The Alvarez theory revolves around two key hypotheses: ('] 65 million years ago, a meteorite struck the earth, and (2) the aftereffects of the impact caused the K-T mass extinction. Since one can accept the first without accepting the second, they need to be kept separate (although the Alvarezes did not). In the rest of this chapter, I will examine the evidence for the first half of the theory. (The second half is covered in Chapters 8, 9, and '0.) Although Luis Alvarez himself identified '5 pre- and postdictions, not all are of equal importance. I will focus on six predictions that if confirmed would be especially corroborative and that can be identified largely by using common sense. If several of the six predictions turn out to be false—certainly if all did—the Alvarez theory would have to be abandoned. On the other hand, if most or all are met, the theory would be strongly corroborated.
An explosion of research effort followed publication of the initial Alvarez paper as geologists around the world set to work, some seeking to confirm its predictions while others tried to refute them (in principle, intent does not matter as long as the rules are followed). Key events in the refinement of the theory were the conferences held at the Snowbird ski resort in Utah in 1981 and 1988, and in Houston in 1994.4 The conferences brought together the leading workers in the new field of impact studies and a variety of other specialists, proponents of the theory and opponents alike, and provided a forum for papers and for debate that went on into the wee hours. For tracing the evolution of the Alvarez theory, the reports from these conferences are indispensable.
Here are six of the most important predictions made by the Alvarez theory, followed in each case by the corresponding findings.
prediction 1: Impact effects will be seen worldwide at the k-t boundary.
A global catastrophe would leave global evidence. Most if not all K-T boundary sites around the world will contain an iridium anomaly, though the concentration might be greater at sites closer to the ground zero of meteorite impact. At some locations, however, subsequent geologic processes might have removed iridium or even eroded the boundary layer entirely away, leaving a gap in the rock record. Thus although the absence of iridium from a few K—T boundary clay sites might not falsify the Alvarez theory, were iridium found nowhere other than in Italy and Denmark, the theory would be in trouble.
By the time of the first Snowbird Conference in 1981, only a year after the original paper in Science, the number of sites with confirmed iridium anomalies had risen to 36. By the end of 1983, it had reached 50; by 1990 it had climbed to 95; today it is well over 100. Iridium concentrations in the boundary clays are the highest ever measured in terrestrial materials. Only a few K-T sections lack iridium.
One site was of critical importance, for it was the first in which the rocks studied had been deposited not in seawater but in fresh. Some had claimed that impact of a meteorite was not the only way to get iridium into a rock layer. Seawater contains trace amounts of the element; perhaps there were processes that could somehow concentrate the iridium from a large reservoir of seawater into a particular rock layer. Iridium might be absorbed selectively on the surfaces of the clay minerals, for example. Or, perhaps the clay and iridium were once dispersed minutely throughout a thick, marine limestone bed that slowly dissolved away, leaving behind only the insoluble clay and iridium. These ideas might have applied to rocks deposited in the sea, but not to those laid down as sediments in freshwater, which contains even less iridium and where there is no opportunity to tap a vast reservoir. The discovery of a strong iridium anomaly in rocks from the Raton Basin in New Mexico and Colorado, rocks recognizable as having formed in freshwater, put the idea of seawater extraction to rest.5 (Luis Alvarez, with the advantage of hindsight, said that the occurrence of the iridium spike in freshwater rocks should have been one of his predictions.] At the exact level of the Raton iridium spike, several Cretaceous pollen species went extinct and ferns—which are opportunistic and move in after other species disappear—proliferated.
prediction 2: Elsewhere in the geologic column, iridium and other markers of impact will be rare.
If high iridium concentrations come from meteorites, they will not be found in most other rocks. If the indicators of shock described in Chapter 3—shatter cones, shocked quartz, coesite, stishovite, and tektites—are produced only by impact, they too will be rare to nonexistent in other geological settings.
