1R-8

WATER BALANCE OF A RESERVOIR

SA TISH CFIANDRA

DIRECTOR

STUDY GROUP

S K JAIN

NATIONAL INSTITUTE OF HYDROLOGY

JAL VIGY AN BHAVAN

ROORKEE
I 986-87
 CONTENTS
PAGE

Abstract
List of Figures ii

INTRODUCTION 1
1.0
1.1 Importance of Water Balance of Reservoirs 1

1.2 Scope of the Present work 2

DATA REQUIREMENTS FOR RESERVOIR WATER BALANCE 3

2.0
5
3.0 COMPONENTS OF WATER BALANCE EQUATION FOR A
RESERVOIR
3.1 Computation of components of Water Balance 8
Equation
8
3.2 Estimation of Surface Inflow
9
3.2.1 The Analogue Method
10
3.2.2 The Water Balance Method
11
3.2.3 Discharge Isoline Maps
11
3.2.4 Heat Balance Method
12
3.3 Estimation of Precipitation
13
3.3.1 Precipitation Network
14
3.3.2 Analysis of precipitation data
3.3.3 Estimations of average precipitation-5
over an area
17
3.3.4 Polynomial Interpolation
17
3.3.5 Least Squares Approach
18
3.3.6 Lagrange Interpolation
19
3.3.7 Distance Weighted Interpolators
19
3.3.8 Multiquadric Interpolation
20
3.3.9 Optimal Interpolation
22
3.3.10 The Technique of Kriging
 3.4 Evaporation
24
3.4.1 Estimation of Evaporation
25
3.4.2 Energy Budget Method
27
3.4.3 Aerodynamic Determination of Evapo-
ration from Reservoirs 29

3.4.4 Combination Methods

30
3.4.5 Water Budget Method
31
3.5 Estimation of Reservoir Outflow
32
3.5.1 Discharge through spillways and
undersluices 32

3.5.2 Discharge through turbines

33
3.5.3 Leakage through dam
35
3.6 Computation of Ground Water flow
35
3.7 Estimation of change of storage
36
3.7.1 Estimation of Volume of Stored Water 38
3.7.2 Setting the gauge datum 40
3.7.3 Calculation of mean reservoir level 40
4.0 ERRORS IN WATER BALANCE COMPUTATION
43
5.0 CONCLUSIONS
45
REFEREMCES
46
 ABSTRACT

The water accounting of reservoirs is necessary for

their proper management. The purpose of water balance of a
reservoir is to estimate the various components representing

inflows, outflows and change in storage. Once these are

determined, the continuity equation can be used to either
test whether the components tend to balance out or to estimate

an element of water balance which could not be determined.

Although all the components of a water balance equation should
balance themselves theoretically, a residual term is obtained
in practice. This happens because the related measurements

are susceptible to various sources of errors.

The water flows into the reservoir through streamflow,

surface runoff, direct precipitation and ground water inflow.
The main components of outflow are evaporation, discharge

through spillway and outlets, turbines and seepage. Various

techniques which can be used for determination of each of these
individuals terms and change in reservoir storage are
described in the report. As far as possible, independent
methods should be used to determine the individual component

so that the errors are not, propagated.

 LIST OF FIGURES

TITLE PAGE NO.

FIGURE

Diagramatic representation of 7
1
components of water balance of

a reservoir

Definition sketch for computation 37

2
of groundwater flow
Computation of fluctuation of water 40

level in a reservoir
Computation of fluctuation of water 40

level due to wind in a reservoir

when wind directions are changing

1.0 INTRODUCTION

The reservoirs are constructed to reduce the variability

in the downstream. The aim is to change the natural
availability in a beneficial way and provide assured water
supply, mitigate floods, and generate hydroelectric energy.
These are the most common purnoses for which reservoirs are
constructed in our country although benefits are also derived
from other incidental uses.

It has been estimated that water bodies, excluding oceans,

occupy 1.4% of world's total land area. The total full
volume of 10000 major reservoirs of the world is about 5000
3
km which is equivalent to about 11% of total annual runoff
from the surface of the land. The total water surface area
of reservoirs is estimated to be about 600,00 km 2. These
figures give an idea as to how big the reservoirs are. The use

of reservoirs in river regulation can not be ignored.

1.1 Importance of Water Balance of Reservoirs

The term 'water balance' in the context used here
signifies quantitative assessment of various components of
water balance equation of a reservoir. While studying water
balance, it is indispensable to adhere to the law of conser-
vation of mass. However, due to a large number of variables

involved which defy an exact quantification, the water balance

of a reservoir can not be wateTtighi, errors occur while
closing the water balance equation and these are to be appropriately

1
considered.

Water, besides being essential for sustaining of life,

is an important input resource in a number of economic
activities. Due to its scarcity in many regions of the world
and increasing depletion in other regions because of growing
population, greater emphasis is being placed on better management
of water resources. To take better reservoir operating
decisions, it is required to have a complete quantitative

understanding of the water cycle of a reservoir. Predictions

of water balance components are also very helpful in design
of reservoirs. A knowledge of these components can give a
significant contribution to the study of extreme events and
climate variability. These studies are also useful in
estimation of components of water balance like seepage etc.

whose direct determination is quite difficult.

1.2 Scope of the present work

In the present report, various components required for

water balance study of a reservoir are discussed. Different
methods to compute the individual components are discussed
in detail. Requirement of data for a typical water balance

computation is also given.

2.0 DATA REQUIREMENTS FOR RESERVOIR WATER BALANCE

The data requirement for computation of various component

of a reservoir water balance is given below:

Catchment man for the reservoir with subbasins marked,

showing location of raingauges, stream gauges, and gauges
for other meteorological variables such as evaporation,
temperature, wind velocity and direction etc.
Map of the reservoir area showing location of dam,
hydromet stations, nearest upstream stream gauges at
all the streams which directly enter into the reservoir.
Contour map of the reservoir area, on a contour inter-
val of 1 m (preferably) and elevation-area-capacity
tables,

Records of streamflow at all the stations on the streams

which directly enter in the reservoir.

Precipitation data at the raingeuge; which are in the

vicinity of reservoir and could be used to compute the
direct precipitation input to the reservoir,
Evaporation data at the reservoir site,

Water level gauge data at all stations measuring reser-

voir water level,

Velocity and direction of wind in the reservoir area,

incoming radiation, and sunshine hours,

Position of water table around the reservoir and soil

permeability,

3
 Spillway discharge tables showing discharge at different
reservoir levels for various gate openings(if spillway

is gated),
Discharge through undersluices at various reservoir

levels,

1) If a power house also exists then following data is

needed:
Details of turbines
power generated
after-bay levels
efficiency of turbines and generators
head loss in penstocks and turbines

For all the data which involve time factor, it is

required to have information for each time interval of computa
tion and this data is needed for the entire duration for which

water balance computations are to be performed.

