Lecture 19: Synthetic Unit Hydrographs (October 21rd, 2024)

Learning Objectives

After this session, students should be able to:

 Define the difference between a data-derived unit hydrograph and a

synthetic unit hydrograph

 Define the Soil Conservation Service (SCS) Dimensionless Unit Hydrograph

(DUH)

 List the properties of the SCS triangular unit hydrograph

 List the properties of the SCS curvilinear unit hydrograph

 Transform an X-hr unit hydrograph to an aX-hr unit hydrograph (where a is

an integer bigger than 1)

 Use the S-hydrograph to calculate an aX-hr unit hydrograph (where a is a real

number bigger than 1)

4.1.4. The synthetic SCS unit hydrograph

The Soil Conservation Service (SCS) has performed a synthesis of a great number of unit
hydrographs for river basins of various sizes located in different regions of the USA. In order
to compare these unit hydrographs their ordinates and abscissas were normalized. Were
divided the ordinates by the peak discharges up while the abscissas were divided by the time
to peak tp. Furthermore, the non-dimensional ordinates and abscissas were averaged
obtaining the non-dimensional synthetic hydrograph (SUH), the abscissas t/tp and ordinates
of which u(t)/up are given in Table 4.2.

Table 4.2 The US-Soil Conservation Service synthetic unit hydrograph

Time SUH Time SUH Time SUH

t/tp u(t)/up t/tp u(t)/up t/tp u(t)/up
0,0 0 1,1 0,99 2,4 0,147
0,1 0,03 1,2 0,93 2,6 0,107
0,2 0,1 1,3 0,86 2,8 0,077
0,3 0,19 1,4 0,78 3,0 0,55
0,4 0,31 1,5 0,68 3,2 0,04
0,5 0,47 1,6 0,56 3,4 0,029
 0,6 0,66 1,7 0,46 3,6 0,021
0,7 0,82 1,8 0,39 3,8 0,015
0,8 0,93 1,9 0,33 4,0 0,011
0,9 0,99 2,0 0,28 4,5 0,005
1,0 1,0 2,2 0,207 5,0 0

The synthetic unit hydrograph and its corresponding "S" curve are presented in Figure 4.3.
Also in the same Figure the triangular approximation of the SCS-synthetic unit hydrograph
and its "S" curve is shown.

Figure 4.3 The SCS-synthetic unit hydrograph and its triangular approximation

The following variables presented in the graph are defined:

 Duration of effective rainfall D ;

 The concentration time tc between the end of effective rainfall and the inflexion point
of the hydrograph;

 The lag time t1 between the mid-point of the effective rainfall and the time of the peak
discharge (approx. t1 = 0.66 tc);

 The increase time of the hydrograph tp;

  The decrease time of the hydrograph tr;

 The base time of the hydrograph tb.

This triangular SCS-synthetic hydrograph is defined by the parameters up (maximum

discharge) and tp. In the triangular hydrograph 37.5% of the volume of the flood wave is
found just before the peak, hence tp/tb = 0.375 or

(4.9)

The volume of the hydrograph is:

(4.10)

From equation 4.9 one obtains the expression of the peak discharge as a function of
increase time:

(4.11)

(4.12)

with

Considering that:

(4.13)

and replacing the relation 4.13 in 4.12 the parameter K = 0,75

Linking the SCS-synthetic unit hydrograph with the reference duration of effective rainfall, D,
one obtains:

(4.14)
As such, SCS consider that the following relation is valid for the United States:

(4.15)

4.1.5. The "S" hydrograph

The "S" hydrograph is defined as the integral of a hydrograph of a continuous effective

rainfall having a duration greater than the concentration time tc of the basin. Since the
concentration time of the basin represents the necessary travel time that a drop of effective
rain that falls at the endmost point of the catchment would take to reach the outlet, the entire
area of the basin contributes to the formation of the maximum peak discharge of the
hydrograph. Consequently, after the moment when the duration of the rainfall exceeds the
concentration time the "S" curve becomes horizontal, assuming uniform temporal and spatial
characteristics of effective rainfall.

Figure 4.4. The construction of the "S" curve

The "S" curve allows for deriving the unit hydrograph of certain duration Tτ of effective
rainfall, starting from a unit hydrograph having the reference duration τ, through the following
steps (Figure 4.5):

 shift two "S" hydrographs with a time interval T,

 abstract the ordinates of one hydrograph from the other and

 divide by the ratio T/τ

 the unit hydrograph of effective rainfall of duration Tτ is derived.

 Figure 4.5 Computation of the unit hydrograph of reference duration

For cases where the duration T is a multiple n integer of τ (T = nτ), the unit hydrograph can
be found more easily by applying the principle of convolution. Thus the unit hydrograph of
duration τ is shifted n times with the duration τ, the ordinates are added and the resulted
hydrograph is divided by n.

