Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
3 A Habitable Planet E arthâs hospitable climateâwith temperatures siders the relationship between geology, climate, and high enough to keep surface water in the liquid life. The picture that emerges from Question 7 is that a state but low enough to keep too much water large number of factors contribute to governing Earthâs vapor from entering the atmosphereâis a special and climate, but how the interplay of these factors results in probably critical feature of the planet. There is growing a particular climate state is still unknown. The answer public awareness that climate can change, and there is to this question is critical for addressing future climate abundant evidence in the geological record that climate change. Question 8 raises the interesting possibility has changed in the past. The history of Earthâs climate, that life itself helps govern climate and other aspects a peculiar combination of both variability and stabil- of Earthâs surface conditions, while at the same time ity, poses challenging scientific questions. Our current we have conclusive evidence that climate change has understanding suggests that many factors can change at times been seriously detrimental to life, occasionally climate, some capable of producing rapid changes and killing off huge numbers of species and often forcing some requiring much more time but also potentially evolutionary change. causing much larger changes. However, despite the many ways that natural forces can change Earthâs QUESTION 7: WHAT CAUSES CLIMATE climate, substantial geological evidence suggests that TO CHANGEâAND HOW MUCH CAN IT Earthâs overall climate, although it has oscillated be- CHANGE? tween relatively warm and relatively cold states many times, has somehow been maintained in a reasonable, Among the systems of planet Earth addressed in this and quite narrow, range that is conducive to the pres- report, climate is the most widely discussed in public ervation of life. The equitable climate conditions have forums. We know that human civilizations developed been present for perhaps 3.5 billion years, despite the during an unusual period of climate stability over the fact that both the Sun and Earth have changed in past 10,000 years or so, but we also know from geo- ways that might be expected to play havoc with Earthâs logical evidence that momentous changes can occur in climate. periods as brief as centuries or even decades in ways This chapter addresses major questions that relate that would disrupt human settlement patterns world- to understanding how Earthâs surface conditions can wide. Moreover, it is widely recognized that Earthâs change and at the same time can be maintained be- mean global surface temperature has risen since the tween limits that are conducive to life over extremely beginning of the industrial age and that emissions of long times. Question 7 is concerned with the geological CO2 and other greenhouse gases are at least partly, if and astronomical factors that affect climate and the not wholly, responsible (IPCC, 2007a). The potentially geological evidence of climate change. Question 8 con- serious consequences of this global warming, ranging 71
72 ORIGIN AND EVOLUTION OF EARTH from inundation of densely populated coasts to ocean or ice; how the continents are arranged; the extent of acidification to the poleward spread of tropical diseases, land vegetation; and the amount of reflective material underscore the need to determine how much of the (clouds and particles) in the atmosphere. It is generally warming is caused by human activities and what can believed that the key determinant of Earthâs ability to be done about it. Earth science has an important role capture energy from the Sun is the amount of green- in answering both questions. house gases, predominantly carbon dioxide, present The immediate grand challenge in climate science in Earthâs atmosphere. Increasing the CO2 content is predicting how climate will change over the coming of the atmosphere stimulates warming, which is then decades. However, the broader challenge is to account amplified by increasing amounts of water vapor that for both the long-term consistency of Earthâs climate can evaporate from the oceans at higher temperature. and its multiple and varied excursions in the context Hence the cornerstone of any broader understanding of a constantly evolving global geological and biologi- of Earthâs climate is the question of what controls the cal framework. Only when we are able to capture past amount of CO2 in the atmosphere. climate changes in models will we have confidence The various processes that contribute to the CO2 in our predictions of future climate. Reliable models content of the atmosphere are referred to collectively have not been available because the conditions that as the carbon cycle. The carbon cycle is a key regulator characterized ancient climatesâsuch as ground surface of climate change. The overarching issue is the fraction temperature, sea surface temperature, and mean annual of Earthâs carbon that is present in the atmosphere in precipitationâvanished thousands or millions of years the form of CO2 or other greenhouse gases like CH4. ago, along with the climate they shaped. Lacking real- For the modern Earth, most of the carbon is stored in time data for ancient events, geologists are assembling rock, and most of that is stored deep within the mantle toolkits of âproxyâ data. The temperature and precipita- and core. Estimates suggest there is 500,000 times more tion of continental regions, for example, can often be carbon stored in Earthâs mantle than in the atmosphere inferred from evidence left in the sediments of lake beds (McDonough and Sun, 1995; Salters and Stracke, or in ancient preserved soils. Earthâs large-scale surface 2004), and there is likely to be more carbon in Earthâs temperature structure, as well as information on ancient core than in the mantle. Most of the carbon not stored ocean currents, is also reflected in fossil and geochemi- in the mantle and core is found in sedimentary rocks as cal records of deep-sea sediments and in records of the mineral calcite or as organic material like kerogen sea-level change. Similarly, atmospheric temperatures and petroleum. Most of the rest is either dissolved in for at least the past 100,000 years or so are recorded in the oceans, stored in soils, or present as living plant and glacial ice and retrievable through deep drill cores in animal tissue. Only a very tiny fraction (roughly one- the ice. However, the further we journey into Earthâs millionth) is present in the atmosphere and acting to past, the more different Earth was from our modern help warm Earthâs surface. The Venusian atmosphere, planet. To understand Earthâs climate in geologically which contains about 200,000 times more CO2 than ancient times, we need to know an enormous amount Earthâs preindustrial atmosphere, is clear evidence that about the geology and geography of the ancient Earth; the distribution of carbon between a planetâs interior this is where geological science and climate science and atmosphere can be very different from that of become inseparable. Earth. Even though the amount of carbon in Earthâs at- What Processes Govern Climate Change? mosphere is small, changes in the amount have a major effect on the surface temperature. Although the carbon The climate system is regulated by how much energy in Earthâs core is not likely to be transferred to the Earth receives from the Sun and how much is radi- atmosphere, there are ways that at least some fraction ated back into space (Figure 3.1). How much energy of the enormous amounts of carbon in Earthâs mantle, is absorbed depends on the reflectivity (or albedo) of crust, and oceans could be. Similarly, there are ways to Earthâs atmosphere and surface. The albedo depends on transfer the carbon in the atmosphere to the oceans and how much of Earthâs surface is covered by water, land, to sediments and then to subduct them into the mantle.
A HABITABLE PLANET 73 INCOMING SOLAR ENERGY - SHORT WAVE RADIATION heat lost clouds reflect short-wave energy Emission of CO2 greenhouse from volcanoes weathering effect CO2 used and consumes given up by atmospheric greenhouse precipitation greenhouse plants and soil CO2 burning effect effect fossil fuels CaCO3 CO2 absorption sedimentation ice reflects SW energy evaporation (cools surface) CO2 exchange evapotranspiration continental crust (cools surface) ocean crust Submarine volcanoes release CO2 lithospheric mantle Melting and lithospheric mantle metamorphism of sediments sends asthenosphere carbon back to surface KEY: Long-wavelength (LW) Short-wavelength (SW) radiation; transfer of movement of movement of radiation; heat includes visible light CO 2 water plates FIGURE 3.1â Schematic view of the global climate system, showing many of the flows of energy, water, and CO2 that control the climate and the processes that play a role in regulating Earthâs greenhouse and determining what happens to the solar energy. Not shown is the warm circulation near midocean ridges, which moves CO2 from the ocean to the shallow oceanic crust. SOURCE: After ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ <http://www.carleton.edu/departments/geol/DaveSTELLA/climate/climate_modeling_1.htm>. Courtesy of David Bice, Pennsylvania State University. Used with permission. Studies of the carbon cycle are aimed at understand- At each timescale, different processes are primarily ing how the atmospheric carbon content is regulated responsible for the changes. by geological and biological processes. Over the past Over the past century and through the next, century, fossil fuel burning has overwhelmed natural changes in the greenhouse gas content of the atmo- processes, quickly transferring a large amount of bur- sphere are the most important factor affecting climate, ied carbon (in the form of organic matter, petroleum, although changes in atmospheric particulates and coal, and natural gas in sedimentary rock formations) clouds are also important. Burning coal, oil, and natural into the atmosphere as CO2. On longer timescales, gas continues to add greenhouse gases and aerosols to natural processes (e.g., volcanism, subduction, chemical the atmosphere, reducing emissions of infrared radia- weathering, sedimentation, metamorphism, glaciation, tion to space and causing Earthâs global mean surface wildfires) also shift carbon between the atmosphere, temperature to rise. The amount of increase depends on oceans, sedimentary formations, soils, plants, and deep feedbacks in the climate system, especially the (poorly interior. These processes produce cycles of increas- known) feedback from clouds. On even shorter times- ing and decreasing atmospheric CO2 that occur over cales (years to decades), changes in atmospheric particle timescales of thousands, millions, and billions of years. loading, notably sulfate aerosols, can affect climate, in
74 ORIGIN AND EVOLUTION OF EARTH FIGURE 3.2â Example of oxygen isotope data measured on carbonate shells of a single species of foraminifer separated from a 10-m core of deep-sea sediment. Glacial-interglacial cycles are evident. Higher Î´18O values represent times when bottom water temperature was lower and the volume of continental glaciers was larger. Modern time (depth = 0, age = 0) corresponds to an âinterglacialâ period. Upper graph shows depths where age can be estimated and the estimated age. SOURCE: Data from SPECMAP, <http://www.ngdc. noaa.gov/mgg/geology/specmap.html>. part countering the effect of increased CO2. The 1991 riods that characterize the past 3 million years of Earth eruption of Mount Pinatubo, for example, caused history (Figure 3.2). Over thousands of years the oceans slightly cooler-than-average global temperatures for are important as well; for example, excess CO2 in the about a year. Despite the uncertainties and feedbacks, atmosphere should dissolve into the oceans after about a doubling of CO2 from fossil fuel burning is now pre- 1,000 years. And in glacial times the increased ice cover dicted to increase the mean surface temperature 2Â°C on Earth changes the albedo. If ice caps start to grow to 4.5Â°C by about the middle of this century (IPCC, as a result of cooling over thousands of years, they can 2007a). reflect more sunlight and enhance cooling. As the period of time under consideration length- The role of tectonic processes (volcanoes, moun- ens, more diverse processes that can affect climate come tain building, continental drift) becomes dominant at into play. Over thousands of years, variations in Earthâs timescales of a million years or longer (Figure 3.3). Vol- orbit around the Sun (Milankovitch forcing) affect how canoes, for example, tend to move CO2 from the deep solar energy is distributed around the globe and lead to Earth to the atmosphere, whereas erosion of mountain changes in mean annual temperature, precipitation, and ranges and the associated chemical weathering of seasonality. Earthâs orbital cycles are responsible in part minerals tend to remove CO2 from the atmosphere for the oscillations between ice ages and interglacial pe- and ocean and bury it as calcite and organic matter in sediments on the ocean floor. Plate motions, which rearrange the continents and oceans, affect atmospheric <http://data.giss.nasa.gov/gistemp/2005/>.