3.0 COMPONENTS OF WATER BALANCE EQUATION FOR A RESERVOIR

The water balance equation for a reservoir is nothing

but the mass balance or continuity equation. This equation
states that the sum of inflow and outflow components and
change in storage ( with appropriate signs) must be zero over
a given time interval. In the simplest form, the equation

can be expressed as :

+IG +P-E-Q-L-6S+6=0 (1)

where,
IS Surface water inflow into the reservoir,

IG Ground water inflow into the reservoir,

Precipitation on the surface of reservoir,

Release from the reservoir

Evaporation from the reservoir
Storage losses including seepage etc.

AS Change in reservoir storage during the period of

computation
6 = error term
The water enters in the reservoir through surface inflow
and direct precipitation; the water that leaves reservoir
comprises of releases through outlets and spillways, evaporation
and losses due to seepage. The reservoir storage increases if
the inflow exceeds outflow and decreases if the outflow exceeds
inflow. All the components of water balance equation should be
independently estimated. The term6 in the above equation (1)
represents the net effect of errors involved in the estimation

5
of different components. In practice it is quite likely that

errors will be present while measuring or computing various

terms involved in the water balance computation and the left
hand side may not sum to zero. Thus a large value of .5 represents
significant error in estimating different variables involved
in equation(l) . However, a small value of6 does not indicate
that the errors are small. The errors may be opposite in sign
and thus may balance themselves. The components of water balance
of a reservoir are diagramatically shown in figure 1.

The water balance equation(1) may be applied for any time

interval. Mean water balance is a term specifically used for
computations which are spread over an annual cycle e.g., a
calender year or a water year. Sometimes this term is also used
for seasonal water balances. The computations of mean water
balance are simplest in nature. However, with the shortening
of computational period, a more detailed accounting procedure
is required. The additional factors which are to be included
in the computations include bank storage during reservoir
filling,water loss due to water and ice left on the banks when
the reservoir is drawn down and return of this water to
reservoir later on. The following equation was proposed to
Vikulina (1970) for water balance computations for a short time
interval (a month)

°S "3- Q -E tQi = S (2)

where,
Qs = temporary water losses by saturation of shores

of the reservoir,
Q. = temporary water losses by the ice left on the
FIG.1 Diagramatic representation of components of water
balance of a reservoir (adapted from Ferguson & Znamensky)

7
 shores after the fall of level of the reservoir.

The accuracy of water balance components as well as the

duration of the design interval are stipulated by the accuracy
of determination of balance components. The most important
components are the surface inflow and change in storage.
Vikulina (1970) proposed the following equation to determine
relative error B (%) of water storage changes compared to
e
the inflows :
4
10 A-14)6 h
Be - 13)
864001s T

where,
Aw = water surface area of the reservoir in km 2
611 = error of mean level estimation(m)
I s = discharge into the reservoir

T = time interval or duration of balance period(in days)

This equation can be used to determine the length of balance

period such that B
e is less than +

3.1 COMPUTATION OF COMPONENTS OF WATER BALANCE EQUATION

3.2 Estimation of Surface Inflow

Surface inflow is the most important component of income

part, of the water balance equation for reservoirs. It can be

determined by direct measurement, can be computed using direct

measurement of related variables or can be estimated indirectly.
The total surface water inflow in the reservoir can be sub-
divided into two components : Contributions of the main rivers
debouching in the reservoir and the runoff from the surrounding
area which directly enters in the reservoir.

8
 The site at which streamflow measurements are carried out
is called a gauging site. The location of the gauging site
may be little away from the reservoir to avoid the back-water
effect and in such cases, the contribution of the area lying
between the gauging site and the rim of the reservoir has to
be considered to arrive at the correct figure. Most commonly
the variables measured at the gauging site are river stage
and discharge. If only river state is measured at the gauging
site, discharge can be estimated using the rating curve at
the site. A large number of methods are available to determine
streamflow at a particular site. These methods include the
velocity area method, slope area method moving boat method,
dilution methods and ultrasonic method. These methods are
described in great detail by Herschy(1978). Selection of a
particular method largely depends upon the site and flow
conditions, equipment available and accuracy requirements.

In case no measured data is available for the drainage

basin surrounding the reservoir, techniques discussed below
may be used to estimate inflow to the reservoir. These are
described in greater detail in Ferguson and Znamensky(1981)

3.2.1 The Analogue method

In this method, the basin is subdivided into sub-basins

based on factors which affect runoff such as topography,soil
type and land use, precipitation etc. Now for each sub-basin
which lacks measurements data, a search is made to find out
the particular sub-basin which has similar characteristics
and has sufficient observations avallaole. The runofr from the wagaugea

9
sub-basin can be estimated using the unit discharges from the
gauged subbasins. The procedure is repeated for all such sub-
basins which lack measurements and discharge in the reservoir
and the summation of discharges of all such subbasins gives the
inflow to the reservoir. In the situations where no sub-basin
with adequate measuring stations is available, it may still be
possible to use this method if a gauged sub-basin with similar
hydromet and topographic characteristics is available in a
neighbouring watershed. The indices which are important in
comparing the subbasins are drainage density,mean slope soil
type and land use. The formulae which could be used to determine

discharge of ungauged basins are described in detail in

Ferguson and Znamensky(1981).

3.2.2 The Water Balance Method

The water balance equation for an ungauged subbasin, which
discharges directly in a reservoir, can be written as:
Q = P - Eu - ASu (4)
u u
where
Qu = discharge of the ungauged subbasin
Pu = Precipitation over the ungauged subbasin

Eu = evaporation from the ungauged subbasin

ASu = storage change in the ungauged subbasin which may

be in the form of snow pack, soil moisture or ground

water
For the long term water balance computations, the storage

10
change is not very important parameter and in mean water balance
computations, it may as well be assumed zero. In case the
measurements of P and Eu are not available, it may be possible
u
to use this method by correlating with physiographic character-
istics such as slope,elevation etc. A number of studies have
been cited in Ferguson and Znamensky(1981) where this has been
successfully achieved.

