Thanks to the activities of plankton, for example, around a third of the carbon-dioxide emissions of industrial man have been pumped down into the oceans, rather than building up in the atmosphere and causing global warming. If photosynthesis by plankton were suddenly to stop, levels of carbon dioxide in the atmosphere would soar, creating near-Venusian conditions on Earth.
The oceans contain some 40,000 billion tonnes (gigatonnes, or Gt) of carbon - as against 750 Gt in the entire atmosphere. While the global biomass of plankton is less than one per cent that of terrestrial life, those plankton are responsible for half of the world's carbon recycling. Any reliable model of global climate must be based on a thorough understanding of what the plankton are up to. But in spite of recent successes, this is still a long way off.
One problem is that plankton are ever on the move, potentially travelling 1,000 or more miles in a single month. We know about the 400 or so types that are regularly encountered, the springtime blooms of phytoplankton (planktonic plants) that can spread over hundreds of thousands of square kilometres, and long-term shifts in the abundance of species - including a general decline in plankton abundance in the north Atlantic from the 1950s onwards. But such information is far from complete. For example the prochlorophytes are probably the single most abundant group of plants on the planet, says Ian Joint of the Plymouth Marine Laboratory (PML), yet they were only discovered in 1984 because they are so tiny, less than a micron across, that they pass unseen through the plankton recorder's filters. They only came to light using the technique of flow cytometry, which causes their photosynthetic pigments to glow in droplets of water exposed to laser light. Another way of observing plankton is by satellite. On average the sea reflects three per cent of sunlight, but a dense phytoplankton bloom can darken the sea so that only one per cent is reflected.
Yet while plankton records have given us some idea of the biodiversity of plankton, they offer few clues as to how different plankton actually behave. For example, we may know that blooms of phytoplankton in the north Atlantic are often of diatoms, phaeocystis (which are responsible for malodorous sea-foams) or coccolithophores (so-called because they make chalky shells, or coccoliths). But what determines the sequence in which one species follows another? The classical theory is that succession depends mainly on environmental factors such as temperature, light and the availability of nutrients. Diatoms have a particular need for silicate, so when the silicate has run out, the diatoms die off and another group takes their place. Ultimately all the nutrients are depleted, and pekoplankton, which have very low nutrient demands, take over.
But there are also the interactions among species to consider. Classical theory alone is unable to explain why a large bloom of diatoms may come to an end literally overnight, only to be followed by another diatom bloom, but of a different species. According to Nick Mann of the University of Warwick, hitherto unsuspected viruses are responsible. He and colleagues in Norway have now isolated 15 viruses that infect synechococcus phytoplankton. "You might have several species in an area, and the population of one gets dense enough to be infected by a virus, then another strain will dominate, so you get a succession of populations," says Mann. "We are finding there is a vast world of biodiversity out there of which we have only the slightest glimpse."
It is becoming clear that nutrients are recycled far more often and actively than previously known. As viruses kill plankton, stored energy and nutrients are returned to the water, providing food for bacteria and other plankton. This diversity and complexity gives oceanic ecosystems resiliency, but also makes their likely response to change - of climate, for example - hard to predict.
Much interest has focused on the coccolithophores, among the most climactically significant groups - in particular, the species Emiliana huxleyii, the most numerous and ubiquitous of the coccolithophores, which can cover areas of 250,000 square kilometres in the north Atlantic spring. Like all other phytoplankton, coccolithophores are constantly fixing carbon dioxide out of sea water. On average barely one per cent of the carbon fixed will sink into deep ocean storage, as the rest is cycled through bacteria and zooplankton in the top few hundred metres of water. But coccolithophores have other important properties too.
In order to stop their internal fluids being sucked out into the salty sea-water by osmotic forces, they synthesise sulphur-based "osmo-solutes" which equalise their own osmotic pressure with that of sea-water. On death, these osmo-solutes are broken down by bacteria, giving off dimethyl sulphide (DMS) gas. The DMS then oxidises in air to make tiny droplets of sulphuric acid, which act as highly efficient cloud condensation nuclei. With a large bloom of E huxleyii giving off hundreds of tonnes of sulphur a day, coccolithophores are the major determinant of cloud formation above the oceans. And as low-altitude clouds reflect sunlight, they help keep the oceans cool.
Coccolithophores play another important role, as they build their chalky shells that whitewash the sea surface with calcium carbonate. A bloom drops roughly a tonne of chalk to the sea floor per square kilometre per day, taking hundreds of thousands of tonnes of mineralised carbon dioxide out of circulation.
