Reflections on the stabilization of Earth’s climate by life.
People frequently believe the claim that basic physics, established in the 19th Century, is sufficient to predict that Earth will warm in response to increasing CO2. However, I argue here that negative feedbacks due to life (‘Gaia’) may have stabilized the planet’s climate — on geological timescales and in recent decades. The biology of any such stabilization is far from settled, with a mechanistic understanding delayed by evolutionary debate. I conclude that even with such advanced biology we have little power to predict global climate changes.
There is a basic flaw in the basic physics argument of climate change: biology. Indeed, just one word should be enough to cast doubt on all models of the atmosphere: “oxygen”. No educated person is unaware of one aspect of Earth’s basic biology: most atmospheric oxygen results from living organisms. Physics and chemistry therefore cannot explain atmospheric composition or properties. Basic chemistry would leave the planet a rusty ball (like Mars or Venus). So, as James Lovelock articulated in his Gaia hypothesis in the 1970s, the properties of our atmosphere result from the tight coupling of living and non living components (biota and abiota). Earth’s obvious and massive departure from chemical equilibrium is unique in the solar system. So, if it’s easy to understand that life is central to atmospheric chemistry, why have many people found it much harder to understand life could be pivotal in atmospheric energy and climate? And if life is so intimately involved, predictive models would need to include it — which I’ll argue they can’t because the biology is too complex.
An initial response, I anticipate, will be that oxygen is not a climatically-active gas, because it is not radiatively active. However, that does not weaken the argument that life changes Earth far from the state which non-biological “basic” science would predict — an example of the planetary power of life. Moreover, few realise that oxygen could have major implications for the long-term temperature trajectory of the planet, if it is helping to keep Earth wet. This controversial idea was discussed in meetings on Gaia in Oxford in the 1990s, postulating that in the absence of life and oxygen, the splitting of water by sunlight would eventually lead to desiccation of the planet (as hydrogen bled away into space). Photo-dissociation might be offset by the presence of atmospheric oxygen, scavenging hydrogen and restoring water. If so, the dominant climatically-active gas in the atmosphere — water — also owes its abundance to life.
Whether the planet is wet due to life requires further study and discussion. Fortunately my argument — that life is largely missing from the models — does not depend on this. What is more important is that people who believe basic physics is sufficient to predict climate should consider cloud condensation.
It is very widely accepted that clouds are hard to model, yet central to understanding climate sensitivity to CO2. It is not even known if the overall cloud feedback effect in a warming world is positive or negative. Indeed, the IPCC (2013) state: “Clouds and aerosols continue to contribute the largest uncertainty to estimates and interpretations of the Earth’s changing energy budget….some aspects of the overall cloud response vary substantially among models…”.
The basic physics of absorption and emission of infrared radiation have been combined with complex and uncertain physics to estimate that doubling of CO2 would warm the Earth by about one degree Celsius. Feedbacks involving water vapour and clouds are required to invoke larger climate changes from a doubling of CO2. Unsurprisingly, cloud feedbacks estimated from models vary substantially. Cloud-related feedbacks could be net positive (because condensed water emits infrared radiation). Cloud-related feedbacks could be net negative (because clouds reflect sunlight back into space). Further, cloud processes and convection induce and modify complex atmospheric motions, from very small scales to planetary scales. The uncertainty of cloud behaviour might eventually be tractable with complex physical models for a lifeless planet (which somehow retained water), but I think that the uncertainty is amplified to unmanageable levels on our biologically-active Earth.
It was James Lovelock who identified a potentially huge impact of life on the climate. No wonder, then, that he now argues that “anybody who tries to predict more than five to ten years is a bit of an idiot, because so many things can change unexpectedly”. Consider this: some unknown fraction of the cloud of this planet, of unknown type and altitude and climate activity, is produced for unknown reasons by unknown numbers of living species with unknowable population dynamics. If there are any modellers who think this is tractable, I hope they will indicate how in the Comments below.
How, how much, and why is life involved in cloud formation? Nobody knows. I’ll outline a few of these unsettled elements of the science of climate change.
