The Green Marble

THE EVOLUTION OF THE BIOSPHERE

What I see is that life produces byproducts and side effects that can shove the environment around into various chemical states. All organisms, linked in the chemical vessel of the biosphere, must adapt to these states or go extinct.

—T. Volk (1998)

As is evident from chapters 1 and 2, the field of Earth system science sprawls across many disciplinary boundaries. From its beginning, there has been a struggle to formulate unifying principles or narratives that are scientifically valid as well as broadly applicable and compelling. Perhaps the most well-known thrust in that direction is the Gaia hypothesis. Most broadly, it suggests that Earth as a whole is analogous to a living system, and correspondingly has homeostatic capability. The concept is attractive because it encompasses all the physical and biological properties of Earth, and proposes a basis for their integration as a system.

As we will see, the homeostatic Gaia concept has been rejected by geoscientists because of its teleological packaging. Nevertheless, the Gaia hypothesis matters to Earth system science because of its scientific implications: It has inspired researchers to look carefully at global-scale processes and at the mechanisms by which Earth’s environment is regulated. It matters politically because whether or not there is Gaian homeostasis impacts the urgency of doing something about human-induced climate change. It matters to global environmental governance because the belief in a planetary-scale living entity has been adopted by many nonscientists and underlies their support for addressing key issues associated with global environmental governance.

THE GAIA HYPOTHESIS

The original Gaia hypothesis of James Lovelock and Lynn Margulis (expounded in the 1970s) held that the biosphere is part of a self-regulating biogeochemical cycling system on Earth that maintains conditions favorable to life (Lovelock 1979; Lovelock and Margulis, 1974). Two observations were key to their thinking. The first was that Earth’s atmosphere is unusual among the atmospheres of planets in our solar system in having significant concentrations of reactive gases, including oxygen and methane. Oxygen is critical for aerobic respiration and is used in the production of metabolic energy by much of the biomass on Earth; methane is a greenhouse gas that significantly warms the global climate. Lovelock was a chemist and realized that since oxygen and methane readily react, their notably high concentrations in Earth’s atmosphere could only be maintained by a high rate of production. The likely source of those gases was the biosphere.

The second observation was that climate conditions on Earth have remained within a habitable range (i.e., above the freezing point of water and below its boiling point) over the past four billion years. At first glance, climate stability may not seem remarkable. However, as we saw in the last chapter, that stability has been in the face of an approximately 25 percent increase in solar luminosity, a large variation in the rate of greenhouse gas inputs to the atmosphere from tectonic sources, and multiple catastrophic collisions with meteorites. Even if all else were constant, the increase in solar luminosity alone would have increased global mean temperature somewhere between 6°C and 25°C (Celsius), thus pushing the planet to more Venus-like conditions.

Lovelock considered the influence of the biosphere on the greenhouse gas concentrations as a key mechanism by which Gaia maintains homeostasis (Lovelock and Watson, 1981). And he challenged the geosciences community to work out other mechanisms. However, the reception of his ideas by the scientific world was generally not very positive.

One problem was that the original formulation of the Gaia hypothesis had a mythological tone to it: Gaia is the Greek name for the mother goddess. The name was suggested by Lovelock’s neighbor, British author William Golding (Lord of the Flies). Lovelock sought to popularize the concept with a book in 1979 (Gaia, a New Look at Life on Earth). In doing so he gave Gaia a rather teleological tone, somewhat more philosophical that strictly scientific.

Another recurrent complaint was that the Gaia hypothesis was not testable. Classically, a scientific hypotheses or theory is “a statement about some phenomenon in nature that in principle can be confronted with reality and possibly falsified” (Bak, 1996, p. 162). Supporters of Gaian homeostasis thus have worked to identify feedback loops (see box 2.1 in the previous chapter) by which the biosphere interacts with the climate. A weak test would be whether biosphere feedbacks in the climate system are always negative (i.e., stabilizing). If they are positive, the biosphere would have to be considered disruptive rather than homeostatic.

The problem with testing specific predictions from Gaia theory is that we are just beginning to be able to monitor global-scale biosphere processes and trends (see chapter 9). We began directly monitoring global mean temperature only about 100 years ago and have monitored atmospheric carbon dioxide (CO2) concentration only since 1958. The record of past states of the Earth system in the paleorecord (e.g., ice cores and the chemical properties of rock formations) can be used to evaluate feedback mechanisms to some degree, but these records are rather sparse, and interpretations are subject to controversy. To compliment observations, Earth system scientists have built simulation models of the Earth system to test their ideas about Gaian homeostasis. Here, too, there is much debate as to validity (Schellnhuber, 1999).

Despite the qualms within the scientific community, the American Geophysical Society (in response to advocacy by scientists including Carl Sagan and Steve Schneider) sponsored two Gaia-themed workshops (Schneider and Boston, 1991; Schneider, Miller, Crist, and Boston, 2004). The topic was popular around 1990 because the issue of human-induced global climate change was beginning to engage the geosciences community. The appeal to geoscientists lay in creating a multidisciplinary forum open to speculations about global-scale process that involved the biosphere. There was a general recognition that the biosphere indeed has a role in global climate regulation. But as we shall see, there was also substantive criticism of making general conclusions about Gaian homeostasis (Kirchner, 2003).

