2
OXYGEN THROUGH TIME

We live today in a world that is quite atypical of Earth over most of its 4.6 billion years long history—and probably atypical as well of much of Earth’s future history. Yet it is human nature to regard the world that we are used to—what might be called “the world as we know it”—as a permanent thing. The oceans, prairies, and mountain chains—even the air we breathe—seem the norm and therefore permanent things. This sense of permanence is especially true for our atmosphere, since it has a multitude of functions that are necessary to allow the continued existence of life: it distributes water, drives critical chemical cycles, feeds plants, provides oxygen for animals, shields the surface from deadly ultraviolet light, and moderates global temperatures. Surely, it would seem, its levels should have remained constant at least from the time that animals first appeared on our planet, some 600 million years ago. But not so far back in Earth’s history, perhaps only 5 million years ago, oxygen levels were significantly higher than now, while less than 100 million years ago, oxygen levels were significantly lower than today. The atmosphere has changed markedly over time, and these changes may be key to understanding major evolutionary changes of life on Earth over time, including changes of the vertebrates. This chapter will look at how our current atmosphere came about and how it was recently discovered that the air has undergone relatively recent (compared to the age of



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere 2 OXYGEN THROUGH TIME We live today in a world that is quite atypical of Earth over most of its 4.6 billion years long history—and probably atypical as well of much of Earth’s future history. Yet it is human nature to regard the world that we are used to—what might be called “the world as we know it”—as a permanent thing. The oceans, prairies, and mountain chains—even the air we breathe—seem the norm and therefore permanent things. This sense of permanence is especially true for our atmosphere, since it has a multitude of functions that are necessary to allow the continued existence of life: it distributes water, drives critical chemical cycles, feeds plants, provides oxygen for animals, shields the surface from deadly ultraviolet light, and moderates global temperatures. Surely, it would seem, its levels should have remained constant at least from the time that animals first appeared on our planet, some 600 million years ago. But not so far back in Earth’s history, perhaps only 5 million years ago, oxygen levels were significantly higher than now, while less than 100 million years ago, oxygen levels were significantly lower than today. The atmosphere has changed markedly over time, and these changes may be key to understanding major evolutionary changes of life on Earth over time, including changes of the vertebrates. This chapter will look at how our current atmosphere came about and how it was recently discovered that the air has undergone relatively recent (compared to the age of

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere Earth) periods of hypoxia or low-oxygen, whereas the bottoms of the ocean have undergone not only hypoxia but, on occasion, anoxia, or the complete absence of oxygen, and also periods of higher than present oxygen. THE ATMOSPHERE AND THE GLOBAL OCEAN There can be no discussion about the atmosphere without including the global ocean. The two are coupled; even small changes in the temperature or chemistry of the global ocean can produce enormous changes in the atmosphere. The composition of Earth’s present-day atmosphere is basically known. Essentially it is made up of two gases: 78 percent of its volume is nitrogen, and 21 percent is oxygen. The remaining 1 percent is made up of trace amounts of other gases. Yet even at this small volume, this 1 percent has a huge effect on the planet, for within this 1 percent are both the important greenhouse gas of carbon dioxide and water vapor (itself a gas). “Greenhouse gas” is the term now used to describe any gas at work to trap heat in the atmosphere, thus warming the planet. How long has our planet had this atmosphere? The atmosphere of our planet is as old as Earth itself. The two originated at the same time somewhere around 4.6 billion years ago—a date that is almost one-third the age of the Universe itself. The planet was molten soon after formation, but rapid cooling set in and as temperatures dropped, the planet rapidly evolved. Once formed, the solid Earth and its gaseous atmosphere evolved in quite different ways, even though each influenced the other over time. Like all planets, Earth formed through accretion of particles in a solar or planetary nebula. The formation of Earth was but one part of the formation of the entire solar system. As our planet accreted, it began to differentiate, with the heavier elements sinking toward the center and the lighter elements staying near the surface. In this fashion the major structural elements of our planet, its dense inner core, middle mantle, and outer crust regions formed. This process led to rapid changes in the atmosphere of the forming earth as well. Enormous quantities of gas were trapped in the differentiating Earth and sequestered far beneath the surface of the

