Hirakud dam causing rather than preventing Orissa floods?

By Ranjan K Panda

The Hirakud dam was originally conceived as a flood control measure on the Mahanadi. The rule curve or storage level was revised in recent times to prioritise the needs of irrigation, industry and power-generation. With serious consequences for flood control in Orissa. Are economic compulsions being put before human lives and livelihoods?

preventing Orissa floods

Orissa is reeling under yet another flood -- called the 'Super Flood', the likes of which has not been seen in recent times. Nineteen of the state's 30 districts are affected. Initial calculations by the state government reveal that almost 4.5 million people -- more than 11% of Orissa's total population -- have been directly and significantly affected. Crops on 4.78 lakh hectares of land -- nearly 7.5% of total cultivable land -- have been destroyed. At least 68 deaths have been reported so far. More than 2,900 km of road have been damaged. Thousands of affected people are still desperate for food.

The government is claiming that the flood was unavoidable. It tries to draw a parallel with the Kosi river which recently flooded Bihar and alleges that the Centre has been unfair to Orissa in the allocation of grants and assistance. While there is near unanimity that the floods have caused unprecedented damage, there are not many takers for the argument that the flood was inevitable or unavoidable.

Water management experts point out that the Hirakud reservoir may have turned a relatively moderate flood into a mega flood, underlining once again that dams are not the answer to managing floods, and that ignoring climate change could place a very heavy burden on the state and its people.

Hirakud in the eye of the storm

The Hirakud reservoir -- one of the largest in India -- was the first major dam to be built in the post-Independence period and the only dam on the river Mahanadi in Orissa. It was primarily conceived as a flood control measure for the area downstream of the Mahanadi, the river delta in particular. The initial project report read: "The aim of the Hirakud dam is primarily flood control; irrigation and power-generation are incidental."

Though the dam was built to control floods, it has never been very effective at this. Its water spread reaches 746 km2 when the reservoir is at its full reservoir level (FRL) -- that's more than half the size of the entire Delhi area. Despite being so large, however, it was able to intercept only 16% of total inflow into the Mahanadi. Hence, the dam's flood control capability has always been suspect.

The government continues to claim that the Hirakud dam has succeeded in moderating 24 out of 30 potentially big floods. But a contributing paper to the World Commission on Dams says: "India's Hirakud dam was first justified in the name of flood control, yet extreme floods in the Mahanadi delta between 1960 and 1980 were three times more frequent than before the Hirakud was built." Experts now believe that floods have increased many times over since then. They argue that the rate of flood occurrences in the period referred to in the report was minor compared to what is happening now.

They have grounds for such allegations. Four out of the eight years in this decade have seen severe flooding in the Mahanadi river system, and on at least three occasions fingers have been pointed at the Hirakud reservoir.

The Hirakud reservoir would have succeeded in managing floods to a large extent had flood management been the dam's only priority. "Nowadays, earning profits from the dam has become the major priority of the dam; saving life and property from flooding has been relegated to the last," alleges Bimal Pandia of Water Initiatives Orissa, a civil society group that works with water and climate change issues. "Engineers have become economists these days. Earlier they justified the dam for its flood control potential, but now they try to justify the dam for the billions of rupees that the reservoir purportedly earns by providing irrigation, generating electricity, facilitating industrialisation, and property saved by controlling floods," he adds. A retired engineer-in-chief in Orissa argues that the dam contributes an average of Rs 1,700 crore annually. It's a different matter that he sharpened his financial skills by working with the World Bank after retiring from his government job. And that these calculations fail to take into account the negative impact of the reservoir including the devastation from floods that the dam has allegedly magnified. Initial estimates of the flood damage have already exceeded Rs 2,500 crore. This must be added to the dam's so-called 'contribution'.

Relegating flood control priority

Experts allege that the Hirakud dam -- in its 50 years of existence -- has undergone a sea change in terms of priorities, particularly within the last two decades. The 'rule curve' -- that prescribes the upper limit of water in the Hirakud reservoir during the monsoon months -- has become the focus of discussions after the flood. The 'rule curve' was framed to keep a balance between the dam's two contradictory objectives - containing less water to control floods and greater water storage for irrigation, power-generation etc. The 'rule curve' suggests a minimum and maximum water level in the reservoir on different dates from July 1 to October 1 so that there is enough space in the dam left to moderate floods, and ensure that the reservoir is full by October 1.

