ECW 557:

ENGINEERING HYDROLOGY

HYDROGRAPH ANALYSIS

Zaizatul Zafflina Binti Mohd Zaki

March July 2013
WEEK 6 HYDROGRAPH ANALYSIS
LEARNING OUTCOMES
At the end of the week, students should
be able to:
1) apply unit hydrograph to determine direct runoff hydrograph
2) derive unit hydrograph from a storm hydrograph by principle of
superposition and lagging (CO2);
3) derive unit hydrographs of other durations from a unit
hydrograph (CO2);
4) derive synthetic unit hydrograph for a meteorologically similar
watershed (CO2).
 Total runoff Direct runoff
hydrograph, TRH - Baseflow = hydrograph, DRH

Q Q Q

t t t

Area below graph = Runoff volume

 Unit Hydrograph (UH) pg 205
Introduced by Sherman (1932) predicting the
flood hydrograph from known storm.
Definition: UH is the hydrograph of direct runoff
resulting from one unit depth (1cm) of rainfall
excess occurring uniformly over the basin and at a
uniform rate for a specific duration (D, hours).
D-hr UH

D
1cm RE * Volume of water in UH = RE
Q (m3/s)
* Average Intensity of RE

D-hr UH = 1/D (cm/h)

D = duration of RE
t
* Distribution of storm considered
tb
uniform all over catchment
 Assumptions in UH
Time Invariance: The direct-runoff response to a
given effective rainfall in a catchment is time-
invariant, implying that the DRH for a given ER in a
catchment is always the same irrespective of when it
occurs.
Linear Response: Principle of proportionality ~
(effective rainfall proportional to direct runoff in
diff D, same t) enable superposition to derive
DRHs

(UH Ordinate) x ER depth = DRH

HENCE, UH can be used to predict runoff from a storm event

 Effect of Storm Duration, D (hours)
2-hr UH 6-hr UH 12-hr UH
2 hr 1cm RE 6 hr 1cm RE 12 hr 1cm RE
Q Q Q

Unit RE
Unit RE Unit RE

A1 A2 A3
t t t

Area under the graph: A1 = A2 = A3;

Volume of effective rainfall: V1 = V2 = V3 = 1 unit of
Effective Rainfall
Differences: Time to peak, time base, peak value
 Principle of Proportionality
DRH due to
D-hr UH P cm RE (P>1) in D hr

1cm RE P cm RE
Q Q

t t
tb tb

Time to peak, tp & time base, tb remain the same

Ordinates & peak value changed
 Principle of Proportionality
DRH due to
D-hr UH P cm RE (P<1) In D hr

1cm RE P cm RE
Q Q

t t
tb tb

Time to peak, tp & time base, tb remain the same

Ordinates & peak value changed
 Method of Superposition & Lagging
DRH for two events of D-hr rain of 1 cm each
D-hr UH occurs consecutive

Q Q
D hr D hr D hr

1 cm 1 cm
1 cm
Total DRH:2 cm
1 cm RE
RE
1 cm RE
1 cm RE

t t
tb tb for 1st rainfall
tb for nd
Total DRH = Sum(DRHi) 2 rainfall

Time base for Total DRH, Lag time tb

tb = tb + (n-1)D D hr
n = number of rainfall

If dif. storms occurred successively, hence, D-hr UH can be used

to derive DRH for storms which has a duration of nD, n=integer
 Method of Superposition & Lagging
DRH for a complex storm event

D-hr UH
Q D D D

D Q cm R cm
1cm P cm
Q

= (P+Q+R) DRH

Q-DRH

P-DRH R-DRH
t
tb

tb for P t
Total DRH = DRHi
tb for Q
Time base for Total DRH, tb for R
tb = tb + (n-1)D tb
n = number of rainfall
 Example: 6.4, pg 207
t(h) Ordinates Ordinates
x RE depth: linear response
6 h-UH DRH
x3.5
0 0 0
x3.5
3 25 87.5
x3.5
6 50 175 3.5 cm DRH
x3.5
9 85 297.5
x3.5
12 125 437.5
6-hr UH

x3.5
60 8 28
x3.5
69 0 0 tb
Example: A catchment has a constant loss rate of 1.5 cm/h.
In a particular storm event, the catchment receive 6 cm
rainfall in 2 hours and 5 cm rainfall in the next 2 hours.
Determine the resulting runoff hydrograph if the average
baseflow is 20 m3/s and the 2 h UH is given.