The question “how” is life involved is the simplest: some species release chemicals that become cloud condensation nuclei (CCN), without which water remains a vapour. Some species secrete a gas, DMS (dimethyl sulphide), which seeds some clouds. Some plants secrete gases with similar properties, including Volatile Organic Compounds such as isoprene and pinene. Clouds are often observed rising over rainforest trees and other forests. It has been known for hundreds of years that some forests create rainfall (and I hypothesize that life in lakes similarly creates some of the clouds associated with them).
Unfortunately, “how much” cloud is created by life is unknown, a problem worsened by paucity of data on how much of each type of cloud cover there is and was (particularly before satellite observations). Some argue that life creates a substantive fraction of the global cloud cover, others less – and the fraction will vary through time.
“Why” does life create clouds remains unknown, but two fascinating evolutionary reasons have been proposed. Hamilton and Lenton (1998) suggested that “microbes fly with their clouds”. This is a proposal I expect many scientists will too-readily dismiss — even if they understand the track record of Hamilton as the biologist central to modern evolutionary theory (through his initially controversial ideas). However, the ‘selfish’ reason microbes of oceans, forests (and lakes?) secrete a cloud-forming gas (at metabolic cost) could be to generate latent heat of condensation, thence uplift of air — and thus dispersal of life to sites with more opportunities. And a plausible reason for plants to generate clouds is that they use rainfall. Predictions that clouds should increase when plankton become stressed (such as by nutrient deficiency or irradiance) will require long-term and large-scale observation.
I guess climate modellers will counter that they have performed sensitivity analyses, and that life and its interations with clouds, are not needed to predict the future climate accurately enough, or have small effects. Such arguments might have convinced me whilst models appeared to fit the unadjusted observations. However, several inexplicable (but biologically evident) warmer periods in the Holocene and Eemian damage climate model credibility. It’s not possible to do sensitivity analysis for an element of a system if there is no reliable benchline against which to measure the effects of manipulations.
Biology is very poorly represented in all of ‘climate science’, be it the mechanisms, ecological effects or policy response. Tellingly, the IPCC Assessment Report (2013) calls its first volume ‘The Physical Science Basis’. As one of the few scientists publishing on the evolutionary mechanisms of ‘Gaia,’ I know that very little attention has been paid to this topic. Perhaps if Bill Hamilton were still alive and researching the stability of the Earth system, things would be different. Because Lovelock’s original version of Gaia has an evolutionary flaw, I redefined Gaia as “planetary stability due to life”, and worked with Hamilton and Peter Henderson to seek mechanisms compatible with evolutionary biology. (Amongst the reasons few biologists have taken an interest in Gaia are that the original theory and models, such as ‘Daisyworld’, had an evolutionary bias, required ‘group-selection’, or implied natural selection amongst communities or planets). Instead, Hamilton, Henderson and I looked for negative feedbacks though two biological processes: i) ecology (density-dependent population growth); and 2) evolution (frequency-dependent selection – a mechanism also postulated by Richard Dawkins in The Extended Phenotype in 1982). The frequency of cloud-producing living organisms (abundance or biomass) is likely to be responsive to CO2, generating positive and/or negative biological feedbacks (Canney & Hambler, 2002, Biological Feedback, in: The Encyclopedia of Global Change).
At the risk of adding yet another failure to the litany of failed climate predictions, I predict climate models will struggle to include biology. No amount of physics, basic or complex, will overcome this deficiency. It is not possible to model population changes of even one species of organism several generations into the future. The unpredictability of complex systems is well known in ecosystems – as Robert May and colleagues demonstrated in the 1970s for multispecies fisheries. Populations of species that influence each other’s survival, reproduction or dispersal in ways related to abundance are likely often to demonstrate ‘deterministic chaos’, in which simple equations including time lags often generate superficially chaotic population changes. Even two species coupled through the Lotka-Volterra differential equations may show such behaviour. Imagine the problems, then, of modelling millions, billions or even trillions of microbial ‘species’ on Earth – when not even the number of species is known, let alone each of their requirements and climatic influences. Whether multi-species systems have more predictable emergent stability remains to be seen; this would make incorporation of ecology into climate models easier. Such stability is being investigated by Peter Henderson in the ‘Dam World’ model of Gaia he created with Bill Hamilton (Canney & Hambler, 2013, Conservation).
