Wednesday, August 27, 2008

Kosi Breaches Embankment, More Water Refugees In Bihar

I wrote sometime back that although there are been much focus in the media and also concerns raised in the National Action Plan on Climate Change (15 mb) about sea-level rise and coastal populations being displaced, the interior of the country will also see its share of water refugees as climate and environments change. A couple of months ago, a prolonged drought in the Bundelkhand region whose effect was amplified by decades of neglect of water conservation measures lead to a massive dislocation of people in the affected areas. Now, Bihar is witness to another wave of people fleeing a water crises, this time a near catastrophic flood of the Kosi river.

Nearly 2.5 million people living in the Kosi flood plains have lost everything, as the river broke through its constraining embankments near Kusaha village in Nepal on August 18 and flooded several districts in Bihar. See map below for a synoptic view.

As you can see the Kosi forms an alluvial mega-fan, a massive triangle shaped body of sediment which you can distinguish from adjacent environments due to the closely spaced meandering network of channels between the present day Kosi river and Basantpur to the east. Such sedimentary deposits form when rivers suddenly change gradients as the Kosi does when it flows out of the Himalayas into the Indo-Gangetic plains. This fan is enormous, about 180 kms long and 150 km wide. Westward tectonic tilting of the entire area has been the major controlling factor for the river shifting its course about 120 km from east to west over the last 250 years. I have labelled the various paleo-channels using this study as a rough guide.

About 10-15 million years ago rivers bringing sediment from the rising higher Himalayas formed similar alluvial fans in a depression where today the Siwalik mountains stand. The Siwaliks are made up of sediments of such ancient alluvial fans that - as the Indian plate pushed against the Asian plate - have been consolidated, hardened, compressed, deformed and uplifted to form undulating mountains. The locus of sedimentation has now shifted in front of the Siwaliks. The Himalayas are growing southwards. There is nothing more awesome than the geological forces that give shape to this dynamic planet.

The Kosi has breached the embankments before but they have been downstream of the barrages and canals that are used as safety valves. Engineers have been able to control floods by diverting water to the canals. But this breach occurred upstream of the barrage and canal system giving engineers no way of controlling the excess flow. The river has occupied one of the abandoned channels to the east of the present day channel as seen in the image below taken on August 24 2008.

Source: NASA Earth Observatory

I couldn't make out exactly which paleochannel from the image but the district of Madhipura has experienced the worst flooding, so could be one of the channel close to that village. It will be months before the embankment is plugged and the river returns to its original course. Meanwhile hundreds of thousand of people have become water refugees. I don't know how many will go back. It is possible that the topography and recognizable landforms will change significantly to make identifying the tiny plot of land that is farmed a messy process. In short the future looks bleak for large number of people especially those with no clear titles to land and there are plenty of those in Bihar.

The Kosi is such a dynamic system that people living along its flood plains will always face this risk. Embankments don't offer a permanent solution to flood control. The danger is that global warming might make the monsoons more erratic at the same time resulting in more concentrated periods of rains. Coupled with increased summer melting due to the accelerated decline of Himalayan glaciers, the situation will strain the capacity of embankments to hold the river back. It is a question of when not if till the next embankment failure.

High dams in Nepal have been proposed as a "permanent" solution. This experts say will result in better flood control. The worry is the extremely high sediment load carried by Himalayan streams. For example the life span of the Tehri Dam in the Gharwal Himalayas is now being reassessed to being only about 30-40 years as against the older estimate of a 100 years. Dams on the Kosi river will no doubt also be short lived. What will we do after the dam fills up with silt?

But then what can be done? I don't know. The problem really appears intractable for the near future. India simply does not have enough excess arable land to relocate people living in high risk areas like the Kosi flood plain. I mean where can these people go? In developed countries like the U.S. high risk areas seem to be serving as a magnet for people to settle in large numbers. Katrina has not stopped people from returning to New Orleans, and people still seem willing to relocate to Florida in the path of hurricanes and California in earthquake prone zones.

Over here people do return to areas of disaster but because they have little choice. It's either farm these plots or live in slums in big cities. Livelihoods, land use and habitation patterns will change slowly but until then the best we can do is set up better early warning and disaster management systems. This will hopefully save lives, but the problem of loss of property and livelihood remains.


  1. Flooding is not the problem, people have dealt with Kosi flooding for centuries.

    In 2006 eight social entrepreneurs in India spent months investigating clues for solutions to the tragic devastation caused by Bihar floods. They identified through extensive field interviews that, beyond levees now triggering massively bigger deluges than smaller floods of past centuries, the causes of deep human disaster include local economics and politics which distort flood control and flood relief and often take deliberate advantage of the victims. The saddest conclusion, of course, is that for years the human devastation in Bihar has been predictable, almost reliable, months in advance of the floods each year, and not because of nature and rain, but because of economics and politics.

    Their assessment reveals how flood relief is foreseen by many as a “third harvest,” with private sector and government middlemen buying, selling, and bribing rights to relief supplies months before the floods even arrive, bargaining away goods, property and even children to the sex trade in exchange for promises of flood relief or access to flood management funds. It is terrible, and it has implications for national and international media, businesses and bankers, relief agencies, citizens, and many others.

    I am eager to share this draft assessment of clues for solutions, in hopes of helping anyone trying to understand and address the deepest roots of this problem. Please contact me at Ashoka,, for a copy of the draft notes from this work if they would be useful.

  2. This comment has been removed by a blog administrator.

  3. The Mithila culture in Bihar and the Nepal Terai may be tied to the Koshi floodplain. It seems to be ambivalent about the river, which gives (fish, water, renewal of the soil with fresh silt), but also takes lives. Perhaps floods were less severe, or unfolded more slowly before there were embankments. On the other hand, this year's reported death level (under 100) doesn't justify the Kosi's nickname "Sorrow of Bihar". That name suggests much greater death tolls historically.

    Undoubtedly people *have* dealt with floods along the Kosi for centuries, but on what terms? There was a news report of a man in his 80s carried to a refuge on his son's back. This man said he was fortunate that his children loved him, because others his age were being left behind as their children and grandchildren saved themselves!

    Perhaps floods have never been benign and local people simply got used to making difficult choices. After all, crop failure used to be deadly too. Perhaps the Kosi floodplain was less vulnerable in this other respect because it had abundant groundwater and renewal of soil fertility through fresh soil deposits. The prudent subsistence farmer would locate so as to minimize risks from both factors. Despite flood risks, life along the Kosi may still be more secure than in other parts of Bihar.

  4. Dear Shriyut Suvrat Kher:I liked your presentation of the disastrous,nay catastrophic event of the dramatic Kosi River Shift of 18th August 2008. I would like you to provide an in-depth comment on a connected subject: It appears at the blogger site at
    Thanks for an informative article on why this breach has been so damaging.

  5. Quite informative…Thanks for sharing the post..
    GIS spatial analysis

  6. Interesting request today, March 2011, from India, for info underlying my comments on this blog in 2008. Point is to draw focus to possible solutions buried within the social and economic systems which create disaster from annual floods, rather than diverting attention to natural/technical aspects of the floods. Underlying hypothesis substantiated by the work of on the ground social entrepreneurs: natural floods throughout history were smaller and people developed adaptive systems. Now (recent decades) because of engineering interventions, floods are devastatingly bigger when they do come, disabling local populations from being able to adapt, and putting in place tremendous incentives for corruption and neglect at higher levels of society and government.