t (h) 2h UH (m3/s) t(h) P (cm) L (cm)ER (cm)

0.0 0
1.0 25
0 0
2.0 40 2 6 3 3
3.0 25 4 5 3 2
4.0 13 For UH, tb = 8h
5.0 6 Time base for Total DRH,
6.0 3 tb = tb + (n-1)D
7.0 1
8.0 0
 x RE depth: linear response
t (h) 2h UH (m3/s) DRH (3cm) DRH (2cm) TOTAL DRH
0.0 0 0 Lag time 0
1.0 25 75 = 2 hr 75
2.0 40 120 0 120
3.0 25 75 50 125
4.0 13 39 80 119
5.0 6 18 50 66.5
6.0 3 9 26 35
7.0 1 3 12 15
8.0 0 0 6 6
9.0 0 2 2
10.0 0 0
 ?????
160
DRH (3cm)
140
DRH (2cm)
120
TOTAL DRH
100 Flood Hydrograph
80
60
40
20
0
0.0 2.0 4.0 6.0 8.0 10.0
 Example 6.5, pg208
t(h) 6hr-UH DRH2cm

0 0
3 25 x2
t = 3h 6 50 0
9 85 50
12 125 100 t = 3h t = 6h t = 9h
15 160 170
18 185 250
24 160 320
t = 6h 30 110 370
36 60 320
. . .
. . .
. . .
 Example 6.5, pg208
t(h) 6hr-UH DRH2cm t(h) UH DRH2
. . .
0 +3 0 . . .
3 +3 25 x2
. . .

6 +3 50 0 x2
48 +6 25 ??
9 +3 85 50 54 +6 16 50
12 125 100 60 +6 8 32
15 +3 160 170
69 +6 0 ??
18 +3 185 250 75 0
21 320
24 +6 160 370
30 +6 110 320 For UH, tb = 69h
36 60 220
. . .
42 36 120 Time base for Total DRH,
. . .
. . . tb = tb + (n-1)D
 Example 6.6, pg 210
For 6hr-UH (same ordinates as EX 6.4), tb = 69h

For Total DRH,

tb = tb + (n-1)D
tb = 69 + 2x6 = 81h
6
 Derivation of UH pg 212
Requires at least five (5) QUALIFIED storm events
with the following characteristics:
Storm must be isolated/ independent
Storm must be uniform over entire duration
Storm must be uniform over entire watershed
Storm duration must be 20%D (preferably less)
Storm duration must be 1/5 to 1/3 of basin lag
Storm magnitude must be significant, preferably
between 1 to 4cm
 Limitations of UH
Storm are rarely uniform over entire catchment,
hence, UH method is only applicable for
2 km2< A < 5000 km2
Large storage in catchment which will affect the
assumption of linear response is not considered.
Precipitation must be rainfall only, snow fall
cannot be considered.
Method invalid if the storm is nonuniform in time.
Accuracy: typically 20% tb , 10% Qp
 Derivation of UH
5nos UH from qualified storm events
-Qp are different
Q
- tpk are different
- tb are different

t
 Average UH
determine average Qp, tpk and tb
Qp Average UH
Q

adjust curve must be smooth

check area below UH - must
represent unit depth

tpk tb t
 Example
Time UH (m3/s) UH (avg)
h storm 1 storm 2 storm 3 storm 4 storm 5 (m3/s)
0 0.0 0.0 0.0 0.0 0.0 0.0
2 1.9 1.2 1.5 1.4 1.6 1.5
4 9.4 6.0 8.5 5.5 9.1 7.7
6 9.3 9.9 8.0 10.1 8.5 9.2
8 7.0 8.2 6.1 8.3 6.5 7.2
10 5.2 6.1 4.5 5.9 5.0 5.3
12 3.3 4.3 2.8 4.3 3.5 3.7
14 2.1 2.9 1.8 3.0 2.4 2.5
16 1.2 1.9 0.8 1.9 1.4 1.5
18 0.6 1.0 0.0 1.2 0.6 0.7
20 0.0 0.5 0.6 0.0 0.2
22 0.0 0.0 0.0

Catchment area = 28.4 km2 RE (cm) = 1.00

Average tpk = 4.8 h Sum UH average ordinates = 39.5 m3/s

Average Qp = 9.4 m3/s
Average tb = 20.4 h
12.0
UH1
10.0 UH2
UH3
8.0
UH4
6.0 UH5
Average
4.0

