With Kivu's rolling green swells and serene coastline, it's hard to imagine why this is called one of Africa's "killer lakes."
Fishermen have known for more than a century about the mysterious gas that occasionally bubbles up, killing fish and sometimes swimmers.
The source, scientists say, is a massive pool of methane and carbon dioxide that lies at the bottom of the deep-water lake on the border between Rwanda and the Democratic Republic of Congo. Gas levels have been steadily rising, and experts say the gases might one day explode or burst to the surface, releasing a deadly cloud similar to one that killed more than 1,700 people at Cameroon's methane-rich Lake Nyos in 1986.
Hoping to avert a catastrophe on the shores of a lake where 2 million people live and to solve its energy woes at the same time, the Rwandan government is embarking on a risky project to extract the methane and use it to generate electricity.
Methane-power generation plants exist elsewhere, but the effort here is the first attempt to extract the gas from underwater and burn it to fuel an electricity plant.
"It's the first of its kind in the world," said Albert Butare, Rwanda's minister of state for infrastructure. "In the beginning, it was a myth. But now the technology is promising."
The government this month launched a $15-million pilot project that will try to power a four-megawatt generator with methane from the lake. A floating platform, installed this year, dropped a pipe more than three football fields deep to reach the methane-rich water.
An American energy investment firm, New York-based Contour Global, is close to signing a deal to build the permanent electricity plant on Lake Kivu's shore, which would eventually produce 100 megawatts of sorely needed power for Rwanda, nearly twice the country's daily production, government officials said.
Previous efforts failed
Only about 5% of Rwandans are connected to the country's national grid, and prices are twice as high as those of other East African nations because of inadequate supply, mostly from diesel-fueled generators.
"This is the only major domestic resource we have," Butare said.
Rwanda has tried to exploit the methane before. Efforts date to the 1960s, and a previous pilot program produced enough electricity to power a local brewery. A line of international contractors came and went without success, most recently a Danish outfit whose contract was terminated.
It's a complicated, potentially dangerous process. Removing the gases could destabilize the lake, leading to an unintended release. Engineers must figure out what to do with excess carbon dioxide, a byproduct of the separation process.
"You need to use gentle methods," said Charles Nyirahuku, a government manager in charge of the project.
Fishermen also worry about leaks, recalling pipeline cracks in the previous pilot program that poisoned the water around the facility.
"From time to time, you would see dead fish floating in the water near the plant," said Semajeri Mussa, 35, head of a local fishing cooperative.
But, he said, many fishermen support the idea of removing the methane, which he believes has inhibited animal life in the water. "We think it will be better for us in the long run."
Doing nothing, government officials point out, carries its own risks.
Fear of boiling lava
In a 2006 report, the United Nations Environmental Program voiced "serious concerns" about volcanic activity around Lake Kivu, noting the 2002 eruption of nearby Mt. Nyiragongo, which blanketed the lakeside city of Goma with lava. Rising water temperatures could ignite the methane.
"Large amounts of boiling lava entering the lake could be more than sufficient to trigger a large overturn, releasing huge amounts of deadly carbon dioxide," the U.N. agency said.
The lake's pool of methane, scientists believe, is a result of ancient volcanic activity that mixed carbon dioxide from decaying organic material with bacteria. Gases became trapped at the lake bottom because of varying water temperatures and sediments. The presence of methane is often an indicator of oil, though no reserves have been found.
It remains unclear whether the extraction process will be efficient and cost-effective enough to merit a large-scale project.
"When you are dealing with methane, you have to put a lot of energy into it to turn it into something else," said Teresa Akenga, a chemistry professor at Kigali Institute of Science and Technology. When burned, methane also produces less energy than similar gases, she noted.
But government officials expressed optimism that, after some false starts, the project would succeed, bolstered by modern technology and economic incentives.
Contour Global, a 3-year-old firm that has also invested in an energy project in Minnesota that turns turkey waste into electricity, is expected to invest as much as $200 million.
"They are very comfortable with the technology," said Richard Mugisha, a Rwandan attorney representing Contour. He said the first phase, a 20-megawatt facility, could be operational in two years.
"It's never be done on such a grand scale before," he said. "But we have to do something. If not, we're sitting on a time bomb."