ENVE 3003 – Runoff Models 1

Runoff is that portion of precipitation that eventually shows up as

stream flow

The system leading to runoff is complex (water budget):

Surface runoff:
 appears quickly as streamflow

Subsurface runoff:
 also called interflow and lags surface runoff

Water budget and surface processes often measured in terms of

depth; whereas streamflow is measured as a volume rate of flow, e.g.
m3/s
ENVE 3003 – Runoff Models 2

 Plot of streamflow versus time at any section of a stream

channel is called a hydrograph
 We can measure hydrographs directly or estimate them based
on watershed and stream characteristics.

In many cases it is desirable to measure the actual stream flow

Weirs: Generally applicable to small flows, a weir with a standard

notch (rectangular, V-notch) is constructed in the stream. The height of
the water in the notch is used along with a calibration curve to
calculate the stream flow
e.g.
ENVE 3003 – Runoff Models 3

Velocity based measurements: The flow in a stream can be

computed using velocity-area methods:
Q  VA
Q = flow [m3/s]
V = velocity in [m/s]
A = cross sectional area in [m2]

However, in most cases the velocity is not constant across the cross-
section and therefore,

Q   VdA  Vi Ai
A

Velocity can be measured by:

 Pitot tubes,
 Impellor
Google “river flow measurement instruments”
Depth can be measured by:
 Wading rod,
 Weighted line,
 Sonic techniques
ENVE 3003 – Runoff Models 4

 Often flows are measured for a range of depths or elevations to

establish a rating curve
 Subsequent to the development of a rating curve, only the
water surface elevation is measured to get Q
Tracer Techniques:
 A solution containing a tracer is continuously injected into the
stream at a known mass flow rate (qC1)
 After a period of time, the concentration is measured at a
downstream point (C2)
 The mass flow of tracer at this point will equal: (q+Q)C2
 Since the mass flows must equal: qC1 = (q+Q)C2
 Solve for Q: Q = q(C1 – C2) / C2
River Stages:
 Generated from actual measurements of flow
 Relates the flow (discharge) in a river at a location to the water
elevation (stage) at that location

Environment Canada’s Water Office” provides a lot of data for the

many river gauge stations in Canada https://wateroffice.ec.gc.ca/
On Carleton’s campus – Station name is “Rideau River at Ottawa” and
it is station number 02LA004

Environment Canada also provides a wealth of climate data from

various weather stations
http://climate.weather.gc.ca/historical_data/search_historic_data_e.html
Just east of Carleton – Ottawa CDS – weather station ID 30578
ENVE 3003 – Runoff Models 5

Hydrograph Components
The components of a hydrograph include:

 Rising limb
 Peak (crest)
 Declining limb
 Baseflow = flow from groundwater or upstream sources which
would be there without precipitation event

Shape is a function of:

 Rainfall intensity and pattern
 Areal distribution of rainfall
 Duration of storm event
 Size and shape of the watershed
 Slope of surface and river channel
 Soil type and distribution
 Roughness of topography
 Storage detention in watershed
 Etc.
ENVE 3003 – Runoff Models 6

Some important hydrograph characteristics include:

1. Time to Peak: elapsed time from beginning of rising limb to

peak discharge T

2. Lag Time (tp or tl) elapsed time from centre of mass of rainfall
and the peak flow of the hydrograph

3. Time of Concentration (tc)

Time of Concentration (tc)

 In physical terms, it is the time required for water to flow from

the most remote point in the watershed to the outlet.
 Discharge increases as water from progressively further away
arrives at point of interest.
 When the duration of the storm equals the time of
concentration, all parts of the watershed are contributing
simultaneously to the discharge at the outlet.
 If the rainfall event continues, the discharge peaks and remains
constant until the rainfall stops.

For smaller watersheds, the intensity (i) using IDF curves is often
obtained by considering the return period of a storm with a duration
equal to the time of concentration of the watershed. Therefore, in
order to reach the peak flow the rainfall duration must exceed the tc
(hours) for which a number of expressions have been developed.
ENVE 3003 – Runoff Models 7

Good Reference NRCS Part 630 Hydrology – Chapter 15 Time of Concentration

https://directives.sc.egov.usda.gov/OpenNonWebContent.aspx?content=27002.w
ba

𝐿
𝑇𝑡 =
𝑣

Travel time or time of concentration is the total length of the travel path
from the most hydraulically distant point to the outlet (𝐿) divided by the
average velocity (𝑣) over that path or length.

