Eccentrics. One sandwich short of a picnic. A screw loose. History is littered with sometimes charming enthusiasts who go a bit over the top, following a line of reasoning beyond the pale, shocking, outraging or just titillating more sober and cautious types. Sometimes they are vindicated as knowledge increases over time and their musings are shown to have been prescient. Hannes Alfven, 1970 Nobel Prize winner in Physics, is one of my favorite examples of this.

for much of his career Alfven's ideas were dismissed or treated with condescension. He was often forced to publish his papers in obscure journals; and his work was continuously disputed for many years by the most renowned senior scientist in space physics, the British-American geophysicist Sydney Chapman. Even among physicists today there is little awareness of Alfven's many contributions to fields of physics where his ideas are used without recognition of who conceived them.

This isn't always or even often the outcome. We are wise to be sceptical, but stupid to ignore them entirely. Sometimes they are adamant and maintain their heresies until they give up the ghost, sometimes they recant such as Einstein did with his cosmological constant, though the idea is still around and new theories of dark energy have revived it somewhat.

As a bio-geek I am naturally attracted to speculation about the size and extent of the biosphere. Thomas Gold's Deep Hot Biosphere thesis is intriguing as is Lovelock's Gaia thesis. Both pay particular attention to microbial life below ground and in the seas, arguing that the amount of life we ignore dwarfs the charismatic surface species, including ourselves, in volume and significance. We do continue to find bacteria and archaea in places we thought to be barren. Some of Gold's ideas have been supported by drilling cores into the earth, and we have made recent discoveries about the seas as well.

Scientists are now revisiting, and perhaps revising, their thinking about how Archaea, an ancient kingdom of single-celled microorganisms, are involved in maintaining the global balance of nitrogen and carbon. Researchers have discovered the first Archaea known to oxidize ammonia for energy and metabolize carbon dioxide by successfully growing the tentatively named, Nitrosopumilus maritimus, in the lab.

"Data from several cultivation-independent, molecular experiments led us to suspect that Archaea could be involved in the marine nitrogen cycle. Subsequently having the organism isolated in the lab allowed us to confirm our suspicions," said David Stahl, professor of civil and environmental engineering at the University of Washington. Stahl's lab group specializes in environmental microbiology and how microbial communities function in diverse locations including the oceans, hot springs, animal intestines and the human mouth.

Archaea have primarily been associated with extreme environments like hot springs and deep-sea vents, but about a decade ago molecular studies proved their abundance in more common environs including the open ocean, freshwater and soil. Subsequent efforts to grow various samples of these organisms led to this cultivation of N. maritimus, or "dwarf belonging to the sea," by Stahl and scientists at the Woods Hole Oceanographic Institution.

Part of our ignorance stems from bad lab techniques. We conclude that there is no life in samples from which we could not grow cultures in petri dishes, but as we become better in the lab we increasingly find that we had previously missed the vast majority of species. This is true for every environment: the seas, the soil and even inside living creatures including ourselves. Some have startling importance, such as Nitrosopumilus maritimus discussed above, because they play crucial roles in nutrient cycles such as nitrogen and carbon.

Greg Retallack, a soil scientist at the University of Oregon in Eugene, is another of those eccentrics that has caught my eye.

After the dinosaurs disappeared 65 million years ago, wiped out by an asteroid impact or other calamity, plants seized their chance. The emergence of the first grasses was the breakthrough. Grass doesn't hold much CO2 itself, but it can create mollisols, soils that are very rich in organic matter and hence carbon. "Typically they are 10 per cent organic matter down to a depth of a metre, whereas forest soils are only that rich down to about 10 centimetres," says Retallack. So a grassland ecosystem can, despite appearances, contain more carbon than a forest ecosystem.

Over the past 40 million years or so, tall grasslands spread across the globe, taking over many formerly forested zones. These ecosystems, Retallack argues, took control of the planetary thermostat, securing lower CO2 levels for their own advantage. New grazing animals evolved to live on and coexist with the grasses. "The co-evolution of grasses and grazer created a carbon-hungry ecosystem of a kind never before seen," says Retallack. "I think mollisols are saving our skins right now. Without them the world would be a lot hotter."

That pressed a few hot buttons. One of my semi-quixotic enthusiasms has been to alert the dim and indifferently educated to the importance of soils for carbon sequestration, and how our agricultural behaviors have so radically diminished soil carbon over the past millenia, accelerating in the 20th century as we industrialized and expanded. We have dumped a lot of carbon from fossil fuels into the air, but we have also dumped a lot of fossil soil carbon into the air by plowing up the grasslands and turning them into field and row crops.

