Engineering Hydrology

(WREN3201)
Course Instructor: Bayisa Tulu (MSc.)

November, 2022
Rainfall-Runoff Relationships
Hydrological Models
o Deterministic hydrological models
o Stochastic hydrological models
Determination of Peak Discharge
o Rational methods
o SCS CN methods
o Time area methods
Stream Flow Hydrograph
o Unit hydrograph
o Synthetic unit hydrograph
o Instantaneous unit hydrograph

Page | 2
Rainfall-Runoff Relationships
Hydrological Models
 What is a Model ?
 A model is a simplified representation of a complex system or reality and its processes.
 Model always describes the basic and most important components of a complex system.
 Model involves similarity but not identity.
 Model simulates some but not all the characteristics of the prototype system.

Real world Model

Page | 3
 Rainfall-Runoff Relationships
Model can be:
1. Physical model (e.g. physical architectural house scale model)
2. Conceptual model (e.g. computer model, mathematical model,…)

o Modeling is the process that uses a mathematics as a tool to explain and make
prediction of natural phenomena.
Page | 4
o It is a process of producing a model.
Rainfall-Runoff Relationships
Hydrological Models
o Hydrological models are a simplified, conceptual representations of the hydrologic cycle.
o They are abstractions of reality that used to simulate natural systems.
o Hydrological models are power full techniques that integrate hydrological systems and comprise a
set of mathematical expressions of the hydrologic cycle to simulate real-world hydrologic processes.
o They are based on a set of mathematical equations to transform physical laws which govern complex
natural Phenomena (kassa, 2009)

Page | 5
 Rainfall-Runoff Relationships
Hydrological models are used for:
☻ Assessment
• land use/land cover/ Climate change impact
• Water Supply and Sewer Systems
• Water Quality
☻ Prediction
• Flood Forecasting
☻ Design
• How much flow will occur in a 50,100 or 500 year return period?

Page | 6
 Rainfall-Runoff Relationships
Classification of Hydrological models
☻ Classification of hydrological models are not exact & different hydrologists
may give different definitions due to the nature of models.

☻ The classification of hydrological models may be vary depending on

justification ( Singh and Frevert, 2006)

Page | 7
 Rainfall-Runoff Relationships
Classification of Hydrological models

Page | 8
 Rainfall-Runoff Relationships
Classification of Hydrological models

Page | 9
 Rainfall-Runoff Relationships
Physical Hydrological Model
o Physical hydrological model is the representation of the real system by another system,
which is similar properties but much easier to work with. It is also called material model.
o Scale model / prototype: represents the system on a reduced scale.
Example: Laboratory watersheds, hydraulic model of dam spillway, rainfall simulator
o Analog model: does not physically resemble the prototype and measures different physical substances
than the prototype.
Example: Flow of electric current which represents the flow of water

Page | 10
 Rainfall-Runoff Relationships
Mathematical (Abstract) Model
o Expresses the system behaviour by a set of equations together with logical statements
expressing relationships between variables and parameters.

Yt = f (Xt Xt−1 Xt−2 ,…, Yt−1 Yt−2 ,…, a1, a2 ,…) + ∑t

Where Xt is the input variable, f (-) is a function of a specified form, ai, i=1,2,… are
parameters, ∑t is residual expressing lack of fit b/n observed output, Yt and fitted output,
f (-)

Page | 11
 Rainfall-Runoff Relationships
Mathematical (Abstract) Model
☻ Deterministic model
☻ Stochastic model
☻ Mixed model

1. Deterministic model
o It is one of the hydrological model in which the outcomes are precisely determined through known
relationships among the states and events.
o No room for random variations
o Two equal sets of inputs always yield the same output.

Page | 12
 Rainfall-Runoff Relationships
1. Deterministic model
o It is one of the hydrological model in which the outcomes are precisely determined through known
relationships among the states and events.
o No room for random variations
o Two equal sets of inputs always yield the same output.
o The vast majority of models used in rainfall-runoff modelling are deterministic

Deterministic hydrological model can be classified as Lumped model, Semi-distributed model and Fully
distributed model based on spatial discretization.

Page | 13
 Rainfall-Runoff Relationships
Deterministic model
A) Lumped model
o The lumped modeling approach considers a watershed as a single unit for computations where the
watershed parameters and variables are averaged over this unit.
o Consider the complete basin as a homogenous, no spatial variations.

B) Semi-distributed model
o The semi-distributed models partition the whole basin into sub-basins or HRUs that partially allow
parameters to vary in space by dividing the basin into a number of smaller sub-basins.

o Attempt to calculate flow contributions from separate areas or sub-basins that are treated as
homogeneous within themselves
Page | 14
o
 Rainfall-Runoff Relationships
Deterministic model
C) Fully-distributed model
o A distributed model can make predictions that are distributed in space by dividing the entire
catchment into small units, usually square cells or triangulated irregular networks so that the
parameters, inputs, and outputs can vary spatially.

o The whole basin is divided into elementary unit areas like a grid net, TIN.

Page | 15
 Rainfall-Runoff Relationships
Deterministic model

Figure 1: A: Lumped model, B: Semi-Distributed model by sub-catchment, C: Distributed model by

grid cell Source: Jan Sitterson, 2017, ESRI, 2015)
Page | 16
Rainfall-Runoff Relationships

Figure 2: Graphic representation of geometrically –distributed and lumped models. (from Jones, 1997). I is
Page | 17
input and O is output
Rainfall-Runoff Relationships
Spatial Scaling of Models

Lumped Semi-Distributed Fully-Distributed

Parameters assigned Parameters assigned to each Parameters assigned to
to each subbasin grid cell, but cells with same each grid cell
parameters are grouped

A3

A1
A2

18
 Rainfall-Runoff Relationships
Deterministic hydrological model can be classified as Empirical model, Conceptual model and
Physically model based on the description of the hydrological processes.

A) Empirical models

o They are known as black box models or metric models or data driven models.
o Empirical models are observation-oriented models which take only the information from the
existing data.
o It involves mathematical equations derived from concurrent input and output.
o Not consider physical processes of the catchment.
o Used the functional relationships b/n input and outputs.

Page | 19
 Rainfall-Runoff Relationships
A) Empirical models
o Models such as Unit Hydrograph, Artificial Neural Network, Fuzzy regression are some
examples of this method.

B) Conceptual Models
o Known as grey box models or parametric models.
o Are intermediate b/n empirical and theoretical models.
o Conceptual models consider the physical laws but in highly simplified form.
o Example: Stanford Watershed Model IV (SWM), HBV model

Page | 20
 Rainfall-Runoff Relationships
C) Theoretical Models
o They also known as white box model or mechanistic models.
o They are physically-based models that include the principles of physical processes.
o Are mathematically idealized representations of a real phenomenon.
o They use state variables that are measurable and are functions of both time and space.
o Theoretical models have logical structures similar to the real world system.
o Data such as soil moisture content, initial water depth, topography, topology dimensions of
the river network, soil data, LULC data etc. are required.
o Example: SWAT, MIKE-SHE, HEC-HMS, Watershed runoff models, Infiltration models,
GW models, Evaporation models

Page | 21
 Rainfall-Runoff Relationships
2. Stochastic model
o Stochastic models allow for some randomness or uncertainty in the possible outcomes due to uncertainty
in input variables, boundary conditions or model parameters.
o Components or results can not be predicted with certainty as the outcome depends on the chance.
o Traditionally, a stochastic model is derived from a time series analysis of the historical record.

o The stochastic model can then be used for the generation of long hypothetical sequences of events with the
same statistical properties as the historical record.
o These generated sequences of data can then be used in the analysis of design variables and their
uncertainties,
o For example, when estimating reservoir storage requirements.
Page | 22
 Rainfall-Runoff Relationships
2. Stochastic model
o Some stochastic models are Autoregressive models, Moving Average models (MA), Auto-Regressive
Moving Average models (ARMA), Daily Data Generation models.

Page | 23
 Rainfall-Runoff Relationships
Estimation of Runoff
1. Rational method
o Is one of the most commonly used method for the calculation of peak flow for small areas.
o It is first proposed by Kuichling in 1889.
o It assumes that the Rainfall is of:
o Constant intensity
o Distributed uniformly over the watershed
o RF occurs for a period equal to or greater than the time of concentration.
o Because of these inherent assumptions, the Rational Formula should only be
applied to smaller drainage areas.
Page | 24
 Rainfall-Runoff Relationships
1. Rational method
𝑪𝑰𝑨
Qp =
𝟑𝟔𝟎

Where Qp = Peak Discharge (m3/s);

𝑰= Rainfall Intensity (mm/hr)
A = Drainage Area (ha)
C = Runoff Coefficient

Page | 25
 Rainfall-Runoff Relationships
1. Rational method
o Runoff coefficient (C) is affected by the ground cover, slope and other hydrologic abstractions.
o The values of C varies from 0.05 for flat sandy area to 0.95 for impervious urban surfaces.
o C also varies for different storms on the same catchment, thus using an average value for C, gives the
rough estimate of Qp in small urban areas.
o If the basin contains different land cover or abstractions, then coefficient of runoff (C) can be calculated
as:

∑𝑪𝒙𝑨𝒙
𝑪=
∑𝑨𝒙
o
Page | 26
Rainfall-Runoff Relationships
Rational Method

Page | 27
Rainfall-Runoff Relationships
Rational Method Surface Coefficient
Concrete or Asphalt 0.8-1
Gravel - Compact 0.7
Clay - Bare 0.75
Clay - Light Vegetation 0.6
Clay - Dense Vegetation 0.5
Gravel - Bare 0.65
Gravel - Light Vegetation 0.5
Gravel - Dense Vegetation 0.4
Loam - Bare 0.6
Loam - Light Vegetation 0.45
Loam - Dense Vegetation 0.35
Sand - Bare 0.5
Sand - Light Vegetation 0.4
Sand - Dense Vegetation 0.3
Grass Areas Page
0.35| 28
 Rainfall-Runoff Relationships
Rainfall intensity (I)
o Rainfall intensity, duration curve and frequency curves are
necessary to use the rational method.
o Regional IDF curves need to be developed for the catchment in
question.

Example of IDF Curve Page | 29

 Rainfall-Runoff Relationships
Time of concentration (Tc)
o Time of concentration is the time required for water to flow from the most remote point (longest distance)
in a watershed to the watershed outlet once the soil has become saturated and minor depressions filled.
o There are a number of methods that can be used to estimate time of concentration. The most commonly
used method is Kirpich formula.

𝑻𝒄 = 𝟎. 𝟎𝟏𝟗𝟓𝑳𝟎.𝟕𝟕 𝑺−𝟎.𝟑𝟖𝟓
Where Tc= time of concentration in minutes.
L = maximum length of flow in meters.
S = the watershed gradient in m per m

Page | 30
Rainfall-Runoff Relationships
Time of Concentration

Page | 31
Determination of Peak Discharge
Time of Concentration

Page | 32
Rainfall-Runoff Relationships
Time of Concentration

Page | 33
 Rainfall-Runoff Relationships
Example 2.1:
Using the following existing and proposed land use; calculate the runoff of coefficient (C) for both existing and
proposed conditions.
Existing land use Area (ha) Runoff Coefficient, C
(unimproved)
Unimproved Grass 8.95 0.25
Grass 8.60 0.22

Proposed (improved) land use condition

Paved 2.20 0.90
Lawn 0.66 0.15
Unimproved Grass 7.52 0.25
Grass 7.17 0.22
Page | 34
 Rainfall-Runoff Relationships
Example 2.1:
Solution:
Step1: Determine weighted “C” for existing (unimproved) conditions
∑𝑪𝒙𝑨𝒙
Weighted, 𝑪 =
∑𝑨𝒙

𝟖. 𝟗𝟓 ∗ 𝟎. 𝟐𝟓 + 𝟖. 𝟔𝟎 ∗ 𝟎. 𝟐𝟐
𝑪=
𝟖. 𝟗𝟓 + 𝟖. 𝟔𝟎
𝑪 = 𝟎. 𝟐𝟑𝟓
Step2: Determine weighted “C” for proposed (improved) conditions
∑𝑪𝒙𝑨𝒙
Weighted, 𝑪 =
∑𝑨𝒙

𝟐. 𝟐𝟎 ∗ 𝟎. 𝟗𝟎 + 𝟎. 𝟔𝟔 ∗ 𝟎. 𝟏𝟓 + 𝟕. 𝟓𝟐 ∗ 𝟎. 𝟐𝟓 + 𝟕. 𝟏𝟕 ∗ 𝟎. 𝟐𝟐
𝑪=
𝟐. 𝟐𝟎 + 𝟎. 𝟔𝟔 + 𝟕. 𝟓𝟐 + 𝟕. 𝟏𝟕
𝑪 = 𝟎. 𝟑𝟏𝟓
Page | 35
 Rainfall-Runoff Relationships
Example 2.2:
Determine the peak runoff rate for 30 years return period to design a gully control structure.
The essential data are as follow
A. Catchment slope is 0.60%
B. Intensity-return period relationship is used as:
𝟓. 𝟖𝟐𝑻𝟎.𝟏𝟔𝟎
𝑰=
(𝒕𝒄 + 𝟎. 𝟒)𝟎.𝟕𝟓
Where, I= rainfall intensity for the duration equal to time of concentration (cm/min)
T= return period (years)
tc= time (minute)
C. Length of longest water course is 1500 m
D. Catchment area is 15 km2
E. Runoff coefficient for the catchment is 0.55

Page | 36
 Rainfall-Runoff Relationships
Example 2.2: Required
Solution:
𝑸𝒑𝒆𝒂𝒌 =?
Given
Solution:
T=30 years Rainfall intensity, I
Time of concentration, tc
S=0.60%=0.006 𝟓.𝟖𝟐𝑻𝟎.𝟏𝟔
𝒕𝒄 = 𝟎. 𝟎𝟏𝟗𝟓 ∗ 𝑳𝟎.𝟕𝟕 ∗ 𝑺−𝟎.𝟑𝟖𝟓 𝐈=
𝐭𝐜+𝟎.𝟒 𝟎.𝟕𝟓
L=1500m
𝒕𝒄 = 𝟎. 𝟎𝟏𝟗𝟓 ∗ 𝟏𝟓𝟎𝟎𝟎.𝟕𝟕 ∗ 𝟎. 𝟎𝟎𝟔−𝟎.𝟑𝟖𝟓 𝟓.𝟖𝟐∗𝟑𝟎𝟎.𝟏𝟔
A=15km2= 1500ha 𝐈=
𝟑𝟖.𝟗𝟗+𝟎.𝟒 𝟎.𝟕𝟓
𝒕𝒄 = 𝟑𝟖. 𝟗𝟗𝒎𝒊𝒏
C=0.55 𝐈 = 𝟎. 𝟔𝟒𝐜𝐦/𝐦𝐢𝐧
𝐈 = 𝟑𝟖𝟒𝐦𝐦/𝐡𝐫
Hence; Peak Runoff, 𝑸𝒑𝒆𝒂𝒌
𝑪𝑰𝑨
𝐐𝐩𝐞𝐚𝐤 =
𝟑𝟔𝟎
𝟎.𝟓𝟓∗𝟑𝟖𝟒∗𝟏𝟓𝟎𝟎
𝐐𝐩𝐞𝐚𝐤 =
𝟑𝟔𝟎
𝐐𝐩𝐞𝐚𝐤 = 𝟖𝟖𝟎𝒎𝟑 /𝐬 Page | 37
 Rainfall-Runoff Relationships
Exercise:
1. A catchment of area 90ha has runoff coefficient of 0.40. A storm of duration larger than the time of
concentration of the catchment and of intensity 4.5 cm/hr. calculate the peak discharge of the catchment.

Page | 38
 Rainfall-Runoff Relationships
2. SCS CN method
o The SCS-CN method (Soil Conservation Service, 1958) is one of the most frequently used methods for all
kinds of watersheds.
o The SCS CN calculates peak flow as a function of drainage basin area, potential watershed storage and the
time of concentration.
o The total rainfall is separated into direct runoff, retention and initial abstraction to yield the following
equations.
𝑷−𝑰𝒂−𝑸 𝑸
= 𝑷−𝑰𝒂
𝑺

(𝑷 − 𝑰𝒂)𝟐
𝑸=
𝑷 − 𝑰𝒂 + 𝑺
Page | 39
 Rainfall-Runoff Relationships
2. SCS CN method
 The initial abstraction can be expressed as a function of S as:
 where a is a fraction; frequently taken as 0.2, implying that 20% of the potential maximum retention
equals the initial abstraction before runoff begins and taking a = 0.2 results in:

(𝑷 − 𝟎. 𝟐𝑺)𝟐
𝑸=
𝑷 + 𝟎. 𝟖𝑺
Where: Q=depth of direct runoff ,P=depth of 24 hour precipitation (mm), S=Retention, mm

Page | 40
 Rainfall-Runoff Relationships
2. SCS CN method
 The potential retention (S) depends upon the soil-vegetation-land use complex of the catchment and on the
antecedent soil moisture condition in the catchment prior to the commencement of the rainfall event.
 For a practical application of the SCS of USA has expressed, S in terms of dimension less parameter CN
(curve number) as:

𝟏𝟎𝟎𝟎
𝑺 = 𝟐𝟓. 𝟒( − 𝟏𝟎)
𝑪𝑵
 Where; S = retention (mm), CN= curve number, 25.4 is a constant that used to express S in mm.
 CN ranges 0 to 100 theoretically but practically ranges from 40 to 98.
 The equation shows that S is zero when CN equals 100 (impervious catchment), and is infinity when CN
equals 0.
Page | 41
 Rainfall-Runoff Relationships
2. SCS CN method
 The CN depend on Soil type, Land use/cover, antecedent moisture.

Page | 42
 Rainfall-Runoff Relationships
2. SCS CN method
Group Minimum Infiltration Hydrologic Soil Group
Rate (in/hr)

A 0.3 – 0.45 High infiltration rates. Deep, well

drained sands and gravels
B 0.15 – 0.30 Moderate infiltration rates.
Moderately deep, moderately well
drained soils with moderately coarse
textures (silt, silt loam)

C 0.05 – 0.15 Slow infiltration rates. Soils with

layers, or soils with moderately fine
textures (clay loams)
D 0.00 – 0.05 Very slow infiltration rates. Clayey
soils, high water table, or shallow
impervious layer
Page | 43
 Computation of Runoff Volume
2. SCS CN method

Page | 44
 Computation of Runoff Volume
2. SCS CN method

Page | 45
 Computation of Runoff Volume
2. SCS CN method

Page | 46
 Rainfall-Runoff Relationships
2. SCS CN method
o The peak flow is then estimated with the following equation:

qp=qu∗Ak∗Q

o Where qp= peak flow, m3/s, qu=unit peak flow, m3/s/km2/mm, Ak=basin area, km2,
Q=runoff depth, mm
o The unit peak flow qu can be computed by using the following equation:

qu=0.000431∗10Co+C1(logtc)+C2(log(tc))2
o Where, Co, C1, C2 are coefficient listed in Table1 below. These are functions of the 24 hr
𝑰𝒂 𝑰𝒂
rainfall distribution type and . ratios are listed in Table2. tc is time of concentration
𝑷 𝑷
(hr), Ia is an initial abstractions (mm) Page | 47
 Rainfall-Runoff Relationships
2. SCS CN method Table 1: Coefficients for SCS peak Discharge Method

Page | 48
 Rainfall-Runoff Relationships
2. SCS CN method Table 2: Ia/P for selected rainfall depths and Curve Numbers

Page | 49
 Rainfall-Runoff Relationships
Example 2.3: Determine the runoff depth and the peak flow from a catchment area of 100 ha for a
15 year return period, during which the maximum rainfall was recorded as 10 cm in 6 hr duration. The
catchment area has two parts, in which one of 25 ha is under pasture range contoured with poor
hydrologic condition and the remaining one is under small grain on straight row farming practice with
poor hydrologic condition. The soil of catchment is related to group-B. The initial abstraction of
catchment is rated as Ia= 0.3S. Take tc =0.5hr
%
Solution Land use
pasture range contoured with 25
Given
poor hydrologic condition
A=100ha, P=10cm, Ia =0.3S
small grain on straight row with poor 75
CN=CNII, AMC=II hydrologic condition
Soil group=B
Total 100
Hydrologic condition= poor
Page | 50
 Rainfall-Runoff Relationships
• Example 2.3:
Required
Runoff, Q=? Land use % CN Product
pasture range contoured 25 67 1675
Peak flow, qp=? with poor hydrologic
condition
small grain on straight 75 76 5700
row with poor hydrologic
condition
Total 100 7375

Average Curve Number, CN

𝟕𝟑𝟕𝟓
𝑪𝑵 = = 𝟕𝟑. 𝟕𝟓
𝟏𝟎𝟎

Page | 51
 Rainfall-Runoff Relationships
Example 2.3:
Potential maximum water retained in the catchment
𝟐𝟓𝟒𝟎𝟎
𝑺= − 𝟐𝟓𝟒
𝑪𝑵
𝟐𝟓𝟒𝟎 Peak flow, qp=?
𝑺= − 𝟐𝟓𝟒
𝟕𝟑.𝟕𝟓
𝑺 = 𝟗𝟎. 𝟒𝟏𝒎𝒎 qp=qu∗Ak∗Q
𝑺 = 𝟗. 𝟎𝟒𝟏𝒄𝒎
𝑷−𝑰𝒂 𝟐
Hence, the runoff, Q 𝑸=
𝑷−𝑰𝒂 +𝑺
But 𝑰𝒂 = 𝟎. 𝟑𝑺
𝑷−𝟎.𝟑𝑺 𝟐
𝑸=
𝑷−𝟎.𝟑𝑺 +𝑺 The total runoff volume over the catchment is:
𝑷−𝟎.𝟑𝑺 𝟐 𝑽𝒕 = 𝑸 ∗ 𝑨
𝑸= , for P>0.3S
𝑷+𝟎.𝟕𝑺
𝟐 𝑽𝒕 = 𝟑. 𝟐𝟓𝟑 ∗ 𝟏𝟎−𝟐 𝒎 ∗ 𝟏𝟎𝟎 ∗ 𝟏𝟎𝟒 𝒎𝟐
𝟏𝟎−𝟎.𝟑∗𝟗.𝟎𝟒𝟏
𝑸=
𝟏𝟎+𝟎.𝟕∗𝟗.𝟎𝟒𝟏 𝑽𝒕 = 𝟑𝟐𝟓𝟑𝟎𝒎𝟑
• 𝑸 = 𝟑. 𝟐𝟓𝟑𝒄𝒎 Page | 52
 Rainfall-Runoff Relationships
Exercise: A watershed has 400ha of row crop with poor hydrologic condition and 100ha of good pasture
land. The soil is of hydrologic soil group B. Estimate the runoff volume for the watershed under
antecedent moisture category III when 2 days of consecutive rainfall of 100mm and 90mm occur. Use the
SCS-CN equation.

Page | 53
 Rainfall-Runoff Relationships
3. Time-Area method
o The time – area method can be considered as an extension and improvement of the rational method.
o The peak discharge Qp is the sum of flow contributions from subdivisions of the catchment defined by
time contours (known as isochrones).
o Isochrones, which are lines of equal flow time to the river section where Qp is required.

o
The flow from each contributing area bounded by two isochrones (T - ΔT, T) is
obtained from the product of the mean intensity of effective rainfall (i) from time T-ΔT to
Page | 54
time T and the area (ΔA). Thus Q4, the flow at X at time 4h is given by:
 Rainfall-Runoff Relationships
3. Time-Area method
The flow from each contributing area bounded by two isochrones (T - ΔT, T) is obtained from the product of
the mean intensity of effective rainfall (i) from time T-ΔT to time T and the area (ΔA). Thus Q4, the flow at X at
time 4h is given by:
Q4 = I3 ΔA1+ i2ΔA2 + i1ΔA3 + i0 ΔA4

Page | 55
 RainArea
fall-Runoff Relationships
3. Time-Area method
We can also determine the peak discharge using this method using table as follows.
Area Intensity Sum
I1 I2 I3 I4 I5
A1 A1I1 - - - - A1I1
A2 A2I1 A1I2 - - - A2I1+A1I2
A3 A3I1 A2I2 A1I3 - - A3I1+A2I2+A1I3
A4 A4I1 A3I2 A2I3 A1I4 - A4I1+A3I2+A2I3+A1I4
A5 A5I1 A4I2 A3I3 A2I4 A1I5 A5I1+A4I2+A3I3+A2I4+A1I5

- A5I2 A4I3 A3I4 A2I5 A5I2+A4I3+A3I4+A2I5

- - A5I3 A4I4 A3I5 A5I3+A4I4+A3I5

Page | 56