NRCS assumes the lag time (𝑡𝑙 ) is given by 𝑡𝑙 = 0.6 𝑇𝑐

Watershed lag method:

𝐿0.8 (𝑆+1)0.7 𝐿0.8 (𝑆+1)0.7

𝑡𝑙 = and with 𝑡𝑙 = 0.6 𝑇𝑐 gives 𝑇𝑐 =
1900 𝑌 0.5 1140 𝑌 0.5

Y is the average watershed slope (%)

S is the maximum potential retention which is a function of curve
1000
number (CN) 𝑆= − 10
𝐶𝑁

The average watershed slope can be calculated a number of ways

 Difference in elevation between hydraulically distance point and
outlet divided by length (L)
 Length-weighted slopes from a topo map along the flow path

https://en.wikipedia.org/wiki/Runoff_curve_number
ENVE 3003 – Runoff Models 8

Velocity Method:

𝑇𝑐 = 𝑇𝑡1 + 𝑇𝑡2 + 𝑇𝑡3 + ⋯ 𝑇𝑡𝑛

0.007(𝑛𝑙)0.8
Sheet flow 𝑇𝑙 = (𝑃2 )0.5 𝑆 0.4

Shallow concentrated flow – velocities for different soils and slopes

from charts and table (see NRCS reference provided earlier)

2 1
1.49 𝑟 3 𝑠 2
Open channel flow 𝑣=
𝑛

See example in NRCS in reference. Manning’s n values on-line.

ENVE 3003 – Runoff Models 9

Upland Method:
Use the following to estimate velocity and then calculate Tc

Above figure and table from NRCS reference

ENVE 3003 – Runoff Models 10

Kirpich’s Method:
tc = 0.019 L0.77 S – 0.385
where tc = time of concentration in minutes.
L = maximum length of flow in meters.
S = average slope along flow path. The watershed
gradient in m per m or the difference in elevation between
the outlet and the most remote point divided by length, L.
[Calibrated for Tennessee]

Kerby equation:
0.467
 Lr 
t c  1.44 
 S 
 O 

where tc = time of concentration in minutes.

L = length of flow in meters (<365m, small watershed).
r = retardance roughness coefficient (0.02 for smooth
pavement, 0.20 for light turf)
So = average slope along flow path. The watershed
gradient in m per m or the difference in elevation between
the outlet and the most remote point divided by length, L.

 Catchments with areas less than 4 ha, slopes less than 1%, and
retardance coefficients less than 0.8 were used in calibrating the
Kerby equation.
ENVE 3003 – Runoff Models 11

Hydrograph: Base Flow Separation

 The problem is to examine a hydrograph (storm) and determine

the relative contributions of direct runoff and baseflow
 Once this is accomplished, the two can be considered separately
 Several techniques are employed to separate direct runoff from
the base flow

An approach that is commonly used for large basins is as follows:

1. Extend the recession curve prior to the storm event until it

intersects a vertical drawn from the peak of the curve (C)
2. Draw a line from the point on the curve where it starts to climb
(A) to (C)
3. Calculate the value N = A0.2 (Linsley et al., 1975)
→ N = time (days)
2
→ A = basin area (mi )
(note: N = 0.8 A0.2 for area in km2 suggested by Linsley et al.
(1975) or to be exact N = 0.827 A0.2 for correct unit conversion)

4. Draw a line from (C) to intersect receding limb at t = (tP + N) (D)

ENVE 3003 – Runoff Models 12

Example: Consider the daily streamflow data for a river

(drainage area=6500 km2)

time d Q m3/s 12000

1 1600
2 1680
10000
3 5000
4 11300
8000
5 8600
6 6500
Flow
7 5000 6000
8 3800
9 2800 4000
10 2200
11 1850 2000
12 1600
13 1330 0
14 1300 0 2 4 6 8 10 12 14 16
15 1280
time (d)