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere planet. Over time this gas began to escape into the atmosphere and in so doing rapidly changed the composition of the planet’s gaseous envelope. We have a clue about the nature of the gas still trapped within Earth by studying the gas composition of volcanoes. Present-day composition of volcano effluents are 50-60 percent water vapor, 24 percent carbon dioxide, 13 percent sulfur, and about 6 percent nitrogen, with traces of other gases, a composition that differs markedly from the current atmospheric composition. Our world ocean (since it is interconnected, even though we give parts of it separate names) has also changed its chemistry, mainly by changes in salinity through time. Most scientists believe the oceans have gradually become saltier through time, although a smaller but vocal group advocates that the oceans have become less salty through time. (The amount of salt in the oceans has no effect on the atmosphere and thus plays no part in our story.) The most characteristic aspect of our planet is its envelope of liquid water, and it would seem reasonable to assume that the voluminous oceans of planet Earth were created as part of the natural evolution of the cooling planet. This may not be the case, however. While the outer planets and moons of our solar system, from Jupiter outward, are rich in water, astronomers modeling how solar systems form have discovered that water should be in short supply among the inner parts of the solar system. Because of this, it is now believed that an appreciable volume of Earth’s surface water was brought here from the outer reaches of the solar system by comets impacting the planet early in its history. If this is the case, it indicates that much of our oceans and perhaps an appreciable portion of our atmosphere are exotic to Earth. Most of this delivery happened in the first 500 million years of Earth’s history, and the rain of comets onto the planet during the period from 4.2 billion to 3.8 billion years ago, known as the Heavy Bombardment period, may have caused Earth’s early oceans to be repeatedly vaporized into steam. The composition of Earth’s atmosphere early in its history is a controversial and heavily researched topic. While the amount of nitrogen may have been similar to that of today, there are abundant and diverse lines of evidence indicating that there was little or no oxygen available. Carbon dioxide, however, would have been present in much higher

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere volumes than today and this carbon dioxide–rich atmosphere would have created hothouse-like conditions through a super greenhouse effect, with carbon dioxide partial pressures (measured as the actual amount of total gas pressure exerted by the atmosphere) 10,000 times higher than today. There is abundant evidence that the present-day atmosphere is very different from that of the past. The most compelling lines are geologic. Today, the atmosphere contains so much oxygen that reduced metal species quickly oxidize: the familiar rusting of iron to a red color or the oxidation of copper to shades of green is evidence of this. In similar fashion, many metal-rich or organic-rich types of sediment quickly bind with atmospheric oxygen to produce oxidized minerals. Long ago in Earth’s history, however, minerals formed that are no longer seen on the planet’s surface. Before about 2.5 billion years ago the formation of “red beds,” sedimentary beds rich in oxidized iron minerals such as hematite did not form. Instead, there was formation of “banded iron formations,” composed of only partly oxidized iron species. Other rock types from this ancient time include uranium oxides and iron pyrites that cannot form in today’s atmosphere. This evidence strongly suggests that prior to 2.2 billion years ago there was no free oxygen in the atmosphere and little oxygen dissolved in seawater. Even though there must have been, at most, only a few percent of oxygen in the gases making up Earth’s atmosphere as late as 2.2 billion years ago, soon after that the amount of oxygen began to climb rapidly. Where did all the oxygen come from? Some oxygen can be generated by photochemical reactions, where water high in the atmosphere is broken by sunlight into hydrogen and oxygen, but this process could account for only a small percentage of the oxygen rise. The most likely explanation is that most came from photosynthesis by single-celled bacteria. Life is known to have evolved on Earth by about 3.5 billion years ago, perhaps hundreds of millions of years earlier than that. Certainly, by 3.5 billion years ago, life had evolved to the point where cyanobacteria (informally and improperly known as blue-green algae) were widespread in the oceans. The cyanobacteria were the first organisms to use carbon dioxide to produce free oxygen. They still exist and use carbon dioxide as a