(This is an appropriate place to note that the K-T mass extinction was one of many times during which substantial numbers of species disappeared. Paleontologists have identified five, including the K-T, that were especially severe. If impact is responsible for any others of the "Big Five," they too might show an iridium spike and impact markers. However, the presence or absence of indicators at those horizons would have no direct bearing on the Alvarez theory, which applies only to the K-T event. The possibility that impact might have caused more than one mass extinction is a related but separate theory that I will address later.)
It is obviously impossible to search for iridium in every rock on the surface of the earth. Frank Kyte and John Wasson of UCLA did the next best thing by measuring iridium content in a long, continuous core of sediment pulled up from the deep seafloor in the Pacific.6 It captured the sedimentary record from about 35 million years ago all the way back to the K-T boundary at 65 million years. They found iridium levels above background only at the K-T boundary. As far as we know, high iridium concentrations are exceedingly rare in terrestrial rocks.
prediction 3: Iridium anomalies will be associated with proven meteorite impact craters.
The Alvarezes started with an iridium spike and inferred an impact; it should be possible to move in the other direction as well. That is, it should be possible to find a crater whose origin by impact is undisputed, predict where the corresponding iridium-enriched ejecta will be located, and go find it. But since it is hard to detect terrestrial craters in the first place, and since erosion will have removed some ejecta layers, the absence of such a connection would not falsify the Alvarez theory.
Two craters have been found to have associated iridium-rich ejecta layers. One is the 600-million-year-old crater at Acraman, South Australia, whose ejecta deposit contains not only iridium but other platinum group metals as well as gold.7 This ancient crater has been so deeply eroded that only a multiringed scar remains. Its ejecta, even though located more than 300 km away, can still be tied confidently back to the crater. The other is the 40-km diameter, 143-million-year-old Mjolnir crater, in the Barents Sea north of Scandinavia, which was detected through geophysical methods.8 A diligent search led by a group of Norwegian geologists found its ejecta layer, which contained both iridium and shocked quartz, in a core taken 30 km from the crater's center.
At first it may seem surprising that it is so difficult to connect iridium-rich ejecta layers to their parent craters. But remember how difficult it is to recognize terrestrial impact craters, and to find the thin ejecta layers, in the first place. Comparing it to the search for a needle in a haystack may be optimistic. In any case, the two examples prove the principle. As the science of crater detection improves, other ejecta layers will be tied back to their parent craters.
prediction 4: The boundary clay layer will generally be thin and of worldwide distribution.
The immediate effects of a giant impact take place in minutes or hours; the secondary ones may last for hundreds or at most a few thousand years. On a geologic time scale, even these are instantaneous. Thus the boundary layer will be thin everywhere except, perhaps, at sites closer to ground zero. The layer ought to be found globally, though erosion might on occasion have removed it. If a thin layer is found worldwide at the K-T boundary, it would be the first universal geologic marker—rock formations ordinarily are no more than regional.
Around the world, the K-T boundary is marked by a thin clay layer, almost always with high iridium levels. (As we will see, in North America there are two boundary layers, with the thicker one on the bottom.] No other rock unit extends over even a single continent, much less over all of them and the seafloors in between. The very existence of this universal layer is evidence of a rare, perhaps unique, geologic event, and is as strongly corroborative a piece of evidence for the Alvarez theory as any.
prediction 5: The K-T boundary clays will contain shock metamorphic effects.
Known markers of impact—shocked quartz grains; coesite or stishovite; glassy, tektitelike spherules—will be found in the boundary clays. The presence of these accepted indicators would provide much stronger corroboration to doubting geologists than the iridium spike, which prior to the Alvarez discovery was unrecognized as an impact marker.