Separate the baseflow from the direct runoff hydrograph and

determine the depth of direct runoff.
ENVE 3003 – Runoff Models 13

Analysis of runoff

 The objective of the hydrologist is to predict runoff, knowing

something about the precipitation which causes it and the
watershed characteristics which affect it.
 Depending on our engineering needs, we could require as little
as the peak flow or as much as the whole hydrograph.
 Depending on what we need, we have a variety of methods:

1. Rainfall-runoff correlations

2. Water balance

3. Rational method

4. Time-area method

5. Unit hydrographs

6. Sheet flow

Peak Runoff models estimate only the peak runoff, not the entire
runoff hydrograph: e.g. Rational Method
 Design for flooding e.g. culvert design

Continuous-runoff models estimate the entire runoff hydrograph from

the rainfall excess (remaining after abs+infiltration+storage): e.g. unit
hydrograph models
 Mostly applicable to designing storage resevoirs in
stormwater management systems.
ENVE 3003 – Runoff Models 14

1. Rainfall - Runoff Correlations

A relationship between rainfall and runoff may be developed for a

specific basin based on historic data

In its simplest form:

Often additional parameters may be used to indicate the soil moisture

conditions prior to the rainfall event to improve predictions
 Antecedent – precipitation index (API)
 Antecedent – flow index
 Weekly, monthly or seasonal adjustments
ENVE 3003 – Runoff Models 15

2. Water Balance

Can be used to predict runoff if the following are known or estimated:

 Precipitation
 Evapotranspiration
 Infiltration
 Interception
 Depression storage

3. Rational Method

One of the most common (and simplest) models

𝑄𝑝 = 𝐶 𝑖 𝐴

Qp = runoff rate, flow

C = runoff coefficient, which is a function of soil type, vegetation, etc.
i = design rainfall intensity for a duration equal to tc
Where tc = time of concentration, time for rainfall at the most remote
portion of the basin to reach the outlet
A = area of the watershed

Watch your units to ensure they are consistent.

You cannot define the Rational Method without stating that the rainfall
duration must equal or exceed the time of concentration

Rainfall intensity can be obtained from IDF (intensity-duration-

frequency) curves.
ENVE 3003 – Runoff Models 16

MTO provided two additional methods to determine tc

Bransby William Formula: for runoff coefficients > 0.4

0.057 𝐿
𝑡𝑐 =
𝑆 0.2 𝐴0.1

Airport Equation: for runoff coefficients < 0.4

3.26 (1.1 − 𝐶)𝐿0.5

𝑡𝑐 =
𝑆 0.33

tr = storm duration
tc = time of concentration

if tr < tc entire watershed area is NOT contributing to peak flow

if tr > tc then in theory we should reach a steady state of outflow (see
Figure 5.32 in text)

Ideally tr = tc when using the Rational Method

ENVE 3003 – Runoff Models 17

Advantages of Rational Method

 Easy to use
 Constants readily available for different soil types and land uses

Disadvantages
 Gives only estimate of peak flow
 Need to meet tc criterion
 Applies to smaller watersheds < 3 km2
 Can not handle very short storms due to tc requirement and
hence hard to estimate 10 yr storm event (e.g. to estimate 10
year storm event you can only use a storm duration equal to the
time of concentration where in reality the 10 year storm event
may correspond to a storm of duration less than tc)

Table 5.21 Chin, David A., Water Resources Engineering, 2nd Edition ©2007, reprinted by
permission of Pearson Education Inc., Upper Saddle River, NJ.

Numerous other sources available in texts, manuals and on-line.

ENVE 3003 – Runoff Models 18

An area-weighted composite value can be calculated for drainage

areas that have mixed uses:
n

C
j 1
j Aj
Cw  n

A j 1
j

In most situations a large drainage basin is subdivided and the

Rational Method is applied separately to each basin

Examples from a Google Search …

Report by McIntosh Perry (2016)
http://webcast.ottawa.ca/plan/All_Image%20Referencing_OP%20Ame
ndment%20Application_Image%20Reference_D01-01-16-
0004%20Stormwater%20Management%20Brief.PDF
Report by Smart Property Advisors (2013)
http://webcast.ottawa.ca/plan/All_Image%20Referencing_Site%20Plan
%20Application_Image%20Reference_Stormwater%20Management%
20Assessment.PDF
ENVE 3003 – Runoff Models 19

