Water-Reservoirs
and greenhouse emissions

Our reservoirs are plumbing the depths of insanity Once reservoirs and dams were the answer to pollution-free power. Now evidence shows greenhouse gases bubbling up from them at an alarming ratee

Fetid, choked with weeds and swarming with mosquitos, the Balbina reservoir in the Amazon rainforest is a billion-dollar boondoggle. The dam soars 50 metres above the trees. But much of the reservoir behind it, which floods an area the size of Warwickshire, is less than four metres deep. "From the air, you can see brown trees beneath the water across huge areas," says Philip Fearnside, from Brazil's National Institute for Research in Amazonia. He has counted 1500 islands and "so many bays and inlets it looks rather life a cross-section of a human lung". Even the introduction of a herd of grazing manatees has failed to staunch the spread of weeds across the surface. Water stagnates in the reservoir's backwaters for years before reaching the dam's hydroelectric turbines, which have a piffling generating capacity of 112 megawatts. That means the reservoir needs to flood the equivalent of two football pitches to power a one-kilowatt air conditioner in the Amazonian capital, Manaus. The depth of the insanity of this hydroelectric dam has only recently emerged. Built 13 years ago to provide "green", pollution-free electricity, it in fact produces eight times more greenhouse gas than a typical coal-fired power station with a similar generating capacity. The rotting vegetation has generated millions of tonnes of two greenhouse gases. These are carbon dioxide and methane - a gas that, molecule for molecule, is 20 times more potent at warming the planet than carbon dioxide.

Balbina is not alone. Recent research by Marco Aurelio dos Santos, at Cidade University in Rio de Janeiro, suggests that that up to half of Brazil's hydroelectric reservoirs have a global-warming capability similar to that of a fossil-fuel power plant. And just across the border in French Guiana, the Petit-Saut dam, which powers Europe's Ariane rocket launch site, produces three times as much gas as a coal-burning equivalent.

