Peter Bosshard Headshot

The Dam That Shook the Earth

Posted: Updated:
Print

Scientists agree that dams can trigger earthquakes. A new research paper presents fresh evidence that the devastating earthquake which killed more than 80,000 people in China’s Sichuan Province in May 2008 was triggered by the Zipingpu Dam. This would be the world’s deadliest dam-induced earthquake ever.

Reservoirs can trigger quakes by adding weight to the Earth's crust, and by lubricating the fissures of faults. There are approximately 100 earthquakes which scientists believe have been induced by dams. In the US, the construction of the Hoover Dam led to several hundred small tremors in the 1940s. California’s Oroville Dam, the country’s largest earthen dam, caused a quake measuring 5.7 on the Richter scale in 1975.

The Zipingpu Dam rises 511 feet above the Min River and is located only a few miles from the Longmen Shan fault, where the Tibetan Plateau collides with the Eurasian Plate. Fan Xiao, a courageous official with the Sichuan Geology and Mineral Bureau, warned against the construction of a high dam at this location for many years, but without success. Less than three years after the Zipingpu reservoir was impounded, the 7.9 Wenchuan earthquake hit Sichuan Province on May 12, 2008. Thousands of buildings collapsed, and 88,000 people were reported killed or missing. Sixty-nine dams, including the Zipingpu Dam itself, were badly damaged and at a high risk of failure, but could be salvaged. If they had breached, millions of more people could have been killed.

After the disaster, Fan Xiao pointed out that “Zipingpu has all conditions that provoke reservoir-induced earthquakes,” and said that “we cannot rule out the possibility that building the Zipingpu Dam induced the earthquake because the epicenter is so close to the dam.” Christian Klose of Columbia University’s Lamont-Doherty Earth Observatory and Lei Xinglin, a geophysicist at the China Earthquake Administration in Beijing, also found correlations between the Zipingpu reservoir and the earthquake. Yet the International Commission on Large Dams rejected their findings. After conducting a study tour of the area, the industry body argued that it was “very unlikely” that the Wenchuan earthquake was triggered by the dam.  

The Chinese authorities have never released the exact seismic data of the Wenchuan quake. Based on the figures which are available, Shemin Ge of the University of Colorado, Mian Liu and Gang Luo of the University of Missouri, Ning Lu of the Colorado School of Mines and Jonathan Godt of the US Geological Survey recently created a two-dimensional model to test how the Zipingpu reservoir is impacting the Longmen Shan fault zone. In a paper published in the Geophysical Research Letters, they demonstrate that by May 2008, the reservoir had increased the stress on Yingxiu-Beichuan fault, one of the faults which ruptured during the quake, by 0.01-0.05 MPa (or 0.1-0.5 bar). Such a stress change, the scientists conclude, “could have hastened the rupture of the fault by tens to hundreds of years.” If the quake had hit a hundred years later, it might have found the local population living in earthquake-proof buildings rather than the tofu-like structures of 2008.

Hundreds of high dams are currently being built in earthquake-prone areas such as China’s Southwest, the Himalayas, Iran, Turkey and Central America. Such projects should only go forward if the seismicity around the dam site is continuously monitored, if water levels are not allowed to fluctuate quickly, and if all buildings in the reservoir area are seismically retrofitted. When I met Fan Xiao in his modest office in June, he warned that these consequences of the Wenchuan quake have not yet been drawn. “Dam builders often avoid such measures because of the high cost,” he warned.