2.0

0.0
0 5 10 15 20 25
Time UH (avg) UH (avg)
Time, h corrected
h (m3/s)
0 0.0
0 0.0
2 1.5
2 1.5
4 7.7
4 7.7
5 9.4
6 9.2
6 9.2
8 7.2
8 7.2
10 5.3
10 5.3
12 3.7
12 3.7
14 2.5
14 2.5
16 1.5
16 1.5
18 0.7
18 0.7
20 0.2
20 0.0
22 0.0
22 0.0
1.0
1.00
 10.0
Qp = 9.4 9.0
8.0
7.0 Average
6.0 corrected
5.0
4.0
3.0
2.0
1.0
0.0
0 5 10 15 20 25

tpk = 4.8 (5.0) tb = 20.4 (20)

Try problems Solving 6.11 in class

 Derivation of UH
DRH for D-hr rain D-hr UH

D hr D hr

Q Q

P units RE
1 unit RE

V=PxA V=1xA

t t
tb tb

(Ordinates DRH) / ER = ordinates UH

Example 6.7, pg 213

Example 6.8, pg 215
 1D 2D 3D
UH of Complex Storm
R1 R2 R3
Total DRH due to two D-hr rainfall
Q
R1U4+ R2U3
R1U5+ R2U4

DRH due to D-hr effective rain of R1 unit

R1U3+ R2U2 R1U6+ R2U5

DRH due to D-hr effective rain of R2 unit

R1U7+ R2U6

R1U2+ R2U1 R1U8+ R2U7

R1U9+ R2U8

R1U1 R1U10+ R2U9

R2U10 t

D-hr
 Convolution Computation
Time Rainfall UH Runoff due to Runoff due to Runoff due to Total direct runoff
(hr) excess Un 1st 2nd 4th Qi
Rm X-hr rain X-hr rain X-hr rain
0 - 0 0 0 0 0

X R1 U1 R1U1 0 0 R1U1

2X R2 U2 R1U2 R2U1 0 R1U2 + R2U1

3X 0 U3 R1U3 R2U2 0 R1U3 + R2U2

4X R4 U4 R1U4 R2U3 R4U1 R1U4 + R2U3 + R4U1

5X U5 R1U5 R2U4 R4U2 R1U5 + R2U4 + R4U2

6X U6 R1U6 R2U5 R4U3 R1U6 + R2U5 + R4U3

7X U7 R1U7 R2U6 R4U4 R1U7 + R2U6 + R4U4

8X U8 R1U8 R2U7 R4U5 R1U8 + R2U7 + R4U5

9X U9 R1U9 R2U8 R4U6 R1U9 + R2U8 + R4U6

10X U10 R1U10 R2U9 R4U7 R1U10 + R2U9 + R4U7

11X 0 0 R2U10 R4U8 R2U10 + R4U8

12X 0 R4U9 R4U9

13X R4U10 R4U10

14X 0 0
 Deconvolution Computation
Q1 = R1U1
Q2 = R2U1 + R1U2
Q3 = R3U1 + R2U2 + R1U3

QM = RMU1 + RM-1U2 + + R1UM

QM+1 = 0 + RMU2 + + R2UM + R1UM+1

QN+M-2 = 0 + 0 + + RMUN-1 + RM-1UN

QN+M-1 = 0 + 0 ++ 0 + RMUN

Q=RxU

Unit hydrograph U can be solved using matrix method.

 Analysis
Input Transfer Function Output

Rainfall Data
?
Unit Hydrograph Streamflow Data

Synthesis
Input Transfer Function Output

Rainfall Data Unit Hydrograph

?
Streamflow
 UH of different durations

Previously, UH from simple isolated storms can be

grouped under average D-hour. Practically, UH in
different D. Can use following methods:
Method of Superposition
S-hydrograph Method
 Method of Superposition
DRH for 2D-hr rain
D-hr UH
Q
2D hr
D hr

Q 1cm RE 2cm RE

t t
tb
tb
tb
Intensity = 1/D cm/h
Lag time

(UH Ordinate) x ER depth = DRH

(DRH Ordinate) / RE depth = UH Ordinate
 D = duration of UH tb' tb (n 1) D
n = number of D-hr rainfall
Intensity = 1/nD cm/h

Limitation: For an UH of duration D, other UH

derived has duration which is multiple of the
original UH only, i.e. nD, where n is an integer.