A HABITABLE PLANET 75 Age Î´ 18 O (â°) Tectonic Biotic Î´ 13 C (â°) Climatic (Ma) 5 4 3 2 1 0 Events Events Events -1 0 1 2 3 0 Plt. Large mammal extinctions N. Hemisphere Ice-sheets Plio. Panama "Great American Interchange" W. Antarctic ice-sheet Seaway closes Hominids appear Asian monsoons C4 grasses expand intensify 10 Miocene E. Antarctic ice-sheet Mid-Miocene Columbia River Climatic Optimum Volcanism Horses diversify 20 Tibetan Plateau uplfit accelerates seals & sea lions appear Antarctic Ice-sheets Mi-1Glaciation Red Sea Rifting Coral Extinction Plate reorganization Oligocene Late Oligocene & Andean uplift Warming Drake Passage large carnivores & 30 opens other mammals diversify Tasmania-Antarctic archaic mammals & Oi-1 Glaciation Passage opens broad leaf forests decline, baleen whales appear Small-ephemeral Ice-sheets appear plate reorganization & 40 reduction in seafloor spreading rates Eocene Ungulates diversify, Partial or Ephemeral primates decline Full Scale and Permanent Archaic whales 50 E. Eocene appear Climatic Optimum N. Atlantic Late Paleocene Rifting & Volcanism Mammals disperse Thermal Maximum Benthic extinction Paleocene India-Asia contact 60 Meteor Impact K-T Mass Extinctions 70 0Â° 4Â° 8Â° 12Â° Temperature (Â°C)* FIGURE 3.3â Global deep-sea oxygen and carbon isotope variations associated with major climatic, tectonic, and biotic events, based on data compiled from more than 40 ocean drilling holes. The temperature scale refers to the temperature of typical water near the ocean bottom and applies only to the time period from 70 million to 34 million years ago. Presently the bottom water tem- peratures are typically about 2Â°C and global mean surface temperature is 15Â°C. Fifty million years ago, bottom water temperatures were about 10Â°C to 12Â°C, which corresponds to a global mean surface temperature of about 25Â°C. From the early Oligocene to the present, about 70 percent of the variability in the Î´18O record reflects changes in Antarctica and northern hemisphere ice volume. The vertical bars provide a rough qualitative representation of ice volume in each hemisphere relative to the Last Glacial Maximum, with the dashed bar representing periods of minimal ice coverage (â¤ 50 percent), and the full bar representing close to maximum ice coverage (> 50 percent of present). SOURCE: Zachos et al. (2001). Reprinted with permission from the American Association for the Advancement of Science (AAAS). and oceanic circulation, which in turn changes the ef- passages as continental drift separated Antarctica from ficiency of heat transport from low to high latitudes. neighboring continents is close in time to the first These connections can be seen where geological events growth of continental glaciers on Antarctica. Factors are correlated with major climate shifts. Volcanic activ- on the Antarctic continental shelf, such as the elevation ity that occurred as North America broke away from of the Vostok and Gamburtsev Mountain regions, may Europe and the large outpouring of lava that produced have played an important role in initiating glaciations as the Columbia River plateau about 15 million years ago well. Continental drift, combined with volcanism, also are both associated in time with globally warm tem- closed the Panama Seaway, which once connected the peratures. The opening of the Tasmanian and Drake Pacific and Atlantic oceans, drastically changing ocean
76 ORIGIN AND EVOLUTION OF EARTH circulation patterns and probably triggering glaciation in the northern hemisphere about 3 million years ago BOX 3.1 A Hospitable Climate (Zachos et al., 2001). Some also think that the continental collision of We know that during the past 4 billion years Earthâs climate has varied enough to contribute to the extinction of many species. India with Asia, which formed the Himalayas, the And yet the variations have been mild enough that life has always Tibetan plateau, and related mountain ranges, has rebounded quickly. So how hot is too hot, and how cold is too cold been a primary cause of Earthâs gradual cooling to gla- for humans? cial conditions over the past 50 million years (Raymo Earthâs nearest planetary neighbors have both stronger and and Ruddiman, 1993). The growth of those massive weaker greenhouse effects, with climates either too hot or too cold mountain ranges is hypothesized to have accelerated for life as we know it. A thick cloak of CO2 heats the surface of Venus to about 470Â°C, whereas the thin atmosphere of Mars keeps the erosion and weathering, yielding dissolved calcium that mean annual surface temperature at about â56Â°C (Consolmagno was carried to the oceans by rivers. This calcium was and Schaefer, 1994). In comparison, Earthâs mean global surface used by organisms to build shells of calcium carbonate temperature is currently about 15Â°C. In our present climate state, (calcite or aragonite), some of which accumulated as the mean annual temperature is about 27Â°C in the equatorial regions sediment on the ocean floor. This well-known example and below freezing (and perennially ice covered) at high latitudes. of sequestering carbon by burying it on the seafloor If we imagine an Earth with a global mean temperature just 10Â°C higher, the equatorial regions might have temperatures as high has drawn broad interest among those searching for as 35Â°C (depending on how much the tropics widen due to water a practical way to reduce atmospheric CO2 (IPCC, vapor feedback), unusually hot by human standards; there would 2005). Also, when sedimentary conditions and ocean be no permanent ice cover in polar regions, and most high-latitude chemical conditions are right, as has happened many precipitation would fall as rain rather than snow. If the global mean times in the past 500 million years, large amounts of temperature were 10Â°C lower than today, Earth would be covered carbon can be held as organic matter within silicate and with ice to the midlatitudes, more extensively than in the ice ages of the past few million years. carbonate sediment on the ocean floor. It is this process The geological record suggests that climate has stayed within that produced the rock formations that we now exploit these extremes throughout Earthâs history, except for geologically for fossil fuel. brief âsnowball Earthâ episodes in the Precambrian. But even much smaller fluctuations in temperature can have a significant impact on human settlement. For example, the Medieval Warm Period (about Why Has Climate Stayed in a Hospitable Range? AD 1000 to 1270) brought extensive drought that may have caused indigenous peoples to abandon the great cliff cities in the western The luminosity of the Sun may be an important regula- United States (Herweijer et al., 2006); at the same time it made tor of climate on timescales of billions of years. Stellar Greenland habitable to the Vikings until the Medieval glaciations evolution models indicate that the Sunâs power output of the early and mid-14th century (Barlow et al., 1997). Even with has increased by about 40 percent since it first became modern technologies, the coldest and warmest areas on Earth sup- a star. The lower solar luminosity 4.5 billion years ago port only small populations. would correspond to an Earth surface temperature about 35Â°C lower than the presentâwell below the freezing point of waterâif other conditions on the early Earth were similar to those of today (Kasting and Catling, 2003). And yet there is evidence from 3.8- al., 1983). According to this model, weathering slows billion-year-old rocks and more controversially from as climate cools, allowing volcanic CO2 to accumulate 4.4-billion-year-old zircons (Valley et al., 2002) that in the atmosphere. The added CO2 warms the climate the earliest Earth had liquid water at its surface (see again, causing weathering to accelerate and prevent Question 2). How can this be possible? further warming. The same feedback loop may have al- How Earth has remained within the temperature lowed more CO2 to accumulate in the atmosphere early limits for liquid water and life for over 4 billion years in Earthâs history, compensating for the lower solar is a central question about our planet (Box 3.1). A luminosity and keeping temperatures above freezing. feedback involving volcanism and weathering may This stabilizing feedback mechanism would operate provide a partial answer (Walker et al., 1981; Berner et slowly and so would be effective only over millions of