3.2.3 Discharge Isoline maps

The runoff from the ungauged basins can also be computed
using isoline maps. To draw the isoline maps, first of all
normal runoff for each individual subbasin is computed. The
data obtained are related to the center of gravity of the
sub-basin. Now the hydrometric stations are marked on a map
on which boundaries of each sub-basin are drawn and normal
runoff values are marked at the center of each basin. Now,
considering topographic and physiographic characteristics into
account, lines are drawn connecting points with similar normal
runoff values. The mean runoff for a basin area A is computed
from
1 n
= (5)
1=1
where
normal runoff
sub area of the basin between two adjacent isolines
with an average runoff depth
n = number of subarea

3.2.4 Heat Balance method

The runoff from a basin may be considered to be the difference

11
between precipitation and evaporation. The normal annual
precipitation may be obtained from climatic maps and the normal
annual evaporation may be computed by heat balance or by
empirical formulae. It may be mentioned that for small basins,
this method may give quite incorrect results.

3.3 Estimation of Precipitation

It has been estimated that precipitation falling directly
over the reservoir surface forms approximately 17% of total
water input to reservoirs in Asia, Unesco (1974). For a
particular reservoir, naturally, the contribution of this com-
ponent increases with increase in surface area of the reservoir.
Precipitation is the most important meteorological parameter
and apart from the water balance computations, it is extensively
used in hydrology such as rainfall-runoff modelling, flood
forecasting and reservoir operation.
The measurement of precipitation is carried out using the
precipitation gauges which give the point values of precipita-
tion. The precipitation, however, is not uniform over a
particular area. Therefore some procedure is required to
estimate total amount of precipitation falling over the given
area using the point measurements available at a number of
gauges scattered over the area. A large variety of instruments
and techniques have been developed for gathering information on
various phases of precipitation. On the instrument side, the

most important ones are those measuring the quantity and

intensity of precipitation although devices for measuring the

raindrop size distribution and for the time of beginning and

12
ending precipitation are also available (Linsley,et al.1975).
The measurements of precipitation are expressed in terms of
vertical depth of water which would accumulate on a level
surface if the precipitation remained where it fell. The gauges
which are commonly used to measure rainfall include weighing-
type, Siphon-type and Tipping bucket raingauge. Of late, radar
is also being increasingly used for estimation of precipitation.

The precipitation gauges are subject to various errors. The

individual error components may be small in magnitude but
the cumulative effect is to yield a low value of observation.
Among the errors, the most serious is the deficiency of measure-
ments due to winds; other components caused by evaporation
adhesion etc. are small. The deficiency increases with the
reduction in raindrop size and thus it is greater for light
rain. A number of shields have been developed and various agenc-

ies have recommended the use of wind shields particularly if

the incidence of light rain or drizzle is high or a portion
of catch is snow. The deficiency of catch varies from place
to place and hence attention must be paid while applying
corrections. The site for establishing a gauge should also
be carefully selected. Preferably, the site should have level
ground in its vicinity with bushes and trees serving as wind
break. These however should not be too close to the gauge
to affect the catch. The obstacles which serve as wind break

should subtend an angle of at least 200 to 300 from the gauge

orifice.

3.3.1 Precipitation network

A precipitation network can be considered as a system for

13
collection of precipitation data with due consideration given
to the needs as well as the economy. Typically purposes
behind setting up a network are management of reservoirs for
irrigation water supply and hydro-electric power generation,
planning of water resources systems, agricultural planning
and meteorology. The optimum density of network which is the
number of gauges per unit area is determined based upon the
purposes of measuring data. A relatively thinner network may
be required for estimation of seasonal figures while a dense
network will be needed for flood forecasting purposes. The
World Meteorological Organisation,WM0(1974) has provided
detailed recommendations for the optimum network for various
climatic zones and type of terrain for general hydrologic

purposes. The guidelines of Indian Standard Institution are

also available in ISI 4987-1968.

In many instances, it may not be possible to establish an

optimum network because of financial, physical or institutional
problems. In such cases a minimum network is established.
Guidelines are alsa available for installation of additional
gauges in an existing network.

3.3.2 Analysis of precipitation data

Before any observed data is used in analysis, it is necessary

to make checks regarding its consistency etc. so that any error

which might have cropped up due to say, instrument failure or

mistake by olserver may be removed. Further, it must be ensured

that the station has not been shifted during the period of
analysis. The precipitation record may also have gaps, i.e. the

values may be missing for one or more periods. One method which is
14
popularly used to fill short data gaps is normal ratio method.
In this method, the precipitation at the station x is estimated
from the observations at three stations which are as close to
station x as possible. The precipitation at station x is
estimated by
N N
P
x
- 1 14x
P + X + X—P (6)
N A NB B N C
A
where
N represents the normal annual precipitation, and P represents
precipitation. Inconsistency in the observed data may be
present due to change in the location of the station. Checks
are available to test this type of inconsistency. A graphical
method, called double mass analysis is in use since a long
time. In this method, the consistency of the record at a station
is tested by comparing its accumulated annual or seasonal
precipitation with the concurrent accummulated values of mean
precipitation for a group of surrounding stations. A change
in slope of the line indicates a change in the precipitation/
regime at the base station. This type of change can not be
attributed to meteorological causes because in that case, all
the stations would be similarly affected.

3.3.3 Estimations of average precipitation over an area

The average depth of precipitation over a particular area is
needed in a number of hydrological applications. The simplest
method is to take arithmatic average of all the gauges located
in that particular area. If the terrain is flat, gauges are
uniformly spread over the area ,and the storm is quite uniform,
this method may give quite accurate results.
One of the most popular method of estimation of areal average

15
precipitation is the Thiessen Polygon method. The method is

based upon the concept of proximal mapping. The nonuniform

distribution of the gauges is accounted by providing a

weighting factor to each gauge. To determine weights for the

gauges, straight lines joining the gauges are drawn.Perpendi-

cular bisectors of these lineslead to the formation of polygons

around these gauging stations. It is assumed that the area

enclosed in a polygon is represented by the station within it.

This area can be measured and when expressed as a function

of the total area, represents weight of that particular gauge.

Weighted average precipitation for the area can be obtained

by multiplying the observed precipitation at each gauge by the

corresponding weight and then summing up. The weights remain

unchanged unless there is change in the gauging network.

Due to simplicity, the method is very popular and widely

used. One big limitation, however, is that the method is

unable to consider orographic effects.

The average precipitation over an area may also be calculated

using isohyets which are nothing but lines of equal rainfall.

Once the location of station and the observed precipitation

values are available, the isohyets can be drawn in the same

manner in which contours are plotted. The average for an area

can be computed by weighting the average precipitation between

successive isohyets by the area between them, summing up these

figures and thondividing by the total area. This is a linear

interpolation method in which the effect of physiography may

be taken into account.

lb
3.3.4 Polynomial Interpolation

In this technique a polynomial function(either algebraic

or trignometric) is fitted to the observed data. The interpo-
lated value at any point (x ,y0) is given by
0
* m
Po = klak fk (xo'Yo) (7)

Where ak is the k th polynomial coefficient,fk(x0,y0) is the

k th monomial in terms of xo and yo and m is the number of

monomials determined from the degree of polynomial function

fitted. There are two approaches for polynomial fitting
least squares approach and Lagrange interpolation.