Modelling changes in plankton becomes even more implausible when one considers the responses to changing CO2: ‘ocean acidification’ might boost plankton through improved bicarbonate availability, and thence even cool the planet through DMS induced clouds. Or it might impact plankton through metabolic costs, thereby reducing calcification and a carbon sink and creating a positive feedback. The population and metabolic consequences of interactions (including those between warming water, CO2 outgassing, pH changes, thermoclines, nutrient and carbon dioxide availability for photosynthesis) are not known for any planktonic species, let alone entire hyper-complex marine ecosystems. Even if population changes could be predicted, we could not predict their cloud production behaviour — or the overall effect on albedo or convection.
It should come as no surprise to scientists and the public that wildlife has climate impacts – yet few realise how large these can be. When and if people accept that life can greatly change the chemistry of the atmosphere, they may be ready for another logical step. In this paradigm, temperature drives life drives CO2 levels. As Murry Salby (2012) deduced (Physics of the Atmosphere and Climate), CO2 lags temperature on a wide range of timescales (including glacial to interglacial oscillations, the last few hundred years, decades, and within a year). About 5% of the CO2 emitted to the atmosphere each year is from human activities, leaving ample scope for minor changes (perhaps in solar activity) to change the major biological sinks and sources of this gas and overwhelm human influences on radiative forcing. Perhaps the paradigm shift required to understand causality in climate is comparable to discovering the ancient nature of fossils, or plate tectonics, or neo-Darwinism, or the inhibitory models of plant succession. I’ve witnessed and taught through some of these shifts, so know how hard they are.
The ecology and evolution of negative feedbacks and Gaia might provide a framework to reconcile climate data and theory – but with very different theory to the basic physics of the climate. Instead, climate becomes — as many others have noted — a perhaps intractable and wicked problem. Prediction and attribution of useful climate detail may be beyond any science. If ‘the pause’ continues, or the world now cools or warms, we may never know why. It might be that negative biological and other feedbacks prevented runaway warming in the past, and have already begun to act. Or solar activity might be driving the carbon cycle, stifling CO2 increase. Or both. If extinction rates continue to rise such feedback may collapse — a perverse outcome of climate policy that destroys habitat. We hear a lot about high risk justifying high expenditure on reducing CO2 emissions, despite low probability of such risk. If we applied those expenditures to protecting the biological component of climate, we would conserve the climatically-active ecosystems — not, perversely, destroy them though renewable energy impacts and opportunity costs.
I anticipate many of the suggestions above will raise calls for publication in journals. Perhaps that’s the way physics works. Yet many key biological advances have been published in books or informal articles. Some of Hamilton’s ideas were published only in less formal articles and in a film on clouds (which very few people have watched). Moreover, conventional peer review demonstrably does not work well in some areas of climate science.
I thank Judith Curry for yet another brave move in hosting this entry. I hope policy makers will focus on no-regrets actions (such as protecting forests and marine life) which are relatively cheap and would work even if I’m wrong.
Link to essay published in the Bulletin of the British Ecological Society: ‘Thank you for Gaia’, by Clive Hambler [hambler-bes-gaia-paper]
Biosketch. Clive has been an Oxford College Lecturer in biology at Merton, St Anne’s, Pembroke and Oriel. He joined Hertford in 1998 and is the college’s director of studies for Human Sciences. He works in Oxford’s faculties of Zoology, Geography and Anthropology. He is coauthor of the acclaimed book Conservation, published by Cambridge University Press (see reviews).
Moderation note: As with all guest posts, please keep your posts civil and relevant.
JC note: You may recall that CE has published previous posts related to this topic, from the team of Makarieva, Gorshkov, Sheil:
- Forest climate and condensation
- Condensation-driven winds
- Water vapor mischief Part II
- Water vapor mischief
Image from Pixabay