A HABITABLE PLANET 77 years; it would not significantly temper the effects of of that time, polar temperatures up to 14Â°C were high rapid CO2 increases over the next 100 years. In addi- enough to support evergreen vegetation, dinosaurs, tur- tion, there is still uncertainty about the effectiveness tles, and crocodiles north of the Arctic circle (Tarduno of this weathering-volcanism feedback because of the et al., 1998). Equatorial temperatures were 3Â°C to 5Â°C competing effect of waterâcrust interactions as a sink warmer than today (Wilson and Norris, 2001), and for CO2 and because of increasing evidence (discussed the deep-ocean temperature may have reached 20Â°C below) that weathering rates do not depend mainly on (Huber et al., 2002) as compared to 0Â°C to 5Â°C today. Earthâs surface temperature. If temperature is not the Models and proxy studies suggest that the atmospheric primary determinant of weathering rates, atmospheric CO2 concentration during the Cretaceous was 2 to 10 CO2 could vary rapidly and the fluctuations may be times higher than it is today (Caldeira and Rampino, even more difficult to predict because they would 1991; Ekart et al., 1999; Haworth et al., 2005), al- depend on global factors such as the rate of mountain though these estimates are still highly uncertain and we building due to continental collisions. do not know how variable the CO2 concentration was Long-term climate regulation may also involve on shorter timescales during the Cretaceous. other processes and other greenhouse gases. During The causes of Cretaceous warming are still un- the first half of Earthâs history, when atmospheric O2 known. Volcanic activity and hence the input of CO2 levels were low (Holland, 1984; Farquhar et al., 2000), to the atmosphere were probably unusually high, as reduced gases may have been more abundant in the suggested by the plethora of volcanic mountains and atmosphere. Methane (CH4), for example, could have plateaus of that age on the western Pacific Ocean floor. been present at concentrations of 1,000 ppmv or more, The weathering that removes CO2 from the atmo- compared to only 1.6 ppmv today (Kharecha et al., sphere may have been reduced by two processes: (1) 2005). At such high concentration, CH4 could have the higher sea level would have reduced the continental contributed 10Â°C to 20Â°C of greenhouse warming area subject to weathering, and (2) this period lacked (Pavlov et al., 2003). Disappearance of much of this the major continental collision zones that make moun- CH4, which must have happened when atmospheric tains, which weather more rapidly than flatter terrain. O2 levels rose at about 2.4 Ga (billion years ago), could The paucity of sea ice would also have decreased al- explain why Earth became glaciated at that time. This bedo. The clustering of continents could have changed hypothesis is attractive, but it has not been tested di- atmosphere and ocean circulation patterns, increasing rectly with data from the geological record. Below we the poleward transport of heat and thus making the discuss what types of information are available from polar regions warmer relative to the tropics. Whatever detailed sampling of this record. the primary causes, the middle Cretaceous is our best example of a greenhouse Earth. However, the geogra- What Caused Exceptionally Warm and Cold phy and ocean circulation are so different today that a Periods in Geological Time? future greenhouse may look very different. The coldest period we know of occurred in the The geological record of climate change, written in ice Neoproterozoic. This period is particularly interest- cores, sediments, fossils, and rocks, provides clues about ing for climate scientists. Conditions then were so how much climate has varied over the past 4 billion drastically different from those today that they strain years (Box 3.2) and the future habitability of Earth. our understanding of how the climate system works. From this record geologists have been able to identify Between 750 million and 580 million years ago, Earthâs some of Earthâs more extreme climates and the factors surface, including all of the oceans, may have frozen that may have triggered them. over completely for several brief intervals (Hoffman et One of the warmest extended periods in the al., 1998), creating a âsnowball Earth.â This hypothesis geological record occurred in the Cretaceous period, is vigorously disputed (e.g., Hyde et al., 2000)ânot the about 120 million to 90 million years ago (Barron and anomalous cold but its cause, duration, and severity. Washington, 1982), when large areas of the continents The cold was almost certainly triggered by transient were flooded with shallow seas (Figure 3.4). At the end lowering of greenhouse gas concentrations, and the
78 ORIGIN AND EVOLUTION OF EARTH BOX 3.2 How Do We Estimate Climate Variables in the Past? Historical accounts of climate are available for only the past few hundred years, so information about older climate events must be gleaned from alternative archives, such as tree rings and isotopic compositions of ice cores and ocean sediments. These records provide an indirect (or proxy) measure of climate variables, such as temperature and CO2. Proxies tend to respond to more than one factor in the climate system, so multiple measures are needed to interpret them. The further back in time we go, the fewer kinds of proxy records are available, the more limited their spatial coverage, and the greater the uncertainty in what they mean. Thus, a major effort is being made to expand the collection of proxy observations in space and time and to develop new kinds of proxies (Henderson, 2002). Proxy Measurements Used to Estimate Climate Variables Variable Age Range Proxy Measurement Mean temperature Centuries Glacier length Ground surface temperature Centuries Borehole temperature measurements Summer temperature Few millennia Tree rings, pollen analysis Land temperature, precipitation Millennia Lake sediments (O isotopes) Mean annual temperature, Millennia Speleothems (O isotopes) precipitation Sea surface temperature Millennia Corals (O isotopes, Sr/Ca, and U/Ca) Atmospheric temperature Hundreds of thousands of years Ice cores (O and H isotopes) Sea surface temperature Millions of years Foraminifera (O isotopes, Mg/Ca) Land or ocean temperature Millennia to hundreds of millions of years Fossils, evidence of ice, sedimentary structures (evidence of water) CO2 and ocean pH Tens of millions of years Foraminifera (B isotopes, Ca isotopes) CO2 Hundreds of millions of years Soil carbonate (C isotopes), stomatal indices in plant leaves FIGURE 3.4â Physiographic representation of North America, Europe, and North Africa 90 Ma when climate was warm and sea level was high. The land area of the continents was substantially smaller because oceans had risen above the edges of the continents and flooded the interiors. North America was still close to northern Europe, and the North Atlantic Ocean was barely connected to the other oceans. The South Atlantic Ocean (not shown) had not yet formed. SOURCE: <http://jan.ucc.nau.edu/~rcb7/090NAt.jpg>. Courtesy of Ron Blakey, Northern Arizona University. Used with permission.