Besides its scientific legacy, the concept of Earth as Gaia has had a significant impact on the environmental movement. Lovelock originally was not attempting to inspire the environmental movement; in fact, his first book was rather hostile toward environmentalists (Lovelock, 1979). However, he did maintain that “As the transfer of power to our species proceeds, our responsibility for maintain planetary homeostasis grows with it” (p. 131). Many people in the environmental movement warmly embraced Gaia despite Lovelock because the concept not only got people thinking about global change issues but also became the platform for development of a global environmental ethic (Lautensach, 2008).

Because of the possible dire consequences for advanced technological civilization associated with human perturbation of Earth’s climate system, there is now tremendous interest among the science, policy, and environmental communities in understanding global climate regulation. If Gaian negative feedbacks to climate change turn out to be strong enough, we might be saved from our own excesses. If the biosphere feedbacks to anthropogenic warming are predominantly positive, there is greater urgency to mitigating climate change. If humanity is the product of a benevolent global entity (Gaia), then perhaps we should learn to love and protect that entity (in a sense our Mother).

THE MEDEA HYPOTHESIS

In a pointedly contrary view to the Gaia hypothesis, Professor of Earth Science Peter Ward developed the idea (in his book The Medea Hypothesis) that the biosphere is not homeostatic, but rather in some ways self-destructive (Ward, 2009). Medea, like Gaia, is a mother figure from Greek mythology. But instead of symbolizing creativity and nurturing, Medea is famous for killing her own children. Ward’s line of reasoning focused on the multiple extinction events in Earth’s history that were not caused by collisions with asteroids.

The Great Oxygenation event and associated extinction event, about 2.5 billion years ago, is a prime example. It was induced by newly evolved bacteria capable of aerobic photosynthesis. The associated transformation of the atmosphere was toxic to most existing life forms. A second example is the end-Permian extinction event about 250 Mya (million years ago). It has been hypothesized (Kump, Pavlov, and Arthur, 2005) that the kill mechanism in this case was biologically based production of toxic hydrogen sulfide (derived from sulfate-reducing bacteria). Ward also emphasized the future fate of the biosphere. He cites global carbon-cycle model outputs that suggest biosphere-driven drawdown of atmospheric CO2 will eventually prove fatal to the biosphere itself (Franck, Bounama, and von Bloh, 2006).

Lovelock built the Gaia hypothesis based on assumed negative feedbacks in the Earth system. For example, as solar luminosity increased over geological time, the biosphere compensated by reducing the concentration of greenhouse gases. Gaian homoeostasis relies in theory on cooperative interactions among biosphere components. In contrast, the Medea hypothesis looks to positive feedbacks, and to competition among biosphere components. In the Great Oxidation, microorganisms capable of aerobic photosynthesis and respiration outcompeted the primitive anaerobes and forced their retreat to the anaerobic ocean depths. Their exit left more resources (e.g., sunlight) for the aerobes.

Like the Gaia hypothesis, the Medea hypothesis has normative implications. Ward makes a connection between the dystopian tendencies of the biosphere and the current human-driven disruption of the Earth system. He pointed to Homo sapiens as a species that has found a winning strategy (use of fossil fuels) and is now laying waste to everything else in the biosphere, i.e., acting in typical Medean fashion. Ward’s take-home message was that if we continue on this path (continuing to “rape and pillage the planet”), we could destroy ourselves. The conclusion is that we must change course and begin rationally managing the Earth system.

The Medea hypothesis did not receive nearly as much scientific or popular attention as the Gaia hypothesis. Ward never brought it to the peer-reviewed literature as Lovelock and Margulis had done with the Gaia hypothesis. Reviews of his book called it thought provoking, and indeed it does inspire one to consider the geological context of the human disruption of the climate system. However, the whole premise is rather weak because the biosphere has not destroyed itself over the 3.5 billion years of its life span, and, in fact, has consistently recovered from catastrophic disturbances and returned to high levels of biodiversity.

MECHANISMS OF BIOSPHERE INFLUENCE ON THE CLIMATE SYSTEM

Greenhouse Gases

As we noted in chapter 2, the terrestrial biosphere speeds up (by a factor of 10–100) the rock weathering process, hence pumping CO2 out of the atmosphere. This amplification drives the carbon cycle to an approximate steady state at a lower atmospheric CO2 concentration, hence a cooler mean temperature, than would be the case with the background purely physical-chemical system (Lenton, 2002). However, there is considerable disagreement about the magnitude of this cooling effect and how it has changed over geological time (Schwartzman, 1999). Presumably as the world gets warmer and wetter, this negative biosphere feedback strengthens.

How about at the cold end of the spectrum? When global climate is relatively cold, atmospheric CO2 is generally low, and the biosphere in fact begins to be starved for CO2 as photosynthesis decreases. As the vigor of the terrestrial biosphere declines, its capacity to enhance weathering weakens, and the CO2 sink is moderated (Beerling et al., 2012; Pagani, Caldeira, Berner, and Beerling, 2009). Again, we have a negative feedback.

