Reservoir triggered seismic (RTS) phenomena can be described as specific responses of the tectonic environment to the impoundment of large reservoirs. They are seismic events which were not quite ‘ripe enough’, and needed the incremental effects of reservoir load and the build up of pore pressure to make them happen.

The problematic association between reservoir impoundment and seismic activity has been considered in the dam engineering community since 1935, when the first documented case of RTS at Lake Mead was created by the 220m high Hoover arch dam in the US. Furthermore, during the 1960s several strong earthquakes in India, China, Greece and Zambia caused considerable damage and were suspected of being reservoir triggered. Consequently, interest in and the quest for knowledge about this phenomenon grew.

Although icold acknowledges that RTS only affects a small proportion of the dam population as seismic events, it has remained controversial and the subject of much study due to associated environmental impacts and risk considerations.

The decision was taken to make RTS the subject for an ICOLD Bulletin under the auspices of the Committee on Seismic Aspects of Dam Design. Bulletin 137 is called Reservoirs and Seismicity: State of Knowledge and aims to help dam engineers understand the specific features of reservoir triggered seismicity, and evaluate the likelihood of facing the phenomenon.

What is RTS?

As ICOLD states, consensus has been reached that impounding can cause triggering of seismic activity only if favourable pre-existing tectonic conditions have already developed. This means that the fault that can produce the earthquake energy release is already near to failure; so that the added weight stresses and pore pressure build up due to reservoir impounding can trigger seismic energy release. Indeed, the added weight of water cannot substantially increase the seismic energy release. The increase of energy potential due to impounding is practically insignificant in view of the size of the actual seismic energy releases.

Rapid reservoir level variations are known to increase the occurrence of event triggering, while the increasing height of dams and size of reservoirs also have an effect. ICOLD recommends that RTS potential needs to be considered at the outset for large dams over 100m in height.

A correctly defined dam which has been designed to current ICOLD practices will be able to withstand the effects of the largest reservoir triggered earthquake. If a dam has been designed against earthquakes using current state-of the-art practices, triggered seismicity cannot increase the seismic hazard at the dam site, or endanger the structure or the people who could be at risk in the case of a dam failure. Current practices require the dam to safely withstand ground motions caused by the maximum credible earthquake (MCE). Therefore RTS seismicity is not a direct safety problem for a well-designed dam as the maximum reservoir triggered earthquake cannot be stronger than the MCE.

However, RTS may still be a problem for other structures, buildings and appurtenant works because they often have a much lower earthquake resistance than the dam. ICOLD emphasises that RTS only concerns the existing infrastructure and buildings in the storage zone as the dam itself must be designed for magnitudes covering the triggered possibilities.

Adequate monitoring of RTS prior, during and after impoundment is of great importance, ICOLD states. Such monitoring provides the only conclusive evidence as to whether or not storage impoundment causes triggered earthquakes. To help distinguish between background seismicity and RTS, monitoring should start at least a couple of years prior to impounding of the reservoir. This is important as lack of changes in background seismicity during impounding means an absence of RTS. A single seismic station is the minimum required during the pre-impounding period where no RTS is expected. As a general rule five seismic stations are a minimum for the successful determination of epicentre locations and hypocentral depths.

Gathering information

The exact number of reservoir triggered cases is not known but ICOLD believes that the number of accepted cases is today somewhere between 40 and 100. The largest recorded reservoir triggered events have had magnitudes ranging from 6.0 to 6.3.

A full review and presentation of known and available case histories would be valuable but such data are not readily available and are rather difficult to collect. Therefore, ICOLD opted to publish a few typical cases within its bulletin:

• Hsingfengkiang buttress dam in China – a large triggered seismicity, causing a strong local earthquake, which significantly damaged the dam.

• Mratinje arch dam in Yugoslavia – a moderate RTS. It is of particular interest as seismic monitoring was introduced prior and after impounding, witnessing re-appearance of RTS after 17 years of service.

• Kurobe arch dam in Japan – was seismically monitored prior and after impounding. The case is of special interest as the dam was reported as an RTS case on the basis of microseismic monitoring after impounding. But later analyses led to the conclusion that Kurobe dam was not an RTS case.

• Takase rockfill dam in Japan – a large dam carefully monitored prior and after impounding, where similar micro-seismic activity was present before and after impounding.

• Poechos embankment dam in Northern Peru – a case of a seismically active environment where RTS was absent or was masked by basic background activity.

Monitoring of RTS phenomena is the best means to obtain further data on the subject. ICOLD recommends that this should be followed and actively developed by all interested entities, especially by dam designers and owners. A more extensive collection of recorded and commented case histories would prove to be of great value.

The above information was collated from ICOLD’s Bulletin 137 on Reservoirs and Seismicity: State of Knowledge. For more information contact ICOLD (secretaire.general@icold-cigb.org) or Dr Martin Wieland, Chair of the Committee on Seismic Aspects of Dam Design (martin.wieland@poyry.com)