A HABITABLE PLANET 79 actual cause may have been the different locations of CO2. The temporarily high atmospheric CO2 would the continents. The continents were all situated at low probably have made the rain especially acidic, enhanc- to midlatitudes where temperatures are warmest, al- ing chemical weathering and causing a large amount of lowing silicate weathering to proceed rapidly and draw calcium to be delivered to the oceans by rivers; this may down CO2 levels, even as the global surface tempera- explain the unusual, rapidly deposited limestone layers ture dropped and polar ice accumulated (Marshall et that cap most Neoproterozoic glacial deposits (Hoffman al., 1988; Donnadieu et al., 2004). Alternatively, CH4 and Schrag, 2000). A recent three-dimensional climate concentrations may have been high during the mid- simulation by Pierrehumbert (2004) has cast doubt on Proterozoic and then dropped as O2 levels increased this scenario, however. The new calculations indicate (for a second time; see Question 8) near the end of this that even 0.2 bars of CO2 (700 times the preindustrial time (Pavlov et al., 2003). In either case, as ice cover level) could not have deglaciated a hard snowball Earth. increased, the albedo and thus cooling would have Given the many uncertainties involved in applying cli- increased until the planet plunged into an extreme âice- mate models to the Proterozoic Earth, it is not yet clear houseâ condition. Surface temperatures calculated for whether the hypotheses or the models are incorrect. this hard snowball Earth are about â20Â°C at the equator Indeed, there are many arguments against the and about â40Â°C averaged over the globe (Pollard and snowball Earth hypothesis. Even supporters of this Kasting, 2004). theory disagree about significant issues. One is the The existence of a snowball Earth must be inferred survival of photosynthetic algae through the plunge from geological evidence. Translation of such evidence in temperatures. How was this possible if the ice was into a hypothesis about Earthâs climate and evaluation a kilometer thick everywhere as some models have of the hypothesis using modern climate models and it? Could photosynthetic life have survived in local concepts provide an interesting example of the scientific volcanic hot spots, like modern Iceland? Or did other challenges inherent in reconstructing Earthâs past con- refuges exist? One variant of the snowball hypothesis, ditions. The rock assemblage now considered indicative the so-called thin-ice model (McKay, 2000), suggests of the snowball period was initially difficult to decipher. that the ice in the tropics was only about 1 to 2 m thick, There are marine glacial deposits that formed near the allowing enough penetration of sunlight for photosyn- equator, suggesting glaciation in the tropics and hence thesis. In addition, there would likely be leads and lanes exceptionally cold conditions; banded iron formations, of open water in very thin ice. This model allows Earth suggesting anoxic conditions in the oceans; and strati- to deglaciate at a much lower CO2 level, only about 30 graphically above and below the glacial deposits there times the present level (Pollard and Kasting, 2005). are limestones, which suggest warm conditions (Figure However, there are questions as to whether such a solu- 3.5; Hoffman and Schrag, 2000). In some cases there tion can be stable, given that sea ice can flow from the are nonmarine deposits, which suggest that sea level poles to the equator, where it would melt (Goodman dropped, and there is carbon isotopic evidence suggest- and Pierrehumbert, 2003). Clearly, much more work is ing that photosynthesis all but stopped. required if the snowball Earth hypothesis is to become The warm conditions following the snowball Earth an established chapter in Earthâs climate history. Nev- period may have arisen because volcanism would have ertheless, even the most moderate of interpretations of continued through the snowball period, contributing the Neoproterozoic evidence for glaciation suggest that CO2 to the atmosphere that could not be removed by it was the coldest period in the past 2 billion years. By rock weathering because the rocks were covered with comparison, the glaciations that have affected Earth in ice. Once extreme levels of CO2 were reached (~400 more recent times have had comparatively little effect times the modern preindustrial level; Caldeira and on the global carbon cycle. Kasting, 1992), the greenhouse effect would have been strong enough to overcome the high albedo, melt the What Triggers Abrupt Climate Change? ice, and swing Earth to exceptionally warm conditions (~40Â°C global average in this model) before weathering Abrupt climate events are unusual, but they provide processes could catch up and remove the atmospheric insights on the rates at which the climate system is
80 ORIGIN AND EVOLUTION OF EARTH FIGURE 3.5â Example from Namibia of the rock record of extreme climate change in Earthâs past. These 750-million-year-old sedi- mentary rocks have been tilted by tectonic movements, but the time sequence is preserved, progressing from lower right to upper left. The Ombaatjie formation is a limestone deposit formed in shallow ocean water; near its top, isotopes indicate that a glaciation was starting, and above that level the rocks are wind-blown sand dunes, indicating that sea level dropped due to glaciation. Above the dune deposits are limestones deposited after the glaciation ended. The âcrystal fansâ are a rare type of limestone that is hypothesized to form when inorganic carbonate is rapidly precipitated from the oceans. The time duration represented by this rock sequence is not known, but estimates suggest a few million years. SOURCE: Halverson et al. (2002). Copyright 2002 American Geophysical Union. Reproduced with permission. capable of change. Much like the extremes of warm years during the last ice age, especially the interval be- and cold discussed above, the rapidity of abrupt climate tween 60,000 and 25,000 years ago (Figure 3.6). Each events provides additional clues about how climate is oscillation is characterized by gradual cooling followed controlled by Earth processes. Abrupt climate events by abrupt warming, typically over just a few decades. also serve as important time lines, enabling the correla- Even though these changes are rapid, their magnitude tion and analysis of fragmentary stratigraphic records is largeâannual temperature swings of up to 16Â°C are from around the world. Examples of abrupt climate recorded in Greenland ice cores. A number of mecha- change include the Permian-Triassic boundary (see nisms have been invoked to explain them, including Question 8), the Paleocene-Eocene Thermal Maxi- solar influences (Bond et al., 2001). Some of the coldest mum, and Dansgaard-Oeschger events in the more events are thought to be related to massive discharge recent Pleistocene Epoch. and melting of icebergs, which would have delivered Dansgaard-Oeschger events, named after the geo- fresh water to the North Atlantic and possibly changed chemists who first documented them, refer to rapid ocean circulation (reviewed in Hemming, 2004). Simi- climate fluctuations that occurred about every 1,500 larly dramatic but temporary events almost certainly
A HABITABLE PLANET 81 -32 of increased CO2 in the atmosphere as well as higher -34 temperature and humidity (Zachos et al., 2001). -36 Can Earthâs Past CO2 History Be Determined? Î´ O -38 18 -40 The connection between atmospheric CO2 levels and -42 climate is generally accepted, but there are still few reli- -44 able data confirming the relationship through Earthâs -46 history. The examples above show that additional or 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 alternative factors, including other greenhouse gases Years before present like CH4, may be required to explain some tempera- Figure 3.6.eps ture changes. For example, estimated concentrations of FIGURE 3.6â Record of Î´18O, a proxy for mean annual tem- perature, of Greenland ice from the GISP2 ice core. A change atmospheric CO2 are too low to explain some of the of five units of Î´18O corresponds to a change in temperature warmest times of the Cenozoic (Fedorov et al., 2006; of 14Â°C at the GISP site. The important features of this record Stoll, 2006), and CO2 concentrations were believed to are the rapid shifts between 60,000 and 25,000 years ago, be very high in the Ordovician and Jurassic, despite when temperatures oscillated by 10Â°C to 15Â°C over periods as short as 100 years, and the unusual stability of climate over the evidence of episodically cool climate (Kump et al., past 10,000 years. SOURCE: Data from Grootes and Stuiver 1999; Veizer et al., 2000). Confirming a correlation (1997). between periods of warm climate and high atmospheric CO2 levels during the Phanerozoic remains a major objective. Other key questions include whether other greenhouse gases were important in the more distant occurred during earlier glacial periods, although high- geological past, and whether other causes of climate resolution ice core and marine sediment records are not change besides greenhouse gas forcing can be inferred available to confirm this. from the geological record. The most extreme abrupt global warming event Much of the work on deep time has focused on recorded in geological history was the Paleocene-Eocene proxy studies of marine sedimentary rocks, which Thermal Maximum, which occurred 55 million years record the evolving chemistry of the ocean. Since the ago (Figure 3.3; reviewed in Zachos et al., 2001). In ocean and atmosphere are roughly at chemical equi- less than 10,000 years, deep-sea temperatures are esti- librium over timescales longer than 10,000 years, and mated to have increased by 5Â°C to 6Â°C and sea surface because most of the available carbon is stored in the temperatures by as much as 8Â°C at high latitudes (Stoll, oceans, reconstructing past changes in ocean chem- 2006). This warming event was associated with changes istry would help establish how atmospheric CO2 has in global carbon cycling, oceanic and atmospheric cir- changed. But because the chemistry of the oceans is culation, and the extinction of many marine organisms. so complicated, available data are still insufficient for Detailed chronology of the interval suggests that it the task. Further complicating the picture are isotopic took about 170,000 years to flush the excess 12C from data suggesting that steady state models of the carbon the ocean and atmosphere through burial of carbonate cycle are applicable in the Cenozoic (0 to 65 Ma), but and organic carbon in deep-ocean sediments (RÃ¶hl et not the Neoproterozoic (1,000 to 543 Ma; Rothman al., 2000). et al., 2003). Some aspects of ocean chemistry at least The cause of this abrupt event is still debated. confirm that the oceans undergo major shifts in com- At least seven possible triggers have been proposed, position. For example, there is evidence that the ratio including a catastrophic release of 1,050 to 2,100 giga- of Ca to Mg and the ratio of carbonate (HCO3â) to tons of carbon from seafloor methane hydrate reservoirs sulfate (SO42â) have changed markedly and systemati- (Zachos et al., 2005). A significant shift in osmium cally (Figure 3.7). isotopes suggests that continental weathering increased Similarly, the rates of past volcanism and weather- substantially (Ravizza et al., 2001), possibly as a result ing cannot be measured directly, and better estimates
82 ORIGIN AND EVOLUTION OF EARTH are needed to determine how closely they are balanced and how much their changing rates influence the cli- mate system. Volcanism rates are commonly estimated from seafloor generation rates, which themselves must be estimated since most of the ocean floor has been subducted. Seafloor generation rates are calculated from plate tectonics reconstructions and ridge or trench lengths or from global sea level determined from shoreline markers. However, uncertainties are large and results vary. For example, scientists disagree on whether the global rate of seafloor generation has changed over the past 100 million years (Rowley, 2002, versus Engebretson et al., 1992). Interpreting sea-level FIGURE 3.7â Variation in the ratio of Mg2+ to Ca2+ in the records is complicated by uncertainties about whether Figure 3.7.eps ocean over the past 550 million years. Red bars represent the volume of ocean water has remained constant over values estimated from measurements of fluid inclusions in halite crystals from salt deposits. The gray line is a model. Shown at the past 500 million years. Figure 3.8 shows deduced the top are summaries of geological evidence consistent with the sea-level variations for the past 500 million years. The model and measurements. When Mg/Ca > 2, aragonite rather double-humped curve (second column of the chart) has than calcite tends to precipitate from the oceans as the primary nonbiogenic carbonate mineral, and MgSO4, rather than KCl, is become a backbone of Phanerozoic climate studies and the first mineral to precipitate when seawater evaporates to form is often regarded as a proxy for the CO2 supply side of salt deposits. SOURCE: Loewenstein et al. (2001). Reprinted with permission of AAAS. Backstripped EPR sea level Flooded continental Acritarchs Calcareous Dinoflagellate Diatoms Sea Level records (m) estimates (m) area (106 km2) nannoplankton cysts -100 0 100 200 -100 0 100 200 300 10 20 30 40 50 0 50 100 0 200 400 600 0 50 100 150 0 Quat. fall rise fall rise Harrison Watts Bond 50 This study Sahagian This study 100 Watts Haq 0 50 150 Sahagian 0 200 (m) 200 Pangea rifting Time (Ma) 250 0 20 40 0 100 200 Sloss Vail species diversity 300 genus diversity 350 Kominz Haq 2005 400 Ronov Western U.S. Great Basin 450 Virginia Levy Bond 500 California Pannotia rifting 100 200 erosion flooding FIGURE 3.8 â Amplitude of sea-level change, in meters relative to modern, extracted from the stratigraphic record. The âbackstrippedâ Figure 3.8.eps values account for the effects of sediment compaction, loading, and variations in water depth on basin subsidence. SOURCE: Miller et al. (2005). Reprinted with permission from AAAS.
