Dams and earthquakes
The theory that reservoir-induced seismicity led to the Koyna earthquake 32 years ago is still used by anti-dam activists.
THIRTY-TWO years after an earthquake that measured 6.5 on the Richter scale rocked the Koyna region in Maharashtra, myths and hypotheses about its origin and cause abound. The most dominant among them is the contention that the dam at Koyna had made the
region seismically unstable.
Before the earthquake, which occurred on December 11, 1967, Koyna was considered largely aseismic. Therefore the quake not only caught everyone by surprise, but generated a lot of intellectual grinding among the scientific community. Besides earthquakes,
reservoirs, lakes and dams became subjects of debate and controversy.
Prior to the Koyna earthquake, geophysical and geotechnical studies had supported the assumption that peninsular India was seismically stable. The only avilable record of an earthquake in the peninsula was that of the Coimbatore earthquake on February 8,
1900, which measured 6.0 on the Richter scale. The Koyna earthquake shook the theory about the seismic stability of peninsular India.
The earthquake caused some repairable damage to the Koyna dam. In the absence of any scientific theory, some people tried to attribute the maiden earthquake in the Sahyadri mountain range of Maharashtra to the dam and the reservoir. That the epicentre wa
s located in the vicinity of the Koyna dam gave rise to speculation that the reservoir was the sole cause of the earthquake. Although the theory of reservoir-induced seismicity (RIS) had no scientific backing, people bought it for want of another plausib
The theory had its critics too. The late Dr. K. L. Rao, an engineer and a scholar, compared the Koyna reservoir sitting on the earth to a fly sitting on an elephant. He dismissed the theory saying that the fly's weight could not disturb the elephant. Ant
i-dam lobbyists, however, are still not convinced.
In the years that followed the quake, the damaged portions of the dam were repaired. Reservoir capability and power generation were restored to pre-seismic levels.
However, the RIS theory has done irreparable damage to new dam projects. Protests against the construction of new dams and reservoirs such as the Tehri and Narmada projects have stemmed from this misconception.
The earthquake in Latur on September 29, 1993 provided an opportunity to anti-dam campaigners to raise the bogey of RIS simply because a small dam about 10 metres in height stood on the Terna river, 20 km away from the site of the quake. The dam was cite
d as the cause of the quake.
The period after 1967 witnessed a few earthquake of magnitude ranging from 6.0 to 6.5 on the Richter scale, the first one being the 6.5-magnitude quake near the Bihar-Nepal border in August 1988. This was followed by an earthquake of magnitude 6.5 in Utt
arkashi in October 1991; it occurred in the vicinity of the Tehri dam. Another was the May 1997 earthquake at Jabalpur, in the vicinity of the Narmada ravine. The RIS theory, as applied to these quakes, did not provide any conclusive proof that there exi
sted a co-relation between them and the dams. The theory was dismissed as a conjecture and the picture it presented was described as nebulous.
Activists protesting against the construction of dams on the Narmada in (Gujarat and Madhya Pradesh), the Tehri (Uttar Pradesh) and Bedthi (Karnataka) and their supporters argue that dams would lead to RIS and subsequently earthquakes. The Uttarkashi and
Jabalpur earthquakes are often cited to support their arguments. But the fact of the matter is that these quakes made the area less vulnerable to further seismic shocks, and therefore it is a safe bet for dam projects.
BY SPECIAL ARRANGEMENT
The Koyna dam.
A DETAILED analysis of seismic data reveals a sequence in the occurrence of earthquakes. First, most earthquakes occur along an established geological or tectonic fault. Second, earthquakes have a certain frequency and a return period. If the return peri
od is, say Y years, then an earthquake of similar magnitude is likely to occur after Y years anywhere along a fault line. The seismic gap perpetually floats on the fault and wherever the situation is seismically suitable, a quake is triggered. Once an ar
ea experiences a quake, its epicentral volume is crushed and it becomes unable to accept any further geo-technical stress. This means that once an area has experienced an earthquake, it will not experience another earthquake of a similar magnitude for a
long time. The time gap is usually about three to five times the return period of the earthquake. Usually the return period of medium-sized earthquakes (of magnitude ranging from 6.0 to 7.0) is of the order of a few decades. For powerful earthquakes (wit
h magnitude higher than 7.0), the return period is usually 100 years or more. Hence the possibility of occurrence of one earthquake immediately after another at the same location or in its vicinity is almost nil. The area may experience earthquakes of sm
aller magnitude, known in the seismic lexicon as micro-earthquakes. The magnitude of a micro-earthquake is less than 4.5 on the Richter scale, and they do not cause any evident damage.
