Dam Safety in Southern Africa: Will the Walls Come Tumbling Down?

Tuesday, March 14, 2006

The Mozambican NGO, JA-Justica Ambiental released the following statement on the dam safety in Southern African region for the Day of Action for Rivers, Water and Life in 2006.

As a developing country, Mozambique, and the whole Southern African region, has a growing demand for electricity. Given Mozambique’s hydroelectric potential, much interest has been shown in building more dams on some of the country’s large rivers. The project that has been most keenly examined is the Mphanda Nkuwa Dam, proposed for the lower Zambezi about midway between the City of Tete and the existing Cahora Bassa Dam. Recently, however, concern has grown regarding the environmental impacts and safety of such large dams.

Kariba Faults
Kariba Faults
Large dams can fail, either because of dam wall failure caused by poor construction or damage from earthquakes, or because of overtopping due to flooding. The earthquake experienced on the 23rd of February has generated concerns about both the structural integrity of existing dams like Cahorra Bassa and Kariba, and the construction specifications for new projects like Mphanda Nkuwa. Dam failure on one of these dams would be disastrous due to the human settlements located downstream of the dams, as well as the riverine ecosystem which supports them. The key settlements in the path of a dam-induced flood include the large cities of Tete and Quelimane, smaller cities such as Marromeu and Mutarare, and large numbers of subsistence farmers who rely on the river for farming and fishing activities.

In terms of seismic risk the situation in Mozambique is rather complicated. The country is located in the vicinity of what is known as the Nubia-Somalia boundary and in fact straddles a highly active fault zone called the Shire Trough, which runs southward down the country from the southern point of Malawi almost all the way to Maputo. Thus the country is considered to be in a seismically active zone, however, due to poor record-keeping in the area scientists’ ability to determine the potential for large earthquakes is severely constrained. This was evidenced by the earthquake on the 23rd of February which was nearly 29 times bigger than what was thought to be possible along that fault!
This raises concern for the Mphanda Nkuwa project which is being designed to withstand
earthquakes that Professor Chris Hartnady considers to be alarmingly small given the
potential for large earthquakes on the plates in close proximity to the proposed dam site – the dam is to be located only around 200 km from the heart of the Shire trough fault zone. This concern is heightened as Professor Hartnady thinks the earthquake on the 23rd might be indicative of an increasing trend in seismic activity for the area. He thinks that it is possible that that earthquake loaded the stresses on the surrounding plates, thus increasing the potential for a subsequent large earthquake. Finally, and also of concern is the fact that due to the shape of the Fault Zone in the Shire Trough, the construction of another reservoir might in fact increase the seismic potential in the area as the stress on the plates is increased by the weight of the reservoir.

The last of the concerns mentioned above relates to what are termed reservoir-triggered earthquakes, which are certainly of concern for Mphanda Nkuwa. This is due to the underestimation of the seismic potential on a local geologic feature called the Estima Fault. This fault has not been active recently, but recent migration of seismic activity southward down the Shire Trough has raised concerns that this fault may now be active and in fact may pose the greatest threat to the Mphanda Nkuwa dam.

Due to the problems of estimating the ”maximum credible” earthquake and the fact that so many people would be put at risk should the dam fail, it is imperative that the dams’ designers take a prudent, not optimistic, approach when designing the dam. This would entail more detailed studies of the area, in which scenarios are analyzed which take into account much larger earthquakes in the area surrounding the dam. Further, regarding reservoir-triggered earthquakes, it can only be considered appropriate that the dam be built to withstand the largest conceivable reservoir triggered earthquake. Since poor record keeping precludes an accurate estimate of the size of this earthquake, a prudent approach is to take the largest reservoir-triggered earthquake ever experienced in the world and build the dam to these specifications.

Finally, there are continuing concerns regarding the appropriate management of the Cahora Bassa Dam for which the seismic records were not available to the team charged with undertaking the seismic assessment for Mphanda Nkuwa. These records were reported to not exist. This is in violation of international best practice which stipulates the necessity for ongoing monitoring of seismic activity in the area surrounding large dams.

Large dams cost enormous amounts of money, and given the potentially disastrous consequences of dam wall failure, the relatively small increases in spending on safety cannot be refused. Further, the difficulty in accurately estimating earthquake magnitudes in Mozambique should serve as a wake up call to dam designers to design dams which can withstand larger earthquakes than are currently expected.

A less prudent approach could place the lives of millions of people at risk, and as such must be considered thoroughly unacceptable.