4.Time-Area Method

Based upon a combination of Rational Method and travel times in

collection system

This approach uses the concept that it takes a known amount of time
for effective precipitation, R (excess rainfall) to reach the watershed
outlet from all parts of the watershed

Ri  CI i
Watershed is subdivided into sub-basins by isochrones (equivalent
time of travel to discharge point)

All of the water from a sub-basin is assumed to reach the outlet of the
watershed at the same time (which must be estimated)

The peak runoff is estimated for each sub-basin for each time
increment of the design storm hyetograph

The runoff from the individual sub-basins are routed to the outlet and
combined

Qn  Ri A1  Ri 1 A2  ... R1 Aj kc

Where,
3
Qn = Hydrograph ordinate at time n (m /s)
Ri = Excess rainfall at time i (mm/hr)
Aj = Time-area histogram ordinate at time j (ha)
3
kc = 0.00278 conversion factor (ha-mm/hr to m /s)
ENVE 3003 – Runoff Models 20

Figures 5.41 and 5.42 Chin, David A., Water Resources Engineering, 2 nd Edition ©2007,
reprinted by permission of Pearson Education Inc., Upper Saddle River, NJ.
ENVE 3003 – Runoff Models 21

Example:
A watershed has been sub-divided into 3 sub-basins of 1, 2.5 and 6.3
ha, all with a runoff coefficient of 0.5. The time for the excess rainfall
from each sub-basin to reach the outlet gauge is 1, 2 and 3 hours,
respectively. If the watershed is subjected to following storm event,
compute the resulting hydrograph.

Time (hr) Intensity (mm/hr)

0-2 8
3-6 4
ENVE 3003 – Runoff Models 22
ENVE 3003 – Runoff Models 23

5. Unit Hydrographs

We don’t always have a hydrograph for all possible storm events and
hence we use the concept of a unit hydrograph to construct the
hydrograph for a specified storm event.

Defined as a hydrograph of direct runoff observed at the downstream

limit of a watershed due to one unit of rainfall excess (e.g. 1 cm)
falling for a unit time tr.
ENVE 3003 – Runoff Models 24

Recall rainfall excess:

 The unit of time for precipitation may be 1 day or less, but must
be less than the time of concentration (< tc)


The basic assumptions are:

 The system response is linear

 Overlapping storms do not interact, their results are
added or cumulative
 There is a unique unit hydrograph for each watershed and
event

Once this unit hydrograph is known, it can be used to determine the

effects (runoff) which would result from precipitation of different
amounts, durations and sequences.

Total volume of direct runoff in a unit hydrograph = 1 cm * A

(watershed area)

Since the basin features do not change from storm to storm,

hydrographs from storms of similar duration and pattern are
considered to have similar shape and time base.

 Theory of superposition: For 2 cm of net precipitation in

time tr, the resulting hydrograph will have the same shape as
the hydrograph from 1 cm of precipitation of same tr except
that all ordinates will be 2x as large.
ENVE 3003 – Runoff Models 25

Figure 5-34 Chin, David A., Water Resources Engineering, 2 nd Edition ©2007, reprinted by
permission of Pearson Education Inc., Upper Saddle River, NJ.

Determining a unit hydrograph for field data:

 Select a storm event(s) of correct duration and uniform intensity

 Determine the direct runoff (DRO) by subtracting the base flow
from the hydrograph
 Determine the area under the DRO hydrograph to estimate the
total flow volume
 Divide the total flow volume by the area of the watershed to
determine the excess rainfall as a depth
 Divide the flows of the DRO hydrograph by the excess rainfall
to generate a “unit hydrograph”
ENVE 3003 – Runoff Models 26

EXAMPLE: Unit hydrograph Determination

The hydrograph (after subtraction of base flow) resulting from a two-

hour storm of constant intensity over a 100 hectare basin area is
presented below. Determine the 2-hour unit hydrograph.
ENVE 3003 – Runoff Models 27

Changing the Unit hydrograph Duration

Rainfall of different durations will produce different shapes of the unit

hydrograph (UH)

Two common techniques used to adjust a UH from one duration to

another.

1. Lagging method

 Used when time duration has to be converted to a longer

duration that is a multiple of the original duration.