In June, the World Commission on Dams warned that the problem extended beyond rainforest reservoirs. It told UN climate-change negotiators that greenhouse gases bubble up from "all 30 reservoirs for which measurements have been made." The message was clear: "There is no justification for claiming that hydroelectricity does not contribute significantly to global warming." The Commission is a blue-chip assembly of scientists, engineers and environmentalists, and is supported by the World Bank, the world's biggest funder of large dams. Its findings have been corroborated by researchers from Canada, home of some of the world's largest hydroelectric projects. Vincent St Louis, of the University of Alberta, has made the first ever calculation of the total contribution of the thousands of reservoirs round the world to global warming. In the September issue of the journal BioScience, he says that they produce a fifth of all the man-made methane in the atmosphere. Add in their emissions of carbon dioxide, and they make up 7 per cent of the man-made greenhouse effect. That is a bigger impact than, for instance, aircraft emissions. While probably only a handful of reservoirs - the likes of Balbina and Petit-Saut - are turning out to be dirtier than coal-fired power stations, almost all make a significant contribution. "Whatever dam builders may say, reservoirs are not greenhouse-gas neutral," says St Louis. He says all governments should start measuring the emissions from their reservoirs. And scientists policing the Kyoto Protocol to halt climate change should insist that the findings are included in national emissions inventories. As of now, that does not happen. Nobody thought it was necessary. Only last year, the UN's chief scientific advisers for the protocol, the Intergovernmental Panel on Climate Change, dismissed reservoir emissions as "a minor source of methane compared to other energy sector or agricultural activities." St Louis's paper should put an end to that notion. His startling calculations rely on three breakthroughs. First, he has recalculated the area of the planet covered by reservoirs. The previous estimate of 500,000 square kilometres included only large hydroelectric reservoirs. But he has estimated the area covered by reservoirs of all sizes and all uses - collecting water for cities and farm irrigation, improving river navigation and preventing flooding. The figure comes to 1.5m square kilometres - six times the land area of Britain. Secondly, he has cross-checked the two standard ways of measuring emissions from bodies of water. One is based on physical measurements, capturing gases in a box above the water. But because this is difficult and time-consuming, researchers have often plumped instead for theoretical calculations of how fast the gases will diffuse into the atmosphere. But according to St Louis, the theoretical calculations are often far too low. This is because they ignore bubbles, which form wherever the fetid reservoir water cannot dissolve methane as it is generated. Bubbles reach the atmosphere independently of diffusion processes. Thirdly, St Louis has looked at how long reservoirs carry on producing greenhouse gases. When the issue first emerged in the early 1990s, scientists believed that the gases came mostly from vegetation trapped underwater when the reservoir filled. They reasoned that the rotting vegetation would soon be gone and emissions would cease. Reservoirs would become like natural lakes, in a rough chemical equilibrium with the atmosphere above. Not so, it turns out. As reservoirs round the world age, it appears that most continue to produce significant greenhouse gases throughout their working life. As Fearnside says, it can take 500 years for a tree to rot in a stagnant Amazon reservoir. And in any case, probably most rotting vegetation in a reservoir is from elsewhere, floating down rivers that drain into the reservoir. Most of this vegetation would have rotted anyway, of course. But, without reservoirs, the decomposition would occur mostly in the atmosphere or in well-oxygenated rivers or lakes. The presence of oxygen would ensure the carbon in the plants formed carbon dioxide. But many reservoirs, particularly in the tropics, contain little oxygen. Under those anaerobic conditions, rotting vegetation generates methane instead. Both gases have a greenhouse effect. But a molecule of methane is 20 times as potent as a molecule of carbon dioxide. Reservoirs thus magnify the greenhouse effect of the rotting of a significant amount of the Earth's vegetation. St Louis found that emissions from tropical reservoirs contain the most methane. "We suspect this is because warmer temperatures in sediments promote faster, more anaerobic decomposition," says co-author Carol Kelly of the Canadian government's Department of Fisheries and Oceans. While only a third of the world's reservoirs are in the tropics, they appear to produce around 80 per cent of greenhouse gas emissions. But in the long run, reservoirs in northern countries could be as dangerous for the world's climate - especially those that flood peat bogs. A thick peat bog, says Kelly, contains "large stores of decomposable organic carbon" - far more than a rainforest. Over the centuries, most of it will rot in a reservoir and bubble into the air. Scientists working for dam-builders say that reservoirs often flood marshes and other land that is already a source of greenhouse-gas releases to the air. So their extra emissions are, at least in part, illusory. But equally, some flood land was absorbing greenhouse gases from the atmosphere, says St Louis. Many healthy peat bogs and forests fall into this category. Some studies suggest that the surviving stands of the Amazon rainforest are major absorbers of carbon dioxide from the atmosphere. The inclusion of reservoir gases could transform national inventories of greenhouse gases. French Guiana used to be one of the world most greenhouse-friendly nations. But according to calculations by Robert Delmas, of the Laboratory of Aerology Observation in Toulouse, the Petit-Saut reservoir has turned it into one of the worst polluters. Per head of population, French Guiana's emissions are three times those of France and greater than those of the US. Another secret emitter is Ghana. It has flooded a twentieth of its land area to create the giant Akosombo hydroelectric dam. If St Louis's average figure for emissions from tropical reservoirs holds for Akosombo, the reservoir will emit five times as much greenhouse gas as all the country's fossil-fuel burning. Industrialised countries that rely on reservoirs for much of their electricity -- such the Canada and Norway -- could also be embarrassed. Including reservoir gases could make it harder for them to meet targets for cutting emissions agreed in the Kyoto Protocol three years ago, and due to be finalised in The Hague in November. But the people with most to fear are the builders of dams. They have endured constant attack from environmentalists for damaging wildlife habitats and flooding people from their homes. They had hoped the drive to cut greenhouse-gas emissions would revive their industry. They have been lobbying hard this year to have hydroelectricity included in the Kyoto Protocol's list of greenhouse-friendly technologies that should get international support. That drive is now in jeopardy.