Hence the damage that would have occurred without dams is put at Rs.(19 X 60 + 8 X 40
- 6 X 400) million =Rs.3860 million. The damage with the dams is estimated to be Rs.
(4 X 60 + 40) million =Rs.280 million. Hence the reduction is damage due to the dams is
estimated as Rs.3580 million.
Since this refers to the period 1958-64, the corresponding average flood benefit per year
becomes Rs. 3580 million = Rs.51 million approximately.
- AN ALTERNATIVE APPROACH TO BENEFIT ESTIMATION
The method of evaluating flood-control benefits described above is based on the
correct idea that these benefits consist in the reduction of flood damage that the project
in question makes passive. However, the method suffers from a number of inadequacies.
Firstly, in using the sample observations during 1958-64 to derive the project benefits of
flood control, the method appeals implicitly to the theory of probability; but it fails to
specify a well-defined probabilistic model which can serve as a basis for prediction.
Secondly, the sample itself is too small.
Thirdly, the method does not lend itself conveniently to an analysis of the variations in
project benefit corresponding to variations in the reservoir operation policy.
Fourthly, the inclusion of all flood discharges above 100,000 cusecs in a given year leads
to double-counting of crop damage and hence to overestimation of flood-control benefit.
Because of these defects in the method used by the authorities of D.V.C. and as the more
conventional analysis based on the ‘stage-damage curve’ (which relates various flood
stages to the associated levels of flood damage) is ruled out by the shortage of date, we
shall attempt to develop an alternative approach in terms of which the date that are a
available can best be utilized.
The basis assumption underlying our approach is that flood damage is a function of the
peak flow only (cf.Maass et al.[1] pp.287-8). In fact, the level of damage depends also
on such considerations as the time of year (which, for example, affects the maturity of
crops), the velocity of flow, the physical and chemical properties of flood waters and the
depth and duration of inundation. The use of peak flow as the determinant of flood
damage only provides a convenient first approximation (By defining ‘peak flow’ as the
yearly maximum of ‘average daily flows’ (rather than as ‘momentary peaks’), one can
also, to some extent, implicitly take duration into account).
If flood damage can be regarded as related predominantly to the level of flood discharge,
as is often the case, the benefits of flood control per period can be written as
B = ƒx p (x){c(x)-c(x^1 )]dx