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere source of carbon for building cells. They cannot use it for energy. They also made nitrogen available to their protoplasm by developing specialized structures (Heterocysts) as locations for nitrogen fixation. The cyanobacteria were eventually co-opted by other, larger cells (the eukaryotic cells that contained a distinct, membrane-bounded nucleus, in contrast to the smaller bacteria without a nucleus). This theory, known as the Endosymbiosis Theory, was proposed by biologist Lynn Margulis. Some members of the cyanobacteria became the modern chloroplast, the part of the plant cell in which photosynthesis is carried out. This transition to larger “plant” cells took place perhaps 2.7 billion years ago, and by 2.3 billion years ago a buildup of oxygen in the atmosphere was taking place. The buildup of oxygen in Earth’s atmosphere led to the formation of an ozone layer thick enough to shield life on the surface of the planet from the harmful effects of ultraviolet radiation. Ozone is another chemical form of oxygen. Because of its different bonds, it cannot be used to “burn” sugars but does screen out harmful radiation that would otherwise hurt organisms on Earth. The amount of oxygen depends in part on the amount of oxygen in the atmosphere. At times of low-oxygen, all the oceans will similarly have little oxygen in them. However, since the amount of oxygen that can dissolve into seawater is also affected by temperature, as shown in the previous chapter, warm oceans might have little oxygen in them everywhere but in a narrow surface zone, despite there being high-oxygen levels in the atmosphere. For instance, this condition exists in the modern Black Sea. ATMOSPHERIC OXYGEN ESTIMATES OVER TIME The amount of oxygen in Earth’s atmosphere is determined by a wide range of physical and biological processes, and it comes as a surprise to most people that the level of oxygen in the atmosphere fluctuated significantly until relatively recently in geological time. The same is true of carbon dioxide. As we read in the press virtually daily, carbon dioxide levels can be raised quickly if sources of the gas start putting more of it into the atmosphere by, say, volcanoes or SUVs. But why do the levels of these two gases change on a slowly aging planet such as Earth?

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere There is no simple answer to questions about the cause of either rises or falls in the amount of oxygen in the atmosphere, much as we would like there to be. For more technical details on the various geochemical and geological issues underlying the discussion that follows, Robert Berner’s book, The Phanerozoic Carbon Cycle, is highly recommended. The major determinants of the changes in atmospheric oxygen levels are a series of chemical reactions involving many of the elements abundant on and in Earth’s crust, including carbon, sulfur, and iron. The chemical reactions involve both oxidation and reduction, processes that involve chemical reactions where certain elements either add or lose electrons. In the case of oxidation reactions, free oxygen (oxygen) combines with molecules containing carbon, sulfur, or iron to form new chemical compounds and in so doing oxygen is removed from the atmosphere and stored in the newly formed compounds. Oxygen is liberated back into the atmosphere by other reactions involving reduction of compounds. This is what happens during photosynthesis as plants liberate free oxygen as a by-product of the break-up of carbon dioxide through a complex series of intermediate reactions. Two important cycles ultimately dictate oxygen levels: the carbon cycle and the sulfur cycle. There may be other elements that are important as well, but currently they are deemed far less instrumental in affecting oxygen levels than are carbon and sulfur. Let’s look first at the sulfur cycle. Sulfur is found in a wide variety of compounds, but the most important for understanding the rise and fall of oxygen over time is pyrite. This gold-colored cubic mineral is familiar to us as “fool’s gold,” and while of little value compared to gold in monetary terms, it is hugely important in dictating the amount of oxygen in the atmosphere and hence the state of the biosphere. Sulfur is added to the oceans from rivers as it weathers out of pyrite-bearing rocks on the continents, or it comes from sulfur-rich sedimentary rocks, such as gypsum and anhydrate. These latter are already in chemical states that do not react with oxygen. Such is not the case with pyrite, however. There are huge quantities of pyrite locked in a variety of rocks, most importantly dark shale that originated in the oceans, which are uplifted onto continents via plate tectonic mechanisms and then weathered under the onslaught of rain, wind, cold, and heat. There are