In 1981, geologist Bruce Bohor of the U.S. Geological Survey decided to look for shocked quartz at the K-T boundary and applied for a Survey fellowship (ironically named in honor of G. K. Gilbert). Turned down by the fellowship panel (which included a specialist in shocked quartz), Bohor reapplied, only to be rejected again. Showing admirable resolve, he went ahead on his own and shortly did locate shocked quartz at the K-T boundary in a 1 cm thick Montana claystone that also contained both a large iridium spike and a pollen extinction.9 Bohor's discovery was crucial in making believers out of many geologists. First of all, one of their own, rather than a know-it-all physicist, had made the discovery. Second, instead of being based on an invisible element, shocked quartz was a tried-and-true indicator that geologists had discovered
FIGURE Q Imprints left by K-T spherules where they fell in soft clay. From a drill core that penetrated the K-T boundary underneath New Jersey. [Photo courtesy of Richard Olsson, Rutgers University.'0]
themselves—it was "invented here," and could be seen with a microscope. Bohor and others went on to find shocked quartz at many other K-T boundary sites around the world. Stishovite, which provides evidence of extreme pressures, has been found at several.
Many K-T sites have yielded millimeter-sized spherules that look for all the world like microtektites externally but that internally are composed not of glass but of various crystallized minerals. Some show beautiful flowlines on their surfaces. They have been studied extensively and have a mineralogy unlike anything geologists have seen before. The pro-impactors interpret them as droplets melted by the shock of impact and blasted into the earth's atmosphere, where they solidified and fell to earth (Figure 9), subsequently recrystalliz-ing into the minerals that we now find.
prediction 6: A huge impact crater formed 65 million years ago. If it has not disappeared, it may yet be found.
If the Alvarez theory is correct, there once existed, and we can hope there still does exist, a huge crater exactly 65 million years old. Failure to find it would not falsify the theory, however, because the crater could easily have escaped detection. The meteorite might have hit somewhere in the two-thirds of the earth's surface that is now covered with water, leaving the crater hidden under younger sediments. It might have landed in the 20 percent of the seafloor that has subducted (carried down beneath an overriding tectonic plate) since K-T time. The crater might be buried under the polar ice caps. It might have struck on land but now be so eroded as to be undetectable, or it might be buried there beneath younger rocks. It might have triggered a volcanic eruption and now be covered with lava. Finding the crater thus would require more than good science—it would require good luck.
Locating the K-T impact crater obviously would provide the most corroborative evidence of all, but prior to 1990, no crater of the right age and size had been discovered. Indeed, the only candidate much discussed was the buried structure at Manson, Iowa, but it seemed too small to create a worldwide catastrophe and, some critics said, was a cryptoexplosion structure formed by underground gas explosions, as Gilbert and Bucher had claimed for Meteor Crater. (Chapter 7 is devoted to the search for the K-T impact crater.)
The six predictions just reviewed are specific to the Alvarez theory, but there is a seventh prediction that can safely be made whenever a theory with far-reaching implications is explored.
prediction 7: Unanticipated discoveries will be made.
As theories are explored, unexpected discoveries almost always turn up. Sometimes these surprising findings turn out to strengthen a theory; sometimes they provide critical evidence that helps to falsify it. With the advantage of 20-20 hindsight, one can often see that a particular discovery could have been anticipated and stated as a prediction.
Prior to the Alvarez discovery, little was known about the clay layer, but now a host of techniques were applied to it. Three scientists from the University of Chicago made one of the most astonishing finds.11 Searching the Danish K-T clay for a possible meteoritic noble gas component, they found large amounts of soot, which was missing in the other late Cretaceous rocks and marine sediments that they analyzed for comparison. If the clay layer had been deposited suddenly, for which there is much independent evidence, then such a large amount of soot could only have come from global wildfires in which possibly as much as 90 percent of the total mass of living matter on the earth burned. Supporters of the impact theory naturally found the presence of the soot highly corroborative. Opponents pointed out, however, that the conclusion depends on the assumption that the clay layer was deposited rapidly; if it were not, the levels of soot would not be extraordinary.
