Collapse of St. Francis Dam in California:
St. Francis Dam was a concrete gravity-arch dam, constructed between 1924 and 1926 to create a reservoir to supply water to the city of Los Angeles. The 59 meters high dam across San Francisquito Canyon, 64 km north-west of Los Angeles, created a reservoir of a storage capacity of 47 million cubic meters. Soon after the construction of the dam, throughout 1926 and 1927, cracks kept appearing in the dam and its abutments. Moreover, some of the cracks started leaking muddy water. On the 7th of March 1928, as the reservoir was filled to capacity for the very first time, more cracks appeared. Only five days later, on the 12th of March 1928, thee minutes before midnight, the dam collapsed, and 45 million cubic meters of water that formed a 38 meters high wall, rushed down the canyon destroying everything on its path. The flood-wave raced south to the Pacific Ocean at Montalvo, 87 km away. The flood was 3 km wide, travelled at a speed of 8 km per hour as it reached the ocean at 5.30 am. The floods killed more than 600 people. The dam had broken into several large pieces. The dam’s failure is attributed to the unstable basement upon which it stood, and poor design of the dam, which was excessively tall without adequate support.
Collapse of Banquiao and Shimantan dams in China:
In 1950, immediately after a severe flooding in Huai River Basin, the Government of China undertook a programme to control the Huai River i.e. the Huang He or the Yellow River system by constructing a series of dams for flood prevention. The secondary purpose of the dams was irrigation and power generation. Under this programme two major dams, the Banquiao Dam on the Ru River and Shimantan Dam on the Hong River, in addition to a large number of smaller dams were constructed. The storage capacity of Banquiao Dam was 492 million cubic meters, out of which 375 million cubic meters were reserved to impound flood waters. The height of the dam was 116 meters. The Shimantan Dam had a storage capacity of 94 million cubic meters, out of which 70 million cubic meters were reserved for flood waters.
The initial construction of Banquiao Dam had some design flaws that led to the appearance of cracks in the dam as well as the sluice gates. Following the advice of Soviet engineers both the dams were reinforced and expanded, and were termed ‘iron dams’, which meant that the dams could never be broken; pretty much like the Titanic was termed ‘unsinkable’. Banquiao Dam now stood to protect ‘once a thousand year flood’, which meant a flooding resulting from a downpour of 53 cm of rain over a period of three days. Shimantan Dam was ready to protect ‘once a five-hundred year flood’, which meant a flooding resulting from a downpour of 48 cm of rain over a period of three days.
After the construction of these two major dams, a large number of smaller dams were constructed. Initially the smaller dams were confined to the mountains. In 1958, however, Chinese Vice Premier Tan Zhenlin, against the advice of the hydrologist Chen Xing, decreed that the smaller dams be constructed on the plains in order to give priority to irrigation. This arbitrary shift in priority by the orders of the politician, despite the protests of the technocrats, led to the construction of a huge reservoir on the plains. The hydrologist Chen Xing was banished to a remote part of China, as a punishment, and experienced further humiliation in 1961, when he was labelled ‘a right-wing opportunist’, and was purged.
In the beginning of August 1975 an unusual weather pattern caused a set of storms that poured about one meter of rain within a space of three days. On the 5th of August, the first storm dumped 45 cm of rain, which was 40% higher than the previous recorded rainfall. The second downpour of 16 hours came on the 6th of August, and was followed by the third downpour that lasted 13 hours on the 7th of August. Banquiao and Shimantan reservoirs that were designed to handle a maximum rainfall of 50 cm over a period of three days were filled to capacity by the 8th of August 1975. A little after midnight (12.30 am) flood waters rose 40 cm above the crest of Shimantan Dam, and it collapsed. Within 5 hours 120 million cubic meters of water went rushing down the Hong River. After half-an hour, around 1 am, flood waters rose above the crest of Banquiao Dam, and it collapsed. A six meters high wall of flood-wave with a twelve km wide front, rushed down the Ru River emptying over 600 million cubic meters of water. The flood waters raced down at a speed of about 50 km per hour, comparable to the speed of fast moving cars on a high-way.
The collapse of Banquiao and Shimantan dams led to a catastrophic collapse of 62 dams downstream. The flood waters ravaged more than a million hectares of farmland spread over 29 districts. According to the Hydrology Department of Henan Province of China, 26,000 people died from flooding, 145,000 died during subsequent epidemics and famine. The actual number of dead was probably much higher than reported. About 6 million buildings collapsed and 11 million people were affected.
The hydrologist, Chen Xing, who had opposed the dam building plans on the plains, was rehabilitated instantly. He was rushed to Beijing to urge the communist high-command to order the bombing of many of the smaller dams for the flood waters to drain.
