Lecture notes Dr. J.K.

Mwangi ECE 2402: Hydrology I

RUNOFF AND FACTORS AFFECTING IT
INTRODUCTION
When a storm occurs, a portion of rainfall infiltrates into the ground while some evaporates. The rest flows as
a thin sheet of water over land surface as overland flow. If a relatively impermeable stratum exists in the
subsoil, the infiltrating water moves laterally in the surface soil and joins stream flow as sub-surface flow
(interflow, Fig.1). If no impeding layer exists in the subsoil the infiltrating water percolates into the ground as
deep seepage and builds up ground water table. Ground water contributes to stream flow, if the GWT is higher
than the water surface level of the stream thus creating a hydraulic gradient towards the stream. While the
three types of flow may contribute to stream flow, overland flow reaches the stream channel first, interflow
being slower reaches later while ground water may reach the stream channel after several days. Runoff is the
flow collected from the drainage basin and appearing at the basin outlet and is influenced by two factors:

i. Precipitation Characteristics
ii. Physiographical Factors.

PRECIPITATION CHARACTERISTICS:

a. Types of Precipitation:
Precipitation occurring in form of rainfall immediately produces runoff while precipitation in the form of
snow produces slow (delayed) runoff.

b. Rainfall Intensity:
If the rainfall intensity is greater than the infiltration rate of soil then runoff starts immediately after rainfall.
While in case of low rainfall intensity runoff starts later. Thus high intensities of rainfall yield higher runoff.

c. Duration of Rainfall:
It is directly related to the volume of runoff because soil infiltration rate decreases with rainfall duration.
Therefore medium intensity rainfall even results in considerable amount of runoff if duration is longer.

d. Rainfall Distribution:
Runoff from a watershed depends very much on rainfall distribution. When rainfall is concentrated over the
basin and the entre basin is contributing runoff is maximum. An intense storm on one portion of the catchment
results in higher runoff than a moderate storm over the entire catchment.

e. Direction of storm movement:

A storm moving in the direction of stream flow results in a shorter period of runoff and a higher peak than a
storm moving upstream.

f. Other Climate Factors:

Other factors such as temperature, wind velocity, relative humidity, annual rainfall etc. affect the water
losses from watershed area.

PHYSIOGRAPHIC FACTORS:
This includes both watershed and channel characteristics, which are as follows,

a. Size of Watershed:

1
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
A large watershed takes a longer time to drain the runoff to the outlet than a smaller watershed. Peak flow
decreases as the area increases because intense storms are usually over small areas and the period of surface
runoff increases with area.

b. Shape of Watershed:
Runoff is greatly affected by shape of watershed. Shape of watershed is generally expressed by the term “form
factor” and “compactness coefficient”. Compactness Coefficient is the ratio of watershed perimeter to the
circumference of a circle whose area is equal to that of the watershed.

Two types of shape:

Fan shape [tends to produce higher runoff very early]
Fern shape [tend to produced less runoff].

Fan shaped catchments are indicative of higher runoff because all tributaries are almost of the same size and
peak from the tributaries tend to reach the outlet roughly at the same time.

Fern shaped catchments on the other hand tend to give less runoff because the tributaries are generally of
different lengths and meet the main stream at regular intervals thus the discharges are likely to be
distributed over a larger period of time.

c. Slope of Watershed:
It controls the time of overland flow and time of concentration of rainfall. Sloppy watershed results in greater
runoff due to greater runoff velocity and vice-versa.

d. Orientation of Watershed:
This affects the evaporation and transpiration losses from the area. The north or south orientation, affects
the time of melting of collected snow.

Geological Factors
Soil type and its permeability greatly influence runoff generated from a catchment.

Meteorological Factors
Temperature, relative humidity, and wind affect runoff. High temperatures, low humidifies and strong winds
increase evaporation thus reducing surface runoff.

2
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I

Soil moisture:
Magnitude of runoff yield depends upon the initial moisture present in the soil at the time of rainfall. If rain
occurs after a long dry spell then infiltration rate is more, hence it contributes less runoff.

Soil type:
In filtration rate varies with soil type, so runoff is greatly affected by soil type.

Topographic characteristics:
It includes those topographic features which affect runoff. Undulate land has greater runoff than flat land
because runoff water gets additional velocity due to slope and little time to infiltrate.

Drainage Density:
It is defined as the ratio of the total channel length [L] in the watershed to total watershed area [A]. Greater
drainage density gives more runoff. Drainage density = L/A

Storm characteristics and direction

Low intensity storms over long spells produce relatively less runoff. If the direction of the storm producing
rainfall is down the stream receiving surface flow, it will produce more flood discharge than when it is up the
stream. Catchments located on the windward side of mountains, receive more rainfall and hence produce
more runoff than those on the leeward side.

Land use pattern

Arable land, grass land, forest or cultivated area all affect runoff generation.

Storage characteristics
Storage in channels and depressions reduce flood magnitudes. Upstream reservoirs, lakes and tanks moderate
flood magnitudes due to their storage effects. They also increase water loss due to evaporation.

Other factors like urbanisation affect runoff generation by increasing the area of impervious surfaces and
altering the natural drainage pattern of an area. Urbanisation also affects peak runoff rates. The net quantity
of water available from a catchment for purposes of water resources utilisation after deducting all losses is
called catchment yield. This is a random variable whose value may vary from year to year and can be estimated
only with some degree of probability. For example 75% dependable yield is the yield of a basin that can be
expected to be equalled or exceeded on average every three (3) years.