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere also great piles of the two sedimentary rocks, gypsum and anhydrate, that in similar fashion weather and release oxidized sulfur compounds into rivers and ultimately the sea. There is a second pathway as well for the oxidation of sulfur compounds, one that takes place deeper in Earth, as pyrite is broken down during burial or subduction of pyrite-bearing rocks in the descending slabs of rock at the deep-sea trenches. Eventually this pyrite is heated to the point that it combines with oxygen, and the products of the reaction are emitted as the familiar and noxious-smelling sulfur gases found in volcanoes and hot springs, the poisonous gases called hydrogen sulfide and sulfur dioxide. When this happens, oxygen levels can drop in the atmosphere, especially if Earth is undergoing a phase of mountain building that exposes vast new reserves of sulfur-bearing rocks to erosion. Even more important in dictating oxygen levels is the carbon cycle. Carbon makes up much of our bodies. Whether large quantities of reduced carbon compounds, such as animal and plant bodies after death, are left on the surface of the planet to react with atmospheric oxygen or are quickly buried has a major effect on oxygen levels. The rate of burial of organic carbon, along with the burial rate of sulfur-bearing compounds, is thus the major determinant of atmospheric oxygen levels. Unfortunately, there is no direct way to measure past oxygen levels. (About a decade ago one such method was thought to be discovered: trapped air in amber was ballyhooed as a direct measure of past oxygen levels—until it was found that the small bubbles were not cut off from later changes in atmospheric levels.) Indirect methods based on an understanding of the relative ability of various minerals to undergo chemical changes in the presence or absence of oxygen have been used to infer relative oxygen levels, as have indirect methods based on biological evidence. For instance, in South Africa a very old mineral deposit was found that contains sedimentary uranium minerals. These minerals quickly change into other kinds of minerals when exposed to oxygen. But their presence in river deposits of more than 3 billion years ago yields powerful evidence that the land of the time (where rivers are) was covered with an oxygen-free atmosphere.

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere Another method involves computer modeling of past oxygen and carbon dioxide levels through time, based on a set of equations and then checking these model values with the mineralogical or paleontological evidence to validate the models. There have been a number of models specifically derived to deduce past oxygen and carbon dioxide levels through time, with the set of equations for calculating levels of carbon dioxide referred to as “GEOCARB” being the most elaborate and oldest. This model and a separate model for calculating oxygen have been developed by Robert Berner and his students at Yale University. GEOCARBSULF The GEOCARBSULF model is a recent combination of the much earlier models for carbon dioxide (GEOCARB) and oxygen (isotope mass balance model). It is a computer model that takes account of the many factors thought to influence atmospheric oxygen and carbon dioxide. A computer model such as GEOCARBSULF must take account of “forcings,” processes that affect the oxygen levels. Chief among these are the rate of metamorphic and volcanic degassing of reduced carbon- and sulfur-containing gases, the rate of mountain uplift, sea level change, burial of organic matter accompanying land plant evolution, and colonization of land by plants. Each of these factors influences the burial rates of reduced carbon and pyrite sulfur (which cause atmospheric oxygen to increase) and the rates of erosion and thermal decomposition to volcanic gases of sedimentary rocks containing significant quantities of reduced carbon and pyrite sulfur (which cause oxygen to decrease). Understanding the history of oxygen through time thus involves understanding the causes for the rise and fall of oxygen, and thus it is imperative to understand when and if burial and weathering rates of organic carbon and pyrite were either enhanced or inhibited. Carbon dioxide levels are also influenced by these factors. They are also influenced by the enhancement of weathering by land plants; the apportionment of carbonate burial between deep and shallow seas; and the change of insolation (the amount of sunlight hitting the planet through

