Biosphere: How Life Alters Climate

People had long speculated that the climate might be altered where forests were cut down, marshes drained or land irrigated. Scientists were skeptical. During the first half of the 20th century, they studied climate as a system of mechanical physics and mineral chemistry, churning along heedless of the planet's thin film of living organisms. Then around 1960, evidence of a rise in carbon dioxide showed that at least one species, could indeed alter global climate--humanity. As scientists looked more deeply into how carbon moved in and out of the atmosphere, they discovered many ways that other organisms could also exert powerful influences. Forests in particular were deeply involved in the carbon cycle, and from the 1970s onward, scientists argued over just what deforestation might mean for climate. By the 1980s, it was certain that all the planet's ecosystems were major players in the climate changes that would determine their own future.

Gaia (1972-1988)

Geoscientists had thought of carbon mainly as something to do with volcanoes and the weathering of rocks. But from the early 1970s forward, they understood that biology was a major player in the global carbon budget. Indeed it dominated the game on the human timescale of centuries. For other chemical elements, for example the cycle of sulfur through the oceans and atmosphere, scientists still felt that simple mineral chemistry must predominate. That changed during a research voyage on the Atlantic Ocean that included James Lovelock, a wide-ranging and exceptionally independent-minded researcher. His Ph.D. was in medicine, but his most notable achievement at this point had been inventing instrumentation for measuring rare gases even at tiny concentrations. On the high seas Lovelock discovered that one such gas, dimethyl sulfide (DMS), was a principal element in the global sulfur cycle. The main source of DMS was ocean plankton. ⁽¹⁾

Lovelock was already convinced that, as he put it, "the atmospheric gases are biological products." His interest was partly stimulated by gases that he found everywhere in the Earth's atmosphere and that were undoubtedly produced by living creatures: pollutants from human industry. But Lovelock based his thinking more deeply on the most fundamental property of biology, the uphill march of life against entropy. ⁽²⁾

Back in the 1960s, Lovelock had proposed measuring gases in the Martian atmosphere as a way to look for traces of life. Living creatures, he realized, emitted gases that would drive their planet's atmosphere into "a state of disequilibrium." Mars lacked the free oxygen of our own planet precisely because Mars was sterile. At this point in Lovelock's thinking, a stable balance gave witness to dead minerals, whereas the system of life plus minerals created a perpetual state of dynamic imbalance. ⁽³⁾

Lovelock ran into trouble when he tried to publish these ideas in 1966. At the time he simply remarked that the physical sciences habitually ignored the physical effects of life "to the point of blindness." Long afterward, he reflected that "Conventional biology and planetary science held the false assumption that organisms merely adapt to their environment. My ideas for life detection acknowledged that organisms change their environment... Neither my critics nor I were aware of this fundamental difference of viewpoint." Lovelock's difficulties illustrated how hard it was to grasp that living creatures could play a huge role in the geochemistry of their planet. ⁽⁴⁾.

In 1974, Lovelock put together a grand generalization in collaboration with Lynn Margulis (who shared a taste for planet-sized speculation with her former husband, Carl Sagan). Their article was entitled, "Atmospheric homeostasis by and for the biosphere: The Gaia hypothesis." Lovelock and Margulis proposed that the ensemble of living creatures had taken "control of the planetary environment" in a way that would maintain conditions favorable for life itself. This pushed to the limit the new way of seeing the atmosphere as something susceptible to biological influence. Under the new hypothesis the atmosphere was altogether "a component part of the biosphere," in fact a "contrivance." The rhetoric and the name, after the Greek Earth-goddess, carried an implication of purposeful and indeed supernatural guidance, which disgusted many scientists. But if you stripped away any implication of conscious purpose, the idea that biology controlled atmospheric content was rationally defensible. ⁽⁵⁾

For more than a decade the Gaia hypothesis led nowhere scientifically. Most scientists considered it visionary at best. Then in 1987 Lovelock, working with Robert Charlson and others, hypothesized that the DMS emitted by ocean plankton could influence climate, much like the smoggy sulfur aerosols produced by human industry. In the clean air over the oceans, particles of DMS were a major source of nuclei for the condensation of the water droplets that would form clouds. This suggested a Gaia-like self-regulation. Perhaps if the oceans got warmer, the plankton would produce more DMS... which would make more clouds and more reflection of sunlight from the atmosphere... which would bring a compensatory cooling back toward normal. Perhaps this biological regulation "has already counteracted the influence of the recent increase in CO₂ and other 'greenhouse' gases." ⁽⁶⁾ However, one could also imagine scenarios where global warming killed off plankton, bringing no beneficial feedback but a vicious circle of increasing warmth.

