On 28 March 2025, an earthquake with a moment magnitude Mw of 7.7 occurred near Mandalay in Myanmar along the well-known North-South stretching Sagaing fault, which has the capability of producing even stronger earthquakes. The damage caused by the earthquake was huge, although reports are still incomplete.
Some embankment dams – most reservoirs are used for irrigation and water supply – located along the Sagaing fault, experienced significant deformations, though none of them failed. The satisfactory behaviour of the dams was also due to the low water level in the reservoirs during the dry season. In Figure 1 transverse cracks near the right abutment of the Chaungmagyi dam, an earth dam located very close to Naypyidaw, the capital of Myanmar, are shown, which were caused due to differential settlements of the dam. The maximum settlement was ca. 1.7m and the downstream movement of the crest was of the order of 0.6 to 0.9 m.
Figure 2 shows a typical longitudinal crack in the 131 m high Lower Paunglaung earth core rockfill dam, which is also located in the vicinity of Naypyidaw. The maximum crack width is about 12cm and the crack can be observed over a length of ca. 800m.
Another magnitude 6.2 earthquake occurred on 23 April 2025, in the Marmara Sea close to Istanbul. None of the dams within a radius of 100km from the epicentre, which were inspected immediately after the earthquake, suffered any damage. This behaviour is expected for well-designed, well-constructed and well-maintained dams. However, only two dams were located within an epicentral distance of 50km and the closest dam was 30km away from the epicentre, so that the intensity of ground shaking was much lower than that used in the design of these dams. Nevertheless, it should be kept in mind that magnitude 6.0 earthquakes could occur in many parts of the world, although such events may be very rare.
Large downstream and centreline tailings dams have survived strong ground motions associated with large earthquakes (e.g., in 2010 the El Maule Mw 8.8, and in 2015 the Illapel Mw 8.4 earthquakes in Chile).
The above news is good news, but there have been recent dam failures (both water storage and tailings dams) that suggests that most likely they were not designed against earthquakes using today’s seismic design and safety criteria, or they were not operated and maintained in accordance with good practice as otherwise they would have survived. It is the responsibility of all dam owners to ensure their dams satisfy today’s seismic safety criteria and they are properly operated and maintained. This means the safety of the dams must be reassessed periodically, especially after strong earthquakes, when the risk classification of a dam has changed or when there are new seismic design and safety criteria.
The seismic safety re-evaluation of existing dams has been a long-term concern of the ICOLD Committee on the Seismic Aspects of Dam Design. Such evaluations have been carried out or are in progress in different countries but eventually such re-assessments must be carried out for all dams, to confirm that they comply with the current seismic safety criteria.
Also much is written and said about sustainability of storage dams, but it must be kept in mind that dam safety, which is often governed by seismic and flood safety, is the prerequisite for any sustainable dam project. This fact has hardly been realised by many people.
Earthquake safety and performance criteria
The main objective of the earthquake-resistant design of any structure, including dams, is to protect the life of people and to protect them from injuries. This is a universally accepted principle. More specifically, for storage dams the following seismic safety and performance criteria must be taken into account:
- Retain the reservoir and protect people from the catastrophic release of water from the reservoir,
- Control the reservoir level after an earthquake as a dam could be overtopped and destroyed if the inflowing water into the reservoir cannot be released through damaged spillways or low-level outlets, and
- Lower the reservoir level after an earthquake (1) for repair works, (2) for increasing the safety of a damaged dam or (3) when there are doubts about the safety of the dam.
The earthquake hazard is probably the most difficult hazard in the natural environment that a dam must be able to withstand, as strong earthquakes are unpredictable in terms of magnitude, time and location. There is no time for warning or taking precursory action, like lowering of the reservoir, as seismic waves can propagate with velocities of several km/sec, so that if an earthquake occurs only post-event rescue and evacuation is possible. For these reasons it is mandatory to ensure that dams are safe with regard to earthquakes. Earthquake safety implies that the dams are also safe against other types of actions.
