Geoengineering: Can 'fertilising' the ocean combat climate change?

Can 'fertilising' the ocean combat climate change?
12 September 2007
NewScientist.com news service
Emma Young
RUSS GEORGE calls it a "voyage of recovery". His opponents call it blatant pollution. Only time will tell who is right.

In May this year, 350 miles north-west of the Galapagos Islands, George's company, Planktos, based in Foster City, California, began the first of six large-scale trials to release more than 50 tonnes of finely ground haematite into the ocean. The company aims to show that fertilisation with iron can safely boost levels of phytoplankton - single-celled photosynthetic organisms responsible for half of the carbon fixation on Earth. More of such plankton, Planktos reasons, means the ability to trap more carbon dioxide from the atmosphere, which will help combat global warming.

Not everyone is convinced. The Charles Darwin Foundation on the Galapagos Islands calls the project "an unwelcome visitor" and says it is "alarmed... because of the unknown effects it could have on marine life". So is this, and other projects like it, a real environmental fix or an eco-disaster in waiting?

Iron seeding is based on the well-accepted idea that plankton growth in the equatorial Pacific, the Southern Ocean and the north Pacific is restricted by low levels of iron. The concept was first proposed in 1990 by John Martin, then director of the Moss Landing Marine Laboratories in California. Since then, 12 international experiments in these waters have shown that adding iron can cause plankton to bloom, increasing the amount of CO2 drawn into the surface of the ocean. By contrast, in sub-tropical ocean regions such as the waters off Australia, nitrogen, rather than iron, is the main brake on plankton growth. Researchers there are experimenting with seeding the ocean with nitrogenous fertiliser.

Now, though, private companies are getting in on the act. They are keen to talk up the benefits. George maintains that iron seeding should be seen as "remediating" the oceans, restoring what some say are falling plankton levels. Ian Jones, head of the Ocean Technology Group at the University of Sydney, Australia, and director of the Ocean Nourishment Corporation (ONC), intends to release 1000 tonnes of urea off the coast of the Philippines later this year. He says that more plankton will ultimately mean more fish, and fewer hungry people.

Critics argue that talk of ocean remediation or boosting fish stocks is simply window dressing and that these groups are racing not to save the Earth, but to carve out a slice of the booming market in carbon-credit trading. "These are very much business projects, not research projects," says Philip Boyd at New Zealand's National Institute of Water and Atmospheric Research in Dunedin. "Planktos, for example, views the ocean as a simple, predictable system that can be readily manipulated. The bottom line of the manipulation is that it's all about carbon offsets and carbon credits."

There are certainly large amounts of money at stake. In February, British entrepreneur Richard Branson launched the $25 million Earth Challenge prize. The award will go to the best scheme for removing at least one billion tones of carbon dioxide from the atmosphere every year, for a decade. A bigger lure, however, is the carbon-offsets market, which allows companies to invest in carbon-reduction schemes to mitigate their own greenhouse-gas emissions. According to World Bank figures released in May, this market virtually doubled in 2006, to $5 billion. Ocean-fertilisation projects are particularly attractive as they could be cheaper than alternative methods, such as renewable energy sources or carbon capture and storage. Urea fertilisation, for example, would cost $10 to $15 per tonne of CO2 sequestered, Jones estimates, whereas George reckons iron seeding could be done for as little as $4 per tonne. By comparison, carbon capture and storage from coal-fired power stations could cost $50 per tonne of CO2.

That assumes, of course, that ocean seeding will work - something that has yet to be shown, according to Boyd. In a review of iron-seeding experiments in February (Science, vol 315, p 612), Boyd and his colleagues found that the amount of carbon drawn into the ocean's surface layer varied widely. One study, which involved adding 1.1 tonnes of iron, found no increase in carbon fixing, but it was conducted in the autumn - possibly too late for plankton to bloom. Another found that 350 kilograms of iron boosted plankton levels sufficiently to fix an extra 1250 tonnes of CO2 - 250 times the average British citizen's annual emissions.

It sounds good, but it's not enough. To lock carbon away for the long term, the plankton has to die of natural causes and sink to the deep ocean, where the carbon may be trapped for hundreds or thousands of years. None of the 12 iron-seeding experiments in Boyd's review showed that adding iron increased the amount of plankton reaching the deep ocean. Part of the problem is that it's very difficult to measure sinking carbon. Recent studies, though, have made some progress.