4.2. The instantaneous unit hydrograph (IUH)

4.2.1. Definition and properties

The instantaneous unit hydrograph is defined as a unit hydrograph produced by an effective

rainfall of 1 mm and having an infinitesimal reference duration (in other words the duration
tends towards zero). In terms of the systematic conception this means that the instantaneous
unit hydrograph represents the response of the catchment to an instantaneous impulse. Of
course, this is only a theoretical concept and cannot be realised in actual catchments, but it
is useful as the IUH can be analytically computed and, on the basis of the "S" curve, a
certain unit hydrograph of any reference duration may be derived from it. Moreover, the IUH
can be related to the catchment geomorphology.

It is reminded that the product of convolution of two mathematical functions is defined as:

(4.16)
where:

g(τ) represents the intensity as a function of time t

f(t) is the transfer function

Let S(t) be the dimensionless ordinates of the time t of the "S" curve of a river basin.
According to Figure 4.5 the ordinates of the unit hydrograph of a duration τ are given by the
relation:

(4.17)

where u(t) has the dimensions [h-1, min-1], S is dimensionless and τ is a duration

The ordinates of the instantaneous hydrograph u0 corresponds to a duration hence:

(4.18)

and

(4.19)

Furthermore, the intensity function should be now defined. Let Ie(τ) represent the effective
rainfall intensity function that is multiplied by the area of the river catchment A and by the
duration τ in order to determine effective rainfall volume.

According to Figure 4.7 the product corresponds to the volume

generated by the intensity Ie(τ). Integrating these elementary discharges over the duration of
rainfall, the product of convolution is given by:

(4.20)
 Figure 4.7 Convolution of the intensity function Ie(t) with the instantaneous
unit hydrograph u0(t) giving the discharge hydrograph Q(t)

The boundaries of integration are:

 τ should not exceed the total duration of effective rainfall, that is ;

 t - τ should not exceed the base time of the unit hydrograph Tb , that is

 hence the following condition will be satisfied:

The product of convolution allows for maintenance of the linearity and invariance hypotheses
of the theory of the unit hydrograph.

In the following paragraphs instantaneous unit hydrographs resulting from different

conceptual models are presented.

4.2.2. Instantaneous unit hydrograph resulting from conceptual models

The conceptual unit hydrographs result from the representation of the integration process of
the effective rainfall as a cascade of reservoirs that fill and deplete according to certain laws.
Let us consider one reservoir as presented in the Figure 4.8 that is characterised by three
functions that are the rate of filling I(t), the storage S(t), and the rate of depletion (drainage)
Q(t).
 Figure 4.8 A linear reservoir

The behaviour of the function of this reservoir is ruled according to the following laws:

Continuity law:

(4.21)

Storage law:

(4.22)

where K is the storage coefficient and it is a constant that confers to the reservoir the
property of linearity.

In effect, replacing 4.22 in 4.21 one obtains:

(4.23)

The solution of this equation is of the type:

(4.24)
where M(d) and N(d) are the differential operators relative to the I and Q, respectively. The
ratio M(d)/N(d) is defined as the transfer function of the system. The methods for establishing
above relations as proposed by several authors is discussed below.

4.2.3. The instantaneous unit hydrograph of Nash

The instantaneous unit hydrograph proposed by Nash (1957) is a conceptual model that
uses the effect of "lamination" of an effective rainfall having the depth of 1mm and duration
that tends towards zero. This lamination is performed by a cascade of n linear reservoirs
having identical storage coefficient K (Figure 4.9).

Figure 4.9 The Nash concept for deriving the instantaneous unit hydrograph

Equation (4.23) may be written as:

(4.25)

Considering the equation of convolution the expression of Q(t) is:

(4.26)
The integration of equation (4.25) is obtained by multiplying both terms from the right and left
side with et/K and one obtains:

(4.27)

Thus the output from the first reservoir becomes the input in the second one and so on.
Therefore, replacing t - τ by τ yields:

(4.28)

Again the convolution equation allows for determining the output from the second reservoir:

(4.29)

or:

(4.30)

Reiterating the procedure for n reservoirs the general expression of the ordinates of the
Nash instantaneous unit hydrograph is obtained as:

(4.31)

where Γ(n) is the gamma function defined as:

(4.32)
.......

with the following iterative relation:

 (4.33)

Relation 4.31 is the probabilistic Γ distribution function with two parameters. It is well known
that the integral of the probabilistic distribution function is always equal to 1, which is the
volume of the unit hydrograph.

The increase time tp and the peak discharge Qp are derived by solving the equation dQ/dt =
0, as follows:

(4.34)

and:

(4.35)

Finally, the ordinates of the instantaneous unit hydrograph might be expressed as a function
of the increase time and the peak discharge, as given below:

(4.36)

The instantaneous unit hydrograph derived by Nash is determined only as a function of two
parameters, namely n and K . The influence of these parameters upon the shape and the
magnitude of the hydrograph are presented in Figure 4.10. One observes that while the
parameter n influences the shape of the hydrograph, the parameter Kdetermines the position
of the peak discharge. The higher the parameter K, the further the peak discharge is shifted
with respect to abscissa. When the parameter naugments the peak discharge Q augments
too, since, under the circumstances that tp is kept constant, the increase of the parameter n
corresponds to the diminution of the storage coefficient; and therefore the wave becomes
sharper.
 Figure 4.10. Sensitivity of the Nash model to the parameters n and K(tp)

Another example of the sensitivity of the Nash instantaneous unit hydrograph to the
parameter n while the storage K parameter is kept constant is presented in Figure 4.11.

Figure 4.11. Sensitivity of the Nash instantaneous unit hydrograph to the parameter
n

By taking first order M1k, M1Q and second order M2k, M2Q moments of the effective rainfall
hyetograph and the resulting discharge hydrograph respectively, and taking into account the
convolution equation 4.20 and the analytical expression for the instantaneous unit
hydrograph 4.25, Nash (1957) found the following relations between the parameters n, K and
these moments:

(4.37)
where the moments are computed as presented in Figure 4.12.

Figure 4.12 The method of moments for deriving the instantaneous hydrograph

Moments of the hyetograph of effective rainfall and the hydrograph of the discharge were
determined as follows:

(4.38)
 (4.39)

(4.40)

(4.41)

where:

Qi ordinate of the surface discharge hydrograph [m3/s] at time i = 1, NQ;

hi depth of effective rainfall [m/s] throughout the time interval Δt between the
ordinates i and i + 1;

NH number of segments of the duration Δt of effective rainfall

N
number of ordinates the hydrograph has been split.
Q

Solution of the system of equations 4.31 gives the values of n and K :

(4.42)

(4.43)

Variation in values of n and K determined by application of the method of moments to many

flood events over a given catchment may be explained as follows:
  The method of moments induces errors in the estimation of M2Q due to the errors of
the discharge along the decreasing limb of the hydrograph: the term

has a significant value because i is large.

 The even distribution of effective rainfall around the catchment is only an

approximation.

 The unit hydrograph is considered a constant function of the basin while actually it
depends on the effective rainfall intensity, which will vary from one flood event to
another.

for this reason the method of moments is jointly applied with an optimisation procedure of the
parameters that involves the following operations:

 Let us consider N pairs of hydrograph-hyetograph, resulting in N pairs of parameters,

n and K by the method of moments.

 For each pair of j parameters the ordinates of the instantaneous unit hydrograph are
computed.

 Applying the "S" curve method, the unit hydrographs specific to the considered
reference duration are determined for all flood events. These unit hydrographs are
multiplied by convolution with the corresponding effective rainfalls, resulting in
computed hydrographs.

 The computed Qij(c) and measured Qij(m) hydrographs are then compared using a
criteria function F as follows:

(4.44)

where NQ(j) is the number of the ordinates of each of j hydrographs (that is different from the
number of effective rainfall segments of τ reference duration).

4.4. Regionalization of the unit hydrograph (Synthetic unit

hydrograph)
The unit hydrograph derived from data recorded on rainfall and the corresponding flood hydrograph
applied only for the basin for which the data was available. The synthetic unit hydrograph is computed on
the basis of data recorded at gauging stations and might be used for assessing the flood wave in ungauged
catchments having determined effective rainfall over the basin.
There are three types of synthetic unit hydrographs (Chow et al., 1988)

 Those relating the unit hydrographs characteristics (peak discharge, time to peak, base time, etc.)
to the morphologic parameters of the catchment. The procedure of such geographical interpolation
is called regionalisation.
 Those based on a procedure that makes dimensionless the unit hydrograph (see the US-Soil
Conservation Service unit hydrograph). This could be used for assessing the unit hydrograph in
the similar regions with those for which the dimensionless unit hydrograph is available.
 Those based on conceptual models of the effective rainfall integration process over the basin (see
chapters 4.2.3 - 4.2.5). This type of approach determines the synthetic unit hydrograph either on
the basis of parameters of the mathematical expression of the instantaneous unit hydrograph or on
the basis of moments of several orders of the unit hydrograph. These are further linked with the
morphologic parameters of the catchment. Thus, the third category represents in fact the first one
to which the mathematical model was added.

4.4.1. Snyder's Synthetic unit hydrograph

Snyder (1938) made a study of the unit hydrograph expressed as a function of several characteristics of the
basin. First, Snyder standardized the unit hydrograph as one those rainfall duration tr is related to the
catchment lag time tlag by

(4.83)

For a standard unit hydrograph, the following relations have been derived:

(4.84)

where:

tlag catchment lag time in hours

L length of the main river course in km from the outlet to the upstream divide line

Lc distance in km from the outlet to the point of the river nearest the centroid of
the catchment

C1 coefficient derived from the gauged catchments in the same region (1 for the
English system)

The specific peak discharge [m3/km2] of the standard hydrograph is given by:

(4.85)
where:

C2 = 2,75 (640 for the English system)

Cp coefficient derived from the gauged catchments in the same region

For a given effective rainfall tR that is different of durations tr of the standard hydrograph the
"derived unit hydrograph obtained from the data on effective rainfall and corresponding
hydrograph" would be defined by several parameters as shown in Figure 4.2:

 Peak discharge of the unit hydrograph ; upR

 Lag time defined as the time difference between the centroids of effective rainfall and the
unit hydrograph peak ; tlagR
 Base time as defined above ; tb
 Widths W50 and W75 at 50% and 75% of the unit hydrograph peak discharge

If then tlaf R = 5.5 tR and tR = tr and upR = up and Ct and Cp are computed applying equations 4.82
and 4.83.

If tlaf R ≠ 5.5 tR then the lag time of the standard basin is:

(4.86)

and values of tlag and tr are obtained from equations 4.83 and 4.85, which are solved
simultaneously. Values of the coefficients Ct and Cp are computed from 4.84 and 4.85 with upR =
up and tlaf R = tlag.

Figure 4.20. Snyder's unit hydrograph; (a) Standard unit hydrograph (tlag= 5.5 tr )
(b) Required unit hydrograph (tlag ≠ 5.5 tr). (Chow et al., 1988)

Once the coefficients Ct and Cp are derived for both cases tlaf R = 5.5 tR or tlaf R ≠ 5.5 tR the unit
hydrograph indicators of an ungauged catchment situated in a region that is similar, from the
point of view of physiographical characteristics of the landscape, with the basins from which the
coefficients Ct and Cp were derived, are given by:

 The peak discharge is

(4.87)

 The base time is derived assuming the shape of the unit hydrograph is a triangle of an unit
area (1mm):

(4.88)

where:

C3 = 5,56 (1290 for the English system)

 The widths (hours) of unit hydrograph corresponding to a discharge of the percentage 75%
and 50% is computed as:

(4.89)

where:

Cper = 1,22 (440 for the English system) for per = 75%

Cper = 2,14 (770 for the English system) for per = 50%

Mention is made that practically one-third of the width is distributed before the peak time and
two-thirds after the peak.

4.4.2. Regionalization of the parameters of the instantaneous unit

hydrograph

The synthetic unit hydrograph for ungauged basins are obtained from relationships 4.87-
4.89, which are derived from the correlation of the parameters of the mathematical
expressions of the instantaneous unit hydrograph and the morphologic characteristics of the
catchments. These relations are conceptually based on the physical meaning of the
parameters while performing the conceptual model of unit hydrograph. For example,
according to relations 4.34 and 4.35 the time to peak and the maximum discharge of Nash
instantaneous unit hydrograph are function of the parameters n and K, which define the
ordinates of IUH given by the function of distribution Γ with two parameters. At their turn the
time to peak tp and the maximum discharge Qp are physically related with the morphologic
parameters of the basin. Good relationships have been found between the time to peak tp

and the morphologic parameter , where Ib, Ir are the mean slopes of the basin and the
river, respectively, and F is the area of the basin. This parameter can be written as

and the first term expresses the velocity along the river and the second one the
velocity along the slope. Both velocities are determinant for the time to peak of the
hydrograph. Thus

(4.90)

Again, the peak discharge is a function of the shape of the basin (see Figure 4.16) and the
concentration time of the basin. Physically, the shape of the basin is dependent on the shape
coefficient of the basin B/L where B and L are the mean width of the basin F/L and the
length, respectively. Also the concentration time depends on both the slope of the river and
of the basin so that finally the peak discharge of the unit hydrograph may be written as a
function:

(4.91)

In Figures 4.21 and 4.22 regionalisation relationships of characteristics tp and Qp of the

instantaneous hydrograph function of morphologic parameters of the basin determined by
Serban (1979) for the Mures basin (Romania) are presented.
 Figure 4.21 Time to peak of the instantaneous unit hydrograph

function of the morphologic parameter

Figure 4.22 Peak discharge of the instantaneous unit hydrograph

function of the morphologic parameter

Further on the determination of the parameters of the instantaneous unit hydrograph function
is made as follows:

 The morphologic parameters are determined on the basis of maps of the catchments

 The system of equations 4.34 and 4.35 is solved by constructing the function

and having the product Qp.tp from the graph of the types presented in Figures 4.21
and 4.22 the value of the parameter n is determined.

 The value of the second parameter is then obtained from equation 4.34:

(4.92)
Figure 4.23 Graph of the function