A HABITABLE PLANET 83 FIGURE 3.9â Comparison of measured dissolution rates for natural samples of soils and sediment versus laboratory measurements. The âageâ scale represents the geological age of the material (age of the soil or sediment; length of the laboratory experiment after producing freshly ground powder). SOURCE: Modified from Maher et al. (2004). Copyright 2004 by Elsevier Science and Technology Journals. Used with permission. the climate equation. Because of this importance, there relate weathering rates to erosion rates and mountain is a major effort to reduce its uncertainties. building, and to evaluate how the age dependence of Weathering rates of ancient rock are not well weathering rates affects models for the regulation of known because of basic uncertainties about the pro- global climate. cess. For example, newly exposed (fresh) surfaces of A promising proxy for globally averaged rates mineral grains weather orders of magnitude faster of weathering in the geological past is the strontium than long-exposed surfaces (Figure 3.9)âa factor isotopic composition of the oceans. The variation of that does not appear explicitly in chemical models 87Sr/86Sr provides a measure of the relative Sr inflows of reaction kinetics. In other words, areas of active to the ocean from hydrothermal fluids and eroded mountain building (e.g., continental collision zones) continental material, with high 87Sr/86Sr indicating a generate a large amount of fresh mineral surface area high influx of continental silicate minerals. There is by erosion and hence should contribute much more evidence that Sr isotope ratios respond to continental to CO2 reduction than stable continental areas. Re- collisions (e.g., Derry and France-Lanord, 1996) and search advances are needed to better understand what that periods of high 87Sr/86Sr are correlated with some controls mineral weathering rates, to quantitatively glaciations. However, Sr isotope ratios also indicate
84 ORIGIN AND EVOLUTION OF EARTH changes in the types of rocks exposed to constant rates from the biosphere and other factors will likely lead to of weathering (e.g., Harris, 1995). The relative contri- a more accurate understanding of Earthâs climate and butions of these factors will have to be sorted out before climate history. we can determine how to translate the Sr isotopic data into a quantitative estimate of global weathering rates. QUESTION 8: HOW HAS LIFE SHAPED Other possible proxies for weathering include Os, Ca, EARTHâAND HOW HAS EARTH SHAPED and Mg isotopes, although these elements are still in LIFE? early stages of study. It is not surprising that many Earth scientists have Summary viewed the geological evolution of Earth as a fun- damentally inorganic processâdominated by titanic The geological record teaches us that Earthâs climate mechanisms such as mantle convection and plate has always been changing, but remarkably the surface tectonics. After all, virtually all of Earthâs organic mass temperature has remained within a range suitable exists as a veneer of frail and short-lived creatures for life for the past 3.5 billion to 4 billion years. The within a few vertical miles of the outermost surface, primary factors responsible for this relatively benign a seemingly insignificant afterthought to this mas- climate are believed to be volcanic emissions of carbon sive planetary body of rock. And yet this multitude dioxide to the atmosphere, removal of CO2 by weath- of organismsâmost of them microscopic packages ering of surface rocks, and more subtle effects, such as composed primarily of carbon, hydrogen, nitrogen, the positions of the drifting continents, the patterns and oxygenâdetermines major features of the atmo- of ocean currents, the orientation of Earthâs rotational sphere, oceans, and continents. Biologically influenced axis and orbit around the Sun, and the luminosity of processes like erosion and weathering, for example, the Sun. Other chemical and biological effects are continually shape and reshape Earthâs surface. And as also likely to be important, such as the oxidation state we have seen in Questions 4 and 5, the erosion and of the atmosphere and the concentrations of other weathering influenced by life forms affect not only the greenhouse gases. Interspersed in this vast and mostly topography and composition of continents but also the life-supporting history are a few periods when Earth chemical composition of subducted crust and therefore was considerably warmer than it is at present, and com- the mechanism of plate tectonics and the composition pletely ice free, and a few times when Earth might have of the mantle. been extremely cold and completely ice covered. Life scientists, in the same spirit, have regarded At present the greenhouse gas content of the atmo- the evolution of life as a fundamentally biological issue, sphere is increasing rapidly. The greenhouse gas content dependent primarily on time, chance, and competition of the atmosphere is the most important determinant to trend toward increasing diversity and complexity. We of climate on geologically short timescales, and models now know that Earth itself is not the mere substrate can be used to predict how climate will change over or background for lifeâs activities as once supposed the next decades and centuries. Over longer geological but rather an active partner in evolution. Geological time periods, natural geological processes control the processes and astronomical events have strongly and greenhouse gas content of the atmosphere, and other repeatedly influenced the story of life on Earth and geological and astronomical factors are influential. We often determine the kinds of life that can survive and have a good qualitative understanding of the factors flourish. that contribute to Earthâs natural climate states, but The interconnectedness of life and the environ- we still lack a comprehensive model that can account ment has been a subject of continuing research and for the climate changes of the past or predict climate debate. An extreme view is that life controls Earthâs changes into the distant future. Better models for both surface environment and does so in ways that are most the volcanic and weathering components of the climate beneficial to the continuation of life (Lovelock, 1979). cycle, more quantitative descriptions of erosion and its But evidence in the geological record, especially of mass relation to weathering, and the incorporation of inputs extinctions, suggests that life cannot always maintain
A HABITABLE PLANET 85 conditions favorable to life. We are far from under- processes of these microorganisms. Innovative isoto- standing how much of evolution is purely biological pic techniques are being used to help understand the and how much has been forced by Earth processes; complicated chemical processing that organisms can nor do we know exactly how much of Earthâs environ- achieve. DNA sequencing methods have brought a ment is determined by the presence of life. And yet new dimension to studies of microbiological processes. these questions have suddenly become more urgent as In the past it was difficult to identify the organisms in we find ourselves in an era whenâpresumably for the natural samples because many could not be cultured. first timeâEarthâs surface environment can be ma- Today, organisms do not need to be cultured; their nipulated by a single dominant life form, Homo sapiens, identity can be determined directly from their DNA. that is capable of making choices about the effects of Computational chemistry (see Question 6) also shines its actions. a strong new light on natural biochemical processes, bringing the possibility of calculating from quantum How Does Life Affect Geological Processes? mechanical theory how atoms and molecules will behave in the microenvironments surrounding tiny Life affects Earthâs planetary processes in several ways. organisms. At the microscopic scale, life is an invisible but power- Soils represent a particularly clear example of how ful chemical force. Organisms can catalyze reactions multiple fields, including inorganic chemistry, phys- that would not happen in their absence, and they can ics, and hydrology, can wrest new insights from geo- accelerate or slow other reactions. The chemical reac- biological processes. Inorganic weathering of minerals tions they enhance have a specific character; in general and organic carbon in the soil environment releases they extract energy from Earth and from sunlight to nutrients and carbon. The rate of release and the types fuel life processes. These reactions, compounded over of nutrients define the environment in which life can immense stretches of time by a large biomass, can exist and control the range and abundance of life forms generate changes of global consequence. An example that can survive. In addition, the roots of land plants, as of this global influence is the processing of carbon and well as bacteria, fungi, and animals such as earthworms, oxygen. Weathering reactions on land, combined with can accelerate the weathering of mineral and organic organic precipitation of carbonate shells in the oceans, matter in soils. Such biological catalysis of weathering remove carbon from the atmosphere and convert it to processes can enhance the suitability of soil for life and carbonate minerals on the seafloor (Question 7). Pho- tosynthesis also extracts carbon from the atmosphere, converting carbon dioxide into oxygen plus organic material. Some of this organic carbon is stored in soils, ocean sediments, and the living biomass of the conti- nents and oceans, while the oxygen is delivered to the atmosphere. Larger animals and plants also have physi- cal effects on Earth, such as promoting soil formation and moderating erosion. Beyond these generalities, we understand little about the details of biologically mediated chemical processes in the environment, especially those of the distant past. Like many fields of science, however, this one is being revolutionized by powerful new analytical tools and computational techniques. For example, new ultrahigh-resolution microscopes can now be used to FIGURE 3.10â High-resolution images of (A) a cell (outer cell observe microorganisms in the environment and in wall indicated by white arrows) and associated mineralized fila- ments (white) and nonmineralized fibrals (gray) and (B) FeOOH- laboratory experiments (Figure 3.10). Synchrotron mineralized filaments filtered from water. SOURCE: Chan et al. X-ray techniques can be used to study the chemical (2004). Reprinted with permission from AAAS.