3.3.5 Least squares approach

This method provides• the estimate of the variables as the
average trend of the true process. This is an approximate
method. Let pt be the estimate of the variable at ith point.Then

= E i=1,2....n (8)
1 x=1af(xiY
kk'i)
where m is the number of monomials and m
Here the aim is to determine parameters ak ,k=1....m such that
the sum of square of errors is minimized,i.e.,

Min Z = E 2
(pi- pt) (9)
1=1

Taking derivatives of Z with respect to a's and equating to zero

yields.
m a. n n
E 1 /
fk(x.,3y.)f.(x. y.)= E .
p.f (x1.,Y) (10)
i=1 j=1 3 1 3' j . i
1-1 1 k
Solution of the system of equations(10) gives
n
ak = J E, Tki pi k = 1, m (11)

17
where
T. = H .f.(x.,y.) k=1...m
kj 1=1 Ia.1 3 i=1...m

and Hki = E (x.,y.)f.(x.,y.) -1

fk3 3 1 3 3
(12)

Substituting the value of a k from equation (11) in equation(8)

k!1 Tkjfk(xo'Yo) Pj (13)

PI =j2:1
A noteworthy aspect of the least squares approach is that
the estimate of the variable at the point of observation is

different than the observed value.

3.3.6 Lagrange Interpolation

Unlike least squares, Lagrange interpolation is an exact
interpolation technique. The coefficients ak are determined
by constraining the estimate to exactly match the observed
values. This leads to the necessary conditions that the
number of monomials is equal to the number of observation
stations(m=n). Thus the linear system of equations given by

(8) is written as
pt
1 = k=1 akf k(x.y
,.) j = 1,2.... (14)
3 3
which upon solving yields
ak = Bki pi (15)

where Bkj is an element of the inverse of matrix formed from

= 1 ..n and j = 1...n. If the interpolation is

to be done over an area with fixed observation points and the

polynomial is not changed then the elements Bkj have to be

evaluated only once. This method is good as long as the variable
p is sufficiently regular in space which is true for precipitation.

The polynomials of high degree often yield sharp and unrealistic

gradients and Pose problems of ill conditioning of matrices.
18
3.3.7 Distance Weighted Interpolations

In the distance weighted interpolations the weights

are only function of distances between the points of estimation

and observation. The weights can be defined a priori. Most

commonly the following weighting functions are used

w(d) = lid

w(d) = 1/(d+1)

w(d) = 1/d 2

w(d) =

w(d) = e-ad
where d represents distance and a is a constant.

This interpolator is an exact interpolator if the

weights are equal to lid or 1/d 2. Further as the distance

goes on increasing, the weights go on reducing and approach

to zero for large distances. . A major shortcoming of this

technique is that the spatial interrelationship of the sampling

points is not considered. The redundant information, when more

than one observation stations are close to each other, is not

properly considered in this method.

3.3.8 Multiquadric Interpolation

In this technique, the effect of each observation

point is represented by quadric cones as a function of

coordinates of these points and the Summation of contribution

of each cone gives the interpolated value. Mathematically

p = a d (17)
t i oi
where ai is the multiquadric coefficient of ith station. The

equation (17) is for each point anIdlwe have

19
 n (18)
z a.d..
P- = i.1 j = 1....n
J 1 31
Solvillgecluati°11gOthe"efficielitsa.are
1 determined

a. =E c..p. (19)
1 j.1 13 j

where cij is an element of the inverse of matrix formed from

i = 1...n , j = 1....n. Substituting values of a i from

(19) into (17) yields

P* E c p di (20)
i=i j=1 1J J o

3.3.9 Optimal Interpolation

In this technique of interpolation, the weights are

determined such that the variance of error is minimum:

defining = Var

= Var w.p. (21)

1 1
where \far stands for variance.

Expanding the RHS for this equation

2 -E w Cov(p P.) (22)

a = cr- - 2 1=1 1 -o 1-1:1
e
where a' = variance of p o

Cov(popi) = Covariance between po and p i

Differentiating equation (22) with respect to the weights

.n and equating to zero yields

(23)
iCovIo) = Cov(po pj) j = 1....n
il w
Making use of homogeneity in the variances, the covariance

terms can be substituted by

(7 1'
Cov(p ipj) = id(PiPit
2
P(PiPi)
2 (24)
and Cov(popj) = a (PoPj)

20
 where p( ) is spatial correlation coefficient. A spatial

correlation function must be defined to estimate these correlation

coefficients which can be written as a function of distance

assuming a homogeneous and isotropic spatial correlation

structure. Hence equation(23) can be written as

i hwi P(d..)= P(doiJ j = 1....n (25)

where d ij is the distance between point i and j.

Solution of system of equations given by (25) yields

weights wi ,i = 1....n.
The estimation of the variable will be unbiased if the

sum of weights is unity,i.e.,

w. = 1 (26)
i=1
Incorporating equation(26) into equation(22) using

Lagrange multiplier,
n n
2 2 w Cov(p o p-) 1:41 1 E 1 ww 1 Cov(n 11)
a = o - 2 1 i

+ 2 x i91 w.-li (27)

where A is the Lagrange multiplier. Multiplication by 2

is just for mathematical convenience. Differentiating. )

with respect to weights and equating them to zero gives

i=1 w.Cov(p.p.) + A = Cov(poPj) = 1...n (28)

which using isotropic correlations yields

11
- wi p(d ii) + X .),j=1...n
= P (doj (25a)
i=1
ki (29b)
i-1.1 wi ' 1
Upon solving (n+1) simultaneous equations(eqns.293 and 29b) the

weights wi ,i = 1...n and Lagrange multiplier X are obtained.

21
. The choices of correlation assuming homogeneity and isotropicity
are :
The reciporocal model p(d) = q/(1+d/c0),
The square-root modelp(d) =1/1/ l+d/co, and
The exponential modelp (d) =

where co is called the characteristic radius and is to be

estimated from the data.

3.3.10 The Technique of Kriging

For estimation of the areal averages of the variables which

are considered to be realization of stochastic processes,
Matheron(1971) proposed the theory of regionalized variables.
A variable which characterises.apbenomenon varying in space
and/or time and shows a certain structure is called a region-
alized variable. Thus the variables describing depth of rain-
fall, water level in observation wells, soil transmissivity
are few examples of regionalized variables.
Given the values of the variable at n observation points,
i=1,2...n, the problem of Kriging is to estimate a

quantity p!,
s which is a linear function of variables. Three
types of problems may arise here:
To estimate the value of the variable at a point,
To estimate the value of the variable over a mesh of
given area centered at a known point, and
To estimate the value of variable over a specified
domain.
The first type of problems are called point Kriging;

second and third types are called block kriging.