Geo-technical engineering as a subject has developed considerably in the past three decades. There has been remarkable development in the fields of design, research and construction of dams. India is one of the world leaders in this area. India is capabl
e of designing and constructing a dam that would withstand a seismic jolt. Institutions such as the Structural Engineering Research Institute, Chennai; the Earthquake Engineering Department and the Building Research Institute at Roorkee have made signifi
cant contributions to aseismic designing of dams. It is possible to simulate with the help of a computer various modes of vibrations of the dam (or any construction) to different levels of ground accelerations. In addition, laboratory models of dams, as
high as six metres, could be subjected to simulated seismic vibrations on a vibration table. Very few countries in the world have large-sized vibration tables. In India, Roorkee University has one.
It could perhaps be argued that despite all these scientific and technological developments, an element of seismic risk could cast a shadow on dam construction. But this anticipation of risk should not stand in the way of progress, especially in a develo
ping country like India. The country needs water and electricity to provide its people good living standards. Hydropower is the solution to the country's requirements, and this can be achieved by storing water in dams.
As of now, there are about 23,000 large dams in the world. A large dam is defined by the International Congress on Large Dams (ICOLD) as one which has a height of 33 metres and above. Of these dams, only four, namely Koyna (India), Kremesta (Greece), Kar
iba (South Africa) and Singfenkiang (China), have experienced earthquakes of a moderate magnitude of between 6.0 and 6.5 on the Richter scale within a few years of building. The data available from these four earthquakes did not compare scientifically or
statistically with regard to the rest of the large dams, disproving the claim of the RIS lobby.
Some seismologists have published a list of about 100 cases of RIS. These cases show that after the completion of a dam, the reservoir area experienced earthquakes of microlevel magnitude - 2.0 or 3.0 on the Richter scale. But a close scrutiny makes it c
lear that such observations are deceptive. These earthquakes, of microlevel magnitude, may not be considered earthquakes in the conventional sense. These are minor fractures, which occur routinely. Only seismological instruments "feel" such earthquakes.
The way an open-ended logarithmic Richter scale functions can be understood by a simple experiment. If a standard brick (5 cmx10 cmx20 cm) is dropped from a two-metre height, the impact of its fall on the ground is almost equivalent to the magnitude of a
n earthquake at that point.
One more factor has significantly contributed to the RIS. Earlier, the deployment of seismological instruments to monitor seismicity was widespread. Now most networks are usually established within a radius of between 20 km and 30 km. All these arrangeme
nts have increased the detection potential of seismological instruments. As the Richter scale is logarithmic in nature, the instruments record and report earthquakes down to magnitude -2 or less. The instrument can record any number of earthquakes in a m
onth. Unfortunately, advanced seismological measurement and recording has paradoxically increased the number of cases cited as instances of RIS. The anti-dam activists are citing the increased number of micro-earthquakes to further their campaign.
If construction of a dam is harmful, as argued by the anti-dam activists, then why are they agitating only against the new dams? Why do they not start another movement demanding the removal of existing dams? The Bhakra dam, located in the foothills of th
e Himalayas, has not reported any cases of RIS in the last four decades of its existence even though the Himalayan ranges form part of the active Alpine Himalayan seismic belt.
The earthquakes that the Indian peninsula has so far experienced should make one wiser and help shed all prejudices and superstitions regarding their occurrence. Instead of attributing an earthquake to a nearby dam, efforts should be made, using all avai
lable facilities such as the computational technique, remote sensing, seismic tomography, seismic data from numerous observatories and data from geo-stationary satellite, to understand the mechanism behind earthquakes. Construction of dams should be done
in such a way as to withstand seismic vibrations. Raising the bogey of RIS repeatedly should be avoided. At present, the seismic code requirements for aseismic design and construction are recommendatory in nature. The code requirements will become more
effective if they are made mandatory.
Dr. Arun Bapat is a senior research seismologist.