Salient features of Hirakud Dam:
Hirakud Dam, which is a composite structure of earth, concrete and masonry is massive, and is the longest dam in the world. The main concrete dam that spans two hills or dungris, is 195 meters high and 4.8 km long, and is flanked by earthen dykes on either side. The overall length of the composite dam is 25.8 km that has created a reservoir of 743 square km at full capacity. This reservoir is the largest artificial lake in Asia with a 640 km long shoreline. The original storage capacity of the reservoir was 5,818 million cubic meters, which in 1988 was revised to 5,375 million cubic meters, and now in 2008 is probably around 4,800 million cubic meters. This 20% or so reduction in the storage capacity is due to accumulation of sediments in the reservoir. The dead portion of the storage capacity, i.e. water that cannot flow out of the reservoir thus is not available, hence technically dead, is 1,800 million cubic meters. The average annual inflow of water to the reservoir, fed by the two rivers the Mahanadi and Ib, is 37 million cubic meters.
What can trigger a collapse of Hirakud Dam:
Unlike St. Francis Dam of the US that collapsed in 1928, Hirakud Dam stands on a geologically stable foundation. Moreover, Hirakud is located in seismically stable peninsular India. It is however, a myth that peninsular India is absolutely free from earthquakes, and let it be known that an earthquake can strike any part of this earth.
On the 11th of December 1967 there was a strong earthquake of magnitude of 7.5 in the Richter scale in Koyna, 241km from Mumbai, located very much in peninsular India. The Koyna earthquake killed two hundred people, injured over a few thousand people, and rendered a good few thousands homeless. The epicentre was within 5km of Koyna Dam, which developed large cracks. It is possible that Koyna Dam is built in a moderately seismically active area, and was hit by a natural earthquake. It is however, equally possible that waters impounded in the reservoir of Koyna Dam triggered the earthquake. On the 13th of March 2005, Koyna again was the epicentre, this time of a moderate earthquake of 5.1 on the Richter scale that was felt in far away Bombay.
This hypothesis of reservoir induced seismicity has been proposed, among others, by an eminent seismologist of worldwide fame Dr Harsh K Gupta, who I had the good fortune of working with in the Ministry of Science of Technology of the Government of India. Dr HK Gupta, a Fellow of the National Academy (FNA) subsequently became the Director of National Geophysical Research Institute (NGRI), and retired as the Secretary of the Ministry of Ocean Development of the Government of India. He was the founder and the president of the Asian Seismological Commission and currently is the president of the Geological Society of India. The erudite views of Dr HK Gupta can not be ignored. We, therefore, must take into consideration the possibility of the 4,800 million cubic meters of water and the 1,000 million cubic meters of sediments impounded by Hirakud Dam triggering an earthquake. An earthquake could develop cracks in the dam, and may even cause an immediate collapse of Hirakud Dam.
I have not come across reports of any detailed investigation of sediments that have accumulated on the floor of Hirakud reservoir. If we accept the guesstimates of 20 to 25% of the original capacity of the reservoir now occupied by sediments, there could even be a possibility of a mudslide taking place. Depending upon the bottom topography of the artificial lake, and the location of the mudslide, a tsunami could be triggered in Hirakud reservoir and that could smash against the dam causing a collapse.
A very heavy downpour in the upper reaches of the Mahanadi and the Ib, at a time when least expected, could rapidly fill the reservoir. If water per chance starts flowing above the crest of the dam, it could collapse, as it happened in China.
Let the top priority ‘flood-control’ be respected:
Ever since the conception of Hirakud Dam, its primary function has been flood-control. Let us all be very clear about the top priority of Hirakud, it is flood-control; power generation even irrigation are only secondary functions. It is therefore, absolutely essential that before the onset of monsoon rains, the reservoir must be kept sufficiently empty to have a higher capacity for receiving the runoff from the upper reaches of the Mahanadi. Unfortunately the pre-monsoon period, when the reservoir should be kept empty, is the peak of summer with maximum demand for electricity. So the political masters, in order to appease their electorates often interfere in the management, forcing the engineers to keep the reservoir sufficiently full rather than empty, so as to have enough water to continue to generate hydro-electric-power. 9 out of 10 such political interferences may not cause any serious problems. It is however, possible that in the past when on 6 out of 30 occasions the sluice gates were forced open, as a last resort to protect the dam, releasing substantial volumes of water to the already overflowing Mahanadi drainage basin, thus exacerbating the flood, the reservoir was sufficiently full because of political interference. The process of management of Hirakud Dam must be respected, and there should never be any interference by the politicians and bureaucrats. In the best interest of Hirakud and the people of Orissa, let the technocrats manage the dam and the reservoir.