If we take a Medean view of the low CO2 situation, the biosphere was largely responsible for driving CO2 so low in the first place by speeding up the rate of mineral weathering. This systematic behavior shows up in the model of Franck and colleagues (2006), which Ward repeatedly cites. That model simulates how the interactions of the geosphere, biosphere, and climate regulate the carbon cycle over geological time and into the future. According to the simulations, the biosphere will face a CO2 crisis approximately one billion years in the future because the area of readily weatherable rock will have increased and the biosphere will continue to speed up weathering rates. The biosphere will eventually become starved for CO2, as well as being subject to very high temperature (driven by increasing solar radiation). Ward (2009) suggests that the biosphere will eventually drive CO2 levels down so low that oxygen production will falter. Then the atmospheric O2 level will decrease to the point of causing another extinction event, or indeed the demise of the biosphere. This scenario is a key argument for the Medea hypothesis.

Other mechanisms by which the terrestrial biosphere affects the atmospheric CO2 concentration are through (1) increase or decrease in total biomass associated with changing climate and the area of ice-free land above sea level, and (2) formation of carbon compounds that are not readily decomposed and returned to the atmosphere (e.g., peat and coal). These processes operate at a range of spatial and temporal scales, which complicates attempts to understand the geological record of atmospheric CO2 concentration.

Life in the ocean (the marine biosphere) also impacts the atmospheric CO2 concentration. We’ve noted the ocean CO2 sink associated with the biological pump, which occurs when marine organisms take up CO2, and a portion settles to the ocean floor as organic carbon or calcium carbonate. Enhancement of this mechanism may have been part of a negative feedback to hyperthermal events such as the Paleocene–Eocene thermal maximum, mentioned in chapter 2. Faster weathering on land delivers more nutrients to the ocean, which increases marine productivity, which tends to draw down the atmospheric CO2 concentration. Again, this is a characteristically Gaian mechanism.

In contrast to these negative feedbacks to climate warming, the approximately 80 ppm (parts per million) decrease in atmospheric CO2 seen in the ice core record during the cold intervals of the glacial/interglacial cycle (see figure 2.4) is clearly a positive feedback. As discussed in chapter 2, it most likely reflects a transfer of CO2 from the atmosphere to the ocean. A variety of processes would change the CO2 in the ocean as the global climate warmed or cooled. As noted previously, cold water holds more CO2; thus, simple diffusion tends to bring down the atmospheric CO2 concentration as the climate cools. The ocean holds vastly more carbon that the land or atmosphere. Biologically driven CO2 uptake in the ocean may also be enhanced. Hypothesized mechanisms for that process include increased delivery of iron and other nutrients to plankton in dust associated with the generally drier climate (Marinov, Follows, Gnanadesikan, Sarmiento, and Slater, 2008), or changes in ocean circulation that bring more nutrients to the surface. Greater ocean productivity means more dead organic matter settling to the ocean floor, with an associated increase in carbon sequestration. It thus appears that the marine biosphere participates in a positive feedback to climate change.

Methane and nitrous oxide (N2O) are other greenhouse gases whose concentrations strongly depend on the biosphere. Although their concentrations are much lower than that of CO2, these gases are respectively about 20 and 300 times more effective per molecule as greenhouse gases than is CO2. They contribute 4–9 percent and 1–3 percent, respectively, to the current greenhouse effect. Microbes generate these molecules, and various chemical and biological processes break them down. By providing sources of these gases as a by-product of their energy metabolism, the associated microbes are providing a warming influence on the global climate.

The story is more complicated, of course, because other types of microbes are largely determining the oxygen concentration, and oxygen is involved in the chemical reactions that consume methane and nitrous oxide. The net effect of production and consumption is that their concentrations rise and fall in parallel with the global mean temperature during the glacial/interglacial cycles. Methane is produced by anaerobic respiration in wetlands, and tropical wetlands expand during warm interglacial periods. Likewise, for nitrous oxide. These greenhouse gases are evidently part of a positive biosphere-driven feedback to climate change.

The contribution of troposphere ozone, another significant greenhouse gas, is also complex. Earlier, we looked at stratospheric ozone because it plays a significant role in shielding the biosphere from ultraviolet radiation. However, ozone is also found in the troposphere (the lower atmosphere) at sufficient concentrations to contribute to greenhouse gas warming (it contributes 3–7 percent to the current greenhouse effect on Earth). The chemistry of tropospheric ozone is complicated, but one of the reactants involved in its production is a class of molecules called nonmethane hydrocarbons (NMHCs). These are low molecular weight gaseous compounds that volatilize from the surface of leaves and bark (Tingey, Turner, and Weber, 1991). The fragrance of a pine tree results from the release of NMHC molecules. Various physiological functions have been proposed for NMHC molecules, including chemical defense of plants, and a role in shunting excess solar energy off the main photosynthetic pathway under high light conditions. In any case, the concentration of ozone in the troposphere would be lower without biogenic NMHCs, and hence global climate would be cooler.

While researching this topic with funding from the U.S. Environmental Protection Agency’s Global Change Research Program, I did an early assessment of possible feedbacks to climate change associated with NMHCs (D. P. Turner et al., 1991). Generally, higher temperatures mean greater NMHC production and more NMHCs mean more tropospheric ozone. Thus, the sign of the feedback to global warming would likely be positive. However, my study did not consider the recent ecophysiological studies that find high CO2 may inhibit NMHC production (Beerling, Hewitt, Pyle, and Raven, 2007). The complexity of the coupling between the biosphere and the atmosphere in this case prevents us from specifying the sign and magnitude of the NMHC-mediated climate change feedback.