22
The third type is most generalized and the first two can be
considered as special cases of the third type when the domain reduces to a

point or a block. It is required to find the set of weights

which give best possible estimation. For the estimation

to be best possible, the weights must be

Unbiased, i.e. there should be no systematic cover

or under estimation,
optimal, i.e., the variance between the observed

and computed values must be minimum.

The condition of minimum variance leads to equation (27)

which is rewritten here

n n
2 = a2 -2 E wiCov(popi) + E E IT) Cov(pipi)
e i=1 i=1 j=1 •

+ 2 X wi -1] (30)
i1
=
The theory of kriging assumes the increments of the
variables to follow the weak stationarity of second order.
Under this assumption, a random function is said to be stationary
if the first two moments of its joint probability distribution
at k arbitrary points are invariant under simultaneous trans
lation of all the points. Now the semivariogram is defined

as
= 4 Var[p.1
- p.]
3.
13
(31)
= a 2 - Cov(d..) i,j = 1,....n
13
where y(d.. ) is the semivariogram which is function of the
13
distance between i and j points. Making substitution in (30)

from (21) results

2 2 n ft 11 iw w. la 2- y (d.13-ii
a = a - 2 E w. [a 2 - Y (d03
- .j) 1, 1 j=1 3 -
e 1=1 1
n
+ 2xLE=.1 wi - d (32)

which upon differentiating w.r.t. weights and equating to zero

yields
23
 E w.y(d..) + = Y (d03 ) = 2,....n
1=1 13 j

and 1=1 . = 1 (33)

Solution of these (n+1) simultaneous equation yields n

weights and the Lagrange multiplier. Substitutions from

equation(24) and (25) into equation(32) give the variance

of error of interpolation as

2
1=1 w.y(d
ae ) +
1 oi
(34)
As mentioned above, the knowledge of variogram is required

for interpolation using kriging. A number of models are

available.

Nugget type Y(d) = C ( 1-$)

, bp
Monomial yCd) =g1d1 Ob<2
3
(c).Spherical y(d) = 1.5 IdVa - 0.51dI /aid <a

d> a
Exponential 1(d) = g {(1-exp(-fdll /a).]
Gaussian Y(d) = g[(1-exp(- idt 2 /a 2j

where is Dirac delta and g and a are the constants to be

determined.

3.4 Evaporation

The term evaporation is defined as the net rate of transfer

of vapor to atmosphere. The degree of evaporation depends

upon the nature of the evaporating surface and meteorological

factors. The present discussion is limited to evaporation

from free water surface.

The evaporation can be thought of as an enregy exchange

24
process. The most important factor in the process is radiation
followed by wind speed and vapour pressure of the air overlying
the surface. The amount of evaporation also varies with
latitude, season, time of day and condition of sky. It is
difficult to categorically express the relative effect of the
controlling meteorological factors, if radiation exchange and

all other meteorological elements are constant over a shallow

lake for a considerable time, the temperature of water and
evaporation would become constant. If the wind speed is then
suddenly doubled then the rate of evaporation would also be
double for some time. However, this rate would start decreasing
as the increased evaporation would extract heat from water at
an increased rate than could be replaced by radiation and
conduction and consequently water would achieve a lower
equilibrium temperature.

The quality of water in a reservoir also affects evaporation

although the change may be marginal. This reduction takes
place because the dissolved solids reduce the vapour pressure
of the evaporation, the temperature of water rises and this
partially offsets the effect of reduction in vapour pressure.
Moreover, any foreign material which affects the reflectivity
property of water surface tends to affect evaporation.

3.4.1 Estimation of Evaporation

The instrument Pan Evaporimeter is most commonly used to.

estimate evaporation from water bodies. The pan is a

shallow( and mostly) circular vessel exposed to atmosphere.

The pans can be installed in three ways: on the land surface,
sunked in ground and floating on water surface. The pans installed

.25
on or above the ground surface experience little higher evaporation since extra

heat is absorbed by the side walls. This can be minimized

by suitably isolating the pan. However, this effect must be
suitably considered while estimating the evaporation from the
reservoir using the pan evaporation measurements. The main
advantages of surface pan are economy and ease of installation
maintenance and operation.

By burying the pan, the objectionable effects due to radia-

tion on the side walls are eliminated. But on the other hand
these pans are difficult to instal, maintain, repair and

observe. It is also difficult to detect the leakage which

may take place from the pan. The heat exchange between pan and
soil is appreciable. The height of vegetation adjacent to pan

must also be limited.

The estimation of evaporation from a reservoir can be most
nearly annroximated by a pan floating on lake surface. However,
the installation and maintenance expenses are quite large.
Observation of data is very difficult and many times, splashing
takes place which renders the records unreliable. Due to these

reasons, these plans are not very common in use.

Among the various types of pans in use throughout the world,
the most widely used is the US Weather Bureau Class A Pan.This
pan is made of unpainted galvanized iron. Its shape is circular
with diamater 122 cm. and depth 25.4 cm. It is recommended
that this pan be mounted on a wooden frame so that air may
circulate beneath it. The pan must be filled to a depth of 20 cm

26
•
and it should be refilled when the depth of water falls to
18 cm. The water level can be measured using a hook gauge.

The evaporation is computed as the difference between the

water levels measured after accounting for precipitation .
The estimate of evaporation can be obtained by multiplying
the pan evaporation by a coefficient called pan coefficient.
The average value of pan coefficient for US Weather Bureau
class A pan is 0.70. The value of this coefficient can vary
regionally, it is low in arid regions and higher in humid.
Many times, it is necessary to cover the pan with a screen
to prevent loss of water due to drinking by animals and birds.
The use of screen changes the pan coefficient. •The change

can be as much as 14%

3.4.2 Energy Budget Method

In the energy budget method determination of evaporation
from the reservoir, the energy input and output from the
reservoir is accounted and the residual is assumed to have
been consumed for evaporation. Alongwith energy balance,
a rough water balance is also required since water storage

and inflow/outflow represent energy values.