86 ORIGIN AND EVOLUTION OF EARTH Technology Ice Age Land Animals Mass Extinction Land Plants Animals ?? ?? Protozoans and Algae ?? Bacteria and Archaea 4567 4000 3000 2500 2000 1000 500 0 100 Time (million years) O2 (% of present day) 0 FIGURE 3.11â The history of life, based on geological evidence, along with long-term oxygen, ice ages, and mass extinctions. Molecu- lar data suggest that eukaryotic organisms (protozoans,Figure 3.11.epsand animals) share a common ancestor with Archaea. algae, fungi, plants, also speed up the weathering that would have gone on CO2 that are 10 to 100 times higher than the modern in the absence of life. The ultimate control on the soil atmosphere. The high CO2 concentrations in soil gas environment is probably climate; insufficient rainfall, act to acidify soil water, which leads to increased rates for example, limits how fast both inorganic and organic of dissolution of minerals. Deeply rooted plants can chemistry can proceed. But on a global basis we now also extract water from well below the surface and re- know that soil chemistry is powerful enough to affect turn it to the atmosphere via evaporation from leaves. climate by helping to regulate atmospheric carbon This evapotranspiration has an important cooling dioxide. effect on the land surface, as does the shade provided Similarly, we know that vascular plants have an by the leaf canopy. enormous effect on Earthâs environment. Life on There is ample evidence that plants and animals Earth originated nearly 4 billion years ago, but land also influence erosion rates, but there is still uncer- plants are found in the geological record only during tainty about how important they are in the long-term the past 400 million years or so (Figure 3.11). Several evolution of continental surfaces and how their effects lines of geological evidence suggest that diversifying should be represented in new landscape evolution land vegetation changed the nature of continental models (Dietrich and Perron, 2006). Erosion itself weathering, erosion, and sedimentation, changing the affects habitat conditions and can strongly influence physical stability of stream banks and even influenc- biodiversity and ecosystem processes. Hence a central ing the composition of Earthâs atmosphere (Berner question is the extent to which life and landscape and Kothavala, 2001). Roots break up rock and help evolution are related. For example, do hillslope shapes transform it into soil. Deep roots also contribute car- and river forms reflect the presence of life, or would bon dioxide to soils, resulting in concentrations of soil Earthâs land surface be more or less the same shape if
A HABITABLE PLANET 87 the planet were lifeless? A related question is whether important cyanobacteria, provide at least an impres- topography, which is created by mountain uplift and sionistic view of evolution and diversity that extends erosion, affects the structure of ecological communities. much deeper into our planetâs early history (see Ques- The prospect of changing climate in the near future tion 3). Not all organisms, however, produce the min- brings up other, possibly more urgent, questions. For eral skeletons and tough organic materials preserved example, we would like to know whether rates of ero- as conventional fossils, and this is particularly true sion will change with changing climate and whether of the microorganisms whose metabolic capabilities climate-induced variations in vegetation will reduce or define much of the interface between the physical and enhance the response of erosion rates to climate change. biological Earth. To answer these questions, we need much better mod- Fortunately, a new set of tools has become available els of the effects of biota on weathering, erosion, and to establish the presence and infer the biological activi- sediment transport rates. Biotic diversity needs to be ties of microorganisms in Earthâs history. Most organic linked directly to changes in material strength (resis- compounds in microbial (and, indeed, all) cells decay tance to erosion), mass loss, and sediment mobilization quickly after death. The exception is lipid molecules from hill slopes. Similarly, ecological theory needs to found in cell membranes. These hardy compounds, include explicit physical effects that influence food web commonly called biomarker molecules, can survive processes. These issues lie at the interface of ecological long-term burial in sedimentary rocks and so record and Earth sciences. Since the locus of life is typically aspects of the diversity, environmental setting, and found in the soil that cloaks the landscape, there is a metabolic workings of microorganisms spanning more great opportunity to integrate the fields of pedology, than 2.5 billion years of our planetâs history. Biomarker hydrology, geobiology, geochemistry, and geomorphol- molecules led geologists to the understanding that ogy into a new understanding of this life-supporting petroleum has a biological origin (Triebs, 1936); they system (NRC, 2001). have shown how microbial communities responded to transient oxygen depletion in Mesozoic ocean basins How Long Has Life Fostered a Habitable Surface (Kuypers et al., 2002), illuminated the nature of life Environment? and environments in Proterozoic oceans (Brocks et al., 2005), and provide our earliest evidence for the Because organisms on Earth help maintain a life-sup- presence of lifeâs great evolutionary branches in late porting surface environment today, it is natural to ask Archean ecosystems (Brocks et al., 2003). Preserved or- whether they have always done so. This question proves ganic molecules have even been reported as biomarkers surprisingly difficult to answer, however, largely because in 3.5-billion-year-old rocks from Australia (Marshall we have so little evidence from the early geological et al., 2007). Much remains to be learned about the record. The parts of early organisms richest in infor- sources, function, and biosynthesis of biomarker mol- mation are organic, namely proteins and nucleic acid, ecules, but new research that combines microbiology, but these are also the most reactive and appealing to a genetics, and emerging technologies for analysis (e.g., gauntlet of other organisms bent on using them as food. Brocks and Pearson, 2005) promises unprecedented Even biomolecules that reach the seafloor after death insights into evolution and environmental history both are usually broken down by decay processes within the in the marine realm (e.g., Grice et al., 2005) and on land sediments. Therefore most of our paleobiological infor- (Freeman and Colarusso, 2001). mation is gleaned from the bones, skeletons, and other hard parts preserved as fossils in sedimentary rocks. How Did Organisms Influence the Oxygenation of For the interval of Earth history that begins with the Atmosphere and Oceans? the Cambrian Period (542 Ma), paleobiologists have abundant fossils of plants, animals, and selected algal Perhaps the most obvious and vital link between life and protozoan groups that preserve a compelling record and Earth systems, at least from a human point of of ancient diversity, ecology, and evolutionary pattern. view, is the maintenance of abundant atmospheric Fossils of microorganisms, including the geologically oxygen, a feature whose development we still do not
88 ORIGIN AND EVOLUTION OF EARTH 40 30 GEOCARBSULF no volc 35 25 GEOCARBSULF volc 30 GEOCARB III 20 25 O2% 20 RCO2 15 15 10 10 5 5 0 0 â600 â500 â400 â300 â200 â100 0 â600 â500 â400 â300 â200 â100 0 Time Ma Time Ma Figure 3.12 top left.eps Figure 3.12 top (my BP) Age right.eps 400 300 200 100 0 6000 5000 4000 pCO2 (ppmV) 3000 2000 FIGURE 3.12â (Top) Phanerozoic history of O2 and CO2 1000 inferred from models. SOURCE: Berner (2006). Copyright 2006 by Elsevier Science and Technology Journals. Repro- duced with permission. (Right) CO2 inferred from chemical 0 analysis of soil carbonates. SOURCE: Ekart et al. (1999). Copyright 1999 by the American Journal of Science. S D C P Tr J K T Reproduced with permission. Geologic Period fully understand. Todayâs atmosphere contains about many bacteria and nearly all forms of eukaryotic life 21 percent oxygen and only about 0.03 percent carbon and is critical Figureconcept of planetary habitability. to our 3.12 bottom.eps dioxide, yet multiple lines of evidence indicate that the Photosynthesis provides the only plausible source of atmosphere contained little or no O2 for the first 2 bil- this oxygen, and so oxygenation of the atmosphere lion years of Earthâs history (Bekker et al., 2004) and and oceans constitutes an essential example of how life may have contained much more CO2. In the past 500 has profoundly influenced Earthâs surface conditions. million years, the O2 content of the atmosphere seems Oxygenic photosynthesis also links atmospheric oxygen to have varied from perhaps 10 to 30 percent and the with atmospheric carbon, in that the O2 comes mostly CO2 content from as low as 0.02 percent to as high from extracting oxygen from CO2 and making reduced as 0.7 percent (Figure 3.12). Oxygen is necessary for carbon in the form of organic molecules.