Example 6.9, pg 216

Given 4-h UH, find the 12-h UH???
 S-hydrograph Method

Overcome the limitation of Method of Superposition.

Can derive UH with ANY DURATION.
 S-hydrograph
Q
S-hydrograph, a.k.a S-curve,
hydrograph produced by
continuous ER at constant
D hr
Maximum
rate & is the summation of
infinite series of D-hr UH,
S-hydrograph spaced D-hr apart.
D hr UH
Maximum equilibrium
discharge at a time
equal to the time base of
t the first UH
D hr D hrtb1
tb2 *Approach constant at tb D
tb3
 S-hydrograph
Intensity = 1/D cm/h Q T hr

Duration = T h (Duration for required UH) Storm B

Depth = Intensity x Duration Storm A
= T/D cm D hr
S-hydrograph A

S
S-hydrograph B

S S
T/D cm
S x D/T

1 cm
t
lagged T hr

tb = tb + (T-D)
(DRH Ordinate) / RE depth = UH Ordinate
 Example 6.10 pg:219
Derive S-curve for the 4-hr UH?
Hint: use Eq 6.8; S(t) = U(t) + S(t-D)
t = time, D = rainfall duration
 D -hr Hint: use Eq 6.8; S(t) = U(t) + S(t-D)
Derive 12-h UH from 4-h UH???? t = time, D = rainfall duration

T-hr SA SB (SA SB)/Depth

 Example 6.11; pg221
Approach constant at tb D = 44 4 = 40h For S-curve
Q (m3/s)

652 679 694 699 699

0 4 8 12 16 20 24 28 32 36 40 44 t (h)
 Example 6.11, pg221
Consider two D-h S-curve, displaced by T-h
tb = tb + (T-D) = 44 + (12-4) = 52h
Q (m3/s)

S1 S2

0 4 8 12 16 20 24 28 32 36 40 44 48 52 t (h)
Example 6.12, pg221

Hint: use Eq 6.8; S(t) = U(t) + S(t-D)

 Example 6.12, 221

tb = tb + (T-D) = 44 + (2-4) = 42h

Q (m3/s)

20 20 10 10
0 +4

0 12 42 -4 t (h)
 Example 6.12, pg221

tb = tb + (T-D) = 44 + (2-4) = 42h

Q (m3/s)

Fairing

0 12 42 t (h)
 UH of different durations
Problem solving: 6.18, 6.19
 Synthetic UH
Gauged Data UH

Ungauged Data UH but HOW??

A synthetic unit hydrograph is derived from

empirical equations of regional validity when
observed rainfall-runoff data are not available
for unit hydrograph derivation. Its uses are:
to anticipate post-development conditions,
to extend the usage of pointspecific unit
hydrograph to other locations of similar
characteristics.
 Synthetic UH
Snyders method
SCS method
Both to utilize existing UH from a neighbouring
meteorologically similar watershed.
Watershed A Watershed B
UH available UH not available

Rainfall duration Rainfall duration

Qp

t ? t
 tc Storm Hydrograph
D
D = rainfall duration
*L Rainfall
tpk = time to peak
tb = time base
Q
tc = time of concentration
Point of
inflection/max L = lag time, basin lag
storage
* tr * centre of mass
tr = recession time

tpk

Beginning of rainfalltb
Components of Hydrograph pg 198
 Snyders Method popular method
Snyder adopted a standard duration tr hours of effective rainfall
given by
tr = tp / 5.5

tr = rainfall duration for standard (h)

tr
tp = Ct ( L.Lca )0.3
tp = basin lag (h)
tp = basin
lag/lag time
L = basin length (km)
Qp Lca = distance from watershed centroid
to outlet (measurement point) along
the main channel (km)
Ct = regional constant (0.3< Ct <6.0)
(Representing watershed slope and storage)
Basin lag/lag time = mean time for water travel from
t all parts of watershed to outlet.
For flood forecasting, the actual storm duration is normally different from tr thus
For any UH, tR = rainfall duration (h), tp = modified basin lag (h)

For standard UH For non-standard UH

tp = tp tp tp tr tR
t r = tR
tR tp = tp + 0.25(tR tr)
Qps = 2.78 CpA/ tp
tp = 21tp /22 + 0.25tR
tp @ tp
Qp = 2.78 CpA/ tp
Qp

A = watershed area (km2)

