The temperatures in the inner part of the solar nebula were too hot for these vapors to condense. Where, then, did the atmosphere and oceans of the Earth originate?

In the outer parts of the solar nebula, far from the Sun, temperatures were much lower (as they are today among the outer planets). Here water ice was abundant, as well as other frozen gases such as methane, ammonia, carbon dioxide, carbon monoxide, and even ethyl alcohol. The volatiles on Earth were probably derived from this distant reservoir in the outer solar system.

The most likely way to bring water to the Earth was in the form of comets. Comets (as well as the distant asteroids) are rich in volatiles and organic compounds. During the first half-billion years of solar system history, gravitational forces scattered many comets from their place of origin inward toward the Sun. This comet bombardment took place at least one hundred million years after the birth of the Earth and Moon, after both bodies had cooled and formed solid crusts. When the early comets and volatile-rich asteroids crashed into the Earth, their water and organics and other exotic compounds vaporized. Part of the vapor was blasted back into space by the force of the explosion, but part was retained to gradually built up a thick hot atmosphere. The same thing presumably happened to Venus, which is nearly the twin of the Earth in size and mass. However, the Moon and Mercury were too small to retain their new atmospheres, which escaped to space, leaving a dry surface exposed directly to the vacuum of space. Mars, being intermediate in size, retained a part of its impact-derived atmosphere, but less than did Earth or Venus.

The bombardment of the inner solar system by volatile-rich remnant planetesimals was critical for the history of the Earth. Most of the material of the biosphere -- and of our own bodies -- is probably comet-stuff, derived from the outer solar system. Were it not for cometary ice and carbon compounds, our planet might well be as dry and lifeless as the Moon. Life is a gift from the comets. But the gift did not come without a price to pay.

As the rain of cometary materials persisted, the Earth (and presumably Mars and Venus as well) built up a thick atmosphere of carbon dioxide and other compounds, and it developed shallow oceans of liquid water, rich in dissolved organic materials. Such an environment is exactly what chemists think was required for the origin of life. The first self-replicating molecules must have formed in such early seas, perhaps on all three planets.

If all of the impacting planetesimals were small, this environment might have approximated the "warm little pond" hypothesized by Charles Darwin for the origin of life. However, the evidence preserved in the densely-packed craters of the lunar highlands suggests otherwise. At least a few of the impactors from that first half-billion years of solar system history were hundreds of kilometers across -- the size of the largest asteroids and comets of today. Kevin Zahnle of the NASA Ames Research Center and his colleagues realized a few years ago that such large impacts were capable of drastically altering the terrestrial environment.

What happened when a several-hundred-kilometer asteroid or comet smashed into the early Earth? Zahnle calculated that the energy of such an impact would melt and vaporize so much of the crust near the point of impact that the planet would acquire a temporary atmosphere of rock vapor at a temperature of about 1000^oC. Under this terrible red-hot blanket, the oceans would boil away, killing any life forms that might have arisen. In effect, such impacts sterilized the planet. After a few decades the hot rock would have cooled and the oceans recondensed, but the clock would have been reset to zero for the origin of life. In an obvious understatement, this is called the "impact frustration" of the origin of life.

Zahnle has calculated that the Earth probably experienced a handful of such sterilizing impacts during its first half-billion years. Venus probably took as many hits, while Mars, being smaller, may have escaped such catastrophe. The last such impact probably took place about four billion years ago, as determined by extrapolating from the lunar cratering record. It is interesting -- and perhaps not coincidental -- that the earliest chemical evidence of life on Earth dates from not too long afterwards, at about 3.8 billion years ago. It appears that our ancestors appeared very soon after the end of this period of "frustration". It is not too great an extrapolation from this evidence to suppose that life had formed several times previously, only to be wiped out by a sterilizing impact.

We are thus led to a remarkable picture of the role of impacts in the history of the Earth, a role recognized by scientists only during the past decade. Our planet was formed by the accretion of impacting debris; the Moon was blasted from the Earth's surface in a giant impact not long after the planet formed; the rain of comets from the outer solar system subsequently carried life-giving water and organic compounds to the inner solar system, but at the same time subjected the Earth to a terrible bombardment of projectiles, the largest of which episodically boiled away the oceans and sterilized the surface. Although life may have arisen several times early in our planet's history, it was only after this heavy bombardment declined that life was able to survive. However, the role of impacts in biological evolution was not over. As we shall see in the following chapters, impacts continued to play a major -- perhaps dominant -- role in the evolution of terrestrial life, even up to the present era.