A HABITABLE PLANET 89 stages 1 2 3 4 5 0.5 0.4 atmosphere PO (atm) 0.3 2 0.2 0.1 ? 0 1 2 3 4 5 500 shallow oceans 400 mO (Âµmol) 300 2 200 FIGURE 3.13â Schematic of the rise of atmospheric O2 concentrations. The two curves indicate the approximate 100 range of values allowed by available data. Photosynthetic ? bacteria evolved no later than 2.7 Ga and perhaps as 0 early as 3.8 Ga. Whether the diversity of early photosyn- thetic bacteria included the oxygen-producing cyanobac- 1 2 3 4 5 teria remains uncertain. Geological evidence indicates 500 that whether or not oxygen-generating photosynthesis 400 evolved in Archean oceans, O2 did not become significant deep oceans in the atmosphere and surface oceans until about 2.4 mO (Âµmol) 300 Ga. Geological evidence also suggests that there was a 200 further delay before O2 levels became significant in the 2 deep ocean. Subsequently, there were times when the 100 deep ocean became oxygen poor, even though there was appreciable oxygen in the atmosphere. SOURCE: Holland 0 (2006). Reprinted with permission. 3.8 3.0 2.0 1.0 0 Figure 3.13.eps Oxygen began to accumulate in the atmosphere becomes possible when rates of oxygen production ex- and oceans 2.3 billion to 2.45 billion years ago, but the ceed those of aerobic respiration and other reactions that abundance remained quite low for another 2 billion years consume O2. For example, burial of organic material (re- (Figure 3.13; e.g., Brocks et al., 2005; Canfield, 2005). duced carbon produced by photosynthesis) by sediments Considering that oxygen-generating photosynthetic inhibits aerobic respiration, paving the way for oxygen bacteria were already present 2.7 billion years ago, the accumulation in the atmosphere and oceans. long delay in oxygenation of the atmosphere is hard to On the early Earth other processes could also have understand (Kopp et al., 2005). Why didnât the radiation contributed to the production of molecular oxygen. of cyanobacteriaâthe only bacteria to evolve oxygenic For example, if the early atmosphere contained much photosynthesis and the progenitors, via endosymbiosis, higher amounts of both CO2 and H2O than it does of chloroplasts in algae and land plantsâspread O2 today, substantial hydrogen could have been lost from rapidly through surficial environments to produce an at- the upper atmosphere to space. This process would have mosphere like the one we have today? Part of the answer had the effect of converting H2O to O2. Considering is biological: organisms that respire aerobically, from the evidence that only tiny amounts of oxygen were bacteria to humans, gain energy from the reaction of present in Earthâs early atmosphere, however, this oxygen with organic molecules, reversing the chemistry process could not have been very efficient (Catling et of photosynthesis. The growth of atmospheric oxygen al., 2001; Tian et al., 2005). Photochemical destruc-
90 ORIGIN AND EVOLUTION OF EARTH FIGURE 3.14â Sulfur isotope data that indicate the atmosphere was effectively devoid of oxygen until about 2,400 Ma. Î33S rep- resents the mass independent sulfur isotope fractionation, which occurs at high ultraviolet radiation levels. Nonzero Î33S before 2 Ga implies that ozone (and therefore O2), which absorbs ultraviolet radiation, had very low concentrations. SOURCE: Farquhar and Figure 3.14.eps Wing (2003). Copyright 2003 by Elsevier Science and Technology Journals. Reproduced with permission. tion of methane in the Archean atmosphere could also environmental record at high resolution. For example, produce hydrogen that would escape from Earth, again the recent discovery that the isotopic composition of at- facilitating the oxidation of Earthâs surface (Catling et mospheric sulfur was subtly different before 2.5 billion al., 2001). years ago (Figure 3.14) confirms that the concentration However oxygen accumulated in the atmosphere, of O2 in the atmosphere back then must have been less its consequences were immense. Some bacteria evolved than 10â5 of the present level (Farquhar et al., 2000; a mechanism to gain energy from the reaction of oxy- Pavlov and Kasting, 2002)âeffectively oxygen free. gen gas with organic molecules (aerobic respiration), The rise of atmospheric oxygen also eventually and the ancestors of modern eukaryotes appropriated produced a rise in the level of atmospheric ozone, which this mechanism by capturing respiring bacteria and shields Earthâs surface from ultraviolet radiation that is reducing them to the metabolic slaves we know as detrimental to life on land. The ozone concentration mitochondria. It has been proposed that the bacterial that is sufficient to provide full ultraviolet shielding is ancestors of mitochondria were only facultative respir- surprisingly smallâan atmospheric O2 concentration ers, conducting anoxygenic (nonoxygen-producing) about 1 percent of the modern level (Kasting et al., photosynthesis in oxygen-free environments (Woese, 1985), a level that was probably reached soon after 2.5 1977). If true, the original basis of ecological interac- billion years ago. tion could have been photosyntheticâbut its lasting We do not understand whether the initial evo- legacy was unquestionably aerobic respiration in nucle- lution of cyanobacteria triggered the first round of ated cells. oxygen accumulation or preceded it by hundreds of We do not yet understand how biological, tec- millions of years; nor do we understand why oxygen tonic, volcanic, and atmospheric processes combined levels remained low through most of the Proterozoic to produce the episodic rise in the amount of oxygen Eon (2,500 to 542 million years ago) or what pro- in the atmosphere. In fact, we are only now developing cesses drove the renewed increase that paved the way the analytical tools needed to read Earthâs long-term for animal diversification. For that matter, we do not
A HABITABLE PLANET 91 understand why the modern atmosphere contains the coincides with geochemical evidence for elevated O2. amount of oxygen it does. Oxygen-related questions Oxygen levels may have reached historically high lev- are sufficiently complex that they will require ex- els (perhaps as much as 30 percent of the atmosphere, panded research interactions among Earth scientists, by volume) some 300 Ma, potentially explaining how atmospheric scientists, and biologists. For example, we pigeon-sized dragonflies could fly above tropical forests need better paleontological resolution of when oxygenic of the day (Dudley, 2000). Sharp, if transient, deple- photosynthesis first evolved and when the eukaryotic tion of oxygen in ocean waters had the opposite effect, organisms that now dominate primary production rose reducing animal diversity and size in widespread areas to global prominence (Falkowski et al., 2004). There is of the seafloorâa particularly widespread episode of still no reliable geochemical proxy for ancient oxygen marine anoxia is associated with mass extinction at the abundances (Berner et al., 2003), and models relating Permian-Triassic boundary, some 250 Ma (Wignall deep-Earth processes to surface conditions do not yet and Twitchett, 1996). take account of historical patterns and feedbacks from physiology, tectonics, and atmospheric chemistry. Other Interactions Between Earth and Life Major questions of oxygen history are not limited to its long-term trajectory. During the Paleozoic and Oxygen provides a compelling example of rapidly Mesozoic eras, wide portions of the oceans beneath unfolding research on the interactions between the the surface mixed layer became essentially âoxygen physical and biological Earth, but it is hardly the only deserts,â a condition known as anoxia. Geologically example. Carbon dioxide is also intimately related to transient but globally distributed oceanic anoxic biological activity, not only through climate and the events are well documented from Early Jurassic, Early carbon cycle (Royer et al., 2001) but also because it af- Cretaceous, and Late Cretaceous rocks ( Jenkyns, fects the ability of marine organisms to form carbonate 2003). In modern oceans, O2 can fall to low levels and, skeletons (Kleypas et al., 2006). The physiological link locally, may decline to zero in the ocean-minimum between skeletons and carbon dioxide may help explain zone just below the well-mixed surface water mass in some major biological changes of the past. For example, which most photosynthesis takes place. What makes accelerating physiological research on the biological the oceanic anoxic events stand out is the large spatial consequences of ocean acidification illuminates Earthâs scale of anoxic water masses. So far, we know that these greatest mass extinction at the end of the Permian pe- events coincide with perturbations in the carbon cycle, riod (252 Ma) when marine ecosystems collapsed (e.g., as deduced from records of the isotopic composition see below). Similarly, current increases in carbon dioxide of marine carbon and the strontium isotopic compo- raise concern about the future of reef corals and other sition of seawater (e.g., Jones and Jenkyns, 2001; see organisms that form carbonate skeletons in the shallow Question 7). But we do not know what tipped the ocean. The other side of this interactionâhow biologi- redox balance, causing anoxia to spread repeatedly cal and physical processes interact to govern CO2 on through Mesozoic oceans; nor do we know why there both short and long timescalesâis also a major issue in are similar (but less well documented) events in Pa- Earth history, one of immense importance as we debate leozoic oceans but none from the Cenozoic Era. And the consequences of current human activities (Question we do not know whether these events were produced 7). As in the case of oxygen, deeper understanding of entirely by inorganic geological processes or whether the feedbacks between life and carbon dioxide levels, on organisms exacerbated, ameliorated, or otherwise scales from the local and ephemeral to those governing responded to these events. the long-term history of the planet, will require better As oxygen levels increased in the atmosphere and geochemical and paleobiological proxies for ancient oceans, new forms of life became possible. Animals CO2 abundances, more nuanced understanding of the that move about in search of food have an elevated biological processes that influence carbon dioxide levels need for oxygen, and so it is not surprising that the (especially those related to microorganisms and plants), first evidence for large animals with high oxygen and increasingly sophisticated models that account for demands met (initially) by diffusion through tissues both biological and physical parameters.