The energy budget for a reservoir may be written as

Rn - R h - R r + R V = 0 (35)

where
Rn Net radiation absorbed by the reservoir,

Rh = sensible heat transfer to atmosphere through

conduction
Re = energy used for evaporation

27
 R = energy stored in the reservoir
r
Rv = net energy content of inflowing and outflowing water

The units used in the above equation are calories per

square centimeter. The term sensible heat transfer can not
be directly observed or computed. Let H represent latent heat
of vaporization and R the ratio of heat loss by conduction to
heat loss by evaporation or Bowen ratio. Thus the above
equation can be written as :

(36)
E = (1(11+R), - lc)/ "v(l+R)
where
E = evaporation in centimeters

n = density of water.
The Bowen ratio can be computed by the following equation

= 0.61 (T0 - Ta)p/1000.0(e0 - ea) (37)

wnere
p = atmospheric pressure,

To water surface temperature

Ta = temperature of air
e0= saturation vapour pressure corresponding to To
ea vapour pressure of air,
The above equation is valid for normal atmospheric conditions.
The limiting values of the constant (0.61) in the above equation
are 0.58 and 0.66 depending upon the stability of the atmosphere.
If the correct value is assumed to be within these limits, the
extreme error is likely to be within + 4 % , Linseley et al. (1975)

The estimation of evaporation very much depends upon accurate

evaluation of net radiation. This can be expressed as

Rn = Rs - Rr + Ra - Rar - Ro (38)

28
where
Rs = sun and sky short wave radiation incident
upon the water surface

Rr = reflected short wave radiation

R a = incident atmospheric longwave radiation,
Rar = reflected longwave radiation
Ro = emitted longwave radiation

The radiation can be measured by radiometers which can

be designed to measure either total incoming or net radiation.
Ideally, it is required to expose the radiometers at
water surface at more than one point. Since it is difficult
to take observations over a reservoir, many times the radio-
meters are exposed over a tank of water assuming that the
emissivity and reflectivity of the water in tank and reservoir
are the same. The incident minus reflected allwave radiation

R. for the reservoir can be measured and the net radiation

1T
for the reservoir can be obtained from

Rn = R. cc; (T0)4 (39)

1T
4
Rn + ca(T0) -
4
R n + ca(To - T0)

where To is the absolute temperature of the tank water surface,

G is the Stefan-Boltzmann constant which is equal to 11.71
-8 Cal/cm 2 k 4d and is a constant which is equal to 0.97.
x 10

3.4.3 Aerodynamic Determination of Evaporation from Reservoirs

The determination of reservoir evaporation using aero-

dynamic concept is based upon turbulent transport concept.

A number of empirical equations have been developed relating

evaporation with atmospheric elements. the general form of

29
these equations is
E = leo -ea) (a+bv) (40)
where e o is the vapour pressure of the water surface, e is
a
the vapour pressure of the over running air at some height,
and a and b are coefficients. Linsley et al (197S) report
some of the equations collected in a study

E = 0.00304 (e - e2)v4 e 2 and v4 over reservoir

o
E = 0.00241 (e - e8)v8 e8 and v8 over reservoir (41)
o
E = 0.00270 (e - e2)v4 e2 upwind and v over reservoir
0 4
where E is reservoir evaporation in inch per day, small e's
represent vapour pressures in incnes of mercury,v's are wind
speeds in miles per day and numerical subscripts designate
heights above water surface in meters.
It has been observed that the vapour pressure of the air
increases downwind across an open water surface and thus the
concepts based upon turbulent transport conclude that evaporation
decreases with downwind . Linsley et al(1975) have quoted a
study conducted by USGS in which the coefficients a and b of
equation (40) were determined after studying a number of
reservoirs upto 120 km 2 in area. The coefficient a was found
to be zero and b was given by
b = 0.00014 A -0.05 (42)
for E in inches and A the reservoir area in acres.

3.4.4 Combination. methods

The combination methods of estimating evaporation make use

of both aerodynamic and energy budget equations. The following
equation has been derived assuming a thin free-water surface,

i.e. without heat storage or conduction from below.

30
 1 + r Ea) (43)
E - A+ r
- (Rn

where A is the slope of the saturation-vaopur-pressure versus

temperature curve at the air temperature Ta,Ea is the evapo-
ration given by equation (40) assuming the water surface

temperature To = Ta, Rn is the net radiation exchange, and

r is the psychrometeric constant in the equation of Bowen
ratio (equation 37):
T-T
a (44)
R - r e oo-e a
Charts are available relating the reservoir evaporation with

solar radiation, air temperature, dewpoint and wind movement.

In the derivation of equation(43), it is assumed that Rn

represents exchange of radiation at water surface.E a is
based upon aerodynamic equation and the correct value of E is
obtained when used with the observed vapour pressure of the

water surface, and A at Ta is good approximation of its

average value between Ta and To.

3.4.5 Water budget method

This method of estimation of reservoir evaporation is based
on determining the various components of water balance(except
evaporation and error term) and then computing evaporation from
equation(1). This method, although simple, is quite inaccurate
because errors in measuring all other components will be
encompassed in evaporation. Tne estimation of various components

of water balance equation is dealt separately in different

sections.

31
3.5 Estimation of reservoir Outflow

The total outflow from a reservoir is sum of discharge

through spillway, turbines, undersluices and leakage through
dam:

Qsp Qtb Qus Q1 (45)

where

Qsp discharge through spillways

Qtb discharge through turbines,

Qus = discharge through undersluices
Q 1 = discharge through lakage from dam

3.5.1 Discharge through spillway undersluices

The discharge through spillway can be computed either

by using hydraulic formulae or by using result of laboratory

model testing. The discharge through spillway can be computed

using following formula.

Q. = C 1.5
sp sp Cq b1/2gh
sp sp (46)
where
C
sp = submergence coefficient,
C = discharge coefficient for spillway,
= width of the spillway
sp

sp head over crest of spillway outside the zone of the

draw down

Under the free flow conditions, the discharge of an ogee spill-

way is given by :
0- 3/2
-sp = C (L - kn H) (H+h) (47)

32
 Where C is the coefficient of the weir, L is the clear

crest length , n is the number of end contractions, H is the head

over spillway, and h v is head due to velocity of approach. In

the metric units, the value of c varies from 2.21 at the

discharge head to 1.71 at very small heads.

Alongwith spillways, reservoirs are also provided with

low le/el outlets for releasing water when the reservoir water

level is low. These outlets behave as orifices and the discharge

through them is given by :

Q Or = or e A i2qh. (48)

whereCOr is the submergence coefficient for semisubmerged

or submerged orifices,e is a factor which accounts for jet

contraction and difference between actual flow veolocity and

idealized flow velocity, A is the corss-section area of the

orifice, and h is the head at the orifice measured from the

center of orifice. The discharge coefficients can be obtained

by hydraulic considerations or they can be determined from

laboratory model tests or field calibrations.-While conducting the

model tests, one should properly considerthe situations like

characteristics of approaching flow, lateral discharge esti-

mation will arise if the actual field conditions are not

considered or the discharge coefficient is wrong.The project

authorities prepare tables giving discharge through gated

spillways for various gate openings. Linear interpolation

is sufficient for intermediate values.

