Hydropower is a frequent target for criticism. Regardless of your views on global warming, turning a serene stretch of river into an artificial lake humming with electrical equipment can make you unpopular, and the announcement of any new hydropower project is often swiftly followed by outcries over habitat disruption and community displacement, among other concerns.
But hydropower's clean energy bona fides are rarely questioned.
In fact, hydropower reservoirs do generate carbon emissions, and some scientists think these emissions could be substantial -- maybe enough to cancel out the system's green benefits.
Steven Bouillon, a carbon cycles researcher at the University of Leuven in Belgium, said the magnitude of emissions depends on the design of the reservoir. Bouillon is leading research in Africa to quantify emissions from inland water systems, including reservoirs.
"We're quite convinced that for certain reservoirs, the effects on greenhouse gas emissions locally will offset the benefits of clean energy production," he said.
When reservoirs are built -- for hydropower or other purposes -- the grass, vegetation and trees submerged underwater begin to slowly decompose, releasing the carbon dioxide they had been storing through photosynthesis for centuries.
How much of this carbon dioxide actually gets "outgassed" into the atmosphere is a product of several factors. The gas can bubble to the surface of the reservoir and escape; it can be released as the water passes through a dam's hydroelectric turbines; and it can be released farther downstream. Some gas could also never make it out at all, buried in sediment in the reservoir or farther downstream, or carried all the way out to the ocean.
The size of the reservoir can make a difference. Shallow reservoirs with a wide surface area can emit more, because they've flooded more carbon-rich land, which can easily escape as gas out of the shallow water. Conversely, deep dams with a small surface area have much lower emissions.
The methane problem
Methane -- a much more potent greenhouse gas than CO2 -- may be a bigger concern, however.
Submerged vegetation emits some methane naturally, but stagnant reservoir water can also create an oxygen-deprived layer of water at the bottom of the lake, and this anoxic environment can turn some of the decomposing carbon into methane instead of CO2.
And to make matters worse, many hydropower stations draw water from the bottom layers of reservoirs to generate electricity, all but ensuring a portion of the methane gas is emitted downstream or as it passes through hydroelectric turbines.
A June 2012 commentary article in the journal Nature Climate Change claimed that the carbon emissions from all of Brazil's hydroelectric reservoirs were equal to or greater than the annual emissions of São Paulo, the country's largest city.
Richard Taylor, executive director of the International Hydropower Association, said these reports exaggerate reservoirs' role in the natural carbon cycle. Indeed, the world over, carbon stored in forests and vegetation routinely seeps through groundwater into rivers and then outgases into the atmosphere (ClimateWire, Nov. 21).
"You have a natural system going on, unrelated anthropogenic activities going on," Taylor said. "We know we can measure emissions on the surface of a reservoir, but the story is much, much more than that."
With carbon constantly cycling through water and forest systems, and constantly flowing in and out of the atmosphere, do reservoirs really change anything? Or do they just emit carbon gas that would've found some other way into the atmosphere?
1 long-term study, mixed results
There has been only one study chronicling pre- and post-flood emissions from the same hydroelectric reservoir, a seven-year study of the Eastmain-1 dam in northern Quebec. Cristian Teodoru, a postdoctoral researcher at the University of Leuven, led the team that spent three years monitoring emissions from the landscape before it was flooded at the end of 2005. They then spent four more years measuring emissions from the flooded landscape.
Teodoru and his colleagues found that in its first post-flood year, the reservoir was a large net source of CO2 but a much smaller source of methane compared with pre-flood levels. In subsequent years, however, net carbon dioxide emissions declined steeply, while net methane emissions remained constant or increased slightly.
Another concern raised by scientists is that, while these emissions may decline over time, the big spike in outgassing early on is much higher than emissions from fossil fuel generation and could take decades to recover from.
Taylor, who's read the study, described the methane emissions as so small they're "negligible." But in its paper, Teodoru's team ultimately concluded that "the reservoir will continue to emit carbon gas over the long term at rates exceeding the carbon footprint of the pre-flood landscape."
Teodoru said net emissions from reservoirs could be even greater in tropical and subtropical climates, since the bacteria in the water breaking down the carbon-rich vegetation work faster in higher temperatures. Northern reservoirs, including Eastmain-1, are also frozen for much of the year, he said, stifling potential emissions.
"If you compare the same surface in Canada to one in Brazil, you'd have totally different emissions," Teodoru said.
The tropics and subtropics could also soon see a hydropower boom. Scientists believe up to two-thirds of the planet's hydropower capacity is still undeveloped -- the majority of it in the Southern Hemisphere -- and forecast hydropower capacity to double by 2050.
With so many factors contributing to the natural carbon cycle, Taylor said research has yet to fully conclude what impact reservoirs have on this cycle.
He also argued that some research up to now has been flawed. Many studies investigating carbon emissions from reservoirs use the Balbina Dam in Brazil as a case study. The dam is wide and shallow, and it's been shown to emit more methane than most coal plants.
Ways to build lower-emitting dams
Taylor called the dam an outlier, with a uniquely poor design and location contributing to exceptionally high emissions. Balbina has a 250-megawatt generating capacity yet has the same surface area as the deeper Itaipu Dam, whose 14,000 MW capacity is second in the world to China's Three Gorges Dam.
Balbina, Taylor said, is "atypical, yet the most intensively studied project on greenhouse gas emissions."
That said, there are steps hydropower developers can take to minimize reservoir emissions.
Developers should try to avoid building dams near major carbon sinks, for example, and should install an off-take system that draws water from the upper levels of the reservoir, not the methane-rich lower levels, Taylor said.
The International Hydropower Association has developed a sustainability assessment protocol that looks at more than 20 topics from the project planning stage to construction and production where emissions could be prevented.
"As we learn, we will evolve that practice," he said. "Certainly, we would look at water quality. If that's well-managed, I think the greenhouse gas issue's well-managed."
Ultimately, Taylor said he hopes the industry can move past this issue and focus on other things, saying he thinks the industry is "very tired of being the butt of this."
"All renewables work together, and the storage of energy in hydro reservoirs is key to the increased utilization of renewables," he said.
"We're managing [emissions] the best we can," he added. "I can't say there's going to be a perfect solution, but I don't see that as the most pressing issue."