Policy for disaster management:
With global warming and climate change, un-seasonal rains will become more common, and the intensity of the storms and down pours will increase. Due to the increase in populations in the towns and cities along the Mahanadi River system, loss of human lives will be greater as years roll by, in case of a collapse of Hirakud Dam. Moreover, as the dam gets older, chances of it collapsing increase. Depending on the time of the year, the lunar cycle, and the fullness of the Mahanadi River system, the floodwaters from a collapsed Hirakud could rush down at a speed anywhere between 10 to 50 km per hour. They could overwhelm Cuttack within four to eighteen hours, and could reach the sea in six to twenty-four hours. Townships on the banks of the Mahanadi like Sambalpur, Sonepur and Banki will be utterly devastated, so will Cuttack be, and all the other habitations on the Mahanadi delta between Cuttack and Paradip.
As of today I have not come across any public policy document for disaster management in case of a catastrophic collapse of Hirakud Dam. A detailed study must be undertaken to determine the likely path of the travelling wall of water from Hirakud, and the length of time it will take to reach the various towns and cities along the Mahanadi River system. A contingency plan that includes an advanced warning system should be in place to warn the townships, urging mass evacuations. Each and every habitation, starting from tiny villages to townships to large cities along the Mahanadi River system should have information available as to the shortest and the safest routes for mass evacuation to safety. A public education programme for the entire population along the Mahanadi River about the dangers of a dam collapse must also be administered.
Important aspects of management of Hirakud:
The very first step in the proper management of Hirakud requires acquisition of an accurate body of knowledge on the capacity of the reservoir. The reservoir capacity of 5,818 million cubic meters at the time of the commissioning of the dam in 1957 is no longer correct as vast quantities of silts and sediments flowing into the reservoir steadily for the last fifty years have substantially reduced the capacity. I have come across some guess estimates of 20 to 25% reduction of the capacity, which could well be the case, but I am not aware of any authoritative study on the subject. Calculation of the capacity requires the construction of a three-dimensional picture of the reservoir that starts with the profile of the bottom. A simple-minded assumption that the bottom of the reservoir is flat is incorrect; in fact it could be anything but flat. The floor of the reservoir could be criss-crossed with numerous channels and may possess complicated topography hosting many interesting sedimentary and hydrological features. Only a detailed investigation with the help of a side-scan sonar can provide an accurate profile of the bottom. A side-scan sonar survey is not particularly difficult (not easy either!), requires a boat to anchor the sonar and a few pieces of equipment, and a lap-top computer to record the bottom profile. A detailed side-scan sonar survey of Hirakud reservoir must be conducted as soon as possible. Moreover, it is important to remember that the bottom profile will keep changing virtually every year after the monsoonal influx. Therefore, survey of the bottom profile may become necessary to be carried out on an ongoing basis.
If the results of side-scan sonar survey reveal the presence of unstable sedimentary deposits which are in danger of toppling, leading to mudslides, dredging will become inevitable. Cutter-section dredging could be carried out, which could also be used either to reinforce the earthern dykes, if necessary, or to land-scape some of the islands within the reservoir for better uses. Bed leveller dredging, which is comparable to the operations of a bulldozer on land, could be done to flatten the floor of the reservoir to eliminate the dangers of a mudslide.
Knowledge of the profile of the reservoir floor when combined with the information on the depth of the water body (bathymetry) will provide accurate information on the capacity of the reservoir. Moreover, knowledge of the amount of water held in the reservoir at any given point in time is crucial for management of the dam. Monitoring of suspended load and sediment load entering the reservoir on an ongoing basis is also important. Results of such a study could be used to design and construct structures to arrest the silt or at least to reduce the amount of silt entering the reservoir.
Monitoring of the seismic activities around Hirakud on an ongoing basis is necessary as it could point out seismic stress in the region, which will help in taking necessary precautions to ensure the safety of the dam. An array of seismometers, installed on the concrete dam could record the seismicity. Another significant aspect in the management of the dam, in this era of terrorism, is round the clock water-tight security operations to ensure that no terrorists ever damage the dam. In 2008, Hirakud is an aging dam comparable to a middle-aged human body that requires proper maintenance. It is needless to say that the engineering structures of Hirakud need proper maintenance, on an ongoing basis.
I must record here that the Government of Orissa have now in 2008 initiated a programme for some limited dredging of the reservoir. Orissa Construction Corporation, the agency entrusted with the dredging, aims at removing some 0.7 million cubic meters of sediments. Although this dredging is limited in scope, designed essentially to facilitate the flow of water in the canal for irrigation, it is a step in the right direction none the less. Hopefully much more work will be carried out in the near future for the management of Hirakud.
I conclude the article by stating that the dam collapses in China killed hundreds of thousands of people; and if Hirakud collapses due to poor management, the devastations will be unprecedented; the Mahanadi delta could be wiped clean. Unfortunately as the dam gets older, the possibilities of a collapse increase.
(Author is an internationally known Geologist now working at Ravenshaw
University, Cuttack (Orissa) as Dean, Administration)