In April, a team led by Ken Buesseler of the Woods Hole Oceanographic Institution in Massachussetts published a study that used automated plankton traps to monitor plankton movement in the water column. The team found that in the north-west Pacific, half of surface plankton managed to sink through the "twilight zone" - the layer between the sunlit surface water and the deep ocean. Near Hawaii, on the other hand, 80 per cent of the plankton was gobbled up by zooplankton or other predators during its downward journey. This means its carbon was recycled, a proportion being excreted and respired by the predator, rather than sequestered. What's more, getting through the twilight zone only means that plankton have made it down to about 500 metres. Boyd says that studies of natural plankton blooms suggest that only a fraction of the carbon that makes it this far falls down into the deep ocean. In fact, field research reveals that just a few per cent of each bloom becomes deeply sequestered, he says.

Then there is the question of how long the carbon will stay there if and when it has sunk. Modelling studies by Jorge Sarmiento at Princeton University and colleagues have addressed this issue. They suggest that one century after a month of continuous iron fertilisation of a given area of Pacific waters, the reduction of atmospheric CO2 would be between 2 and 44 per cent of the tiny amount of CO2 that made it to the deep ocean. The rest would be recycled by predators or microbes and potentially re-released to the atmosphere rather than being sequestered. It is practically impossible to confirm the actual figure, and indirect verification would require long-term monitoring of the ocean depths - something that no company is currently proposing.

Undeterred, Planktos commenced the first of six large-scale iron-seeding trials in May, 550 kilometres from the Galapagos archipelago. George estimates that each test site will sequester between 3 million and 5 million tonnes of CO2 per bloom. If all goes to plan, the company will apply for certification with various emissions-reduction programmes, such as the European Union's Emission Trading Scheme or Australia's planned carbon-trading scheme.

In each trial, Planktos will release between 50 and 70 tonnes of haematite over an area 100 kilometres squared and monitor the results for three to four months to assess the amount of carbon sequestered. The team will keep track of the health of the ocean by recording changes in pH, macronutrients, the concentrations of different species in any plankton blooms and any changes to predator populations. They will also measure the precipitation of particles in the water, and the carbon levels below 500 metres, George says, and take water samples at various depths down to 1000 metres. "This will ensure an accurate measure of the quantities of carbon reaching the deep ocean," he says.

Boyd is far from convinced. Demonstrating that iron seeding has increased the amount of plankton reaching the depths requires measurements of sinking particles, not just stationary particles, he says. It also requires the ability to show that those sinking particles came from an area of water that had been fertilised, rather than drifting in from a neighbouring patch of ocean. The technical challenge is immense. "Even with experiments where in some cases we've had multiple research ships with aircraft and helicopters and up to 50 scientists involved," says Boyd, "we have still not been able to show definitively that there had been a carbon increase to a depth of 300 metres, never mind carbon sequestration into the deep ocean."

Boyd and others also take issue with Planktos's claim that it will be "restoring" phytoplankton levels in the ocean to what they once were. George cites a NASA study based on satellite data from the early 1980s and late 1990s that concluded plankton levels declined by 6 per cent over this period. Yet questions have been raised about the quality of the early satellite data. "There is not an established belief that productivity levels are declining," says Dave Siegel of the University of California, Santa Barbara, who uses satellite data to study the ocean.

Despite these concerns, the ONC in Australia is moving ahead with its plans to use urea as an ocean fertiliser. Jones envisages factories making 2000 tonnes of urea per day from natural gas. This would then be dissolved in seawater and pumped through a pipe laid on the seabed at the edge of the continental shelf. Vertical risers attached to the end of the pipe would then release the urea at a depth of 50 metres, from where it would diffuse into the sunlit layer. Each factory could maintain an area of about 20 square kilometres of plankton, at densities of about 200 micrograms per litre, says Jones, which is much less than the density produced in a toxic plankton bloom caused by pollution or nutrient run-off from land.

Taking into account the CO2 created in the production of the urea, Jones estimates that 1 tonne of nitrogen could sequester 12 tonnes of CO2 - so the output of each plant could sequester 8 million tonnes of CO2 each year, at a cost of US$10 to US$15 per tonne. Jones hopes that the company could be taking part in carbon-trading schemes by 2008.

The ONC team is currently working on a number of small test sites, attempting to demonstrate that adding a macronutrient like urea, in some cases in combination with phosphate, really can boost plankton levels. Later this year it plans to conduct its first large-scale field trial, releasing 500 tonnes of dissolved urea off the coast of the Philippines. This will be followed by a trial involving 1000 tonnes of dissolved urea off Malaysia.