Qp = peak flow rate (m3/s)
Cp = regional constant (0.31< Cp< 0.93)
Indicate retention and watershed
t storage capacity
 Snyders Method
Time base of a unit hydrograph by Synders
Tb = 72 + 3tp (unit in hours)
Sketching UH, width of UH based on W50 & W75

Required UH W75
Qp

0.75Qp W50 = 5.87/ q1.08 q = Qp/A

W75
0.5Qp
W50 W75 = W50 / 1.75

Tp

Tb t
 Snyders Method

Qp

W75 /3
0.75Qp
2/3 W75

0.5Qp
W50 /3 2/3 W50

tpk
t
 Snyders Method
Step 1: check whether source UH is standard
Step 2: determine regional constants (Ct & Cp) of source UH
Step 3: check whether required UH is standard
Step 4: determine parameters of required UH
Step 5: determine time base of required UH
Step 6: finalizing required UH using S-curve method

Source UH Required UH

Qp

t
? t
 Snyders Method
Step 1: check whether source UH is standard

Assume tr = tR
Source UH

tR tr = tp / 5.5 tp

If tp = tp standard
tp

Qp
If tp tp non-standard

Tp = tR/2 + tp

t
 Snyders Method
Step 2: determine regional constants of source UH
If not standard
tp = tp + 0.25(tR tr)
Source UH
tp
tR tp = 21tp /22 + 0.25tR

tp = Ct ( L.Lca )0.3 Ct
tp

Qp
Qp = 2.78 CpA/ tp Cp

t
 Snyders Method
Step 3: check whether required UH is standard

Required UH tp = Ct ( L.Lca )0.3

tR
tr = tp / 5.5

tp Assume tp = tp
Qp If tr = tR standard

If tr tR non-standard

t
 Snyders Method
Step 4: determine parameters of required UH

tp = tp + 0.25(tR tr)
Required UH
tp
tR tp = 21tp /22 + 0.25tR

Qp = 2.78 CpA/ tp Qp
tp

Qp
tpk = tR/2 + tp

tpk t
 Snyders Method
Step 4: determine parameters of required UH

Required UH W50 = 5.87/ q1.08 q = Qp/A

W75
Qp W75 = W50 / 1.75

0.75Qp

0.5Qp
W50

t
 Snyders Method
Step 5: determine time base of required UH

For large catchment:

Required UH
tb = 72 +3tp (h)

Qp
For small catchment:

tb = 5(tp +0.5tR) (h)

To the nearest larger integer divisible by tR

tb t
tb is always least accurate
 Example 6.14, pg228
Q (m3/s)

62
Select suitable tb

46.5 W75 = 25h

31
W50 = 44h

t (h)
4 8 11.4 33 48 58
 Example 6.14, pg228
Q (m3/s)

62 Or use a curve

46.5
W75 = 25 h

31
W50 = 44 h

t (h)
11.4
 Snyders Method
Step 6: Finalising UH using S-curve method

Use the synthetic UH derived to plot an S-curve

Use the S-curve to derive a UH of duration tR
 SCS Method
SCS = Soil Conservation Service, US

Use dimensionless UH derived from large number

of UH on variety of watershed.
t/tpk Q/Qp
0.0 0
0.1
0.2
0.015
0.075
SCS Method
0.3 0.16
0.4 0.28
1
0.5 0.43
0.6 0.6 0.9
0.7 0.77 0.8
0.8 0.89
0.7
0.9 0.97
1.0 1 0.6
1.1 0.98 0.5
1.2 0.92
0.4
1.3 0.84
1.4 0.75 0.3
1.5 0.66 0.2
1.6 0.56
0.1
1.8 0.42
2.0 0.32 0
2.2 0.24 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
2.4 0.18
2.6 0.13
2.8 0.098
3.0 0.075
3.5 0.036
4.0 0.018
4.5 0.009
5.0 0.004
6.0 0
 SCS Method - SCS Triangular UH
Using Snyders method, calc. tpk and Qp
Multiply t/tpk by tpk =t
Multiply Q/Qp by Qp =Q

Plot Q vs t
10
9
8

Example: 7
6
tpk = 20h 5
Qp = 10m3/s 4
3
2
1
0
0 20 40 60 80 100 120
 SCS Method - SCS Triangular UH
UH in triangular shape (FIGURE 6.21(b)
where Tb = 2.67xTp
Qp = 2.08A/Tp
Tp = tr/2 + 0.6tc
Example 6.15 page 231