Surface Energy Balance

The biosphere also plays a poorly understood role in how clouds affect global climate. Clouds mostly reflect incoming solar radiation, thereby cooling the climate. The connection to the biosphere is in the provision of cloud condensation nuclei (CCN). Clouds form because water molecules condense into small droplets under the influence of decreasing air temperature. CCN are required to initiate cloud droplets, and on a non-biospheric Earth the CCN would most likely be dust particles. However, under some circumstances, dust particles can be in short supply, and this shortage may limit cloud formation. One illustration of this phenomenon is the observation of cloud trails behind oceangoing vessels. The air over the ocean usually is close to its saturation point (i.e., high relative humidity) and the provision of CCN from engine exhaust is sufficient to induce formation of clouds in the wake of a ship.

How does the biosphere produce CCN? One source is planktonic algae in the ocean that produce a molecule called dimethyl sulfide (DMS) as part of their osmoregulatory metabolism. When DMS molecules diffuse into the atmosphere, they interact with atmospheric oxygen to produce sulfate molecules. Aggregation of sulfate molecules results in aerosol particles that act as CCN. In fact, a large proportion of CCN over much of the ocean is derived from DMS. Thus, DMS has a significant overall cooling effect on global climate.

If there was a mechanism associated with global warming that favored DMS production, then biosphere-mediated negative feedback would occur (Charlson, Lovelock, Andreae, and Warren, 1987). However, recent modeling exercises do not support the case for strong global-scale negative feedback mediated by DMS, and the idea is still under debate (Levasseur, 2011). The observation that ocean acidification, associated with high anthropogenic CO2, reduces DMS production (Six et al., 2013) does not bode well for a negative feedback mechanism to current CO2-induced warming.

Over land, the shortage of CCN is not as acute as over the ocean, but biosphere production of CCN is important there, as well. The source of CCN in this case is the same NMHCs from plants that contribute to ozone production. In the atmosphere, NMHCs aggregate into aerosol particles that can act as CCN. They are often the dominant CCN over tropical rain forests (Chatfield, 1991). If global warming favored more tropical rain forests, there would likely be more clouds, and hence a cooling effect. Through this mechanism, biogenic NMHCs could thus participate in a negative feedback to global warming. To date, however, no one has worked out what the net sign of the NMHC feedback to global warming would be, using a method that considers both greenhouse gases and CNN effects. Efforts to understand the role of NMHCs in climate regulation exemplify the need for interdisciplinary research in Earth system science (Beerling et al., 2007).

The albedo (reflectance) of vegetation cover itself can influence climate significantly. Dark vegetation absorbs solar radiation, and the warming-driven conversion of tundra to conifer forests provides a positive feedback to climate warming (Beringer, Chapin, Thompson, and McGuire, 2005). In the temperate and tropical zones, the dissipation of absorbed solar energy by vegetation evapotranspiration provides a significant surface cooling effect.

There are also influences of vegetation on precipitation. A high rate of evapotranspiration by forests tends to keep the local atmosphere moist, and when the moisture condenses to form precipitation, this reaction “actively creates low pressure regions that draw in moist air from the oceans, thereby generating prevailing winds capable of carrying moisture and sustaining rainfall within continents” (Ellison et al., 2017, p. 53). Recycling of precipitation is estimated to support 20–35 percent of total precipitation in the Amazon River Basin (Eltahir and Bras, 1994).

BIOSPHERE INFLUENCE ON GLOBAL CLIMATE OVER GEOLOGICAL TIME

The Four-Billion-Year View

Both Lovelock and Ward point to the whole arc of biosphere history in support of their respective models. The pertinent facts are as follows: (1) solar irradiance has increased approximately 25 percent over Earth’s nearly four-billion-year history, and (2) the concentrations of greenhouse gases have fallen by a factor of ten or more. The net effect has been a climate continuously able to support the biosphere.

Lovelock emphasizes that the biosphere has largely controlled the greenhouse gas concentrations in a way that compensated for the increase in solar radiation, and thus has acted in a homeostatic manner. At the time he proposed the Gaia hypothesis, Lovelock did not have a very clear idea about the specific mechanisms by which the biosphere could bring down greenhouse gases. But recent interest in biosphere-enhanced mineral weathering now fills that gap (Schwartzman, 1999). Lovelock’s teleological Gaia seems to be most concerned about the planet overheating.

Ward’s Medea hypothesis emphasizes the cold, low productivity threat to the biosphere. He suggests that the biosphere is practically on a death march, and will eventually (a billion years hence) kill itself by drawing down greenhouse gas concentrations too far.

The Gaian case for homeostasis is more solid on the face of it because it is trying to account for something that has already occurred. The Medean model is less compelling because it relies on projections a billion years into the future (Franck et al., 2006). Of course, neither view really explains how integration could be achieved across something as multifaceted as the biosphere.