Quite a number of embankment dams and a few conventional concrete dams have experienced strong ground shaking, and so there is a basic understanding of their seismic performance. This is not the case for new types of dams such as RCC dams, concrete face rockfill dams, asphalt core rockfill dams, rock-filled concrete dams, hardfill dams, et. as very limited information exists on their performance during strong earthquakes.
The above list of topics shows that the earthquake safety is an important safety concern for most storage dams. The same is also true for nuclear power plant facilities.
2008 Wenchuan Earthquake
The most important earthquake that has occurred since the Montreal 2003 Congress is the Wenchuan earthquake in China. This magnitude Mw 7.9 earthquake of 12 May 2008, damaged over 1000 dams, run-of-river power plants, power plants and reservoirs, most of them in Sichuan Province. A special feature of this earthquake was the large number of rockfalls, rockslides and landslides in the epicentral region, which also affected dams and reservoirs. This is a hazard, which up to now has not received adequate attention in dam projects as the focus on the seismic safety of dams is still ground shaking. But it is very important to note that for large dam projects, the seismic hazard is a multi-hazard that includes ground shaking, fault movements in the footprint of the dam and/or the reservoir, mass movements at the dam site and the reservoir banks and even in the catchment, ground movements, liquefaction, etc. All these hazards must be considered.
The Wenchuan earthquake has also damaged the face slab of the 156m high Zipingpu concrete face rockfill dam (Figure 3). As the reservoir was only 30% full at the time of the earthquake, the damaged dam was safe and mass movements into the reservoir did not cause any dangerous impulse waves. It would still be difficult to assess the behaviour of this large dam if the reservoir were full. The damaged joints were repaired immediately after the earthquake. A damaged vertical joint is shown in Figure 4.
Among seismologists there was speculation that the Wenchuan earthquake may have been a reservoir-triggered event. This is not correct, as since 2008, when this earthquake happened, no observational evidence has been found that supports this claim. Eleven months after the Wenchuan earthquake 13 international dam and earthquake engineering experts participated in a joint ICOLD- CHINCOLD mission and visited several damaged dams, including the Zipingpu CFRD. The issue of reservoir-triggered seismicity was also discussed as such hypothetical theories appeared in the international press shortly after the earthquake. The Zipingpu CFRD and Wenchuan County are close to Chengdu where the 28th ICOLD Congress will be held.
ICOLD Congress in Chengdu
During the recent ICOLD Congress in Chengdu, China, from 16-23 May 2025, four technical questions were discussed, with Earthquake Performance and Safety of Dams being one of them. The General Reporter and the President for this question are the authors of this paper. The last time an earthquake question was discussed was at the ICOLD Congress in Montreal, Canada, in 2003. Since then new experience has been gained on the seismic performance of dams and new seismic design and safety criteria were published by ICOLD.
During the Congress the General Report of Question 111 was presented by the General Reporter Trevor Matuschka from New Zealand, a country that is also prone to earthquakes. The General Report provides an overview and discussion of the papers submitted to the ICOLD Congress, summarises earthquake publications and ongoing work of the ICOLD Committee on Seismic Aspects of Dam Design, and presents on the main changes and developments in seismic design and safety evaluation of large dams since 2003. Topics covered include seismic hazards, significant earthquakes since 2003, dam safety and seismic safety criteria, dam risk classification and seismic design criteria, structures and elements that need to be designed for earthquake effects, safety evaluation of existing dams, dynamic properties of mass concrete and rockfill, shaking table tests, numerical analysis, new types of dams, developments for tailings dams, and the outlook on future developments.
The sub-themes discussed at the Congress are as follows:
- Seismic Design of Dams,
- Earthquake Performance and Safety Evaluation of Existing Dams,
- Seismic Design and Safety of Concrete Dams, and
- Seismic Design and Safety of Embankment Dams.
Separate sessions were held for each sub-theme with presentations selected from submitted papers along with invited presentations from eminent industry experts.
The highlights of the discussions were presentations on the effects on dams caused by:
- The Mw 7.7 earthquake in Myanmar of 28 March 2025.