Counting carbon

Jones believes urea fertilisation has advantages over iron, in that while some of the added iron probably sinks before it can be used by the plankton, the tight chemical relationship between carbon and nitrogen means that, in theory, every added atom of nitrogen in the ocean will trap and hold seven atoms of carbon, even if deep ocean waters eventually return those bound molecules to the surface.

This, says Jones, helps get around one key criticism of other schemes in which trapped carbon may eventually be released - such as when trees in carbon-sink forests die. Nitrogen added to the oceans will always be available to lock away carbon, he argues.

Jones reckons that satellite images providing an indication of plankton volume would be all they need to work out how much carbon has been sequestered. "We argue that if you create organic carbon, all of that organic carbon is eventually exported to the deep ocean." He says some will be remineralised in surface waters, some will be exchanged back into the atmosphere and some will go up the food chain into fish and be respired. In this way, within a decade or so, Jones says, the vast majority of the carbon will be sequestered.

Boyd, for one, is sceptical. "So where are the peer-reviewed papers showing this? If people are going to have confidence in schemes like this, they have to demonstrate their claims."

For Sallie Chisholm, principal investigator in biological oceanography at MIT, urea fertilisation is a scarier idea than iron seeding. The ocean regions deficient in nitrogen are classed as "desert" regions. "But they are not barren, they are teeming with life - life that is exquisitely adapted to these low-nutrient situations," she says. "Thousands of species depend on this ecosystem. When you fertilise it, you disrupt all that, just as you do when fertiliser runs off the land into streams and causes 'dead zones' in coastal water."

Jones says the urea concentrations will be too low to create dead zones. While he admits that ecosystems will change, he insists this is a price worth paying to boost global fish stocks. For every tonne of reactive nitrogen added to the ocean in the form of urea, he estimates that 5.7 tonnes of phytoplankton will be produced, ultimately leading to roughly an extra tonne of local fish. "We transform the land to provide food for people. This is just like practising agriculture in the sea," he says. Chisholm disagrees. Agriculture on land happens in discrete regions, which are easy to control and monitor. The ocean, she warns, is an entirely different matter.

There are also concerns that fertilising the Southern Ocean could change global patterns of plankton growth, robbing ecosystems elsewhere of nitrogen and phosphorus. Changes in nitrogen and phosphorus levels can be monitored around a fertilisation site, but it is impossible to predict what knock-on effects there might be by extrapolating from studies of small patches, says Chisholm. Contrary to Planktos's claims, she says, the ocean system is just too complicated.

The most alarming possibility of all, perhaps, is that fertilisation might actually produce greenhouse gases. The process of breaking down dead plankton requires oxygen, which must come from surrounding water. If oxygen levels dip too low, the microbial community could shift towards those creatures that produce greenhouse gases, such as nitrous oxide and methane.

Despite this, some independent scientists think the potential risks have been overplayed. Ken Johnson of the Monterey Bay Aquarium Research Institute in Moss Landing, California, is one. He reckons that large-scale fertilisation would produce one of several scenarios: "One: the ocean will turn greener, atmospheric CO2 will decrease and not much bad will happen, or two: the ocean won't turn green and CO2 won't decrease but nothing much bad will happen - other than companies losing money."

Until some of the nagging questions have been answered, many researchers believe that commercial ocean fertilisation should be discouraged. "Ocean fertilisation is predicated on there being a policy need to reduce greenhouses gases in any way that we can," says Siegel. "I'm not sure we're at that point. And there are many other ways we could do this besides changing the ocean without much knowledge of the consequences."

Related Articles

Alarm over Galapagos carbon offset plan
http://environment.newscientist.com/article/mg19426103.900
30 June 2007
Company plans 'eco' iron dump off Galapagos
http://environment.newscientist.com/article/dn12111
25 June 2007
First claim for CO
http://environment.newscientist.com/article/mg19325914.300
17 February 2007
Letter: Save us!
http://environment.newscientist.com/article/mg19626240.400
06 October 2007
For the record
http://environment.newscientist.com/article/mg19626241.000
06 October 2007
Weblinks

Planktos
http://www.planktos.com/
Ocean Nourishment Corporation
http://www.oceannourishment.com/
Ian Jones, University of Sydney
http://www.civil.usyd.edu.au/people/jones.shtml
Sallie Chisholm's research lab
http://web.mit.edu/chisholm/www/
From issue 2621 of New Scientist magazine, 12 September 2007, page 42-45