The 500-Million-Year View

Considering only the past 500 million years, Earth system scientists have been able to assemble a reasonably coherent picture of the temporal variations in the geosphere, the biosphere, the atmosphere, and the climate. At the coarsest scale, there appears to be a rough oscillation between a warm state and a cold state. Under warm conditions (“Greenhouse Earth” or “Hothouse Earth”), CO2 concentration is high (usually greater than 1,000 ppm) and glaciation is low. As a correlate of high CO2, water vapor density would also be relatively high. The high greenhouse gas concentrations largely determine the global warmth. Conditions are reversed during the cold periods (“Icehouse Earth”).

Two cycles of about 300 million years from Greenhouse Earth to Icehouse Earth are reasonably clear (figure 3.1). Some geologists also emphasize four cycles of approximately 150 million years. The most general explanation for the apparent cycles is based on plate tectonics. Fischer (1984) suggested a cycle between periods of mountain building and periods of quiescence. During mountain-building phases, volcanism is intensive and CO2 levels are high (mostly greater than 1,000 ppm). Eventually large areas of unweathered minerals are exposed, and the associated high rates of mineral weathering (probably increased by the warm climate) draw down the atmospheric CO2 concentration.

The most striking Icehouse phase is centered in the Permian period around 300 Mya. About that time, there is evidence in the geological record of massive glaciations and in the CO2 record (e.g., from stomatal abundance) of a drop in CO2 concentration (Retallack, 2002). Ward (2009) and other scientists point to a major change in the biosphere at that time, which likely accelerated the decrease in CO2. Vegetation had been present on land since at least 500 Mya, but around 400 Mya, the fossil record indicates the evolution of plants with vascular systems and mycorrhizal (i.e., symbiotic with fungi) root systems. These developments had the effect of creating the pedosphere (soil), which greatly increased the surface area of minerals interacting with roots. The effect was an increase in the mineral weathering rate and a drawdown of atmospheric CO2, i.e., a strengthening of the rock weathering thermostat (see box figure 2.1 in chapter 2). In addition, there is evidence in the geological record of increased coal formation (a carbon sink on land). This propensity to sequester carbon in coal can be attributed to the increasing productivity of land plants, and possibly the evolution of lignin synthesis. Lignin provided structural support for trees, but was not yet readily digestible or decomposable by heterotrophic organisms.

FIGURE 3.1

Comparison of estimated atmospheric CO2 concentration and periods of glaciation throughout the Phanerozoic eon. Panel A has plots for the central tendency based on various proxies in the paleorecord, along with results and shaded uncertainty for GEOCARB III, a model that simulates the global carbon cycle over geological time. Panel B shows intervals of glacial (dark shade) or cool (light shade) climate. Mya = million years ago. Adapted from Royer, Berner, Montanez, Tabor, and Beerling (2004).

The Permian Icehouse phase was apparently not catastrophic for the biosphere. The climate did not get as cold as during a Snowball Earth state, there was not a mass extinction, and CO2 probably did not fall below approximately 200 ppm. If Medea was going to kill itself, as Ward proposes, why not do it then?

The next cold phase, around 180 Mya, is also believed to have been driven by a CO2 decrease associated with a high rate of silicate rock weathering. In this case, it was weathering of the supercontinent Pangea (Schaller, Wright, and Kent, 2015). The continent was moving north (driven by tectonic forces) into warm humid latitudes about this time, so the weathering rate sped up.

The other obvious Icehouse phase in the global mean temperature over the past 500 million years is relatively recent. It began about 30 Mya and continues to this day. In chapter 2, the various factors contributing to late Cenozoic cooling were discussed, in particular, thermal isolation of the poles, which favored glaciations, and the low CO2 concentration (less than 500 ppm), which was probably brought down by vigorous weathering of the uplifted Himalayan Mountains (Ruddiman, 1997).

During the past million years, the climate has been about as cold as anytime in the past 500 million years (figure 3.1). The planet is apparently saved from going all the way to a Snowball condition by the strength of solar radiation (i.e., the Goldilocks effect) and negative feedback mechanisms such as low levels of water vapor and perhaps less cloud cover (associated with cold, dry air). Terrestrial biosphere enhancement of the rock weathering CO2 sink is certainly diminished during the coldest periods (the glacial maxima) because the low CO2 (approximately 200 ppm) reduces terrestrial biosphere vigor and the low temperatures reduce chemical reaction rates. The Milankovitch orbital forcings are evidently sufficient to initiate the glacial/interglacial cycles, but not strong enough to push the Earth system to Snowball Earth or Hothouse Earth states.

The four obvious Icehouse Earth phases do not support the hypothesis of strong biosphere control of global climate. Rather, tectonic processes that control CO2 emissions (volcanoes, seafloor spread) and CO2 sinks (the area of silicate rock available for weathering and its location) are dominant in these slow oscillations.

Climate crises such as the end-Permian extinction event and the Paleocene–Eocene thermal maximum may have more to say about the Gaia/Medea hypotheses. These periods appear to be induced by the geosphere, rather than a Medean biosphere. The termination of the hyperthermal episodes in the early Cenozoic period seems to have a significant Gaian aspect in that terrestrial biosphere–enhanced mineral weathering, and perhaps marine biosphere–enhanced carbon burial, both contributed to drawing down the atmospheric CO2 concentration and returning the Earth system to cooler conditions. The same is true for the end-Permian extinction event.