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere time), which is as much as a 4.4 percent increase of solar energy hitting Earth from the earliest Cambrian to the present-day. The end result of all of this is an estimate of oxygen level. If this is done for various times in Earth’s history, a graph is produced. Results of the computer model for oxygen as a percentage of the atmosphere over time are shown in the figure that begins this chapter. The so-called Berner curves of oxygen levels through time come from the iterations of the GEOCARBSULF and earlier oxygen models. They are not the only models, though. Recently, geochemist Noah Bergman and his colleagues Timothy Lenton and Andrew Watson published a different estimate for oxygen and carbon dioxide through time and at the same time included calculations of global temperatures through time. As with GEOCARBSULF, their model, called COPSE, depends on the input of values for many variables that are known to or are suspected of controlling oxygen and carbon dioxide levels in the atmosphere. Their results, while showing somewhat different levels of oxygen and carbon dioxide through time compared to Berner’s results, do, in fact, show the same shape of curve and thus seem to corroborate the Berner results. GEOLOGICAL TIME If oxygen has changed through time, when was it high and when low? To discuss the model results for atmospheric oxygen levels over time, we thus need to refer to the geological timescale. Because this book is about history, at this point we need to briefly digress from our story and look at how geological time has come to be known. The development of a geological timescale was the product of two centuries of investigation. The geological timescale began as a table of strata, with older strata beneath younger. Gradually the fossil content of the strata defined the units. Only in the past half-century, when radiometric age dating methods were discovered and applied, did the geological timescale acquire ages in absolute time units of years. One of the fundamental divisions of the timescale is made by the presence or absence of common animal fossils. We now know that animals first appeared about 540 million years ago. The entire sweep of

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere The Geological Time Scale. time since then is called the Phanerozoic. That in turn is subdivided into three great eras, named Paleozoic, Mesozoic, and Cenozoic, or old life, middle life, and new life. Each of these three is further subdivided into periods that correspond to particular events that make up the “greatest hits” of the time of animals. The Paleozoic Era includes, in ascending order, oldest to youngest, Devonian, Mississippian, Pennsylvanian (which in Europe are combined into a single period named the

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere Carboniferous because of all its coal deposits), and Permian. The Paleozoic is followed by the Mesozoic and starts with the Triassic, followed by the Jurassic, and ending with the Cretaceous. Finally, the Cenozoic is divided into two periods, called the Tertiary and Quarternary. So what do we see in the GEOCARBSULF results? Again, from oldest to youngest, the model estimates that there were lower than present-day oxygen levels in the early Paleozoic (Cambrian-Ordovician). Assuming that atmospheric pressure was the same as that today, which seems to be a reasonable assumption, according to climatologists, this would be equivalent to between 15-16 percent oxygen, compared to 21 percent present today. The GEOCARBSULF results then indicate that there was a significant rise in oxygen, peaking about 410 million years ago, followed by a fall in oxygen levels in the middle to later part of the Permian. Mesozoic levels were also different from those of today, with GEOCARBSULF predicting lower than current oxygen levels, gradually rising to present-day values in the latter part of the Mesozoic. It thus seems that Cambrian values were lower than present day and that the Permian witnessed the greatest drop in oxygen in Earth’s history. CARBON DIOXIDE THROUGH TIME While GEOCARBSULF has given us a new view of oxygen through time, a second pertinent question concerns carbon dioxide. The level of this gas has also varied through time, and, like oxygen, it has had important and far-reaching effects on the biosphere and on evolution. The main effect produced by varying carbon dioxide comes from its well-known greenhouse effects: during periods of relatively higher carbon dioxide levels, Earth will be warmer than during lower level times, as explained several pages above. Carbon dioxide through the last 600 million years has been modeled by GEOCARB III, and the most recent results, including characteristic climate for different times in Earth history, can be seen in the figure beginning the chapter, where R is the ratio of the mass of carbon dioxide to that of the present.

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere EVOLUTION AND OXYGEN Chapter 1 posed the iconic question of what determined the body plans that we observe in the fossil record and on Earth today. Evolution has created each of the body plans, but what specific factors were involved? Today, we can observe natural selection in action among microbes and among organisms with short generation times. Organisms are adapted to their environments and undergo evolutionary change if changes in their environment, either physical or biotic, affect their survival rates. Increasing temperature or pH are examples of physical environmental changes, while examples of biotic changes might be increased, or a new source of, competition or increased, or a new source of, predation. Evolution can occur in response to new opportunities and resources or as a defense against some new and deleterious condition. If we are to understand why the first animals recorded in the fossil record during the Cambrian period evolved the shapes and morphology that they did, we will have to have an understanding not only of the physical and biological characteristics of their Cambrian environments but of the changes to those environments as well. The history of life subsequent to the Cambrian Explosion can likewise be understood if we have a reasonable understanding of the physical and environmental changes that have occurred since the Cambrian. Evolution of animals can thus be understood as being caused by two different effects: modernization, where body plans increase in fitness through increases in morphological efficiency, and changes in response to environmental change, either physical or biological. What have been the environmental factors that have changed the most since the Cambrian, thus stimulating evolutionary change? Most stocks of organisms have increased their efficiency of design over time and have found new ways of utilizing new resources, such as the vast stock of plant material on land following the evolution of land plants, and have responded to new types of predation, such as the Mesozoic Marine Revolution, proposed some years ago by University of California biologist Gary Vermeij. His hypothesis supposes that there was a great increase in predation during the Mesozoic Era, compared to the Paleozoic Era, brought about by the evolution of shell-breaking or -boring adaptations in many separate groups of animals. These adap-

