With the revelation that hydro reservoirs can produce greenhouse gases, more research is now needed to determine whether hydro is still among the best developmental choices a country can make in order to produce power and reduce global warming. Gemma Newman reports

Clean energy and hydro power have a long standing relationship. As an alternative to burning oil and coal, hydro avoids the emission of pollutants like sulphur dioxide and nitrous oxide which cause acid rain. And, it has always been assumed that all hydro would reduce net emissions of greenhouse gases (GHGs) — until now.

According to Hydro-Quebec, traditional environmental impact assessments only look at the direct impacts of power plants on the environment. With this approach, hydro power plants produce zero emissions. But recent assessments are based on life-cycle analyses which include all significant activities such as building and maintaining a power plant and transporting and extracting fuels. With this approach, two indirect sources of GHG emissions from hydro power must now be considered:

•Emissions during construction.

•Net emissions from reservoirs — after impoundment a portion of the biomass in a reservoir will decompose and generate GHG emissions.

Thematic review.

In February 2000, the World Commission on Dams (WCD) in co-operation with Hydro-Quebec, held a workshop in Montreal, Canada as part of its thematic review process which addresses issues related to dams and global climate change. With guidance from 19 climate scientists from Brazil, Canada, Finland, France and the US, a draft report was presented to WCD, which stated that GHGs (mainly carbon dioxide and a small amount of methane) are emitted for decades from all dam reservoirs in the boreal and tropical regions, as evidenced by measurements made through studies conducted on about 30 reservoirs. This conclusion is in contrast to the widespread assumption that such emissions are zero.

The draft also concluded that GHG emissions cannot be directly explained by the volume of submerged biomass or its carbon content, and that a range of factors directly influence GHG emissions from reservoirs.

With these considerations in mind, WCD met in Germany in June to present the UN Framework Convention on Climate Change (UNFCCC) with the latest scientific evidence regarding GHG emissions from hydro reservoirs, comparisons with emissions from other energy sources and assessment methods for future dams. (UNFCCC was formed in 1992 after UN countries agreed to tackle the problem of global warming and climate change.)

Giving advice

Based on a draft report on GHG emissions from reservoirs; submissions from interested parties; scientific review by NGOs, scientists and industry; the Montreal workshop of 19 scientists; and a case study on the Tucurui hydro project in Brazil; the following findings were presented:

•All 30 reservoirs studied so far emit GHGs.

•Natural lakes studied emit GHGs.

•Natural habitats may be either sources or sinks of carbon.

•A reservoir causes a net change in GHGs from pre- to post-impoundment and it is this net change that should be assessed for its contribution to global warming.

•Flooded biomass alone does not account for observed GHG emissions.

•Emissions are variable over time.

•Carbon is flowing into the reservoir from the catchment (and perhaps from the atmosphere). This implies that less carbon is released to the downstream river than under natural conditions.

•Long term studies are essential to look at full life-cycle emissions over reservoir lifetimes.

Based on these findings, the following advice was given:

•Hydro power cannot automatically be assumed to be a cleaner technology than thermal alternatives with respect to GHG emissions. Research is needed on a case by case basis to make this claim.

•In boreal climates like Canada and Scandinavia, available studies suggest so far that emissions from hydro power reservoirs are very low.

•In Brazil, of the ten dams studied, emissions varied from dam to dam with a 500-fold difference between the lowest and highest. The lowest emissions are similar to Canadian lakes and reservoirs, while the highest gross emissions may be in the same range as the most efficient thermal energy plants (gas fired combined cycle), but still about three times as good as coal fired generation.

•More research is needed on a wider range and diversity of reservoirs. This includes more research into turbine emissions: GHGs from the water going through the turbine may be missed by researchers if they only study emissions from the water surface of the reservoir. More measurements

are also needed on a wider range of natural environments in countries that are currently building dams, and the understanding of transient carbon in reservoirs and natural lakes also needs to be improved.

•The role that oceans play as repositories of carbon in sediments and how this affects dams, as well as the fate of carbon in an undammed catchment compared to a dammed catchment, requires research.

Jamie Skinner, a senior advisor to WCD who chaired the meeting in Germany, said: ‘In short, WCD found that while dams produce net GHG emissions, that fact alone cannot be removed from the context of place, scale, time or how the dam compares to the other options available to any given country.’ He

added that what should matter to governments and the international community concerned about climate change and GHGs is the critical net change that their decisions will bring, and whether the project selected to meet energy needs will in fact emit less, than other alternatives considered.

