(January 26, 2009) Fan Xiao, Chief Engineer of the Regional Geology Investigation Team of the Sichuan Geology and Mineral Bureau, says scientists must investigate if Zipingpu dam triggered devastating 2008 earthquake, describes massive quake-damage to dams, rebuts recent Science Times article.
Three Gorges Probe asked Fan Xiao, Chief Engineer of the Regional Geology Investigation Team of the Sichuan Geology and Mineral Bureau, China for his reaction to Chinese Academy of Sciences experts’ dismissal of the possibility that the Zipingpu dam induced China’s devastating earthquake. Read his rebuttal and detailed description of the extensive damage done to China’s dams by the quake, and why officials must address the urgent question: Did the Zipingpu dam induce the May 12 earthquake?
By Fan Xiao
Major damage to dams as result of May 12, 2008 earthquake
The Zipingpu and many other hydro dams on the Min, Jian, Shiting, Mianyuan, Fu and Jialing rivers in the earthquake-affected region were seriously damaged or totally destroyed by the powerful earthquake and aftershocks that occurred on May 12, 2008, in China, including dams, power plants, power generation equipment and water diversion facilities. The damages were so serious that it was difficult for the reservoirs to work normally and release water properly. In addition, geological disasters, such as landslides, mud and rock flows, and hillside collapses in the affected region also seriously damaged various hydroelectric power facilities and caused reservoirs to rise to dangerously high levels. In some cases, water levels in the reservoirs were rising so fast, resulting in the formation of huge lakes, causing several reservoirs to flow over the dams – called overtopping. Therefore, the situation was much more serious than what Professor Pan Jiazheng portrays as minor damages.
Take the Zipingpu dam, for example. As the official Xinhua reported on May 13, 2008, cracks were found on the dam structure, walls of power plants and other buildings collapsed, and generators in the power station were forced to shut down.
The national Ministry of Water Resources sent a team of experts to carry out investigations and examinations after the earthquake. These experts discovered that, as a result of the powerful May 12 earthquake, the Zipingpu dam subsided a maximum of 100 cm and was displaced from upstream to downstream by a maximum of 60 cm. In addition, several cracks were found on the dam structure, with the longest being about 350 meters across the surface on the upstream side of the dam.
The situation in other hydro dams on the Min River valley, especially those located upstream of the Zipingpu dam, was even worse: immediately upstream, for instance, the Yingxiuwan Power Station’s structure was severely damaged, its power plant destroyed and covered by landslides, and its power distribution facilities totally destroyed. 
Meanwhile, the Tongzhong reservoir, still further upstream, collected as much as 4.5 million cubic meters of flood waters, much more than its normal storage capacity of 3.3 million cubic meters, because its sluice gates were seriously damaged and unable to open to release water downstream. The flood water overtopped this dam as well, seriously putting the dam at risk of catastrophic collapse. The same problem of overtopping occurred also at the Taipingyi dam and power station upstream of Tongzhong.
Hydro dams on the tributaries of the Min River and other affected rivers, also suffered severe damage: the Gengda Power Station was totally buried by landslides, the Yuzixi power station was seriously damaged by the collapse of a mountain, and generators in Zhongba Power Station were flooded and distorted.
A number of other hydro dams upstream of Zipingpu, such as Yingxiu, Jiangsheba and Gengda, were also damaged and overtopped. Because they were in critical condition, officials came up with a plan, shortly after the earthquake, to blast the dams if necessary. It is fortunate that through urgent and timely rescue work, they were able to open the sluice gates of Zipingpu, Taipingyi, Tongzhong and others. But not every dam was so fortunate: the Hongsong Power Station on the Shiting River, tributary of the Tuo River, for example, was blasted, thereby releasing the flood waters in the earthquake-choked lake and reducing the threat to the downstream region from an uncontrolled dam collapse.
Professor Pan argues that most of the damaged dams can be repaired or, if they are too damaged for repair, they should be rebuilt. But the issue should not be whether they can be repaired or rebuilt, but whether they should be built in this area at all. Given that this area is a seismically active one, and given that these dams, first, might induce earthquakes and, second, might fail catastrophically in the event of an earthquake, we must question the wisdom of building a facility that puts sizeable populations at risk of death.