The 100-Year View

In the past 100 years (the blink of the eye in geological time), the CO2 concentration has increased from about 280 ppm to more than 400 ppm (a gain of 43 percent). The increase is undoubtedly driven by CO2 emissions associated with fossil fuel combustion. This change in the atmosphere will have profound impacts on the biosphere, and from a Gaian perspective we would predict some sort of negative feedback, e.g., an increased net uptake of CO2 to compensate for the increased emissions. As we shall see in chapter 4, this has largely not happened. Nor has the rate of silicate rock weathering significantly increased. Homeostatic Gaia has not shown up (Schlesinger, 2013).

If humans are considered part of the biosphere, then the ongoing upsurge in CO2 concentration does appear to have a Medean character to it. As we shall see in chapter 5, the abruptness of the increase will take a toll on the biosphere by way of rapid climate change and ocean acidification. Note, however, that the biosphere is in no danger of annihilation. Much of its metabolism is microbial and will persevere even in the worst-case scenarios of human-induced environmental change.

Given the ambiguity of the historical relationship of the biosphere to the global environment over geological time, it is worth asking if we can identity any mechanism that might shape the biosphere into a superorganism-like entity capable of regulating Earth’s climate.

BIOLOGICAL EVOLUTION AND BIOSPHERE EVOLUTION

One of the strongest scientific objections to Gaian claims of a homeostatic biosphere is that the concept does not fit well with contemporary interpretations of how biological evolution works (e.g., Dawkins, 1999). Traditionally, the unit of selection in biological evolution is the organism. In contrast, Lovelock’s view points to the biosphere components of the Earth system as selected for how they contribute to the overall homeostatic capacity of the planet as a whole. Teleological Gaia would have to evolve a mechanism for coordination among its parts.

Lovelock and Margulis did not specifically address the evolutionary mechanisms by which Gaia was formed. They did point to cooperation within symbiotic relationships among species as an indicator that life can, in fact, be integrated above the organismic level. Margulis was the biologist on the Lovelock–Margulis team and she is best known as an early advocate for the theory that in the primitive biosphere, coevolution is likely to have been responsible for joining of multiple types of single-celled prokaryotes to form the larger, more complex eukaryotic type of cell (Margulis, 1970).

Formally, coevolution is a gene-for-gene interaction between two species; i.e., a mutation within the gene pool of one species favors or defends against a specific gene in the gene pool of another species. This kind of relationship is frequently found between pathogens and their hosts. The relationships can also be positive, as in the case of mutualisms in which both species benefit from the interaction. In the Margulis model of early microbial evolution, what had been free-living organisms became organelles within larger cells. However, these organelles kept their own genetic material (albeit with some losses). Selection thus began acting at the level of the genome—all genetic material within the larger organism.

The lichen is an often-cited example of mutualism. Here, two organisms are symbiotically linked, with the fungal and algal parts of the lichen dependent on each other for survival. (There are some exceptions; occasionally the algae can live independently.) Algal cells provide energy by photosynthesis and fungal cells serve both a structural function (providing a physical framework) and a metabolic function (promoting mineral weathering and nutrient uptake). This strategy has been solidly successful, and lichens have thrived for billions of years in a wide range of environments. Symbiosis between nitrogen-fixing microorganisms and plants verifies that nutrient exchange is a potent force in biological evolution.

Relationships between nonspecialized flowers and their pollinators provide examples of weaker coevolution. Many flowers depend on insect pollinators for their reproduction and they correspondingly invest considerable energy into attracting and rewarding those pollinators. Sometimes these weak relationships become quite strong. Famously, Darwin discovered a tubular flower with an extraordinarily long distance between the top of the flower and the base (containing the pollen and stigma). There were no know pollinators with a proboscis that long, so he predicted that such a creature must exist and would eventually be discovered. Indeed, many years later he was proved right!

One of the best-known examples of mutualism is the mycorrhizal association. In this case, a plant provides carbohydrates to a fungus that has infected its roots. The plant benefits because the fungal hyphae efficiently explore the soil for water and nutrients and transfer those resources to the plant. The intimacy of the interaction is such that in some cases the haustoria (root-like structures) of the fungus anatomically penetrate the fine root cells of the plant. Scientists have hypothesized that what started out as pathogenic relationship between species gradually evolved into a mutualistic relationship.

When microbes perform a nutrient cycling function, such as nitrogen mineralization (release of nitrogen from dead organic matter), they are ultimately providing nutrients to another ecosystem component that, in turn, nourishes them. Consider a soil arthropod (insect) that develops (through genetic mutation) an innovation allowing it to digest a previously indigestible component of tree litter. It uses the carbon-to-carbon bonds in the litter component as an energy source and excretes excess molecules (e.g., of nitrogen) for which it has no use from the litter. The tree life form then takes up the nitrogen and ultimately produces more litter that will eventually feed the soil arthropod.

This kind of nutrient cycling loop is the basis for the concept of an ecosystem. The term originated in the 1930s and is considered to include interacting biotic and abiotic components in a circumscribed location. Ecosystems require energy flow (usually a solar source) and are characterized by materials cycling. Among other pursuits, ecosystem ecologists develop nutrient budgets for the cycles of carbon, nitrogen, and other nutrients. As understanding of nutrient cycling improves, the sense of an ecosystem as a functional entity strengthens; e.g., ecosystem ecologists study mechanisms by which leakage of essential nutrients is accelerated after disturbances (Likens, Bormann, Pierce, and Reiners, 1978).