Some people hoped the Gaia hypothesis could put a scientific foundation under the traditional belief in ecological self-regulation, the beneficent "balance of nature." Over the long run, species that damaged their ecosystem were automatically laid low (a troubling thought, given that humankind was such a species). To others the hypothesis was misleading, not science but mysticism. If the Earth's atmosphere had remained favorable for life over the past billion years, most scientists saw no logic or evidence compelling them to think that it was due to anything but sheer good luck. Lovelock himself admitted that the hypothesis might never be proved definitively. In any case, he later added, human interference might be large enough to force the global system beyond the point where nature could maintain a balance. What the Gaia hypothesis did accomplish was to encourage scientists to investigate how biology could show up in every corner of atmospheric chemistry. For both scientists and the public the debate promoted an understanding that life interacts with climate in ways unforeseeable and disturbingly powerful.

Everything is connected to everything else: from a high-minded but nebulous philosophy, this viewpoint had evolved into a scientific requirement for analyzing the planet. The final answer to the question of climate change would be a set of predictions for the levels of gases, temperatures and precipitation, and their impacts on ecosystems and human society. That could come only through calculations with a model that incorporated all the significant factors and their interactions. A start at mapping such a model was made at a workshop held in Jackson Hole, Wyoming in 1985. The panelists projected sketches on a wall and scribbled over it until they got a consensus on what the most important subsystems of the model would be. The result, which became known in the modeling community as the "wiring diagram," had more than three dozen arrows connecting an even larger number of boxes. Similar diagrams sketched a decade earlier had ignored biology, but here it was at the center. The boxes were highly simplified ("cloudiness," "nutrient recycling," "human land use," "marine biological production," and the like), and the community was a long way from knowing how to calculate what happened in most of them. Even if scientists had known all that, computers that could handle the calculations were decades in the future. ⁽⁷⁾

From the 1980s forward, a variety of experts did extensive work on how human agricultural and economic systems as well as natural ecosystems might interact with global warming and a rise of CO₂ levels. Those studies mostly fall outside the scope of these geophysical essays and are not discussed here.

After 1988

Like clouds drifting in from the horizon heralding the possibility of a storm, the prospect of global warming increasingly caught the attention of scientists far afield from traditional meteorology. They began work to organize big, long-term field studies in dozens of specialized topics of agriculture, forestry, and so forth, to see how climate might interact with the planet's many ecosystems. There was far too little money to support all those studies, but some important questions were at least partly answered.

The oldest question was whether a change in vegetation, especially a change caused by humans, could alter regional climates? The answer was now certain: Yes. At several locations, overgrazed grasslands with dried-out soils had become demonstrably hotter than less-used pastures (and the heating would make it all the harder for grass to return). Some rain forests that had been cut down showed a measurable decrease in rainfall, since moisture was no longer evaporated back into the air from the leaves of trees--in Brazil, rain fled from the plough. [Work published in 2004 gave a more complex picture: under some circumstances, deforestation could bring more rain storms when air rose from the hotter ground.] Scientists also pointed out that if global warming made forests grow farther north, the dark pines would absorb more sunlight than snowy tundra, heat the air, and add to global warming. Such regional studies were too few to paint a clear picture of how the many types of vegetation in total could affect global climate. The studies did show for certain that wherever vegetation was altered, whether through direct human action or by a shift of climate, there could be serious feedbacks with a potential for a lasting, self-sustained change. ⁽⁸⁾

A 1989 review of computer climate studies concluded that the next generation of models would have to include detailed representations of vegetation. By the mid 1990s, biologists and modelers were discussing such details as the way increased levels of CO₂ would affect the evaporation of moisture from leaves. And since nothing influenced vegetation so much as humans, the models must also somehow include social and economic forces. ⁽⁹⁾