3.5.2 Discharge through turbines

Basically, these are two methods of determining discharge

33
through turbines. Each of these is discussed Delow.

3.5.2.1 Flowmeter method

This method, as the name suggests, makes use of flow-meter

to measure discharge passing through a turbine. It uses the

following equation to compute discharge.

A Mf
Qt = Gf (49)

where M is the difference in the flowmeter readings for the

f
time period t and Cf is the flowmeter constant. To avoid
errors in discharge estimation, the fiowmeter calibration
should be checked at specified time intervals.

3.5.2.2 discharge characteristics method

This method makes use of the relation between generated
power and turbine discharge. These two are related by:
W
tb (50)
Q tb =
9.81 htl

where Wtb is the power generated in kilowatt,htb is the effec-

tive water head in meters, Q tb is discharge through turbine
and fl is combined efficiency of turbine and generator. The
effective head is obtained after deducting the head losses
from the total head. The total head is obtained by subracting
the water level of forebay from the level of lower pool. Head
loss takes place because of friction and bends in penstocks
and losses at entrances and exists. Since the hydropower plants
are mostly used for peaking purposes, the discharge through

turbines will have considerable fluctuation during a day

and hence measurements may have to be taken at a shorter time

interval.
34
 The term n in equation(5U) represents combined efficiency

of turbine and generator and is obtained by multiplying the

individual efficiencies.

In case the plant consists of more than one turbine

then the total discharge can be obtained by summing the dis-

charge through all turbines. It has been mentioned by Ferguson

and Znamensky(1981) that the estimation of discharge from

turbines are generally on the lower side and the error in daily

discharge may be of the order of 3-5%. A major source of

error in such cases is the incorrect estimation of head losses.

hrrors may also crop in because of unstable operation character-

istics, wearing of elements etc. and hence the results should

be periodically checked with other methods.

3.5.3 Leakage through dam

This component of outflow consists of loss of water

from the reservoir on account of leakage through the body of

the dam as well as through gates and spillways. It is not

easily possible to relate these losses with a measureable

quantity. For example the losses through the gates or valves

of undersluices depend upon their design, installation and main-

tenance. A simplifying assumption which is usually made in

practice is that these losses linearly vary with the reservoir

level. In general, the amount of water lost due to these

reasons varies between 0.5% to 4% f the total discharge

through the structure.

3.b Computation of Groundwater flow

A reservoir also experiences subsurface flow from or

towards the aquifers though the magnitude is very small compared

35
to the surface water inflow. The amount of this flow depends

upon the physiographical features and soil characteristics

in the vicinity, and the position of water table. Assuming
homogeneous condition, the flow can be computed by the Darcey
Law:
h1-h
I =bdk 2 (51)
G 1
where,
b = base width of flow
d = depth of flow
k = horizontal permeability coefficient (m/day)

hh2= water levels at two sections across the under-ground

current at a distance 1 apart (ref. fig.2)

3.7 Estimation of change of storage

The change of storage component of water balance equation
represents the change in the reservoir storage during the 'period
of computation. As explained by Ferguson and Znamensky(1981)
this term can be expressed as a sum of four components.

AS + AS +AS (52)
TM bs

where,
AS = change in storage in reservoir.
w
A SYM = change in channel storage of all those streams
which directly debouch in the reservoir between the
gauging site which lies just upstream of reservoir

and the rim of reservoir.

A Sbs = change in storage in the banks of the reservoir.

A S = change in storage because some ice is left on the
reservoir banks during winter which melts and flows

back in summer.

36
 High level
— -

5, •

It •h
h>2
Lii .
--ze.
."-*: "Ih. • d `,
d
>1/4. <

FIG. 2 Definition sketch for computation of groundwater flow

37
 Out of these four, the first component is most

important. The last component may have to be considered only

for very few Indian reservoirs.

3.7.1 Estimation of volume of Stored Water

To estimate the volume of water stored in the reservoir

at any time, it is necessary to estimate the stage of the

reservoir. The stage can be used to determine volume by using

stage-volume curve. To prepare 'this curve, first of all, a

detailed surveying is done for the reservoir area and a contour

map with a small contour interval is prepared. Using this map,

the surface area of the reservoir at any particular elevation

can be easily determined. The volume of water between any two

successive contours is the average area at these countours

multiplied by the contour interval. Thus starting at the

bottom of reservoir, it is easy to prepare elevation-area-

capacity curves or table. These figures keep on changing with

time because of instability of shores and deposition of

sediments. Hence, to maintain the required degree of accuracy

in the computations, it is necessary to undertake reservoir

surveys from time to time to have uptodate information.

The water level gauges are installed to measure the

mean water level of the reservoir. The main points to consider

while installing these gauges are the shape of the reservoir

the types of fluctuations that it experiences one to winds

etc. and the ease in installation observation and maintenance

of these gauges. Generally, it is required to locate the

gauges along both banks ( in the upstream direction from dam)

of a reservoir. Special care must be taken of the area where the

.38
reservoir influences the stage of the river unless the storage
in this zone is less than 5-10% of the total accumulation.
When wind blows over a reservoir, it applies shear stress
on the water surface and thereby it tries to carry water along
with it. This leads to a redistribution of water in reservoir,
there will be greater storage in the down wind direction and

lesser water in the upwind direction. Changes in the wind

direction and/or magnitude of wind leads to fluctuations in
the water level. The following equation(Ref.figure 3) can be

used to compute the change in stage due to wind.

8 (53)
A h w = (3+aw) um (CosA ) 10
where u is the wind speed in m/s, 1 is the distance in meters

between two points for which hw is to be computed, d m is

the mean depth of reservoir in meters between these two points.

xis the angle between wind direction and the line joining

these two points, and d w is the mean wave height.

To compute the mean water level in presence of these fluc-

tuations it is necessary to determine the location of

equilibrium axes where the water level fluctuations due to
wind generated shear are minimum. Once the change in water
storage has been determined using the equation (53), the
changes in water volume in a subarea can be determined by
multiplying by the corresponding areas. The equilibrium
axis is determined at the division of sub-areas of positive
and negative change of water volume. The direction of this

axis is perpendicular to the wind direction. The position

of this axis should be found for eight main directions i.e,
N,NE,E.... The point where the longitudinal axis of the

39
 (a)

FIG. 3(a) Computation of fluctuation of water level in a

reservoir

FIG. 3(b) Computation of fluctuation of water level due to

wind in a reservoir when wind directions are changing

40
reservoir crosses the equilibrium axis perpendicular to a given

wind direction is the best location for a stage gauge. At

these points the stage will be closest to mean reservoir level.