Ecosystems have been characterized as complex adaptive systems (Levin, 1998). In that view, the characteristics of the ecosystem (e.g., its trophic structure and nutrient conservation capabilities) are products of interactions among its components. Over time, these components become functionally linked by way of biological evolution because each is part of the selective environment of the others. The interactions of the components open the possibility of positive and negative feedback loops that reinforce the coupled system.

Studies involving artificial selection of miniature ecosystems hint at how ecosystems may change over time. In these experiments (Swenson, Wilson, and Elias, 2000), scientists constructed miniature ecosystems in small cup-like containers. To each container, a dollop of sterile soil was added that had been inoculated with a slurry of nonsterile soil (containing thousands of species of microbes). A fixed number of seeds of a simple plant were added. After the seeds germinated and were exposed to a month of sunlight, the experiment ended and aboveground biomass production—i.e., net primary production (NPP)—in each miniature ecosystem was determined by clipping and weighing. Soil lines that produced the most biomass were then combined (selected) to set up the next generation of containers, which were reseeded and allowed to grow again. After selection over multiple generations, some soil lines were consistently superior in the amount of NPP they could support. Note that the specific mechanisms of growth enhancement are not yet known; but most likely the mechanism involves enhancement of nutrient cycling by means of interactions among a specific set of microbes. In simulated microbial ecosystems, selection at the ecosystem level can be an effective adaptive force given a networked spatial structure (H. T. P. Williams and Lenton, 2008).

Experiments (intentional and otherwise) with introduction (Ehrenfeld, 2010) and removal (Ripple and Beschta, 2005) of species from more natural ecosystems also suggest strong interdependencies. In chapter 5 we will examine the phenomenon of trophic cascades, e.g., the process by which extirpation of wolves changes the behavior of grazers, which alters the composition and structure of associated forests. The emerging theory of niche construction is beginning to provide a basis for reconciling evolution and ecosystem dynamics. This concept (further discussed in chapter 4) refers to “organismal alteration of ecological patterns and processes in ways that confer heritable advantages and/or disadvantages to individuals or populations” (Ellis, 2015, p. 292). In other words, the operation of natural selection is based on both genes and environment impacts of those genes.

Ecologists traditionally think in terms of a levels of organization hierarchy (i.e., cells, tissues, organisms, populations, communities, and ecosystems). As we progress up that hierarchy, we leave behind the “myth of individuality” (Margulis, 2006), recognizing that higher organisms are indeed born as individuals, but if they are not highly integrated with other organisms in their vicinity, and with microorganisms within their own body, they may not last long. Part of the perhaps hyperbolic rejection of the Gaia concept by most evolutionary biologists was an ongoing tension within the discipline about the level at which natural selection operates (D. S. Wilson, 2001) and even what constitutes heritable variation and selection (Ellis, 2015). The controversy starts with disagreements about the “selfish gene” versus the organismic level (Noble, 2008), and extends much farther up the levels of organization hierarchy (Matthews et al., 2011). The Modern Synthesis for evolution accounts for Darwinian natural selection of organisms, Mendelian genetics, and population genetics. And later extensions accommodate the discovery of the biochemical basis for genetics in deoxyribonucleic acid. But there is now serious discussion of a fundamental extension of the Modern Synthesis paradigm for evolution; i.e., the new Extended Evolutionary Synthesis (Laland et al., 2015; Pigliucci, 2007) in which selection operates simultaneously at multiple levels of organization and evolutionary theory expands “beyond genetics to explain the evolution of complex phenotyptic traits across a variety of taxa” (Ellis, 2015, p. 291).

Integration of Earth system science with evolutionary theory is only beginning to be thought about. Schwartzman (1999) refers to feedbacks between biotic evolution and biosphere evolution. Here’s how it might work. Say a new biochemical pathway or trait, e.g., photosynthesis, is favored by natural selection. The photosynthesis trait is adaptive in its own right at the organismic level because it taps a new source of energy (i.e., compared to chemoautotrophs). The trait spreads throughout the biosphere such that the biosphere now begins to alter the global climate by sequestering CO2, e.g., in marine sediments. The associated cooling of the global climate changes the selective regime, which may drive further biotic evolution toward life forms better adapted to the new climate. The evolution of the biota is driven by natural selection; the “evolution” of the biosphere is based on its functional role in the Earth system (box 3.1).

As we noted in discussing the early biosphere, no single way of making a living (i.e., type of metabolism) can persist for long because the available substrates become depleted, and rare nutrients tend to accumulate in living and dead organisms. Specialized organisms that can take advantage of the waste products of other organisms will be favored evolutionarily (Volk, 1998). The earlier noted evolution in the Paleozoic era of fungi capable of decomposing lignin, much of which at the time was turning into coal, is an interesting case study (Floudas et al., 2012; J. M. Robinson, 1990). Lenton and Watson (2011) refer to a series of biogeochemical revolutions in Earth history, generally associated with new forms of chemical recycling.

In Earth system science, we can consider Earth itself as the global ecosystem. The energy source is mostly the sun. Volk (1998) refers to biogeochemical cycling “guilds” as the living parts of the global ecosystem. These various types of organisms (e.g., nitrogen fixers) participate in the synthesis and breakdown of organic materials and maintain the global biogeochemical cycles. Thus, it is meaningful to talk about the “metabolism” of the biosphere, and about the biosphere as a component of the Earth system.