- The Mw 7.8 and 7.7 earthquakes in the Turkey-Syrian border region of 6.2.2023.
- Several strong earthquakes in Japan during the last 20 years.
- Strong earthquakes in China such as the 2008 Mw 7.9 Wenchuan earthquake and the 2022 Mw 6.7 Luding earthquake.
The effects of the 2016 Mw 7.8 Kaikoura earthquake in New Zealand, which resulted in an estimated 80,000 to 100,000 landslides (Figure 5), was also focused upon, plus developments in the seismic analysis, design and safety evaluation of large concrete dams in China during the last 60 years, and the effects of strong earthquakes on large tailings dams in Chile, including the 2015 Mw 8.4 Illapel and the 2010 Mw 8.8 El Maule intraplate earthquakes. During these major earthquakes several dams were damaged and during the magnitude 9.0 Tohoku earthquake in Japan in 2011 the 18.5 m high Fujinuma embankment dam failed. The flood wave created by the release of the reservoir caused the loss of eight lives.
Seismic safety
The seismic safety of dams is important because earthquakes can result in catastrophic effects on people, property, infrastructure, historical and cultural sites, and the environment if they fail. To prevent uncontrolled release of water a dam must be able to withstand the seismic hazards, including ground shaking, associated with an extreme earthquake which is referred to as the Safety Evaluation Earthquake.
The subject of the seismic safety of dams and levees continues to evolve. The awareness of the importance of seismic safety and the technical knowledge base has expanded considerably since a question on seismic aspects of dams was discussed at the ICOLD Congress in Montreal in 2003 where the first author was the General Reporter.
The main recommendations of the General Report are as follows:
- Large dams should be instrumented with strong motion recorders (free-field, base, mid-crest and abutments) to measure accelerations. Deformation monitoring is also important. Measurements of response during earthquake events allows calibrating estimates of seismic hazard (ground shaking) as well as models that predict response of the dam to earthquakes. The information can also add to the body of knowledge for all dams.
- Monitoring and observations of dam and safety-critical equipment behaviour during earthquakes should continue and be carefully interpreted. Observations should be shared to allow benefit to all stakeholders. People with responsibility for dam safety should study the lessons learnt and apply them to their work.
- There are multiple features of earthquake hazards (i.e., ground shaking, surface fault movement, mass movements, waves in reservoirs, etc.). They all need to be considered.
- The seismic safety of dams needs regular review because of the evolving understanding of seismic hazards, changes in the risks associated with potential failure of a dam, deterioration of the dam, the understanding of potential failure modes and the measures available to mitigate risks.
- There is continuing evolvement of the dynamic properties and constitutive models for dam materials and foundations and advances in numerical modelling of the stress-strain characteristics of materials and estimation of the deformation of dams when subject to earthquake ground motion. However, there are still significant uncertainties associated with numerical analyses. Sensitivity studies are always recommended to understand the effects of various assumptions and to properly consider uncertainty.
Final comment
It is obvious that the earthquake safety of dams will remain an important issue for years to come and, therefore, future congress questions on the earthquake safety of dams can be expected.
Finally, it should be pointed out that the seismic design criteria that are based on the seismic hazard, are not related to climate change. The uncertainties in the seismic hazard are orders of magnitude larger than any theoretical effect of climate change. However, climate change effects (increased snow melt and rainfall) may affect the seismic triggering of mass movements in the reservoir region and the catchment as well as glacial lake outburst floods. The latter is of main concern for dams located in the Himalayan region.
The authors are Martin Wieland, Chairman, ICOLD Committee on Seismic Aspects of Dam Design, Honorary Member of ICOLD, Dam Consultant, Dietikon, Switzerland. Email: martin.wieland48@gmail.com
Trevor Matuschka, Vice Chairman, ICOLD Committee on Seismic Aspects of Dam Design, Director, Engineering Geology Ltd, Auckland, New Zealand. Email: trevor.matuschka@egl.co.nz