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere tations caused a resultant evolution toward more shell defense in the prey organisms. But of the various nonbiological factors affecting life, changing oxygen levels were highly significant and the only factor that was global rather than regional. In addition to changing oxygen levels, the significant nonbiological factors affecting evolution would include global temperature, the chemical makeup of seawater (such as pH and salinity), and the amount of sunlight hitting Earth. Of these, the swings in oxygen levels have been relatively greater than the changes in the others. While global temperatures have swung from relatively hothouse conditions during the Cambrian and Triassic through Eocene to glacial conditions during the Ordovician, early Permian, and Pleistocene, in reality there was never a time where some part of Earth, even during the most extreme temperature swings, did not maintain temperatures not only suitable but also favorable for animal life. Unlike the Snowball Earth (when the planet seemed to have undergone a very cold episode, perhaps freezing the oceans), episodes of 2.7 billion and 0.6 billion years ago did not involve pole-to-pole ice cover; there were still tropics at low latitudes. In similar fashion, the hottest periods were never so hot as to threaten the existence of animal life. Swings in ocean salinity and pH have been even relatively less extreme. It is only oxygen content that has been both a global phenomenon and a parameter undergoing swings wide enough to affect not only life but also the evolutionary history of life. This factor perhaps has been rivaled only by biotic interactions of competition and predation in producing the makeup of animal body plans and their changes through time. A TEST OF THE HYPOTHESIS The major hypothesis of this book is that changing atmospheric oxygen levels over the last 600 million years have caused significant evolutionary changes in animals. There has obviously been great change over this period of time, but are there times of greater and lesser evolutionary change and, if so, can these be correlated with oxygen levels? Restated, can the hypothesis that changes in atmospheric oxygen content over time have spurred evolutionary development (to form new spe-

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere cies) and changes in body plan (or, that is, new kinds of morphology compared to the ancestral case) be tested with real data? The answers here are yes and yes. In a 2002 article in the Proceedings of the National Academy of Sciences, James Cornette and colleagues Bruce Lieberman and Robert Goldstein showed a remarkable correlation between atmospheric carbon dioxide levels and rates of generic diversification (a measurement of new species formation: as we saw earlier a genus is the taxonomic rank above species) of marine animals. They took the extensive database records of species origination and extinction compiled by the late Jack Sepkoski and correlated these rates with atmospheric carbon dioxide levels as computed by Robert Berner and others. (It was in this earlier work that Sepkoski demonstrated that the rate of evolutionary change among animals, as measured by either the rate of new species origination or the rate of species extinction, has not been constant over the last 600 million years but has fluctuated.) Cornette and his colleagues found that the high levels of new species formation occurred in the early Paleozoic, most importantly during the Cambrian Explosion, which, it turns out, was a time with high levels of carbon dioxide. But they noticed that at other times with high carbon dioxide the data also showed high rates of new species formation. It seemed to Cornette and his coinvestigators that, somehow, high levels of carbon dioxide in the atmosphere triggered an increase in the rate of new species formation. But why? Animals do not use carbon dioxide in any way—just the opposite. Thus it is very puzzling to see these results. Cornette and his colleagues explained their observations in the following way: The simplest hypothesis is that macroevolution is directly affected by carbon dioxide levels. Alternatively, paleotemperature may be an intermediary between the two systems. But the first of these seems improbable; carbon dioxide, even at its highest levels since the evolution of animals in the Cambrian Explosion, was still at such minute concentrations that it was biologically neutral to animal life (although certainly it affected plant life, since higher carbon dioxide stimulates more growth—but not necessarily more new species), and even if it were at higher levels, animals do not