Wise words

Despite these wise words, the amount of GHGs emitted varies considerably for each reservoir in the world, depending on geography, altitude, latitude, size, depth, temperature, depth of turbine intakes, dam operations and construction procedures. Unless all of these factors can be assessed to determine the quantity of GHGs emitted by specific reservoirs over time, it will be difficult for WCD to achieve its goal: to accurately predict GHG emissions from reservoirs to allow governments to make the best possible decisions about dams compared to other development choices.

The scientific research suggests that authorities will need to establish the baseline level of emissions produced by a given river basin before a dam is introduced, and then compare the net change with emissions and impacts of other energy and water supply options. Governments will also need to evaluate climatic impacts against social, environmental and economic priorities. Unless governments devote time, resources and scientific attention to these variables, an accurate verdict

on the impact of dams on climate change will remain unattainable, and hydro’s critics will continue to misinterpret the little science currently available on this issue.

Based on the information contained in the Montreal report, the UK’s New Scientist published an article called ‘Raising a stink — rotting vegetation in hydro dams stokes global warming’.

It stated that ‘many hydroelectric power schemes release more greenhouse gases into the atmosphere than large coal fired power stations, because of the rotting vegetation they contain.’ Such a comment is highly misleading as the draft report is based on only 30 reservoirs in tropical and boreal regions, and only two projects (atypical projects with a large area of reservoir compared to the installed capacity) of those produced as much GHG as a gas fired combined cycle plant. These are still one-third of a coal plant’s emissions and indeed are not net emissions, but only direct emissions from the reservoir.

WCD makes it clear that there are a number of factors which influence GHG emissions, not just the amount of submerged vegetation they contain. Also, most of the reservoirs observed produced very low emissions.

WCD expressed its concern about getting the right information across in a letter it sent to the editor of the New Scientist. ‘The article gives the unduly simplistic impression that the observed GHG emissions from a few Brazilian dams therefore makes all dams bad,’ Skinner wrote. ‘This misleads those seeking to tackle global warming.

It is incorrect (and scientifically lazy) to conclude “a reservoir will produce more methane than the river did before the dam was built”. The Commission has worked with scientists, industry and non governmental organisations to discuss a common platform for future policy, informed by good science. The workshop report (which incidentally reflects the views of participants, not those of the Commission) emphasises the scientific complexities involved with increasing power production while reducing life cycle emission from dams or other energy options.’

WCD must have been aware of the controversy that would be caused by sharing the current knowledge on GHG emissions from reservoirs.

In the Commission’s defence, Skinner says:

‘This issue is no secret, all WCD’s information is available on the website.

It is an open debate. It is not our role to come

out in favour of any particular option — hydro or thermal, nuclear or renewables. GHG emissions from reservoirs exist and our task therefore is to ensure that the data and experience is made available to practitioners in a way that scientists involved can agree on. It is about informed decision making.’ WCD also recognises that further research is critical, particularly in temperate regions and on reservoirs which are not used for hydro power.

Based on the available science, it is believed that four pieces of information are critical to assess whether or not hydro power dams emit more or less GHGs than their thermal equivalents:

•A pre-dam assessment of GHG emissions/sinks from the natural habitat and basin.

•An assessment of GHGs released from water passing through the dam (turbines and spillway).

•The observed GHG emissions from the reservoir surface.

•Inflows of carbon from upstream in the river basin.


Under the Kyoto Protocol of the United Nations Convention on Climate Change, industrial countries have committed to reduce GHG emissions. One way of doing this is through the Clean Development Mechanism (CDM), where countries can reduce their emissions by purchasing emission credits from other countries that invest in projects and programmes which avoid GHGs and create a net global reduction in emissions. In other words, trading carbon emissions could lead to ‘dirty’ fuels subsidising ‘clean’ power.

Therefore, the evidence presented on GHG emissions from reservoirs will not only help countries to evaluate their options for development to reduce global warming, but it will also affect the future of hydro power which could benefit from these subsidies.

Despite these worries and his own findings and research into GHG emissions from hydro reservoirs,

Luc Gagnon, the senior advisor on climate change for Hydro-Quebec is optimistic. ‘Hydro power is still a powerful tool to reduce GHG emissions,’ he says. According to Gagnon, emissions measured from reservoirs systematically overestimate the level of emissions for which reservoirs are accountable, because the measurements cannot factor in the emissions that would have occurred even in the absence of a hydro plant. In some cases, gross emissions have overestimated true emissions by 35%.

In order to quantify net GHG emissions over a long period of time, Hydro-Quebec will be carrying out another five years of research which will involve studying the entire watershed of the La Grande complex in northern Quebec. Research into net emissions is still in the preliminary stages as the ecological processes of the whole watershed need to be understood before they can be determined with confidence.