 If a UH of duration tr, is added to another (identical) UH

lagged by tr, the resulting hydrograph represents the
hydrograph of 2 units of storm occurring in 2 tr time.

 If the ordinates of the hydrograph are divided by 2, a UH

will result:
ENVE 3003 – Runoff Models 28

EXAMPLE: Lagging Method

The unit hydrograph (below) results from a 2 hour storm. Determine

the hourly ordinates of a 6 hour UH (unit hydrograph).

Time---- hours Q---- m3 / second

0 0
1 1.42
2 8.5
3 11.3
4 5.66
5 1.45
6 0

Solution:

Time hr 2 hr UH 1xUH 1xUH 1XUH total 6 hr UH

m3/s
0 0
1 1.42
2 8.5
3 11.3
4 5.66
5 1.45
6 0
7
8
9
10
ENVE 3003 – Runoff Models 29

Time
0 0 0 0
1 1.42 1.42 0.473333
2 8.5 0 8.5 2.833333
3 11.3 1.42 12.72 4.24
4 5.66 8.5 0 14.16 4.72
5 1.45 11.3 1.42 14.17 4.723333
6 0 5.66 8.5 14.16 4.72
7 1.45 11.3 12.75 4.25
8 0 5.66 5.66 1.886667
9 1.45 1.45 0.483333
10 0 0 0
ENVE 3003 – Runoff Models 30

2. S-curve method
 A UH is converted to any other duration, .
 Results when the unit rate of excess rainfall continues indefinitely
 To construct:
(a) Add together a series of tr (t) duration UHs each
lagged by tr (t). The curve will assume an S shape. At
equilibriumconstant outflow rate equal to net rainfall.
(b) The S-curve is lagged by the desired duration ().
The difference is taken between the two S-curves, and
multiplied by t/.

Figure 5.35 Chin, David A., Water Resources Engineering, 2 nd Edition ©2007, reprinted by
permission of Pearson Education Inc., Upper Saddle River, NJ.
ENVE 3003 – Runoff Models 31

Consider an Example: UH Determination (S-CURVE)

Convert the following 2-hour unit hydrograph to a 3-hour unit

hydrograph using the S-curve method.

Time Q
3
(hr) (m /s)
1 0
2 75
3 250
4 300
5 275
6 200
7 100
8 75
9 50
10 0

time 2-hr UH S curve lag S curvesubtract x(2/3)

(hr) (m^3/s) 3hrs
0 0 0 0 0
1 0 0 0 0
2 75 0 75 75 50
3 250 0 250 0 250 166.6667
4 300 75 0 375 0 375 250
5 275 250 0 525 75 450 300
6 200 300 75 0 575 250 325 216.6667
7 100 275 250 0 625 375 250 166.6667
8 75 200 300 75 0 650 525 125 83.33333
9 50 100 275 250 0 675 575 100 66.66667
10 0 75 200 300 75 0 650 625 25 16.66667
11 50 100 275 250 0 675 650 25 16.66667
12 0 75 200 300 75 0 650 675 -25 -16.6667
13 50 100 275 250 0 675 650 25 16.66667
14 0 75 200 300 75 650 675 -25 -16.6667
15 50 100 275 250 675 650 25 16.66667
16 0 75 200 300 575 675 -100 -66.6667
17 50 100 275 425 650 -225 -150
18 0 75 200 275 675 -400 -266.667
19 50 100 150 575 -425 -283.333
20 0 75 75 425 -350 -233.333
21 50 50 275 -225 -150
22 0 0 150 -150 -100
23 75 -75 -50
24 50 -50 -33.3333
25 0 0 0
ENVE 3003 – Runoff Models 32
ENVE 3003 – Runoff Models 33

Storm Flow Hydrographs from Unit Hydrographs

Unit hydrographs can be employed to estimate hydrographs for storms

of variable duration and intensity

UH ordinates (U) are multiplied by rainfall excess P and added and

lagged in a sequence. This procedure is called: Convolution.