BOX 3.1

Semantic Issues

Earth system science is a transdisciplinary field, and as such is subject to occasional semantic issues when different disciplines use words in different ways. There is not necessarily a right and wrong here, but at times it is worth being especially alert to context.

From the perspective of studying the global biogeochemical cycles, the biosphere is considered the actual biomass of all life on Earth (Hutchinson, 1970). Its mass can be quantified, and it has a distinct functional role in maintaining the global biogeochemical cycles. For geographers, who may be more concerned with the spatial arrangement of their objects of study, the biosphere is considered the spherical space around Earth’s surface that is occupied by life (Gillard, 1969).

An Earth system scientist speaks of the evolution of the biosphere, meaning changes in the array of species on Earth and their corresponding effects on the atmosphere and climate, e.g., the change from a biosphere dominated by anaerobic microorganisms to one dominated by aerobic microorganisms (Schwartzman, 1999). For an evolutionary biologist, the process of evolution refers to changes in gene frequencies within a population.

Ecologists speak of the biosphere as a complex adaptive system, meaning it is composed of coupled components—and the composition of the components as well as the nature of the couplings change in response to environmental change (Levin, 1998). For an evolutionary biologist, an adaptation is a genetically based trait that has been fixed in the gene pool by natural selection.

This issue of conceptual differences among disciplines plays out more broadly. Leemans (2016) refers to the difficulty in developing a conceptual framework for a transdisciplinary research program, “especially if it must be accepted by all different disciplines, who all, for example, differ in how they use central concepts, such as time. Physicists have adopted a continuous notion of time in their differential equations, ecologists and geographers use discrete time steps (years, seasons, generations) in their difference equations, while economists fold the future into the present using a discount rate” (p. 106). These sorts of differences can make transdisciplinary research trying at times, but certainly interesting.

Although, there is only one biosphere and hence no possible analogue to Darwinian competition and selection at the global scale, it is worth noting that quasi-regulatory effects of the biosphere (or a particular species) on global climate might nevertheless be expected (Lenton, 1998, 2002). If the biosphere drove the global environment in a direction detrimental to itself (simply as a byproduct of its normal metabolism), then the biosphere as configured would lose vigor and hence be less effective in continuing to drive the environment in that deleterious direction. That seems to be the case when atmospheric CO2 concentration falls below about 200 ppm (driven down in part by biosphere-enhanced weathering, which is a CO2 sink). This type of negative feedback in the climate system relies only on the inherent property of living matter to grow under favorable conditions.

ESSENTIAL POINTS

The Gaia and Medea hypotheses are not truly scientific propositions. Neither makes sense from a traditional evolutionary perspective. However, both serve a purpose with respect to addressing global environmental change issues. The Gaia concept evokes the Earth system holistically—a planetary body that can be characterized in terms of energy flows and biogeochemical cycles. Conceivably, humanity can build a scientific understanding of the Earth system, which would provide a basis for assessing how it is being impacted by the technosphere and how a sustainable relationship of technosphere to biosphere might be constructed. The Medea concept also encourages a global-scale perspective. It reminds us that the Earth system as we know it is not permanent. Geosphere-induced and biosphere-induced changes can radically alter the chemistry of the atmosphere and oceans, as well as the climate. These changes are not always to the benefit of the dominant life forms.

IMPLICATIONS

1. The biosphere strongly influences the chemistry of the atmosphere, the oceans, and the soil; hence it participates in regulation of global climate. When geological or solar forces alter the climate, the response of the biosphere contributes to the net change in climate. The net biosphere response may be a negative or positive feedback. To understand and anticipate how the Earth system will respond to human-induced climate change, it is thus necessary to understand the response of the biosphere. Earth system science is pursuing that goal.

2. The biosphere strengthens the negative feedback built into the silicate rock weathering thermostat. However, the temporal framework of that feedback is on the order of thousands of years. Thus, it is not relevant in the near term to addressing the current pulse of climate warming associated with anthropogenic CO2 emissions.

3. The influence of the biosphere on the climate is not always beneficial with respect to biosphere productivity and biodiversity. Humanity has come to expect the quasi-regulatory services performed by the biosphere, but they are not guaranteed as we shift the Earth system away from the state in which we inherited it.

4. Biological evolution changes the biota, in some cases introducing new biogeochemical cycling pathways. The spread of such changes throughout the biosphere can alter how the biosphere as a whole affects the chemistry of the atmosphere and the climate system, thus altering the selection regime for the biota. This integration of the evolution of the biota and the “evolution” of the biosphere is going on even now as Homo sapiens, a recent product of biological evolution, has begun (by way of the technosphere) to radically alter the selection regime (e.g., by impacts on biodiversity and changing the global climate).

5. In the evolutionary history of the biosphere, “cooperation” has been as important as competition in driving the increase in complexity from bacteria, to eukaryotic cells, to multicellular organisms, to symbiotic associations among different species, and to ecosystems with their characteristic nutrient-cycling regimes. The cooperation metaphor points toward the needed integration of the technosphere with the rest of the Earth system to address major global environmental change issues.