Professor Pan has also argued it is not fair to ignore the fact that these water projects have played a significant role in the economic development of the region by supplying water for human use and for irrigation and so forth. But counting the ‘profits’ to the dam companies and governments only, while ignoring the risks and costs to the population affected by life-threatening earthquakes, lost livelihoods due to resettlement, altered environments, and the loss of cultural heritage would not only be profoundly unfair, it would be an incomplete cost-benefit assessment and bad economic policy too.
I would suggest that Professor Pan and other hydro dam proponents visit some reservoir areas, such as the Pubugou Power Station on the Dadu River or the proposed Longpan Power Station, upstream of the Tiger Leaping Gorge on the Jinsha River, if they are truly concerned about the effect of hydro dams on the people. It would be good for Professor Pan and other experts who favor dam projects to really know how the affected people are feeling and what the displacees want.
Officials avoid the urgent question: Did the Zipingpu dam induce the May 12 earthquake?
While the author of the Science Times article raises the issue of large hydropower projects inducing earthquakes, Professor Pan insists that “it’s impossible for the reservoirs to ‘make’ earthquakes,” which, either intentionally or unintentionally, confuses scientific concepts. As matter of fact, experts have asked the question of whether the Zipingpu induced the May 12 earthquake, but nobody has suggested that the dam project “made” or “caused” the incident. Normally there are two widely-used terms that describe the concept: RIS (reservoir induced seismicity) or RTS (reservoir triggered seismicity). Although there are differences between them, both involve reservoir-related seismic activity in a geological context in which earthquakes will naturally and inevitably occur.
Here’s the difference between the two terms.
According to Zhou Xin, and others, reservoir induced seismicity occurs when the filling of a reservoir partially changes the natural environment in which seismic activities are likely to occur, creating an interactive dynamic feedback mechanism that induces earthquakes within several years of the filling of the reservoir (see Zhou Xin and others: “Reservoir seismicity: induced or triggered?” published in Seismicity Research in South China (Huanan dizhen), 2005, 25(2)). So, it is appropriate to name such an earthquake as ‘reservoir induced seismicity’ rather than ‘reservoir triggered seismicity’.”
Alternatively, reservoir triggered seismicity is used more often by the water conservancy engineering community (or hydraulic engineering community) and Professor Pan quite correctly defines a reservoir triggered earthquake as one where reservoirs “influence weak links to enable seismic activity to occur earlier.” But, he adds, Zipingpu could not have played a role in the May 12 earthquake because a) there is wide scientific agreement that the May 12 earthquake was a huge natural disaster caused by massive crustal movement, and b) history shows that there has never been a reservoir triggered earthquake of magnitude eight. However, he adds bizarrely and inexplicably that, “if the Zipingpu dam reservoir did play a role in the big earthquake, it was actually to reduce the energy released by the earthquake in general.”
The most crucial difference between the two concepts of “induced” and “triggered” earthquakes, is that induced earthquakes may have a greater magnitude, even higher than the largest in historic record in a given area, while triggered earthquakes are thought to enable seismic activities to occur earlier and more frequently, but with a magnitude not as large as the largest in historic record in a given area. This distinction, however, is not substantiated by the worldwide cases for which data is available. That data shows that it is possible for either induced or triggered earthquakes to be greater in magnitude than the largest on record in a given area.
So far there have been at least four earthquakes with a magnitude of six and above, which are widely recognized as reservoir related, and are either called reservoir induced seismicity or reservoir triggered seismicity, including Xinfengjiang in China, Kariba in Zambia, Kremasta in Greece and Koyna in India. It is worth noting that in each of these four cases, the earthquake that occurred had a greater magnitude than the largest on record for the area.
Professor Pan insists that “no reservoir triggered earthquake with a magnitude eight has ever occurred in history.” But it is neither logical nor scientifically rigorous, and certainly not prudent, to conclude therefore that no reservoir triggered earthquake with a magnitude eight will ever occur in the world in future. Nobody would have imagined, for example, that a reservoir triggered earthquake as powerful as magnitude six would have occurred at the Xinfengjiang reservoir in China, before the earthquake with a magnitude of 6.1 did occur in 1962.
Professor Pan also argues that “the earthquake research community outside and inside China has widely accepted the notion that the May 12 Wenchuan earthquake was a huge natural disaster caused by massive crustal movement.” It would be true to say this if they were referring only to Wenchuan’s geological background and conditions; but it’s far from safe to rule out the possibility that the Zipingpu triggered the earthquake. In fact, a number of scientists inside and outside China, experts in the Chinese Academy of Sciences included, are holding open the possibility that Zipingpu played a role in the May 12 Wenchuan earthquake, and many of them are still conducting studies on the issue.
