Responding to the Ganga-Brahmaputra Floods

Social – and not – civil engineering will help floodplain societies to benefit from floods, while reducing their vulnerability to them.

Recent years have seen growing media attention directed to the floods in the lowlands of Ganga-Brahmaputra basin – the home of one-tenth of humanity. The misery and insecurity of this large, and predominantly poor population, in the aftermath of heavy monsoons has rightly drawn sympathetic attention of the world. Looking beyond immediate relief measures, the challenge is to attain an understanding of the hydro-ecology of these floods, to study their devastation, as well as their benefits. In this task, however, environmentalists have often taken a simplistic yet alluring stand, blaming hill deforestation while ignoring other factors at work. There is need for a holistic ecological analysis of the hydrological processes of floods and the socio-biological adaptations that have evolved. There must be a more scientific and objective basis for the environmental analysis of the mega-disasters in this large basin.

A flood is defined as the subservience of land not normally inundated. Surely, one cannot describe as flood the regular inundation of lowlands from high tides. One should also not describe as flood, the inundation of floodplains during the monsoon downpour, when the rivers of the basin are normally expected to carry a volume of water several hundred times greater than, in the dry months. The behaviour of the monsoon has never been uniform, with spatial and temporal variations in the rainfall an integral part of its normal behaviour. Hydro-ecological analysis of floods must distinguish between this normal pattern of behaviour of nature from the collective, yet subjective, expectations of human beings who are increasingly colonising the floodplain and bearing the resultant risks

Matter dislodged from steep and unstable Himalayan slopes, forested or otherwise, together with water, is what moves downwards along the drainage corridors of the Himalaya. Floods in the mountains are linked to this displacement of matter. In the high Himal, the failure in the lip of glacial moraines allow the icy lake formed behind it to burst out in a violent flood. These are known as glacial lake outburst floods (GLOP). In the mid-hills, landslides regularly block rivers and then collapse to release devastating flash floods into the lower reaches. This phenomenon has periodically occurred in the Himalaya and is known among the Nepali hill people as bishyari. The water level goes down, often accompanied with fish kills, and if it is daytime and the villagers are alerted, they head quickly for high ground. Unlike in the plains, there is no lead time to prepare for catastrophe. Bishyaris and GLOF release tons of debris and sediment, which ultimately find their way to the plains and raise the level of the river-beds. They constitute one more reason for the flood prone nature of the Ganga-Brahmaputra basin.

In the hills, as long as the river gradient is steep, suspended silt, pebbles and even boulders move with the flow of the river. As Himalayan rivers emerge from the mountains and enter the Tarai plains, they encounter a very low gradient along the river bed. Jhapa, in Nepal´s eastern Tarai, is only 60 meters above sea level. Downstream, the sea port of Calcutta is due south about 500 km. This means an average gradient of only one centimeter in 100 meters! Such a low gradient not only incites rivers to overflow their banks at the slightest pretext, but forces them to part with the matter load they carry. The result is that the river beds rise continuously and make the rivers even more prone to spill over their banks. Because more matter is flowing down from the Himalaya than can be carried to the sea by the Ganga and Brahmaputra, their tributaries, such as the Kosi, have the characteristics of inland deltas with channels that shift as the river bed rises. Flooding, in such cases, is normal when judged with geological eyes against a time span of decades.

In the deltaic region, inundation of land occurs diurnally and fortnightly in the form of tides; seasonally, in the form of high river discharges during the monsoons, as well as when sudden cyclones in the Bay of Bengal whip up tidal waves. The Brahmaputra and the Ganga, generally have peak floods in August and September, respectively. In the case of a coincidence of flood peaks in these two rivers, devastating inundation can occur. If this coincidence is simultaneously reinforced by high tides or cyclonic tidal waves, flood waters will find their access to the sea impeded. Then, floods can be disastrous. This phenomenon of coincidence is probable and has occurred with regularity over decades.

Land scarcity (over 60 percent of the people of Bangladesh are said to be landless) will have forced many of the poor to move to highly risk-prone areas during years of calm between disastrous years. When the coincidence occurs, the destruction wrought will be massive. Unfortunately, the poor and marginal hill farmers in the uplands, who do not have an organized voice to articulate their grievances, are convenient scapegoats for the floods. Deforestation in the hills thus becomes a convenient, albeit untenable, cause of the floods in the plains. Even the climatological and hydrological reasons already cited above are forgotten in the rush to blame the hill farmer. The large impact of constricting natural drainage in the floodplain has not been openly discussed. The consequence of building highways, railways, embankments, bridges and urban landfills are conveniently ignored.

Our argument is not that loss of forest cover has no impact on the hydrological characteristics of hill slopes. The point being made is that even if the whole of Himalaya were forested in its primeval state, floods would not be prevented. Certainly, they could be delayed. The water absorption capacity of the forest floor would of course, reduce instant run-off significantly, in the early stages of the monsoon, during July. But, by the middle of August, the forest floor will not have the capacity to absorb spare water. At this stage, intense rainfall will lead to high surface runoff, in turn leading to floods in late August or September. There is, thus, no scientific reasoning to substantiate the contention that the reforestation of the Himalaya would solve the Ganga-Brahmaputra flood problem. Reforestation, however, is direly needed for the other vital needs of the hill people; and it must be undertaken just on that count alone.

There have also been arguments that deforestation causes an increase in silt load. Here, too, a minor factor is being made central. "Mass wasting" is an intrinsic Himalayan process which, in volume, is on an order of magnitude higher than surface erosion. If human-induced factor is to be considered significant, it may be primarily the rampant "cut-and-dump" road construction and then only the encroachment of agriculture on steep slopes.

Unless there is a drastic change in the behaviour of the monsoons, floods will continue to occur. Human societies in the floodplain need to make the best use of the floods, while reducing their vulnerability to them. An old peasant in north Bihar commented that "earlier, the floods used to come like a small cat, give you some fertile silt and go in a few days. Now it comes like a hungry tiger, takes away everything, and does not want to leave." The wisdom in this statement may guide future action in the flood plains. Programmes should try to tame the floods – not stop them. Domesticating the flood is what should be aimed for, so that the local economy reaps the benefits. Drainage constriction should be minimized, so that the "hungry tiger" can easily depart.

People in the floodplain have learned over centuries to take advantage of floods. Indeed, if one calculated the cost of alternative chemical fertilisers to replace the fertility brought by floods, it would be far more than what the peasant societies of Bengal or Bihar could ever afford. The environment, thus, provides goods and services, such as silt and water, which societies have always taken advantage of, to create wealth.

The aim is to fight floods, to minimise risk to life and property, while allowing the advantage of fertile silt to accrue. Whether these twin objectives can be met by simple civil engineering of embankments and dams, or whether other social sciences should enter the fray to decide on the proper way to take advantage of floods is the pivotal question. Indeed, experience of large dams in other less risk-prone areas of the world has shown that in actual operation, electricity and irrigation interests override those of flood protection.

An ecologically and economically sound response to floods will emerge from focusing primarily on the floodplain society and only secondarily, on physical interventions into the natural regime of rivers. Questions such as who owns the land, who are the landless who benefit from the sharing of products, and who bears the brunt of the risks, have to be answered before river regime modifications are attempted. This calls for the primacy of "social engineering" over civil engineering, of hydro-ecology over hydro-technics.

Jyanta Bantiyopadhyay is an ecologist with a special interest in natural resource conflicts and environmental management.

Dipak Gyawali is an engineer-economist with Interdisciplinary Analysts, Kathmandu. This article is based on their joint presentation before a floods conference in Calcutta last May.