Since the mechanization of agriculture began a few hundred years ago, scientists estimate that some 78 billion metric tons – more than 171 trillion pounds – of carbon once trapped in the soil have been lost to the atmosphere in the form of CO2...

Retallack jazzes on this theme.

As the Earth cooled under the influence of grasslands, it seemed to hit an era of abrupt swings into and out of ice ages, beginning about 5 million years ago. Could this too be explained by the battle between plants and animals? Retallack thinks it could, but many scientists disagree. . .

The composition of the atmosphere clearly plays a role. Air bubbles trapped inside ice cores show that the rise and fall of temperatures corresponds to fluctuating levels of greenhouse gases such as CO2. Levels of CO2 jump by around 30 per cent as the world emerges from each glaciation. This seems to suggest that these gases amplify the Milankovitch wobbles.

But Retallack goes further. He wonders if the fluctuations in CO2 could be a result of natural battles between the Earth's living kingdoms. "I am not impressed with the Milankovitch theory. I think ecosystems are in control," he says. For one thing, the wobble in the Earth's tilt does not obviously match actual temperature changes, says Retallack. Equally troubling, the wobble produces a smooth change in solar radiation, whereas the glaciations are not smooth. The world cools gradually over tens of thousands of years, and then warms again in less than a tenth of the time. "Something snaps," Retallack says. But what? And would it snap anyway, without the planetary wobbles? . . .

Retallack is at pains to say he does not discount the power of geology. He is no Gaian purist. He even admits that the geology-based theories are right now "probably closer to proof" than his own. "Meteorite impacts, volcanic eruptions, hot-spring degassing and Milankovitch control are all well accepted by most scientists," he says. "But I think there is a middle way." He believes that biology has played a crucial role in the switchback of climate change over much of our planet's history. And, being a pedologist, he is convinced that the evidence lies in the soil.

The evidence he relies on is the cyclical transition of mid-continental grasslands to sagebrush scrublands, and back, in concert with cycles of glaciation. Which is the cause and which is the effect?

The conventional view is that these changes merely represent the response of ecosystems to climate change. But Retallack believes it may be the other way round: the ecosystems drive the glaciations, as carbon enters soils when grasslands dominate and leaves again in sagebrush eras. The jury is still out on this idea. But Retallack is not alone in thinking biology could be important in ice ages. Lars Franzen of Gothenburg University in Sweden has argued that the formation of peat bogs, which draw large amounts of CO2 out of the atmosphere, could pull the world into glaciation--and push it out again as the bogs decay under the ice. In any event, with evidence growing that biology can drive climate change on other timescales, it cannot be dismissed.

Retallack's ideas were unveiled at the Geological Society of America's annual meeting in Reno last November. It was a difficult audience: geologists are, after all, used to seeing the processes they study as the drivers of the planet's environment. Even Gaians are playing hard to get. One, Lee Klinger of the National Center for Atmospheric Research at Boulder, Colorado, says the story cannot be as simple as Retallack suggests: "He makes no mention of the ocean biota, which certainly impact CO2. Nor does he properly consider changes in peatlands, which contain one or two orders of magnitude more carbon per unit area than either forests or grasslands." Lovelock sees another gap: "He has ignored Gaia's farts. Methane was crucial to the health of the early atmosphere and still plays an important role." But both Klinger and Lovelock think Retallack should pursue his idea further. "I wouldn't want to put him off," says Klinger.

This is an important issue. Whether Retallack's ideas eventually come to dominate or not the fact that soils can sequester massive amounts of carbon, and that we have released a great deal of this fossil carbon in the past 200 years, may be of far greater significance than forest cover. We would be wise to restore grasslands and encourage ruminant livestock operations that explicitly address this issue by managing such grasslands for the deep rooted perennial grassland species that sequester the most carbon.

Update: Mollisols

These carbon rich grassland soils deserve a closer look.

Mollisols are the soils of grassland ecosystems. They are characterized by a thick, dark surface horizon. This fertile surface horizon, known as a mollic epipedon, results from the long-term addition of organic materials derived from plant roots.

Mollisols primarily occur in the middle latitudes and are extensive in prairie regions such as the Great Plains of the US. Globally, they occupy ~6.9% of the ice-free land area. In the US, they are the most extensive soil order, accounting for ~21.5% of the land area.

The image to the right shows world wide distribution of mollisols in red. Notice that they are dominant across Eurasia and N. America, the US is over 1/5 mollisols. It's interesting that these also tend to be the most developed and wealthiest parts of the world. I wonder if Jared Diamond ever learned anything about soil science?