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere use it for any aspect of living. But the second part of their statement is more plausible. This latter statement links the known correlation between carbon dioxide levels and planetary temperature; because carbon dioxide is such an efficient “greenhouse gas,” when the level of carbon dioxide rises in the atmosphere, the planet warms. Thus, Cornette and his team suggested that temperature change has been the most important factor. They stated: One might even hypothesize that high temperatures directly increase marine diversifications or that low temperature and specifically glaciations inhibit marine diversification…. Additionally one might pose a hypothesis that some factors that enhanced plant diversification inhibited marine diversification…. Yet another hypothesis is that enhanced carbon dioxide levels may be associated with increased seafloor spreading rates that could encourage biological diversification. Thus, the three possibilities advanced by Cornette et al. are that the rate of new species formation rises when carbon dioxide levels are high because (1) high levels of carbon dioxide somehow cause new marine animals species to form; (2) higher temperatures somehow cause new animals species to form; or (3) when plant diversification slows during times of dropping carbon dioxide, so too, somehow, does marine animal diversification. It is evident that during much of the past 400 million years, periods with high carbon dioxide were times of low oxygen and that the reverse was true as well. It is only for the Cambrian and early Ordovician that this inverse relationship does not seem to hold. Thus, the alternative to the Cornette et al. hypothesis is that it was not high carbon dioxide (and thus warm temperatures) that stimulated high speciation rates, but low oxygen. It turns out that by doing a statistical test comparing oxygen vs. species formation rates, a highly significant correlation can be found. These results are striking. Carbon dioxide values may have had little or nothing to do with the changes in animal diversification rates. Instead, it can be proposed that oxygen levels, or perhaps the rapid change of oxygen levels from those less than somewhere around 15 percent by volume to present-day levels, stimulated new speciation rates, in response to animals attempting to cope with a reduction (or already low level) of oxygen in the early Paleozoic, but high levels of oxygen are evolutionarily neutral. This is because animals that have

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere adapted for low levels of oxygen become more efficient in higher levels and are not required to make any morphological or physiological changes. But animals that evolved in high levels of oxygen are severely affected when oxygen drops. They must make major morphological and/or physiological changes or become extinct. Niles Eldredge and Steve Gould, more than three decades ago, in their now famous Theory of Punctuated Equilibrium, demonstrated that most morphological changes occur during speciation events. Most morphological changes happen quickly when a new species forms, not gradually. The changes necessary to adapt to low oxygen levels (here somewhat arbitrarily chosen as <15 percent, based on the results of the correlations described above) involve morphological adaptation in existing lineages (such as size decrease) but more probably required such extreme morphological change that a new species would be created, incorporating new adaptations to lowered oxygen. Fluctuating oxygen levels, rather than carbon dioxide levels, proposed here are a more likely explanation for the interesting periods of higher and lower evolutionary change shown by marine animals. If periods of low (or lowering) oxygen are times when the rate of new species formation is high, we might expect that these times would also show high diversity—that is, the number of taxa present at a given time would be high. This can be examined by plotting oxygen concentrations through time against diversity through time, and again there is a significant correlation. These results are being published in scientific literature. A terrestrial vertebrate data set is also available. Once again, if we plot atmospheric oxygen over time, we see the same relationship. By far the greatest origination rates for land animals during the Paleozoic occurred after the Permian extinction. This high peak is in response to the elimination of most land life tetrapods (four-legged animals had to virtually start anew, thus stimulating a very high origination rate). A more telling finding is just before this peak: the late Paleozoic interval of high oxygen shows a low rate of origination on land. Thus, in the sea and on land, the time of high oxygen was a time of stagnant evolution. Beyond affecting diversification rates, how else might oxygen levels have affected land life? I propose that terrestrial vertebrates can be roughly divided into two large assemblages. One group evolved and