Convolution Equation:
Qn  m1 PmU n  m1
nm

or:
Qn  PnU1  Pn 1U 2  Pn2U 3  ...P1U nm1

Where:
Qn= storm hydrograph ordinate
Pm=rainfall excess
Un-m+1=unit hydrograph ordinate
[Similar spreadsheet setup to time area method example]
ENVE 3003 – Runoff Models 34

Example: Consider the following rainfall excess hyetograph

Pn= [0.5, 1.0, 1.5, 0.0, 0.5] ;
and 1 hr UH  Un=[0,100,320,450,370,250,160,90,4,0] cfs
Q0=(0.5)(0)=0
Q1=(1.0)(0)+(0.5)(100)=50 cfs
Q2=(1.5)(0) +(1.0)(100)+(0.5)320=260 cfs
Q3=
Q4=
Time P1Un P2Un P3Un P4Un P5Un Qn
n hr
0 0 0
1 50 0 50
2 160 100 0 260
3 225 320 150 0 695
4 185 450 480 0 0 1115
5 125 370 675 0 50 1220
6 80 250 555 0 160 1045
7 45 160 375 0 225 805
8 20 90 240 0 185 535
9 0 40 135 0 125 300
10 0 60 0 80 140
11 0 0 45 45
12 0 20 20
13 0 0
ENVE 3003 – Runoff Models 35

“Deconvolution method” --- If this procedure is reversed, a UH can be

determined from a direct storm hydrograph produced by a multiperiod
rainfall excess. Consider a four period rainfall:

Q1 = P1 U1
Q2 = P2 U1 + P1 U2
Q3 = P3 U1 + P2 U2 + P1 U3
Q4 = P4 U1 + P3 U2 + P2 U3 + P1 U4
Q5 = P4 U2 + P3 U3 + P2 U4 + P1 U5
Q6 = P4 U3 + P3 U4 + P2 U5 + P1 U6
Q7 = P4 U4 + P3 U5 + P2 U6 + P1 U7
Q8 = P4 U5 + P3 U6 + P2 U7
Q9 = P4 U6 + P3 U7
Q10 = P4 U7
This can be expressed by:
[Q] = [P] [U]
Where [P] is a matrix that can be solved for [U].
ENVE 3003 – Runoff Models 36

EXAMPLE: UHs Deconvolution Method

Given the following resultant hydrograph and rainfall information,

compute the ½-hour unit hydrograph.

Time Excess Direct Runoff

3
(min) Rainfall (m /s)
(cm)
30 1.06 428
60 1.93 1923
90 1.81 5297
120 9131
150 10625
180 7834
210 3921
240 1846
270 1402
300 830
0 313
ENVE 3003 – Runoff Models 37

Q5 = P3U3+P2U4+P1U5 U5=(Q5-P3U3-P2U4)/P1
U5=(10625-1.81(2343)-1.93(2506))/1.06
U5=1461 m3/s

Q6 = P3U4+P2U5+P1U6 U6=(Q6-P3U4-P2U5)/P1
U6=(7834-1.81(2506)-1.93(1461))/1.06
U6=453 m3/s

Q7 = P3U5+P2U6+P1U7 U7=(Q7-P3U5-P2U6)/P1
U7=(3921-1.81(1461)-1.93(453))/1.06
U7=378 m3/s

Q8 = P3U6+P2U7+P1U8 U8=(Q8-P3U6-P2U7)/P1
U8=(1846-1.81(453)-1.93(378))/1.06
U8=279 m3/s

Q9 = P3U7+P2U8+P1U9 U9=(Q9-P3U7-P2U8)/P1
U9=(1402-1.81(378)-1.93(279))/1.06
U9=171 m3/s

Q10 = P3U8+P2U9

Q11 = P3U9
ENVE 3003 – Runoff Models 38

Synthetic Unit Hydrographs

Often information on stream flows does not exist for the portion of a
stream that is of interest

Synthetic unit hydrographs are identical in application (lagging, etc.) as

regular unit hydrographs, their difference is that they are derived
from basin characteristics only (slope, size, shape, etc.)