Soot was not the only unexpected substance. Amino acids, the building blocks of proteins, and ultimately of life, are ubiquitous on the earth but also occur in a class of meteorites called the carbonaceous chondrites. Canadian scientists reported that the boundary clay contains 18 amino acids not otherwise found on the earth.12
Osmium is a platinum metal almost as rare in crustal rocks as iridium. Karl Turekian, a geochemist at Yale, noted that the ratio of two isotopes of osmium, Os 187 and Os 186, in meteorites is approximately 1:1, but in rocks of the earth's crust it is higher than 10:1. Although chemical and geological processes concentrate some chemical elements and deplete others, the ratios of the isotopes of heavy elements such as osmium tend to remain constant. This resistance to alteration is best illustrated by the enormous effort required in the Manhattan Project to separate fissile U 235 from U 238, which is 100 times as abundant naturally. Even the heat and shock of meteorite impact would not change the ratio of isotopes as heavy as those of osmium, and therefore they can be used as a tracer and proxy to reveal the origin of the iridium in the K-T layer. If the Os 187:Os 186 ratio in the boundary clay turned out to be close to the meteorite ratio of 1:1, then the osmium would likely be extraterrestrial, as would the iridium, an almost identical element. If the osmium isotopic ratio were much higher, then both the osmium and the iridium would likely be of crustal origin, weakening the Alvarez theory. According to David Raup, at Snowbird I, Turekian "made it abundantly clear that he expected to find ordinary crustal isotope ratios and that his study would show that the impact theory was neither necessary nor credible."13 A year and a half later, Turekian and a colleague reported that the osmium isotope ratios in the boundary clay were closer to meteoritic than to terrestrial levels.14 The osmium test was less definitive than had been hoped, however, because an osmium ratio of approximately 1:1 turned out to mark not only meteorites but volcanic rocks from the earth's mantle. Thus a low osmium isotope ratio could indicate a mantle source as well as an extraterrestrial one. A recent study of samples from across the last 80 million years of earth history, however, turned up a low osmium isotopic ratio only at the K-T boundary, again strengthening the possibility that the osmium and iridium came from space.15
Spinel is a rare mineral that sometimes forms a variety of ruby. Many of the K-T clays contained a nickel-rich variety of spinel previously found only in material worn off of meteorites. Furthermore, the highest spinel abundances occurred at exactly the same place in K-T sections as the iridium spike—at some locations each gram of boundary rock contained more than 10,000 spinel spherules.
Until recently, diamonds occurred in nature only in rocks believed to have originally formed deep within the earth (where heat and pressure are high), and that subsequently were elevated to the surface. Within the last few decades they have been produced in explosion experiments and found in meteorites, where the diamonds are so tiny as to be barely detectable. The hope that diamonds would also show up at terrestrial impact sites led Canadian scientists David Carlisle and Dennis Braman to search the K-T boundary clay in Alberta, where they immediately found them.16
Now the story gets even more interesting.17 As the Soviet Union began to collapse, reports started to emerge that scientists there had not only found diamonds at several of their impact sites, but in numbers reaching into the millions. The most thoroughly tested crater was the 3 5-million-year-old, 100-km-wide Popigai Crater in Northern Siberia, which the Soviets probed with over 500 boreholes. Most of the diamonds there were tiny, but some were as large as peas. (Although none are of gemstone quality, they may prove useful for industrial purposes.) A British team searched the Ries Crater and soon found diamonds by the billions in the melt rock, which had been the source of the stone for the town hall and the church in Nordlingen, the medieval German town located within the crater. The citizens of that town, unbeknownst to them, had been surrounded all their lives by innumerable diamonds formed 15 million years ago by a giant impact. The Ries diamonds occur in association with silicon carbide, like diamond a rare and hard mineral. From this association and various other chemical indicators, the British scientists concluded that the diamonds and silicon carbide had not formed directly as a result of shock but rather had crystallized in midair from the white-hot impact fireball. If they are correct, then diamonds should be found at other impact craters and provide an excellent marker of impact. The search was immediately extended, and by mid-1996 diamonds had been found in each of the eight impact sites studied. No diamonds occur in rocks immediately above and below the K-T boundary—only right in it. The find ing of billions of diamonds at impact sites and K-T locations must rank as the most surprising and important of the unexpected discoveries triggered by the Alvarez theory.
Was this article helpful?