While some experts have criticised the dam managers for not following the 'rule curve', others believe that the 'rule curve' hasn't got its equilibrium right and is tilted against the objective of flood control. Himanshu Thakkar of South Asia Dams, Rivers and Peoples Network squarely blames the dam authorities for not following the 'rule curve'. But Water Initiatives Orissa maintains that the dam authorities have not violated the 'rule curve'; instead the 'rule curve' has become so obsolete in the present context that it invites more floods than it controls.

Earlier, when flood control was the first priority, the dam was kept relatively empty until August and was then slowly filled to full reservoir level by November 1. But for many years the dam failed to attain its full level, hampering both irrigation and power-generation. So the Orissa government accepted the 'rule curve' suggested by the Central Water Commission (CWC) which recommended that instead of waiting for the reservoir to fill up by November 1, it should be full by September 1. The CWC suggested extremely high levels of water in the month of September. The 'rule curve' suggests a water level between 619 and 627 feet in the beginning of September. That's between 11 and 3 feet lower than the reservoir's FRL of 630 feet. This means that the reservoir cannot be kept empty by more than 30% of its total utilisable storage even at the start of September which is peak monsoon season. Different districts of Orissa receive 17-21% of their annual average rainfall in the month of September.

The 'rule curve' has therefore been undermining the dam's flood control objective since 1988 -- something that was clearly evident this year. When the first warning of an intensifying low pressure area came on September 13, the dam was only 3 feet less than full. Three feet of storage is nothing for the reservoir's vast catchment area; water levels in the reservoir can rise by three feet in just a few hours. That's precisely what happened between September 17 and 18. The level in the dam rose by nearly 3 feet even with a moderate inflow of around 1.5 lakh cusecs of water. So when it started raining heavily, increasing inflow into the reservoir to 7 lakh cusecs, the dam authorities were forced to release more water than was entering the reservoir. By then the priorities had changed to save the dam itself. The release of huge amounts of water caused havoc by aggravating the already heavy flow in the Mahanadi due to incessant rain.

This is not the first time the dam authorities have been forced to safeguard the dam itself. In 2001, the reservoir exceeded its peak in its effort to control floods. In the end the dam released more water than was flowing into it, causing tremendous flood damage downstream.

Over time, therefore, the Hirakud dam has been causing more floods than it has been moderating.

There are a number of reasons for this. Water levels in the dam are now not managed by the dam authorities at the dam site but by a team chaired by the principal secretary of the department of water resources at the head office in Bhubaneswar. There are allegations that this leads to complications and wastage of crucial hours. Ex-minister Trilochan Kanungo says that the dam managers may justify the fact that they are still working within the prescribed 'rule curve' limit. "But how can they justify their inaction in not emptying the reservoir to create more flood cushion even after learning about an intensifying low pressure?" he asks. Prima-facie evidence suggests that the dam management ignored the opportunity that was available to them. "From September 13 to 16, the dam management either kept the status quo or accumulated water storage. By then there was sufficient scope to empty the reservoir before below-stream flooding began from September 18. The high floods could have been totally avoided at least in the delta region," argues Pandia.

The table below explains this further. Arun Kumar Upadhyay, a senior IPS officer who is still in service, terms the dam management's inaction "criminal negligence" and has demanded that a case be filed against them for the "crime" they have committed.

Table: Precious opportunity to empty the dam was lost between September 13 and 16

Date Status of weather/prediction Hirakud water level (ft) Trend of water level in reservoir
September 14
Mostly dry weather    
September 13 Initial information on low pressure formation    
September 14 Information that low pressure was getting stronger 627.02 Same
(inflow: 34,393; outflow: 34,393)
September 15 Prediction that low pressure could take the shape of a cyclone and heavy rainfall within the next few days Rainfall starts in coastal districts Forecast of heavy-to-very-heavy rainfall on the 16th and 17th   Same
September 16 Heavy rainfall. Prediction that the low pressure formation would cross the shore by evening. Heavy rainfall predicted for the 17th. Rainfall was not particularly heavy; Paradeep recorded the highest rainfall (89mm)   Increasing
(inflow: 43,108; outflow: 30,463)
September 17 Low pressure intensifies. It is still centred at 50 km east of Chandbali

Highest rainfall recorded again at Paradeep. Heavy rainfall all over Orissa. Though many rivers in south Orissa flowed over the danger mark, the Mahanadi was still flowing below the danger level but was rising rapidly owing to floods in the Tel, Ong and other rivers below the Hirakud