As for the lessons learned from the Wenchuan earthquake, Professor Pan favors assessing the maximum magnitude earthquake that existing dams and those under construction can withstand. This is necessary, but not sufficient, because in most cases the designed capacity for a hydro dam to withstand seismic activity is based on the recorded earthquakes in a given area. As long as a reservoir can induce seismic activity with a greater magnitude than what is known or predicted, the dam will be at greater risk of damage and catastrophic failure than predicted. Moreover, even if a dam structure itself was unaffected by seismic activity, geological disasters, such as hill or mountain collapses, landslides, and mud-rock-flows caused by the earthquakes upstream of the dam, and within the reservoir area, would create a threat to the dam itself and thus to the downstream area. Therefore, there is not only an urgent need for further study on whether the Zipingpu induced the earthquake but more importantly, to reexamine and even adjust the development planning for the use of water resources in the area as a whole.
Regarding the possibility that the Zipingpu dam induced the earthquake, Professor Chen Houqun has pointed out that “not all high dams and big reservoirs are able to trigger earthquakes.” True, but this does not mean that Zipingpu did not induce the May 12 Wenchuan earthquake.
Professor Chen is emphatic that “Zipingpu had nothing to do with the Wenchuan earthquake at all” because its reservoir is located between the Beichuan-Yingxiu Fault Belt and Jiangyou-Guanxian Fault Belt and because it’s highest operating water level was 875 meters, which is lower than the natural water level of 877 meters in the Min River, where the Beichuan-Yingxiu Fault Belt is located. Thus, he argues, the water in the Zipingpu reservoir had no impact on the Beichuan-Yingxiu Fault Belt, because the hydrogeological conditions within it had little change at all.
In fact, however, the normal pool level did reach as high as 877 meters in the Zipingpu reservoir. But this isn’t the crucial issue. Rather, the crucial is that geological investigations and geophysical surveys show that a system of fault belts, including the Beichuan-Yingxiu Fault Belt (the May 12 Wenchuan earthquake occurred mainly along this fault line), the Jiangyou-Guanxian Fault Belt, and a series of sub-fault belts mingle beneath the reservoir. Because the main part of the Zipingpu reservoir is on the top of this system of fault belts, which are highly non-uniform, the water stored in the reservoir not only penetrates deep in the system of fault belts but has significant effects on the fault belts, and on the Beichuan-Yingxiu Fault Belt in particular.
Second, Professor Chen has argued that no changes were discovered in the recorded seismic activity before and after the Zipingpu reservoir was filled, despite the fluctuations in the reservoir’s water level, so there is no correlation between filling the reservoir and seismic activity. And he adds, “Based on the monitoring data available from October 2005, when the reservoir started filling to April 2008, before the earthquake occurred, the location, frequency and intensity of seismic activity was within the normal range. There were small changes during that period of time, but these changes would have occurred even if the dam had not been built.”
But his statement is inconsistent: it seems that seismic monitoring data showed that changes in frequency and intensity of seismic activity did occur before and after the filling of the Zipingpu reservoir. Arguing that these changes would have occurred even if the dam hadn’t been built, from a scientific research perspective, is qualitative and unconvincing. It is difficult to conclude that there was no correlation between filling the reservoir and seismic activity. Of course, here Professor Chen was not presenting his research paper but responding to questions raised by media reporter. Therefore, we are looking forward to Professor Chen’s articles with detailed analysis and related data in the future.
Back to Post:  From upstream to downstream, the Min River has the following dams: Tianlonghu, Jinlongtan, Jiyu, Tongzhong, Jiangsheba, Futang, Taipingyi, Yingxiuwan, Zipingpu.
Back to Post:  See footnote 1.
Back to Post:  A tributary of the Yangtze.
Back to Post:  The Pubugou dam was hotly contested by affected citizens and, in a violent clash over inadequate compensation, the army allegedly opened fire on more than 100,000 citizens.
A news blackout followed. Later, one of the protesters was executed without trial for his role in the protests.
Meanwhile, at the proposed Longpan hydro and water supply dam 70% of the people affected by this dam would be ethnic minorities and at risk of losing their livelihoods and communities.
Back to Post:  Chinese officials have not yet released data for Zipingpu’s reservoir levels since the initial flooding, nor seismic readings for the area.
Three Gorges Probe, January 26, 2009