We will consider 2 methods: Snyder’s and SCS (NRCS)

1. Snyder’s Method

Based on studies of 20 watersheds located in Appalachian

Highlands, Snyder observed that the peak flow per unit (1 cm) of
rainfall excess is given by
Cp A
Q p  2.75
tl

where Cp is the peaking coefficient

A is the area of the basin in km2
tl is the lag time in hours

The duration of excess rainfall, tr , and the lag time, tl , are related by

tl
tr 
5.5

According to Snyder, the lag time is related to the catchment

characterisitics by:

t l  0.75C t LL c 
0 .3

where L is the length along the main stream from the outlet to
the boundary of the catchment in km
Lc is the length along the main stream from the outlet to
a point nearest to the catchment centroid in km
Ct is the basin coefficient that accounts for slope, land use
and storage characteristics of the basin
ENVE 3003 – Runoff Models 39

Cp and Ct are best found via calibration and are not physically based.
The (LLc)0.3 term is often called the shape factor of the basin.

Tabular values for Ct (0.4-4.0) and Cp (0.26-1.22) are given in Table

5.25 of Chin.

If the duration of the desired unit hydrograph is significantly different

tl
from that calculated using t r  , the following equation can be used
5.5
to adjust the lag time;
t lR  t l  0.25t R  t r 

where tR is the desired duration of excess rainfall

hence the flows need to be adjusted as well using:

tl Cp A
Q pR  Q p  2.75
t lR t lR

The time base of unit hydrograph (TB) in hours is estimated by

TB  72  3t lR

The general rule of thumb is TB / tlR should be on the order of 5 and if

TB / tlR >> 5, then it is recommended that TB / tlR be set equal to 5 or
determine the area under the hydrograph.

Formulas from the Corps of Engineers give additional assistance in

plotting the hydrograph. The time width, W50 in hours, at the discharge
point = 50% of the peak flow, and the time width, W75 in hours, at the
discharge point = 75% of the peak flow.

2.14
W50 
Q pR/ A
1.08

1.22
W75 
Q pR / A
1.08

A in km2 and QpR in m3/s. Typically the widths are proportioned such that
one-third is before the peak and two-thirds after the peak.
ENVE 3003 – Runoff Models 40

Example:
For a basin of 500 km2 having L=25 km and Lc =10 km, derive the 4 hour
unit hydrograph based on Snyder’s method. Assume that Ct=1.6 and
Cp=0.6.
ENVE 3003 – Runoff Models 41
ENVE 3003 – Runoff Models 42

2. NRCS Method (previously SCS)

Natural Resources Conservation Service (NRCS) was formerly called
the Soil Conservation Service (SCS)

NRSC developed a dimensionless UH that represents the average

shape of a large number of UHs from small agricultural watersheds
throughout US.

Figure 5.40 Chin, David A., Water Resources Engineering, 2 nd Edition ©2007, reprinted by
permission of Pearson Education Inc., Upper Saddle River, NJ.

Where
A
Q p  2.08
Tp

tr
Tp   tl
2

The value of tl is a function of the basin properties and can be

estimated as with Snyder’s Method or tl = 0.6 tc.

Once the values of Tp and Qp are known, the ordinates of the

synthetic unit hydrograph above can be scaled. Tabular values
follow.
ENVE 3003 – Runoff Models 43

t/Tp Q/Qp t/Tp Q/Qp

0.0 0.000 2.6 0.107
0.2 0.100 2.8 0.077
0.4 0.310 3.0 0.055
0.6 0.660 3.2 0.040
0.8 0.930 3.4 0.029
1.0 1.000 3.6 0.021
1.2 0.930 3.8 0.015
1.4 0.780 4.0 0.011
1.6 0.560 4.2 0.008
1.8 0.390 4.4 0.006
2.0 0.280 4.6 0.004
2.2 0.207 4.8 0.002
2.4 0.147 5.0 0.000

The NRCS UH can also be estimated by a triangular UH as follows:

Figure 5.40 Chin, David A., Water Resources Engineering, 2 nd Edition ©2007, reprinted by
permission of Pearson Education Inc., Upper Saddle River, NJ.

It is based on this triangular UH and the fact that the area under the UH
A
must equal and excess rainfall of 1 cm that Q p  2.08 T is derived.
p
ENVE 3003 – Runoff Models 44

Example:
For a basin of 500 km2 having L=25 km and Lc =10 km, derive the 4 hour
unit hydrograph based on NRCS method. Assume that Ct=1.6 and
Cp=0.6.
ENVE 3003 – Runoff Models 45