627.81 Increasing
(inflow: 146,341; outflow: 25,636)
September 18 Flow in the Mahanadi growing. Fear grows that 1982 might be repeated. Danger level crossed at Jobra. 14,64,000 cusecs was flowing at Mundali in the evening. Jujumura in western Orissa received 355 mm of rain 629.97 Same
(inflow: 390,000; outflow: 390,000)
September 19 14,86,000 cusecs flow through the Mahanadi. River is rising. Fear that by midnight, 15,82,000 cusecs of water will be flowing 629.62 Increasing
(inflow: 504,000; outflow: 460,000)
September 20 No heavy rainfall reported in the state. But water flow at Mundali peaks to 1982 levels. 15,82,000 cusecs flowing. Ferocious flood causes breaches all over the delta 629.87 Increasing (inflow: 740,000; outflow: 702,000)

At this time, 40 gates were opened after six gates were closed to ease the flood situation in the delta

September 21 Floods continue as excess water discharged from the Hirakud reaches the delta (Note: It takes around 24-34 hours from Hirakud to Mundali) 628.96 Same
(inflow: 332,610; outflow: 332,610)
Source: Compiled from the official website of DoWR, Government of Orissa, and newspapers

The crime of neglecting climate change

There is no denying that the dam managers clearly ignored or undermined the meteorology department's prediction that heavy rainfall would occur in Orissa and in the Hirakud's catchments. But even this pales in comparison with the fact that we have largely ignored the issue of climate change. The State Water Plan (SWP), prepared by the government in 2004, makes only passing reference to "variable runoff" with regard to the Hirakud and flood vulnerability. It observes that "variability of monthly rainfall is increasing, which means that rainfall is concentrated in a particular period". The SWP admits that the "volume of flood events is increasing; the 2001 and 2003 floods are examples".

Many notable changes have been observed with regard to rainfall, river flow and flooding over the past few years. The number of rainfall days is decreasing. A study by noted atmospheric scientist Professor U C Mohanty reveals that the number is dropping by one every five years. On the other hand, the number of days of very heavy rainfall has considerably increased. Now even the Tel and Ong rivers -- both tributaries of the Mahanadi downstream of the Hirakud -- cause severe flooding in the Mahanadi. New records of very heavy rainfall are being reported in areas below the Hirakud catchment. In 2003, Kalahandi recorded 473 mm of rain in less than 20 hours. This year too, heavy rainfall was reported in many parts of Orissa causing flooding in the Tel, Ong and other rivers. Below-stream flood vulnerability has therefore increased substantially. In 2003, 27 of a total of 30 districts were affected by floods. This year's flood affected 19 districts of the state, many of which have previously been unaffected by major flooding.

Changes in rainfall and water utilisation patterns have brought about significant changes in river flow both upstream and downstream of the Hirakud. But the dam authorities have only taken partial note of these changes. Because the number of rainfall days has declined and the monsoon began retreating earlier than usual, the CWC and dam management decided to keep the reservoir at its FRL much before the normal retreat of the monsoon. This decision was aimed at mitigating water scarcity, but it ended up aggravating the flood risk. The dam managers did not take cognisance of the fact that heavy rainfall could still occur during the months of September and October, and that the dam should stay prepared. As they say, a little knowledge is dangerous. In this case, partial knowledge led to disaster.

Lessons still not learnt?

Engineers admit that the Hirakud reservoir is incapable of moderating floods with the present management model. But they express their helplessness. "It's an irony that the Hirakud reservoir pledges a fixed portion of its water to industries but it has not earmarked any space for flood control," says octogenarian social activist Professor D P Nayak. "That tells us where their priorities lie," he adds. The SWP too mentions that the Hirakud is expected to control flooding, yet "there is no storage earmarked for flood control. The FRL and maximum water level are kept the same (630 ft RL)".

The Orissa government appears to have neither learnt from the past nor been updated on the latest potential dangers. The government has not taken note of the serious flooding that has occurred in half of the eight years of this decade. All the flooding has been caused by very heavy rainfall downstream of the Hirakud, and later been aggravated by flood water release from the dam. When a flood occurs, politics reigns supreme and the core issues are pushed under the carpet. The government cannot seem to look beyond compensation, grants and visions of building more dams. This year too, proposals for new dams in Manibhadra, Salebhata, etc, have surfaced despite the Hirakud proving time and again that major dams are not a panacea for all evils. Construction of new dams will create further social and ecological problems. The government must learn to think beyond them.

(Ranjan K Panda is an Orissa-based researcher and writer)

InfoChange News & Features, October 2008