Custom Search


A new model for flood mitigation in the Kosi basin


Dr Sumit Jha
Speciality Registrar, Anaesthetics and Intensive Care
Belfast, Northern Ireland

Floods in the Kosi basin has been a recurrent problem since the 18th century. This has, over the years, caused innumerable suffering and vast amount of damage both in terms of lives and property. The flood delta has its unique advantages such as creating a fertile alluvial land which has the potential to change the lives of millions of ordinary people by lifting them out of the vicious cycle of poverty and malnutrition.

Previous studies:

There have been many technical studies into the practicalities of flood control in North Bihar. Various feasibility studies have looked at building dams, embankments along the river shoreline etc. Efforts have been ongoing to control the devastating consequences of flooding in the successive years with an alarming regularity.

Present scenario:

Flood control measures have been continuing with a renewed vigour under the present governement and able leadership of Honourable Chief Minister Shri Nitish Kumar. Given his background in civil engineering, he is well placed in understanding the special predicament of the Kosi especially the extremely high sediment load which would cause an inevitable short dam life with its associated costs.

Our proposal:

We acknowledge the fact that land is a scare resource; any large-scale reservoir is impractical with its huge implications with respect to displacement of huge numbers of people and associated problems with land acquisitions and social, financial, logistical problems.

We propose a multiple reservoir model for mitigating the water stagnation and flood damage in low-lying areas. I do not think we will be able to control flooding entirely, but the focus should be on decreasing the impact of flooding and recharging the ground-water resources at the same time.

The whole model is backed up by rough mathematical calculations, but I acknowledge that these are rough calculations and borrowed from previously published studies. These are best approximations and should be re-checked by certified professionals.

Quality of water in the proposed ‘multiple pond model’ (MPM) is paramount. We propose gabion-lined 1-acre structures of a triangular/funnel-shaped configuration to ensure platform stability and a longer shelf-life.

Each 1-acre project should have a detailed soil analysis to confirm subsoil quality to prevent pond failures. A mechanical boring machine for soil sampling and a geological or agricultural scientist’s opinion would be vital.

Each 1-acre project should have a strategic location with a watershed area and ideally linked to further projects along downstream by a canal system. The costs for these may be prohibitive. Hence standalone measures should be initiated for now. These could be stepped up at a later date subject to funding.

Each 1-acre project should have active community participation locally, strictly maintained to ensure optimal water quality. Fishing should be promoted as a source of local revenue for the gram panchayats. This revenue should be used to further improve drainage structures in/around the village, establishing rainwater harvesting structures, improving organic vermicompost measures for each village, improving plantations to improve soil quality and prevent soil run-off.

Each 1-acre project should cater to trees and plantations in/around the pond. This will help improve the local ecosystem and also increase the amount of water retention in the soil/subsoil. Care should be taken to avoid deep-rooted trees very close to the banks of the pond. In growing roots can cause pond bank failure.

Each 1 -acre project should have an elevated area lining the pond. This should have pole-mounted solar-powered lamps along the fence (Estimated 3-4 at the corners). These would also serve as warning signals to the site of the pond at the time of crisis if there is a potential devastation with major inundation,  in spite of all these measures.


The three different methods give approximate rather than exact volumes, but these approximations are close enough to the true values.  When using acres for area values and feet for depth values, volume will be in acre-feet. An acre-foot of water is 1 acre of water 1 foot deep, i.e., 43,560 cubic feet. If 1 cubic foot= 30 Litres.  (Rounded up from 28 liters)

As and example of pond/Lake area: 48 acres. (Outer shoreline: 0 feet, maximum depth: 23 feet). Total water volume: Approx 500 acre feet (21825000 cubic feet). Rounded up to= 20 lac cu feet. (approx 600 lac liters= 6 crore liters)

If we re-configure this to our need of 48 ponds of 1 acre.

If we have 100 ponds of 1 acre each= 12 crore litres.

If 1000 ponds of 1 acre each= 120 crore litres.

If 10,000 ponds of 1-acre each= 1200 crore litres.

100,000 ponds of 1-acre each= 12,000 crore liter.

150,000 ponds of 1-acre each=18,000 crore liters.


Approx 45,000 villages in Bihar= Approx 3-4 such projects in each village.

Area of lakes, a basic requirement for volume calculations, has traditionally been estimated with a mechanical planimeter. In recent years, computer programs can compute area (such as Arcview) and volume (such as Tecplot®).]

The Formula for water measurement is

Q = V · A


Q = volumetric flow rate

V = average water velocity

A = cross-sectional area

The Unit for Water measurement is Cusecs

Cusecs = A volumetric unit for measuring the flow of liquids, equal to one cubic foot per second

Cusec is a measurement of *flow* of water (volume / time), defined as:

"A volumetric unit for measuring the flow of liquids, equal to one cubic foot per second."

1 cubic foot = 28.3168466 liter

Say 1 cubic foot= 30 Liters

So, 1 cusec of water = 28.32 liter water flowing per second.

75,000 cusecs per second=  21,22,500 liters per second.

                                             Round up to 21 lac liters per second.

21 lacsx 60 (min) x60 (hour) x24 (hours per day)

504 lacs x3600 litres per day.

5 croresx 3600 litres per day.

18,000 crore litres per day.

If 150,000 ponds of 1 acres each (CONSIDERING EMPTY): 18,000 crore liters

Hence, delay to flooding likely to be of the order of 0.5-1 day. (12 hours- 24 hours), depending on half-full/empty.


  1. Multiple small reservoir may be more amenable to implementation.
  1. Community-participation is key to upkeep.
  1. A conservative estimate of 1000 such 1-acre projects may be able to delay significant flooding by 24-48 hours. This is considering the entire peak flow of the river flows into these reservoirs (calculation purposes only, not in practice). There will, of course, be ongoing rains filling up the reservoirs and catchment area inflow. But delay in devastating flooding is very likely.
  1. Groundwater recharge: After the floodwaters have receded, these reservoirs would provide local communities with a significant resource.
  1. Revenue generation: Inland Fishery has the potential to bring significant revenues every 3-6 months.
  1. Plantations/fruits and vegetables around these water bodies could be a source of cash for the local communities.


  1. In the long-term, these self-sustainable ecosystems should provide the local communities enough resources to build a successful agro-business model.
  • Flood mitigation: Emphasis should be on reducing the impact and devastation of flooding rather than controlling the river per se. Not only is controlling going to be artificial, it will be financially crippling and short-term measure. The river will simply change its course after 10 years.





traffic analytics