When Wolfgang Schwanghart went to Nepal for the first time, he was shocked to see many hydropower dams perching on the steep Himalayan terrains. “They looked rather precarious,” says the geologist, who works at the University of Potsdam in Germany.
That was in early April 2015. Two weeks later, a devastating magnitude-7.8 earthquake struck, killing nearly 9,000 people. It also damaged 31 projects of the type he had seen. Now, Schwanghart and his colleagues have revealed in a study that it was not the ground shaking that damaged most of these projects, as most people had assumed, but landslides triggered by the quake.
The work, published in Geophysical Research Letters1 last month, highlights the risk of quake-triggered landslides to hydropower development in the Himalayas — a global hotspot for this type of energy source. “Our study points to an urgent need to re-evaluate hydropower development in the Himalayas,” says Schwanghart.
The team sifted through reports on damaged hydropower plants and noted that the steepness of river banks seemed to be a good indicator of the severity of destruction at places where the quake's ground shaking was not particularly strong. They then developed a model that overlaid river steepness in the Himalayas onto a map of ground-shaking intensities from the 2015 quake. This produced a pattern in which a combined effect of ground shaking and river steepness produced the greatest damage to hydropower plants. This mimicked the damage to plants in the aftermath of the quake — confirming researchers’ hunch that it was landslides that damaged those sites.
The damaged hydropower plants, which cost the country US$200 million to repair had been built in line with the design criteria as far as earthquakes were concerned, says Schwanghart. “They survived the quake, but got wiped out by moving debris.”
The team applied the model to 273 hydropower projects that are already in operation, under construction or are being planned in the Indian, Nepalese and Bhutanese Himalayas, and for which they had data. The results show that a quarter of them are likely to face severe damage from quake-triggered landslides. So are more than 10% of the potential hydropower sites along Himalayan rivers.
Yet this phenomenon is rarely adequately addressed when deciding where to build these projects, say the researchers — despite quake-triggered landslides being the dominant type of landslide that kill people globally. Between 2004 and 2011, for instance, 49,000 fatal landslides — nearly 60% of the global total — were caused by earthquakes2.
The new study shows that there is a relatively simple way of determining whether quake-triggered landslides pose a major problem to a hydropower project, says Dave Petley, a geologist at the University of Sheffield, UK. “It’s quite elegant."
The model will also help developers to make a more accurate assessment of the risks at candidate locations for hydropower plants, says David Gernaat, a computer modeller at the PBL Netherlands Environmental Assessment Agency in The Hague.
Last year, Gernaat and his colleagues reported3 that nearly 40% of the hydroelectric power that could potentially be generated at low cost is in the Asia Pacific region. But the analyses factored in only risk from earthquakes, not from landslides. This means that “we have overestimated the hydropower potential in the region and underestimated the cost”, says Gernaat. Some of the locations that the study suggested were suitable for developing hydropower might be highly prone to landslides — with or without an earthquake, he says. He plans to include landslide risk in future assessments.
There are also many examples in which the cost of the infrastructure ends up being much higher than first estimated because landslide risks have not been properly assessed or managed, says Petley.
The lack of appreciation of landslides keeps scientists such as Petley awake at night, at a time when the push for hydropower development in the high mountains of Asia and other regions, such as South America, is growing.
He fears a repeat of the major disaster at the Vajont dam in Italy in 1963, when landslides caused a megatsunami of 50 million cubic metres of water — enough to fill 20,000 Olympic-size swimming pools — that overtopped the dam in waves up to 250 metres high. Almost 2000 people died and several villages and towns downstream were destroyed.
“There are so many dams that suffer landslide problems that sooner or later we are going to have another big tragedy,” says Petley. “It’s a disaster waiting to happen.”