Some scientists stuck by the old view that biological feedbacks were reassuring rather than alarming. They held that fertilization from the increased CO₂ in the atmosphere would benefit agriculture and forestry so much that it would make up for any possible damage from climate change. ⁽¹⁰⁾ The fertilization effect was confirmed by field measurements of the exchange of carbon in various forests, and by studies of the consequences of blowing extra CO₂ across crops, grasslands, and so forth. For the planet as a whole, biomass did seem to be absorbing more CO₂ than in earlier decades. However, the studies turned up some unsettling results. The numbers were often very different from what the handful of earlier, more primitive studies had suggested. And the consequences of fertilization were not straightforward. For example, under some circumstances the extra CO₂ might benefit weeds and insect pests more than desirable crops. In any case, as the level of the gas continued to rise, plants would reach a point (nobody could predict how soon) where they would be unable to use more carbon. The increase in plant growth would level off. There was a good chance that more warmth would eventually foster decay, with a net emission of greenhouse gases.

A few people suggested solving the problem by using biology deliberately. Perhaps we could manipulate the "biological pump" of dead plankton that snowed down upon the ocean floor, taking carbon with them? The plankton did not flourish without trace minerals, which are scarce in mid-ocean. For decades there had been talk about improving the biological productivity of barren ocean regions by adding nutrients, something like the traditional nitrate and phosphate fertilizers used by farmers. Studies in the late 1980s and 1990s suggested that iron was the keystone fertilizer. By dumping iron compounds where the element was lacking, we might be able to stimulate plankton to bloom. Could the biological pump bury carbon as quickly as our industries emitted it? The pioneer of the theory, John Martin, joked in a Strangelove accent, "Give me a half tanker of iron and I will give you an ice age!" ⁽¹¹⁾

Scientists began planning experiments to see just how much carbon they could send to the sea floor with a shot of fertilizer. [Quite a lot, under the right circumstances, according to studies completed after 2001. But the details of these circumstances were as obscure and complex as everything else in the oceans.] Many people warned that in view of how little we knew about ocean ecosystems, this sort of meddling might just make things worse. For example, what if fertilizing plankton made them emit extra methane or other potent greenhouse gases?

Meanwhile Broecker and a few other dedicated specialists tried to unravel the tangled biological and chemical history of the oceans through glacial periods, by following tracer minerals such as cadmium. Broecker's initial ideas were in error, as he realized "almost before the ink had dried on the publication." That was the story of much that followed. As he admitted in 2000, "The prize has yet to be grasped." Oceanographers were starting to realize that the drifting plankton formed communities as complex as a rain forest, in which only a tiny fraction of the species had even been identified. Exploring the interactions among these creatures, and the important consequences for the movement of carbon, was a project that would take many decades. ⁽¹²⁾

Attempts to balance the current carbon budget continued to hold center stage through the 1990s. Debate persisted over such issues as whether tropical forests were a net source or sink for carbon. Meanwhile some continued to present arguments that excess CO₂ was mostly sinking into the oceans, opposed by others who came up with equally persuasive argument that the gas was mostly going into plants. Only more data could resolve these questions. Particularly helpful were regular measurements of CO₂ levels at many locations, made by the U.S. government (to be precise, NOAA, with analysis chiefly under Keeling at Scripps). Flasks of air were gathered at a string of stations running from the South Pole up to an ice floe in the Arctic Ocean. The variations from season to season said much about the movements of the gas. Another powerful way to interpret these numbers came from new and precise data on oxygen in the atmosphere. The oxygen level is fractionally altered wherever burning fuel emits CO₂ and wherever plants emit or take up the gas, but the oxygen level is unaffected when CO₂ is taken up in the oceans. The ingenious and painstaking measurements were the work of Ralph Keeling, Charles David's son. ⁽¹³⁾ Over the course of the 1990s, the various numbers tended to converge, suggesting that none of the debaters was entirely right or entirely wrong.

The reasons for the long-standing confusion were explained in part by new studies, which showed that the uptake of carbon by forests and soils was varying erratically and massively. A region that had absorbed carbon overall during one decade might be a major source of carbon in the next. In particular, it seemed that much of the "missing carbon" had been absorbed by Northern Hemisphere forests in some decades, but not in others. The uptake might depend on various things, such as the weather fluctuations brought on by El Niño events. ⁽¹⁴⁾