It has been recommended to locate the gauges near the

equilibrium axis when the stage gradient due to wind effects

exceeds 15-20 cm.

3.7.2 Setting the gauge datum

For ease of comtiutations, consistency of results and
because of several other reasons, it is necessary to reduce all
the gauges to a common datum. In India, the Survey of India
has established very large number of bench marks. by levelling
from the nearest bench mark. Thus the datum of a particular

gauge can be expressed in terms of mean sea level.

For large reservoirs, it may be difficult to carry out
levelling because of inhospitable territory. The technique
makes use of the fact that the surface of still water is hori-
zontal. Using this property, information is transferred
from one place to another. The best time to do this exercise
is the low flow period when the discharge from the reservoir
is quite small. Automatic stage recorders are installed many
times to achieve better accuracy. Further, adjustments are
made using long term observations to weed out cycles of smaller

frequency.

3.7.3 Calculation of mean reservoir level

The mean water level of a reservoir at any time is used to

determine the volume of water stored in the reservoir at that

instant. Using this information at the beginning and end of a
time period the change in storage can be worked out. Several

41
 methods are available for this purpose which could be used
depending upon the slope of the water surface and degree of
knowledge of morphometric characteristics of the reservoir.
If the water surface of the entire reservoir is more-or-less
horizontal, the mean water level can be used to determine
the storage. Otherwise volume may have to be determined
for each subarea but this requires individual elevation-
storage curves for each subarea.

The mean weighted water level can be computed in a

manner which is similar to Thiessen polygon method for
rainfall estimation. In this method the weights for each
area are determined by dividing the area of each subarea with
the total area for reservoir. These weights are then multi-
plied by the corresponding stage to determine mean weighted
water level. Mathematically,
hm = h A1
1 A— + h 2 A2 A
-A h n
-11 n — (54)
AR
where h m is the mean stage, 111 hn are the stages at
the gauges 1,...n respectively. A's are the partial area of
the reservoir associated with these gauges and A R is the surfac
area of complete reservoir.

A graphical method known as smoothed graph method

is quite helpful for reservoirs where long term stage fluc-
tuations take place due to wind. The water level fluctuations
at all gauges are plotted on a graph. The points where the

stage curves intersect are connected with a smooth curve from

the zone of reservoir influence upon the river stages to the
dam. This line shows the position of the water stage undis-
turbed by the fluctuations due to wind. The mean stage can then
42
be used for determination of volume of stored water.

The other components of .equation(52) are not very

significant. Their magnitude is either very small or negligible
in most of the cases. The determination of these components
is described in Sokolov (1974) and Ferguson and Znamensky

(1981).

43
 4.0 ERRORS IN WATER BALANCE COMPUTATION

From the theoretical point of view, the various compo-

nents of water balance equation should sum up to unity.

However, it is not possible to exactly estimate or measure

the various components and thus the term dwas introduced in
equation (1) to take care of the residual error. To avoid the
propagation of errors, it is necessary to estimate the

individual components of the water balance equation independently.

The errors in individual components may be positive or
negative and hence they may also tend to balance. Therefore,
a small value of the error component does not indicate that
the errors in estimation of individual components are small.
The purpose of error analysis is to assess the correctness
of the estimates and their sensitivity.

If the error in estimating individual water balance

components are 61,62,
6n then it is recommended that
the maximum value of error should not exceed the square root
of sum of error of individual components, or

j„ 2 ÷ 62 ... (55)
6<
If this criterion is not satisfied then it is required
to reevaluate the estimation procedure and measurements of
individual components. Since the magnitude of different
comporents vary widely, percentage errors in them will also

vary over a large range. This variation also depends upon

44
upon the time period of computation. As this period increases
the magnitude of error in various terms which represent inflow
and outflow to and from the reservoir also increases. However,
the error in the term representing change of storage tends to

reduce with increase in time period. The aim of any water

balance study is to minimize the errors associated with
different components of water balance equation. This requires
an assessment of the sensivity of various water balance
components. This will depend upon duration of computation
period, climatic conditions, physiographic factors and season
of year. Different components may become significant in
different seasons of year. For example, during summer months,
inflows to the reservoir may be very small and evaporation
quite large while during the monsoon period, the situation
may be just reverse. Hence, the existing measurement network
may have to be expanded in many cases. This requires a care-
ful study and a final decision should be based upon the net-
work analysis for required degree of accuracy and the finances

available.

45
5.0 CONCLUSIONS

The various methods which are used to determine the

different components of water balance of a reservoir have been
described. From the review of literature the following points
emerge:

As far as possible, independent methods should be used

to determine the individual components of the water
balance equation.

The relative magnitude of components of water balance

equation varies from season to season and this fact
should be considered while deciding about the accuracy
of a particular measurement.

The above point 'b' is also important in determining

the number and location of additional stations to be

established if the existing network is to be strengthened

This would also depend upon the purpose of carrying out
water balance computations.

It is felt that in India, extra attention must be paid

for estimation of reservoir inflow from the ungaged
basin and seepage losses from the reservoir.

46
 REFERENCES
Baumgartner,A., and E.Reichel(1975), World Water Balance,
Elsevier Publishing Company,Amsterdam.
Ferguson,H.L., and V.A.Znamensky,' Methods of Computation
of the water balance of large lakes and reservoirs',
Studies & Reports in Hydrology461,Unesco 81.

Herschy,R.W.(editor)(1978), Hydrometry,Principles and

Practices, John Wiley and Sons,New York.
IASH(1970),' Symposium on World Water Balance,Publication
No.92,IAHS-Unesco-WMO.
Linsley,R.K.,et al(1975), Hydrology for Engineers,
McGraw Hill Book Company,Auckland.
Sokolov.A.A. and T.G.Chapman(1974), Methods for Water
Balance Computations,The Unesco Press,Parisw
UN(1974),Manual for the Compilation of Balances of Water
Resources and Needs,The United Nations.
Unesco(1974),' Survey on Water Balance of Lakes and
Reservoirs of the World, Unesco,Paris.
Vander Beben,A., and A.Herrmann(ed)(1985),' New Approach
in Water .Balance computations', IAHS Publication No.148.

Vikulina,Z.A.(1970),' Methods for computation of water

balance of reservoirs', in Symposium on Water Balance,
Publication No.92, IASH.
WM0(1974),' Guide to Hydrological Practices,WMO Publication
No.168,Geneva.

47