So will global warming enhance coccolithophore activity, causing a cooling effect, or the reverse, exacerbating global warming? John Woods, Professor of Oceanography at Imperial College London, fears the worst. Stronger solar radiation, he says, will mean the surface layer of the ocean is warmer, and so less dense. As a result it will mix less with the colder, heavier waters below. So once plankton have depleted the nutrients in the warm surface water, less nutrient-rich cold water will rise to the surface, and growth will slow down.
The response of the ocean to subtle shifts in insolation was shown following the Mount Pinatubo eruption, says Woods. The removal of a mere 5 to 10 watts per square metre by high-altitude dust "was enough to provoke a massive increase in phytoplankton", by reducing surface temperatures, and so increasing ocean turbulence and nutrient availability. This "plankton multiplier", he argues, is also responsible for the rapidity of the shift from ice age to interglacial.
Patrick Holligan, Professor of the Southampton Oceanography Centre, is less certain. "The white cliffs of Dover were laid down by coccolithophores during the Cretaceous, and we know the sea was very warm at that time," he points out. "As an ecologist I might expect the warm oceans to be a more powerful source of DMS than cold oceans, as the cold ocean is full of diatoms [which make nitrogen-based osmo-solutes], and the warm ocean is full of secondary species that tend to be DMS producers." As for indications from Arctic ice cores that there was five times more atmospheric DMS during ice ages than there is now, this is inconclusive, says Holligan. "We know that ocean productivity has not changed that much, so we must look for other factors. For example, winds were much stronger so it may be that DMS precursors were transported that much more efficiently."
Whatever the outcome, it may be wise to look at ways in which the oceans might be encouraged to absorb more CO2 - especially in those areas of ocean with an abundance of the major nutrients such as phosphate, nitrate, sulphate and silicate, plenty of light, but little plankton growth. The surmise of Andrew Watson of the University of East Anglia and PML's Phil Nightingale is that iron might be the missing nutrient - a thesis put to the test in the equatorial Pacific near the Galapagos Islands, by the addition of dilute ferrous sulphate.
In the first IRONEX experiment two years ago, the iron-enriched water was unexpectedly submerged so no conclusive result emerged. But in a second experiment (Nature 10 October), the iron triggered a major diatom bloom and caused a significant drawdown of CO2, with a 50 to 60 per cent reduction in ocean-to-atmosphere flux of the gas, while also stimulating DMS production. In this instance little carbon was put into long term storage as the early use of the nutrients simply prevented plankton growth elsewhere. But large areas of the Southern Ocean have a similar nutrient status, and the cold surface waters sink down as they move north into warmer regions. If these waters also prove to be iron-deficient, says Watson, fertilisation with iron over the period 2000 to 2100 would cause a "significant" CO2 drawdown. But he is wary of rushing into action. "I can see that there may be a great deal of pressure to do this but we should not act without a great deal more research," he warns. "Greenhouse gases are also produced by biological activity, such as methane and nitrous oxide. Their production might also be increased and that could wipe out any advantage."
Such uncertainty is typical of oceanography - hardly surprising given all the difficulties of data gathering in the vast, ever-changing oceans. According to John Woods, this is not likely to change in a hurry. Consider a square metre of ocean, to a depth of one kilometre: it will typically contain one billion plankton, while the water is characterised by temperature, salinity, velocity, turbidity, degree of illumination, and concentration of nutrients. To describe the daily and seasonal variations of the water column for a year as it moves across the oceans, including the life cycles of the plankton, says Woods, would be a massive task requiring almost unimaginable amounts of data. So he has constructed a virtual-ocean ecosystem in his computer, which mimics the main phenomena of natural ecosystems. Using this approach, he has modelled carbon drawdown for 50 years ahead under conditions of increasing solar irradiation.
"What we get is a marked positive feedback effect," he says. "Carbon drawdown to the deep ocean will decrease by 10 per cent by 2045. All the models of the Inter-Governmental Panel on Climate Change (IPCC) assume that half of our CO2 emissions will be absorbed to land and sea. But as we continue to pollute, less CO2 will enter the ocean, more will stay in the atmosphere and this will speed up th process of global warming."
Of course a model such as this is only as good as the data it is based on, and its results must be tested against reality and adjusted accordingly. But Woods insists that ecological modelling is the way forward. Research cruises of the future, he predicts, will be aimed not at data collection but at the verification of phenomena encountered in virtual ecosystems. And with this comes the prospect of tackling the "grand challenges of biological oceanography" - such as fisheries recruitment, ozone depletion, and the influence of climate change. !Reuse content