92 ORIGIN AND EVOLUTION OF EARTH FIGURE 3.15â Number of marine animal genera through time, showing the five major times of âmass depletionâ in biological di- Figure 3.15.eps versity. Only three dropsâend-Ordovician, end-Permian, and Cretaceous-Tertiaryâare driven primarily by increases in extinction rates, rather than declines in rate of origin. SOURCE: Bambach et al.Image Copyright 2004 by the Paleontological Society, Inc. Fixed (2004). Reproduced with permission. Problems as diverse as the influence of rainforests them. But what specific events or environmental on Earthâs hydrological cycle, the role of vegetation in changes precipitated the great mass extinctions, and stabilizing the land surface, the relationship between what aspects of biology influenced the patterns of nutrient availability and diversity, and the oceanwide survival and recovery, are not known. Most Earth biogeochemical consequences of deep-water anoxia scientists agree that a meteorite impact caused the engage a wide range of Earth scientists because they end-Cretaceous extinction of dinosaurs, ammonites, have both deep-time evolutionary componentsâhow and myriad other plant, animal, and microscopic spe- did the diversification of woody plants change Earthâs cies (Alvarez, 1997), but the actual kill mechanisms surface?âand topical applicationsâwhat will be the unleashed by this trigger remain poorly understood. consequences for the Earth system of rainforest clear- The relative importance of coincident environmental cutting, increased soil erosion, and seafloor anoxia perturbations, including an interval of oceanographi- linked to fertilizer-spiked nutrient flows from agricul- cally driven global change, extensive extrusion of flood tural lands to the ocean? Earth scientists have almost basalts, and the particular location of the impact on a limitless opportunities to join with biologists to fashion tropical continental platform, are simply not known. both a new picture of our planetâs history and a clearer Although a single plausible event may account for picture of our future. the end-Cretaceous extinctions, the cause of the end- Permian mass extinction, which may have erased as What Caused Mass Extinctions? many as 90 percent of marine species and many terres- trial species (Erwin, 2006), is still debated. Support for Nothing illustrates how heavily life depends on a favor- an extraterrestrial cause is limited, with growing inter- able surface environment as clearly as a sharp change est in direct and indirect effects of massive volcanism in in that environmentâwhich has occurred several times what is now Siberia. An emerging view is that massive during the past 500 million years, causing the mass flood basalts, intruded through thick carbonates and extinction of species (Figure 3.15). In particular, the extruded onto thick peat deposits, produced unusually great extinctions at the end of the Permian (252 Ma) high emissions of carbon dioxide and thermogenic and Cretaceous (65 Ma) periods influenced the course methane, resulting in global warming, acidification of of biological evolution as much as all the accumulated the oceans, depletion of oxygen in ocean waters below genetic changes during the 187 million years between the mixed layer, and enhanced production of hydrogen
A HABITABLE PLANET 93 sulfide by bacteria living in those oxygen-depleted rapidly emerging insights into the physical history of water masses. Physiological research on modern ma- the Earth surface system provide, for the first time, the rine organisms, aimed at understanding current envi- proper environmental framework to address the issue. ronmental change (e.g., PÃ¶rtner et al., 2005), allows Has primary production increased through time, and Earth scientists to predict the biological consequences if so what have been its consequences? What are the of such an event on end-Permian biological diversity. consequences of sea-level change, episodically flooding Indeed, paleobiological data show that extinctions did and exposing continental interiors, on species origina- not affect all Permian animals equally. For example, tion and extinction in the marine realm (Peters, 2005)? groups whose living relatives were vulnerable to the Did the rules of community construction change when physiological consequences of sea acidification disap- flowering plants evolved the capacity to use animals to peared at rates much higher than those physiologically ensure the faithful spread of pollen from one plant to well buffered against such environmental perturbations. the next? How did the ecological relationships that un- Extinctions on land are consistent with the predicted dergird community diversity reform following episodes effects of rapid climate change (summarized in Knoll of mass extinction? Detailed analyses of community et al., 2007). Continuing research on Earthâs great in- organization in systems as disparate as Pleistocene tervals of biological upheaval will increasingly integrate coral reefs, Cenozoic mammals, and Carboniferous insights from paleobiology, stratigraphy, high-precision forests promise important insights into ecology and geochronology, and geochemistry with physiology and evolution that cannot be made solely on the basis of models generated to help understand current issues of the short-term observations and experiments available global change. to biologists ( Jackson and Erwin, 2006). What Governs the History of Biological Diversity? Summary Major extinctions have clearly influenced the history Earthâs surface environment is obviously altered by of plant and animal life, but what, fundamentally, con- large-scale geological processes (Questions 4 and 5), trols the observed pattern of diversity increase from but it is also affected continuously and pervasively by the Cambrian to today (Figure 3.15)? Quantification the activities of life forms. Likewise, Earthâs geological of diversity change through time on land and in the evolution and infrequent catastrophic events, such as oceans remains a subject of active research and debate, meteorite impacts, have clearly affected the evolution but many Earth scientists would agree that the modern of life. But even when we can document extinctions world (at least in preindustrial times) harbors more and major evolutionary changes, we cannot yet sort species of land plants, more species of land animals, out the causes. To what extent were they caused by and more species of marine animals than any previ- geological as opposed to biological processes? Which ous moment in our planetâs history (e.g., Benton and environmental conditions were responsible for which Emerson, 2007). Attempts to model diversity history extinctions or changes in biological form and function? employ logistic equations, which imply biologically We know that the composition of Earthâs atmosphere, or physically imposed limits to diversification (e.g., especially its high concentration of oxygen, is a major Sepkoski, 1984), or exponential equations, which imply consequence of the presence of life, one that made persistent diversity increases, episodically knocked back possible the evolution of more complex organisms. But by mass extinctions (Stanley, 2007). exactly how other geological events have affected evolu- The tension between these classes of models tion, and how much control life has had on climate, are focuses attention on a great and unsolved problem. still topics of debate. What are the relative roles of genetic innovation, Life processes and Earth processes also interact ecology, and physical Earth history in governing the locally. Erosion rates, climate, and weathering rates long-term history of life? The answer certainly requires affect the habitability of specific regions of Earth, and macroecological insights from biologists, but the ques- the ecosystems themselves in turn affect erosion rates, tions are necessarily framed by paleontologists. And climate, and weathering processes. Understanding the
94 ORIGIN AND EVOLUTION OF EARTH interrelationships between surficial processes that shape lectual challenge. Meeting this challenge will help the land and the life that inhabits it presents a critical us understand how life will respond to present-day challenge for managing land resources and becomes environmental change, but Earth scientists will have even more important as we attempt to forecast the ef- to develop new research and educational partnerships fects of future climate change. with biologists and atmospheric scientists. The search Understanding how Earthâs life and geological for life on extrasolar planets will similarly depend on environment arrived at their present state, and how better understanding of biogeochemical influences on they interacted in doing so, constitutes a major intel- atmospheric composition here at home.