A 1998 review estimated that overall, humanity was emitting seven billion metric tonnes of carbon each year by burning fossil fuels and another one or two by clearing tropical forests. About half of this stayed in the atmosphere, and the oceans absorbed a quarter, which left roughly two billion tonnes per year that terrestrial ecosystems must somehow absorb. Measurements suggested that much of this was taken up in the Northern Hemisphere--most likely by rapidly growing forests in temperate zones, but perhaps also by peat bogs, or by something else entirely. [A massive study published in 2004 emphasized that this uptake was liable to change. The oceans had absorbed about half the total carbon produced by humanity since 1800, with the rest remaining in the air. Thus over the past two centuries, emissions due to deforestation and other changes had roughly matched absorption by the terrestrial biosphere.] ⁽¹⁵⁾

Looking to the future, experts had still not resolved such basic questions as whether tropical forests, by absorbing or releasing carbon dioxide, were more likely to retard global warming or hasten it. In every ecosystem, the carbon balance would depend heavily on what humans did about deforestation, reforestation, and fertilization (including global fertilization not only by CO₂ but also by our rising emissions of nitrate gases). The uncertainties raised severe problems for international negotiators when they tried to assign responsibility to particular nations for how much they added to the greenhouse effect or subtracted from it. In 2000 a group of researchers managed to couple computer models for the atmosphere, oceans, vegetation, and soils all together, and their preliminary results were ominous. Their best guess was that around mid-century the warmer biosphere would turn from a net absorber to a major emitter of carbon, speeding up climate change. [Already in 2003, a measurement of an Arctic bog showed a sharp rise in methane emissions since 1970.] In a 2001 report, an intergovernmental panel of experts concluded that the net effect of feedbacks from global warming "is always to increase projected atmospheric CO₂ concentrations." ⁽¹⁶⁾

Meanwhile cores drilled from ancient ice revealed that such changes had happened in the past. The levels of CO₂ and methane in the atmosphere had lurched up and down as ice ages came and went. It seemed that the gas levels had changed after the temperature shifts. If so, warming and cooling had raised and lowered emissions of CO₂ and methane. It was a strong confirmationthat these gases played a potent role in climate change through feedbacks, whose engine was located in the biosphere.

This essay does not discuss the extensive scientific work that has been done on expected "impacts" of global warming on living creatures (including us). In brief: heat waves, the spread of tropical diseases, storm surges along with sea-level rise, and regional changes in precipitation and temperature, are expected to kill many people and displace many more, decrease agricultural productivity in most areas (although with temporary improvements in some), and threaten or extinguish a wide variety of species if not entire ecosystems.

Related:

The Carbon Dioxide Greenhouse Effect

Simple Models of Climate

1. Lovelock et al. (1972).

2. "products:" Lovelock and Margulis (1974), p. 9.

3. Hitchcock and Lovelock (1967), "disequilibrium" p. 150, "blindness" p. 158.

4. "Conventional" Lovelock (2000), p. 235, for Gaia, see ch. 9.

5. Lovelock and Margulis (1974), p. 5. The more usual spelling was Gaea.

6. Charlson et al. (1987), p. 661.

7. Fisher (1988); earlier diagram: Kellogg and Schneider (1974).

8. The pioneering demonstration that the Amazon Basin generated much of its own rainfall was Salati and Vose (1984); Couzin (1999). More rain (in dry season only): Negri, 2004.

9. Rowntree (1989), p. 174; IPCC (2001), pp. 440-43.

10. Especially Idso (1989).

11. The importance of iron was demonstrated by Martin and Fitzwater (1988); Martin (1990); Martin made his famous "half tanker" remark at a Woods Hole 1988 conference, according to John Weier, obituary at http://earthobservatory.nasa.gov/Library/Giants/Martin/.

12. The Cd pioneer was Edward Boyle, e.g., Boyle (1988b); Coate et al. (1996) was a successful fertilization experiment; "ink had dried... prize:"Broecker (2000); for more on this rapidly developing topic, see news reports in the journals Nature and Science, e.g., Chisholm (2000).

13. Keeling et al. (1989) (this is C.D.); Tans et al. (1990); oxygen work by Ralph K.: Keeling and Shertz (1992); Keeling et al. (1993).

14. Among the many publications: Battle et al. (2000); Bousquet et al. (2000); Schimel et al. (2001).

15. Among the many publications see, e.g., Prentice and Lloyd (1998); Schindler (1999). Ocean study: Sabine et al. (2004). Altering since Neolithic: Ruddiman (2005)

16. Coupled models (emphasizing soil emission): Cox et al. (2000); IPCC (2001), p. 186, including physical ocean effects as well as land biota. Methane: Christensen et al. (2004).