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere flourished in high-oxygen conditions, the other in low-oxygen. With this perspective, changing oxygen conditions led to ecological replacements of one group by another. For example, a large pattern of diversification throughout the Triassic (a time of dropping oxygen to low levels) was the result of high-oxygen terrestrial vertebrates, such as the mammal-like reptiles (the group that was the direct ancestor of we mammals, an event taking place in the Triassic Period of the Mesozoic Era) being replaced by low-oxygen organisms, most importantly the saurischian dinosaurs, which were the first-ever dinosaurs. Let’s summarize this section. While it cannot yet be demonstrated that it was the change in oxygen that actually stimulated the evolutionary change (correlation does not imply causation), our understanding of the importance of respiration to all animals leads to the conjecture that the change in oxygen values was indeed the major stimulus. This can be formalized as follows: Hypothesis 2.1: Reduced levels of oxygen stimulate higher rates of disparity (the diversity of body plans) than do high levels of oxygen. Hypothesis 2.2: The diversity of animals is correlated with oxygen levels. The highest diversities are present during times of relatively high-oxygen content. For terrestrial vertebrates, oxygen levels of less than about 15 percent seem to promote the formation of new taxa, stimulated largely by the anatomical and perhaps physiological needs of organisms in lower-oxygen environments. Another aspect of the influence of low-oxygen levels on animal disparity and diversity is that times of lowered oxygen also produced increased partitioning of land surface by what Ray Huey and I, in a 2005 paper in Science, have named “altitudinal compression.” Our hypothesis suggests that during times in Earth’s history of lowered-oxygen values, even modest elevations would have become effective barriers to gene flow and thus would have stimulated new species formation by isolating populations. Thus, there would have been more endemism (animals found in small geographic areas) during the low-oxygen times.

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere Why, then, would there be a difference in evolutionary rates between low-oxygen and high-oxygen times? Very simply, animals that evolved in low oxygen not only survive but also in some cases thrive in high oxygen (birds are a notable example). In other cases, however, the use of low-oxygen lung systems no longer needed in high-oxygen conditions may have caused some competitive replacement and thus stimulated new evolution. The replacement of many saurischian dinosaurs by the group known as ornithischians, or “bird-hipped dinosaurs,” and mammals in the Cretaceous may be an example of this as well. But, more significantly, times of low oxygen seem to have been major intervals of evolutionary innovation, leading not only to new species but also to new kinds of body parts, such as larger or even new kinds of lungs. Perhaps it is more correct to posit that periods of lower oxygen served to foster increases in disparity (the measure of morphological rather than species diversity) and diversity of species. MASS EXTINCTIONS AND LOW OXYGEN A final aspect of oxygen’s effect on the history of life must be noted. As we will see in the pages to come, the history of life was punctuated by a series of mass deaths (at least 15 over the past 500 million years) when significant numbers of animals and plants rapidly went extinct. Five of these involved the extinction of more than half the world’s fossil-producing species and were named “The Big Five” by extinction specialists David Raup and Jack Sepkoski in the 1980s. At the time of four of these five mass extinctions there was a common environmental condition. All occurred either in a time of very low oxygen (generally <15 percent, as was the case for the Ordovician, Devonian, and Triassic mass extinctions) or happened after at least a 10 percent drop in oxygen (as was the case for the Permian mass extinction). And it was not just the major mass extinctions. Even relatively minor drops in oxygen level were coincident with wholesale species disappearances. In most cases of low oxygen there was high carbon dioxide and thus high temperature and physiologists have repeatedly observed that higher temperatures increase the stress of low oxygen on aerobic organisms. Exactly why this happens is not well understood, but happen it does.

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere Without exception, these mass extinctions were associated with a dip in oxygen levels. It may not be the low oxygen per se but the change in the level of oxygen that caused the extinctions. BACK TO THE PAST With this background on animal respiration and the geological oxygen record, we are now ready to tackle the history of animal life as recorded in the rock record. The next chapters will examine in more detail major episodes in the history of animal life, their relationship to oxygen levels, and morphological adaptations that appear to have been brought about in response to changing oxygen. Thus, each chapter will not only deal with diversification and extinction trends but will also feature as well new hypotheses on specific morphological adaptations that appear to be related to oxygen levels, which can be viewed as the evolutionary attempts of various animals to maximize respiratory efficiency.

OCR for page 31
Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere