LECTURE NOTES - VIII

« WATER RESOURCES »

Prof. Dr. Atıl BULU

Istanbul Technical University

College of Civil Engineering
Civil Engineering Department
Hydraulics Division
 CHAPTER 8

FLOW ROUTING

Flow routing is a procedure to determine the time and magnitude of flow (i.e., the flow
hydrograph) at a point on a water course from known or assumed hydrographs at one or
more points. If the flow is a flood, the procedure is specifically known as flood routing.
In a broad sense, flow routing may be considered as analysis to trace the flow trough a
hydrologic system, given the input. (Chow, et.al., 1988)

8.1. ROUTING EQUATION

Flow routing can be introduced using the black-box concept of systems theory. The input
and output functions consist of the runoff hydrographs at the upstream and downstream
sections of a channel; the transfer function is the channel routing procedure that is used to
translate and attenuate the upstream runoff hydrograph into a downstream hydrograph.
The transfer function usually consists of two parts, the routing method and the physical
characteristics of the stream reach. In theory, we usually assume that the channel
characteristics are known, so the routing method becomes the transfer function.

Figure 8.1. Channel routing

For a hydrologic system, input I(t), output Q(t), and storage S(t) are related by continuity
equation:

= I (t ) − Q(t )
dS
(8.1)
dt

If the flow hydrograph, I(t), is known, Equ. (8.1) cannot be solved directly to obtain the
outflow hydrograph, Q(t), because both Q and S are unknown. A second relationship, or
storage function, is needed to relate S, I, and Q.

155 A.Bulu
S = f (I , Q ) (8.2)

Figure 8.2. Relationships between discharge and storage.

The relationship between the outflow and the storage of a hydrologic system has an
influence on flow routing. This relationship may be either invariable or variable, as
shown in Fig. (8.2).

An invariable storage function has the form of,

S = f (Q ) (8.3)

and applies to a reservoir with horizontal water surface. When a reservoir has horizontal
water surface, its storage is a function of its water surface elevation, or depth in the pool.
Likewise, the outflow discharge is a function of the water surface elevation, or head on
the outlet works. By combining these two functions, the reservoir storage and discharge
can be related to produce an invariable, single valued storage function, S = f(Q) as shown
in Fig. (8.2.a). For such reservoirs, the peak outflow occurs when the outflow hydrograph
intersects the inflow hydrograph, because the maximum storage occurs when,

= I (t ) − Q(t ) = 0 → I (t ) = Q(t )
dS
(8.4)
dt

For this case, the storage and outflow are related by S = f(Q). This is indicated in Fig.
(8.2.a) where the points denoting the maximum storage, R, and maximum outflow, P,
coincide.

A variable storage-outflow relationship applies to long, narrow reservoirs and to open

channel or streams, where the water surface profile may be significantly curved due to

156 A.Bulu
the backwater effects. The amount of storage due to backwater depends on the time rate
of change of flow through the system. As shown in Fig. (8.2.b), the resulting relationship
between the discharge and the system storage is no longer a single-valued function but
exhibits a curve usually in the form of a single or twisted loop, depending on the storage.
Because of the retarding effect due to backwater, the peak outflow usually occurs later
than the time when the inflow and outflow intersect, as indicated in Fig. (8.2.b) where the
points R and P do not coincide. If the backwater effect is not very significant, the loop
shown in Fig. (8.2.b) may be replaced by an average curve shown by the dashed line.
Thus, level pool routing methods can also be applied in an approximate way to routing
with an invariable discharge discharge-storage relationship.

The effect of storage is to redistribute the hydrograph by shifting the centroid of the
inflow hydrograph to the position of that of the outflow hydrograph centroid in a time of
redistribution.

In very long channels, the entire flood wave travels a considerable distance and the
centroid of its hydrograph may then be shifted by a time longer than the time of
redistribution. This additional time may be considered as time of translation. As shown in
Fig. (8.3), the total time of flood movement between the centroids of the outflow and
inflow hydrographs is equal to the sum of the time of redistribution and the time of
translation. The process of redistribution modifies the shape of the hydrograph, while
translation changes its position.

Figure 8.3. Conceptual interpretation of the time of flood movement

157 A.Bulu
8.2. FLOOD ROUTING ON A HORIZONTAL POOL

Flood routing on a horizontal pool is a procedure for calculating the outflow hydrograph
from a reservoir with a horizontal water surface, given its inflow hydrograph and storage-
outflow characteristics.

The time horizon is broken into intervals of duration Δt, indexed by j, that is, t=0, Δt, 2Δt,
……, jΔt, (j+1)Δt,…….., and the continuity Equ. (8.1) is integrated over each time
interval, as shown in Fig. (8.4).

For the j-th time interval,

S j +1 ( j +1)Δt ( j +1)Δt

∫ dS = ∫ I (t )dt − ∫ Q(t )dt

Sj jΔt jΔt
(8.5)

Figure 8.4. Change of storage during a routing period Δt.

The inflow values at the beginning and end of the j-th time interval are Ij and Ij+1,
respectively, and the corresponding values of the outflow are Qj and Qj+1. Here, both
inflow and outflow data a sample data. If the variation of inflow and outflow over the
interval is approximately linear, (Sj+1 – Sj), can be found by rewriting Equ. (8.5) as,

I j + I j +1 Q j + Q j +1
S j +1 − S j = Δt − Δt (8.6)
2 2

158 A.Bulu
The values of Ij and Ij+1 are known because they are prespecified. The values of Qj and Sj
are known at the j-th time interval from calculation from the previous time interval.
Hence, Equ. (8.6) contains two unknowns, Qj+1 and Sj+1, which are isolated by
multiplying Equ. (8.6) through by (2/Δt), and rearranging the result produces;

⎛ 2 S j +1 ⎞ ⎛ 2S j ⎞
⎜⎜ + Q j +1 ⎟⎟ = (I j + I j +1 ) + ⎜⎜ − Q j ⎟⎟ (8.7)
⎝ Δt ⎠ ⎝ Δt ⎠

In order to calculate the outflow, Qj+1, from Equ. (8.7), a storage-outflow function
relating (2S/Δt+Q) and Q is needed. The method for developing this function using
elevation-storage and elevation-outflow relationships is shown in Fig. (8.5).

Figure 8.5. Development of the storage-outflow function for pool routing on the basis of
storage-elevation and elevation-outflow curves.

The relationship between water surface elevation and reservoir storage can be derived by
planimetering topographic maps or from field surveys. The elevation-discharge relation is
derived from hydraulic equations relating head and discharge, such as those shown in
Table (8.1), for various types of spillways and outlet works. The value of Δt is taken as
the time interval of the inflow hydrograph. For a given value of water surface elevation,
the values of S and discharge Q determined [parts (a) and (b) of Fig. (8.5)], then the value
of (2S/Δt+Q) is calculated and plotted on the horizontal axis of a graph with the value of
the outflow Q on the vertical axis [part (c) of Fig. (8.5)].

159 A.Bulu
In routing the flow through time interval j, all terms on the right side of Equ. (8.7) are
known, and so the value of (2 S j +1 Δt + Q j +1 ) can be computed. The corresponding value
of Qj+1 can be determined from the storage-outflow function (2S/Δt+Q) versus Q, either
graphically or by linear interpolation of tabular values. To set up the data required for the
⎛ 2 S j +1 ⎞
next interval, the value of ⎜⎜ − Q j +1 ⎟⎟ is calculated by,
⎝ Δt ⎠

Table 8.1. Spillway discharge equations

160 A.Bulu
⎛ 2 S j +1 ⎞ ⎛ 2 S j +1 ⎞
⎜⎜ − Q j +1 ⎟⎟ = ⎜⎜ + Q j +1 ⎟⎟ − 2Q j +1 (8.8)
⎝ Δt ⎠ ⎝ Δt ⎠

The computation is repeated for subsequent periods.

EXAMPLE 8.1: The river is diverted to a tunnel by a cofferdam during the construction
of a dam. The relationship between the water stage at the entrance of the tunnel, the
storage volume S and the tunnel capacity is given in Table (8.2). The 25-yr flood
hydrograph to be used in the design of the cofferdam is shown in Table (8.3). During a
flood, the flood wave is dampened in the pool before it enters the tunnel. Determine the
maximum water surface elevation in the pool.

Solution : Using the relationship between the water volume stored in the pool S and the
outflow Q, the values of (2S/Δt+Q) are computed and given in Table (8.2) for Δt = 6hr=
6×3600= 21600 sec.

Table 8.2. Storage-outflow function

Water Stage Discharge Storage (2S/Δt+Q)

(m) Q (m3/sec) S ( 106m3) (m3/sec)
(1) (2) (3) (4)
395 900 1.00 993
400 1350 2.00 1535
405 1850 2.75 2105
413 2550 20.00 4402
420 3100 44.50 7220
430 3950 121.00 15150
440 4650 285.20 31060
450 5300 567.00 57800

80000
(2S/Δt+Q)

60000
40000
20000
0
0 2000 4000 6000
3
Outflow (m /s)

Figure 8.6. Storage-outflow function

161 A.Bulu
Assuming that the reservoir is empty initially, Equ. (8.7) is applied with time steps Δt = 6
hr. Computations are shown in Table (8.3) with the following steps.

1) Inflow value at time j, Ij, is added to the Ij+1 of the column (3) and written in column
(4).
2) Using Equ. (8.7),

⎛ 2 S j +1 ⎞ ⎛ 2S j ⎞
⎜⎜ + Q j +1 ⎟⎟ = (I j + I j +1 ) + ⎜⎜ − Q j ⎟⎟
⎝ Δt ⎠ ⎝ Δt ⎠
⎛ 2S 2 ⎞
⎜ + Q2 ⎟ = 1060 + 0 = 1060 m 3 sec
⎝ 6 × 3600 ⎠

and written in column (6).

⎛ 2 S j +1 ⎞
3) The outflow value of Qj+1 corresponding to ⎜⎜ + Q j +1 ⎟⎟ is calculated from Table
⎝ Δt ⎠
(8.2) by interpolation as,

2S 2
+ Q2 = 1060 m 3 s = x
Δt
Q=y

2S
Two pairs of values around + Q = 1060 m 3 s are selected as (x1, y1) and (x2, y2) from
Δt
Table (8.3). They are (x1, y1) = (993, 900) and (x2, y2) = (1535, 1350). The value of y =
Q2 for x = 1060 is, by linear interpolation,

Q2 = 900 +
(1350 − 900) × (1060 − 993) = 956 m 3 s
(1535 − 993)
⎛ 2S ⎞ ⎛ 2S ⎞
4) 2Q2 is subtracted from ⎜ 2 + Q2 ⎟ to obtain ⎜ 2 − Q2 ⎟ by using Equ. (8.8).
⎝ Δt ⎠ ⎝ Δt ⎠

⎛ 2S 2 ⎞
⎜ − Q2 ⎟ = 1060 − 2 × 956 = −852
⎝ Δt ⎠

The calculations for subsequent time intervals are performed in the same way with the
results tabulated in the Table (8.3).

162 A.Bulu
 Table 8.3. Routing of flow through a detention reservoir by the level pool method.

Time Time Inflow Ij+Ij+1 2S j +1 2S j +1 Outflow

Index j (hr) I (m3/sec) (m3/sec) − Q j +1 + Q j +1 Q
Δt Δt 3
(1) (2) (3) (4) (5) (6) (m /sec)
(7)
1 0 0 - 0 - 0
2 6 1060 1060 -852 1060 956
3 12 2410 3470 -1394 2618 2006
4 18 4240 6650 -202 5216 2709
5 24 5930 10170 3178 9968 3395
6 30 8290 14220 9300 17398 4049
7 36 8840 17130 17538 26430 4446
8 42 8290 17130 25192 34668 4738
9 48 7410 15700 31114 40892 4889
10 54 6560 13970 35102 45084 4991
11 60 5540 12100 37118 47202 5042
12 66 4580 10120 37152 47238 5043
13 72 3670 8250 35404 45402 4999

Largest discharge is 5043 (m3/s), which corresponds to the stage using the values of the
Table (8.3),

(z1, y1) = (440, 4650) , (z2, y2) = (450, 5300)

z = 440 +
(450 − 440) × (5043 − 4650) = 446m
(5300 − 4650)
Thus, the highest water elevation in the pool to be used in the design of the cofferdam
should be greater than 446 m.

10000
8000
6000 Inflow
4000 Outflow
2000
0
1 3 5 7 9 11 13

Figure 8.7. Routing of flow through a detention reservoir.

163 A.Bulu
EXAMPLE 8.2. A reservoir for detaining flood flows is 4000 m2 in horizontal area, and
has a 1.50 m diameter reinforced concrete pipe as the outlet structure. The headwater-
discharge relation for the outlet pipe is given in columns (1) and (2) of Table (8.4). Use
the level pool routing method to calculate the reservoir outflow from the inflow
hydrograph given in columns (2) and (3) of Table (8.4). Assume that reservoir is initially
empty.

Table 8.4. Development of storage-outflow function for a detention reservoir. Time

interval, Δt=10 min.
Elevation Discharge Storage (2S/Δt)+Q
3 3
(m) Q (m /sec) S (m ) (m3/sec)
(1) (2) (3) (4)
0 0 0 0
0.15 0.09 600 2.09
0.30 0.23 1200 4.23
0.45 0.48 1800 6.48
0.60 0.85 2400 8.85
0.75 1.22 3000 11.22
0.90 1.70 3600 13.70
1.05 2.21 4200 16.21
1.20 2.75 4800 18.75
1.35 3.31 5400 21.31
1.50 3.88 6000 23.88
1.65 4.42 6600 26.42
1.80 4.90 7200 28.90
1.95 5.38 7800 31.38
2.10 5.80 8400 33.80
2.25 6.17 9000 36.17
2.40 6.54 9600 38.54
2.55 6.85 10200 40.85
2.70 7.16 10800 43.16
2.85 7.48 11400 45.48
3.00 7.79 12000 47.79

Solution: The inflow hydrograph is specified at 10-min intervals, so Δt=10×60=600 s.

For all elevations, the horizontal area of the reservoir water surface is 4000 m2, and the
storage is calculated as S = 4000× (depth of water). For example, for a depth of 0.15 m,
S=0.15×4000=600 m3, as shown in column (3) of Table (8.4). The corresponding value
2S
of ( + Q ) can then be determined. For a depth of 0.15 m, the discharge is given in
Δt
column (2) of Table (8.4) as 0.09 m3/sec, so the storage-outflow function value is,

164 A.Bulu
2S 2 × 600
+Q = + 0.09 = 2.09 m 3 s ec
Δt 600

as shown in column (4) of Table (8.4). The storage-outflow function is plotted in Figure
(8.8).

60
2S/t+Q (m3/s)

50
40
30
20
10
0
0 5 10 15 20 25
Outflow (m3/s)

Figure 8.8. Storage-outflow function for a detention reservoir

The flow routing calculations are carried out using Equ. (8.7). For the first time interval,
S1=Q1=0 because reservoir is initially empty; hence,

2S1
− Q1 = 0
Δt

also. The inflow values are I1=0 and I2=1.70, so (I1+I2)=0+1.70=1.70 m3/s. The value of
the storage-outflow function at the end of the time interval is calculated from Equ. (8.7)
with j=1,

⎛ 2S 2 ⎞ ⎛ 2S ⎞
⎜ + Q2 ⎟ = I1 + I 2 + ⎜ 1 − Q1 ⎟
⎝ Δt ⎠ ⎝ Δt ⎠
⎛ 2S 2 ⎞
⎜ + Q2 ⎟ = 1.70 + 0 = 1.70 m 3 s ec
⎝ Δt ⎠

165 A.Bulu
 Table 8.5. Routing of flow through a detention reservoir by the level pool method.

Time Time Inflow Ij+Ij+1 2Sj/Δt-Qj 2Sj+1/Δt+Qj+1 Outflow

Index j (min) (m3/s) (m3/s) (m3/s) (m3/s) (m3/s)
(1) (2) (3) (4) (5) (6) (7)
1 0 0 - 0 - 0
2 10 1.70 1.70 1.55 1.70 0.073
3 20 3.40 5.10 4.32 5.34 0.51
4 30 5.10 8.50 9.76 12.82 1.53
5 40 6.80 11.90 14.88 21.66 3.39
6 50 8.50 15.30 19.88 30.18 5.15
7 60 10.20 18.70 25.48 38.58 6.55
8 70 9.10 19.30 30.02 44.78 7.38
9 80 7.90 17.00 31.71 47.09 7.69
10 90 6.80 14.70 31.21 46.41 7.60
11 100 6.60 13.40 29.89 44.61 7.36
12 110 4.50 11.10 27.25 40.99 6.87
13 120 3.40 7.90 23.13 35.15 6.01
14 130 2.30 5.70 19.05 28.83 4.89
15 140 1.10 3.40 15.33 22.45 3.56
16 150 0 1.10 11.91 16.43 2.26
17 160 - 0 9.21 11.91 1.35
18 170 - - 7.39 9.21 0.91
19 180 - - 6.15 7.39 0.62
20 190 - - 5.27 6.15 0.44
21 200 - - 4.57 5.27 0.35
22 210 - - 4.03 4.57 0.27

⎛ 2S ⎞
The value of Qj+1 is found by linear interpolation given ⎜⎜ j +1 + Q j +1 ⎟⎟ . If there is a pair
⎝ Δt ⎠
of values (x,y) with known pairs of values (x1,y1) and (x2,y2), then the interpolated value
of y corresponding to a given value of x in the range (x1≤ x ≤ x2) is,

y = y1 +
( y2 − y1 ) × (x − x )
(x2 − x1 ) 1

2S 2S
In this case, x = + Q , and y=Q. Two pairs of values around + Q = 1.70 are
Δt Δt
selected from Table (8.5); they are (x1,y1) = (0, 0) and (x2,y2) = (2.09,0.09). The value of
y for x=1.70 is, by linear interpolation,

166 A.Bulu
y = 0+
(0.09 − 0) × (1.70 − 0) = 0.07 m33 s ec
(2.09 − 0)
⎛ 2S ⎞
So, Q2=0.73 m3/s, and the value of ⎜ 2 − Q2 ⎟ needed for the next iteration is found by
⎝ Δt ⎠
using Equ. (8.8) with j=2,

⎛ 2S 2 ⎞ ⎛ 2S ⎞
⎜ − Q2 ⎟ = ⎜ 2 + Q2 ⎟ − 2Q2
⎝ Δt ⎠ ⎝ Δt ⎠
⎛ 2S 2 ⎞
⎜ − Q2 ⎟ = 1.70 − 2 × 0.073 = 1.55
⎝ Δt ⎠

Proceeding the next interval,

(I 2 + I 3 ) = 1.70 + 3.40 = 5.10 m 3 s ec

and the routing is performed with j=2 as,

⎛ 2S 3 ⎞ ⎛ 2S ⎞
⎜ + Q3 ⎟ = (I 2 + I 3 ) + ⎜ 2 − Q2 ⎟
⎝ Δt ⎠ ⎝ Δt ⎠
⎛ 2S 3 ⎞
⎜ + Q3 ⎟ = 5.10 + 1.55 = 6.55 m 3 s ec
⎝ Δt ⎠

For x=6.55, from Table (8.4). (x1,y1) = (6.48,0.48) and (x2,y2) = (8.85,0.85),

Q3 = y = 0.48 +
(0.85 − 0.48) × (6.65 − 6.48) = 0.51m 3 s ec
(8.85 − 6.48)
⎛ 2S 3 ⎞
⎜ − Q3 ⎟ = 5.34 − 2 × 0.51 = 4.32 m 3 s ec
⎝ Δt ⎠

The calculations for subsequent time intervals are performed in the same way, with the
results tabulated in Table (8.5) and plotted in Fig. (8.9). The peak flow is 10.20 m3/sec
and occurs at 60 min; the detention reservoir reduces the peak flow to 7.69 m3/sec and
delays it until 80 min. The outflow is maximized at the point where the inflow and
outflow are equal, because storage is also maximized at that time, and there is a single-
valued function relating storage and outflow for a reservoir with a level pool.

The maximum depth in the storage reservoir is calculated by interpolation from Table
(8.4) as 2.95 m at the peak discharge of 7.69 m3/sec. If this depth is too great, or if the
discharge of 7.69 m3/sec in the 1.50 m outlet pipe is too large, either the outlet structure
or the surface area of the basin must be enlarged.

167 A.Bulu
 12

10

8
Discharge

Inflow
6
Outflow
4

0
1 4 7 10 13 16 19 22
Time

Figure 8.9. Routing of flow through a detention reservoir.

EXAMPLE 8.3. The following data pertains to an inflow hydrograph whose flows have
been recorded at 6 hours interval on a certain stream.

Table 8.6
Time Inflow
(h) (m3/sec)
0 42
6 45
12 57
18 88
24 147
30 210
36 272
42 340
48 350
54 338
60 314
66 288
72 263
78 240
84 198
90 170
96 143
102 120

This flood approaches a reservoir with uncontrolled spillway with elevation area and
elevation outflow data as shown in Table (8.7).

168 A.Bulu
 Table 8.7

Elevation Area Outflow

(m) (km2) (m3/sec)
100 4.05 0
100.3 4.12 15
100.6 4.20 42
100.9 4.25 77
101.2 4.28 119
101.5 4.36 167
101.8 4.45 217
102.1 4.53 272
102.4 4.60 334
102.7 4.69 405

The flood water level reaches the crest level (elevation 100.00m) of the spillway at 4
hours after the inflow recording station.

Determine the maximum reservoir level and maximum discharge over the spillway. Draw
inflow and routed hydrograph indicating the reduction in peak flow and peak lag
introduced due to routing.

Solution:

1) Draw the elevation–outflow curve as shown in Figure (8.10),

500
Q (m3/s)

400
300
200
100
0
99.5 100 100.5 101 101.5 102 102.5 103
Elevation (m)

Figure 8.10. Elevation-outflow curve.

169 A.Bulu
2) The storage between elevations (100-100.3) m is calculated by,

A100 + A100.3
S= × (100.0 − 100.3)
2
(4.05 + 4.12) ×106
S= × 0.30 = 1.23 ×106 m3
2

Storages between elevations are computed and given in column (4) at Table (8.8).

3) Computing the storage for the elevations given in Table (8.8), calculating the storages
as in step 2, taking the outflow discharges from the Table (8.7), (2S Δt + Q ) values are
calculated and given in Table (8.8) in column (6) for Δt=6 hr= 6×3600=21600 sec.

4) Assuming that there is no flow over the spillway, inflow at time j, Ij, is added to the
Ij+1 of the column (3) and written in column (4) at Table (8.9).

I1 + I 2 = 42 + 45 = 87 m 3 s ec

5) Using Equ. (8.7),

⎛ 2 S j +1 ⎞ ⎛ 2S j ⎞
⎜⎜ + Q j +1 ⎟⎟ = (I j + I j +1 ) + ⎜⎜ − Q j ⎟⎟
⎝ Δt ⎠ ⎝ Δt ⎠

⎛ 2S 2 ⎞
⎜ + Q2 ⎟ = 87 + 0 = 87 m 3 s ec
⎝ 6 × 3600 ⎠

and written in column (6).

⎛ 2 S j +1 ⎞
6) The outflow Qj+1 corresponding to ⎜⎜ + Q j +1 ⎟⎟ is calculated from table by
⎝ Δt ⎠
interpolation as,

2S 2
+ Q2 = 87 m 3 sec = x
Δt
Q=y

Two pairs of values around x = 87 m3/s are selected as (x1,y1) = (0,0) and (x2,y2) =
(129,14.9). The value of y = Q2 for x = 87 is, by linear interpolation,

Q2 = 0 +
(14.9 − 0) × (87 − 0) = 10 m 3 s ec
(129 − 0)

170 A.Bulu
 ⎛ 2S ⎞ ⎛ 2S ⎞
7) 2Q2 is subtracted from ⎜ 2 + Q2 ⎟ to obtain ⎜ 2 − Q2 ⎟ by using Equ. (8.8),
⎝ Δt ⎠ ⎝ Δt ⎠

⎛ 2S 2 ⎞
⎜ − Q2 ⎟ = 87 − 2 ×10 = 67 m 3 s ec
⎝ Δt ⎠

The calculations for subsequent time intervals are performed in the same way with the
results tabulated in Table (8.9).

Table 8.8. Development of storage-outflow function for a detention reservoir. Time

interval Δt = 6hr.

Storage between Cumulative

Elevation Outflow Area contours Storage
(m) (m3/sec) between
S=
( A1 + A2 )
× ( H 2 − H1 )
above (2S / Δt + Q )
Contours 2 crest
A (km2) (106m3) (106m3) (m3/sec)
(1) (2) (3) (4) (5) (6)
100 0 4.05 - - -
100.30 14.9 4.12 1.23 1.23 129
100.60 42.2 4.20 1.25 2.45 269
100.90 77.3 4.25 1.27 3.72 422
101.20 119 4.28 1.28 5.00 582
101.50 167 4.36 1.30 6.30 750
101.80 217 4.45 1.32 7.62 923
102.10 272 4.53 1.35 8.97 1103
102.40 334 4.60 1.37 10.34 1291
102.70 405 4.69 1.39 11.73 1491

2000
(2S/t+Q)

1500
1000
500
0
0 100 200 300 400 500
Outflow (m3/s)

171 A.Bulu
 Figure 8.11. Storage-outflow function
Table 8.9. Routing of flow calculations.

Time Time Inflow Ij+Ij+1 2S j +1 2S j +1 Outflow

Index (hr) (m3/sec) (m3/sec) − Q j +1 + Q j +1 Q
Δt Δt 3
j (2) (3) (4) (5) (6) (m /sec)
(1) (7)
1 0 42 - - - 0
2 6 45 87 67 87 10
3 12 57 102 123 169 23
4 18 88 145 184 268 42
5 24 147 235 265 419 77
6 30 210 357 362 622 130
7 36 272 482 456 844 194
8 42 340 612 546 1068 261
9 48 350 690 604 1236 316
10 54 338 688 624 1292 334
11 60 314 652 618 1276 329
12 66 288 602 598 1220 311
13 72 263 551 575 1149 287
14 78 240 503 550 1078 264
15 84 198 438 514 988 237
16 90 170 368 472 882 205
17 96 143 313 431 785 177
18 102 120 263 392 694 151

Maximum discharge over spillway from column (7), Qmax =334 m3/sec,

Maximum reservoir level corresponding to Qmax = 334 m3/sec by using Table (8.8) is
102.4 m.

Reduction in peak discharge, I max − Qmax = 350 − 334 = 16 m 3 s ec

Lag time between the maximum inflow and outflow discharges

Peak lag = 54 – 48 = 6 hours.

172 A.Bulu
 400

350

300

250
Inflow
200
Outflow
150

100

50

0
1 3 5 7 9 11 13 15 17

Figure 8.12. Routing of flow hydrographs.

8.3. HYDROLOGIC RIVER ROUTING

The Muskingum method is a commonly used hydrologic routing method for handling a
variable discharge-storage relation. This method models the storage volume of flooding
in a river channel by a combination of wedge and prism storages. During the advance of a
flood wave, inflow exceeds outflow, producing a wedge of storage. During recession,
outflow exceeds inflow, resulting a negative wedge. In addition, there is a prism of
storage which is formed by a volume of constant cross section along the length of
prismatic channel.

Figure 8.12. Prism and wedge storages in a channel reach

173 A.Bulu
Assuming that cross-sectional area of the flood flow is directly proportional to the
discharge at the section, the volume of prism storage is equal to KQ where K is a
proportionality coefficient, and the volume of wedge storage is equal to KX(I-Q), where
X is weighing factor having the range (0≤ X ≤ 0.5). The total storage is therefore the sum
of two components,

S = KQ + KX (I − Q ) (8.9)

which can be arranged to give the storage function for the Muskingum method.

S = K [XI + (1 − X )Q] (8.10)

and represents a linear model for routing flow in streams.

The value of X depends on the shape of the modeled wedge storage. The value of X
ranges from 0 for reservoir–type storage to 0.5 for a full wedge. When X = 0, there is no
wedge and hence no backwater; this is the case for a level-pool reservoir. In this case,
Equ. (8.10) results in a linear-reservoir model, S =KQ. In natural streams, X is between 0
and 0.3 with a mean value 0.2. The parameter K is the time of travel of the flood wave
through the channel reach. For hydrologic routing, the values of K and X are assumed to
be specified and constant throughout the range of flow.

The values of storage at time j and j+1 can be written, respectively, as,

[
S j = K XI j + (1 − X )Q j ] (8.11)

and

[
S j +1 = K XI j +1 + (1 − X )Q j +1 ] (8.12)

Using Equs. (8.11) and (8.12), the change in storage over time interval Δt (Fig.8.12) is,

[ ] [
S j +1 − S j = K { XI j +1 + (1 − X )Q j +1 − XI j + (1 − X )Q j ]} (8.13)

The change in storage can also be expressed, using Equ. (8.6) as,

S j +1 − S j =
(I j + I j +1 )
Δt −
(Q j + Q j +1 )
Δt (8.14)
2 2

Combining Equs. (8.13) and (8.14) and simplifying gives,

Q j +1 = C1 I j +1 + C2 I j + C3Q j (8.15)

174 A.Bulu
which is routing equation for the Muskingum method where,

Δt − 2 KX
C1 = (8.16)
2 K (1 − X ) + Δt

Δt + 2 KX
C2 = (8.17)
2 K (1 − X ) + Δt

2 K (1 − X ) − Δt
C3 = (8.18)
2 K (1 − X ) + Δt

Note that,

C1 + C2 + C3 = 1 (8.19)

If observed inflow and outflow hydrographs are available for a river reach, the values of
K and X can be determined. Assuming various values of X and using known values of
the inflow and outflow, successive values of the numerator and denominator of the
following expression for K, derived from Equs. (8.13) and (8.14), can be computed as,

K=
[
0.5Δt (I j +1 + I j ) − (Q j +1 + Q j )] (8.19)
X (I j +1 − I j ) + (1 − X )(Q j +1 − Q j )

The computed values of the numerator and denominator are plotted for each time
interval, with the numerator on the vertical axis and denominator on the horizontal. This
usually produces a graph in the form of a loop. The value of X that produces a loop
closest to a single line is taken to be the correct value for the reach, and K, according to
Equ. (8.19), is equal to the slope of the line. Since K is the time required for the
incremental flood wave to traverse the reach, its value may also be determined as the
observed time of travel of peak flow trough the reach.

EXAMPLE 8.4: The inflow hydrograph to a river reach is given in columns (1) and (2)
of Table (8.10). Determine the outflow hydrograph from this reach if K = 2.3 h, X = 0.15,
and Δt= 1 h. The initial outflow is 85 m3/sec.

Solution: Determine the coefficients C1, C2, and C3 using Equs. (8.16) – (8.18):

1 − 2 × 2.3 × 0.15 0.31

C1 = = = 0.0631
2 × 2.3 × (1 − 0.15) + 1 4.91

175 A.Bulu
 1 + 2 × 2.3 × 0.15 1.69
C2 = = = 0.3442
4.91 4.91
2 × 2.3 × (1 − 0.15) − 1 2.91
C3 = = = 0.5927
4.91 4.91

Check to see that the sum of the coefficients C1, C2, and C3 is equal to 1.

0.31 + 1.69 + 2.91 4.91

C1 + C2 + C3 = = = 1.0
4.91 4.91

For the first time interval, the outflow is determined using values for I1= 93 m3/sec and I2
= 137 m3/sec from Table (8.10), the initial outflow is Q1 = 85 m3/sec, and using Equ.
(8.15) with j=1.

Q2 = C1 I 2 + C 2 I1 + C3Q1
0.31×137 + 1.69 × 93 + 2.91× 85 446.99
Q2 = = = 91 m 3 s ec
4.91 4.91

as shown in columns (3) to (6) of Table (8.10). Computations for the following time
intervals use the same procedure with j =2, 3,……to produce the results shown in Table
10. The inflow and outflow hydrographs are plotted in Figure (8.13). It can be seen that
the outflow lags the inflow by approximately 2.3 h, which was the value of K used in the
computations and represents the travel time in the reach.
Discharge (m3/s)

800

600
Inflow
400
Outflow
200

0
1 4 7 10 13 16 19
Time (h)

Figure 8.13. Routing of flow through a river reach by Muskingum method.

176 A.Bulu
 Table 8.10. Flow routing through a river reach by the Muskingum method
Routing Inflow C1Ij+1 C2Ij C3Qj Outflow
period j I C1=0.0631 C2=0.3442 C3=0.5927 Q
(h) (m3/sec) (m3/sec)
(1) (2) (3) (4) (5) (6)
1 93 - - - 85
2 137 8.64 32.01 50.38 91
3 208 13.12 47.16 53.94 114
4 320 20.19 71.59 67.57 159
5 442 27.89 110.14 94.24 232
6 546 34.45 152.14 137.51 324
7 630 39.75 187.93 192.03 420
8 678 42.78 216.85 248.93 509
9 691 43.60 233.37 301.68 579
10 675 42.59 237.84 343.17 624
11 634 40.01 232.34 369.84 642
12 571 36.03 218.22 380.51 635
13 477 30.10 196.54 376.36 603
14 390 24.61 164.18 357.40 546
15 329 20.76 134.24 323.61 479
16 247 15.59 113.24 283.90 413
17 184 11.61 85.02 244.79 341
18 134 8.46 63.33 202.11 274
19 108 6.81 46.12 162.40 215
20 90 5.68 37.17 127.43 170

EXAMPLE 8.5: The infow flood hydrograph observed at the entrance of a stream reach
is given in Table (8.11). Determine the outflow hydrograph at the exit, 18 km from the
entrance, by Muskingum method. The flood wave propagates in the stream with a
velocity of 2 m/sec. Muskingum coefficient X will be taken as 0.25 and time step is Δt =
2hr.

Table 8.11.

Time 0 1 2 3 4 5 6 7 8 9 10
(hr)
Inflow 10 12 18 28.5 50 78 107 134.5 147 150 146
(m3/sec)
Time 11 12 13 14 15 16 17 18 19 20 21
(hr)
Inflow 129 105 78 59 45 33 24 17 12 10 10
(m3/sec)

177 A.Bulu
Solution: The computation step will be taken as Δx = 6 km. Since K is the time it takes
the flood wave to pass through the reach,

Δx 6000
K= = = 3000 sec
V 2

The coefficients C1, C2, and C3 are calculated by using Equs. (8.16)-(8.18);

Δt − 2 KX
C1 =
2 K (1 − X ) + Δt
2 × 3600 − 2 × 3000 × 0.25 5700
C1 = = = 0.487
2 × 3000 × (1 − 0.25) + 2 × 3600 1170

Δt + 2 KX
C2 =
2 K (1 − X ) + Δt
2 × 3600 + 2 × 3000 × 0.25 8700
C2 = = = 0.744
2 × 3000 × (1 − 0.25) + 2 × 3600 11700

2 K (1 − X ) − Δt
C3 =
2 K (1 − X ) + Δt
2 × 3000 × (1 − 0.25) − 2 × 3600 − 2700
C3 = = = −0.231
2 × 3000 × (1 − 0.25) + 2 × 3600 11700

Check to see that the sum of coefficients C1, C2, and C3 is equal to 1.

5700 + 8700 − 2700

C1 + C2 + C3 = =1
11700

For the first reach Δx = 6 km and Δt = 2 hr, the outflow is calculated using values for I1 =
10 m3/sec, I2 = 18 m3/sec, and Q1 = 10 m3/sec, and using Equ. (8.15) with j=1.

Q2 = C1 I 2 + C 2 I1 + C3Q1
Q2 = 0.487 ×18 + 0.744 ×10 − 0.231× 10 = 14 m 3 s ec

The hydrograph ordinates are calculated at Δx = 6 km reach with Δt = 2 hr steps and

given in the column (3) of Table (8.12). This is taken as the inflow hydrograph and the
outflow hydrograph at section Δx = 12 km is derived similarly and tabulated in column
(4). The same steps are carried on for the reach Δx = 18 km and the values written in
column (5).

178 A.Bulu
 Table 8.12

Time Δx (km)
(hr)
(1) 0 6 12 18
(2) (3) (4) (5)
0 10 10 10 10
2 18 14 12 11
4 50 34 24 18
6 107 81 59 42
8 147 132 111 88
10 146 150 145 133
12 105 125 139 145
14 59 78 99 118
16 33 42 56 74
18 17 23 30 39
20 10 12 16 21
22 10 10 10 12
24 10 10 10 10
26 10 10 10 10
28 10 10 10 10

The peak discharge of the flood hydrograph at section Δx = 18 km is 145 m3/s, a

reduction of 5 m3/s with respect to the peak of 150 m3/s at the entrance section. Lag time
is,
Lag time = 12 – 9 = 3 hr.

160
140
120 Inflow
100 Q (x=6km)
80
60 Q (x=12km)
40 Q (x=18km)
20
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Figure 8.14. Flow routing with Muskingum method

179 A.Bulu