RAINFALL - RUNOFF RELATIONSHIPS

In hydrological analysis and design, it is often necessary to develop relationships between rainfall and runoff
that may be used to extrapolate/ interpolate runoff records from rainfall records which are generally available
for longer periods. Through these relationships runoff from ungaged catchments can be estimated.

RUNOFF ESTIMATION
Runoff from rainfall may be estimated by any of the following methods:

(i). Rainfall runoff coefficient

This is the ratio of the volume of stream flow above base flow to the volume of rainfall and is expressed as
R = KP where R = rainfall and k = runoff coefficient which depends on soil type.

3
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
(ii) Empirical formulae, curves and tables – applicable only in regions where they were developed. Based
on R-R relationships with factors that account for climatic / catchment characteristics. Example are Binnie,
Strange and Barlow tables which help to compute runoff volumes from rainfall amounts. Rainfall Vs runoff
curves enable computation of runoff from rainfall amounts.

Example of Binnie’s table.

Annual rainfall (mm) Runoff volume (Percentage) runoff
50 15
60 21
70 25
80 29
90 34
100 38
110 40
The above percentages may vary from region to region and are obtained from catchment studies

(iii) Infiltration methods:

This involves deducting infiltration losses from total precipitation or using infiltration indices.
If rainfall intensity is always more than infiltration capacity of the soil, this involves defining an infiltration rate
curve and superimposing it on the rainfall hyetograph of the storm. The area above the infiltration curve gives
the volume of runoff which when divided by catchment area gives runoff depth. Antecedent soil moisture plays
an important role in this approach and if rainfall intensity fluctuates below and above the infiltration rate curve
then infiltration indices such as Φ and W-indices can be used. The methods are largely empirical and derived
values are applicable only when rainfall characteristics and initial soil moisture conditions are identical to those
for which they were derived.

Infiltration curve method

4
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I

Φ Index method
1
(iv) Rational Formula method- Q p  CIA where Qp is peak runoff (m3/s), I is rainfall intensity
3.6
(mm/hr.) for the duration equal to the time of concentration tc, A is the catchment area in km2 and C is the
runoff coefficient which varies from 0.3 to 0.82 depending on soil type and cover. Choice of C value is an
important factor in this method.

(v) Hydrograph analysis

Hydrograph analysis to compute runoff

(vi) Unit hydrograph

Unit hydrograph method

Unit hydrograph is a hydrograph due to a unit depth (1cm) of runoff or rainfall excess after deducting losses.
Area under the unit hydrograph represents the total volume of rainfall due to 1 cm of net rainfall. If a storm
produces x cm of net rainfall, volume of runoff = (area of UH*x)

Streamflow hydrograph: A streamflow hydrograph is defined as a graph of flow rate against time at a given
location on a stream. It is an integral expression of physiographic and climatic characteristics that govern the
relationship between rainfall and runoff of a drainage basin.

5
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Although a hydrograph defines the relationship between total rainfall and watershed runoff, it is easier to
visualize the linear relationship between excess (effective) rainfall and direct runoff hydrograph due to the
short time interval between input and response functions. Direct runoff hydrograph is the hydrograph that
remains after deducting base flow from the total hydrograph.

Excess rainfall and direct runoff: Excess (effective) rainfall is that portion of rainfall that is neither retained
on the land surface nor infiltrated into the soil. After flowing over the watershed, excess rainfall becomes direct
runoff at the outlet. The graph of excess rainfall against time (excess rainfall hyetograph), is a key component
in the study of R-R relationships. The difference between total rainfall hyetograph and excess rainfall
hyetograph is abstractions (losses) and is the water lost through infiltration, interception and surface storage.

Methods of estimating ERH: depend on data available. If a streamflow hydrograph is available and base
flow has been separated to obtain a direct runoff hydrograph, ERH can be estimated using infiltration indices.
Infiltration index is the average rate of water loss such that the volume of rainfall excess at that rate equals
direct runoff. It expresses infiltration as an average rate throughout the storm, although infiltration capacity is
high at the beginning of a storm and low towards the end. Common infiltration indices are Ф and W-indices.

i. Ф-Index - is the constant rate of abstraction (cm/hr.) that yields an excess rainfall hyetograph
whose total depth equals the depth of direct runoff over the watershed.

Although the shaded area in the graph below is measured as rainfall, it does not appear as runoff. It represents
losses (evaporation, infiltration, interception and surface detention). Runoff measured over the watershed is
represented by the unshaded area.

Φ Index
ii. W-Index is a refined version of the Ф-Index that allows for depression storage, short rainless periods
during a storm and eliminates rainy periods when rainfall intensity was less than the assumed infiltration
capacity. It is the average infiltration rate when rainfall intensity exceeds infiltration capacity.

In this method average infiltration rate is given by: W  F  P  Q  S where

t t
F- is the total infiltration
t - Time during which rainfall intensity exceeds infiltration capacity
S - Volume of depression storage and interception
P - Total precipitation corresponding to t
Q - Total storm runoff

6
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
W-Index essentially equals Ф-Index less depression and interception storages. Rainfall excess can be estimated
if infiltration characteristics of a basin are known. Relative amounts of infiltration and runoff are determined
by nature of rainfall, season of the year, land slope, local soils and rocks and nature and type of crops grown.

Example 1
A 6 hour storm produced rainfall intensities of 7, 18, 25, 12, 10 and 3 mm/hr. in successive one hour intervals
over a basin of area 800 km2. If the resulting runoff observed was 2,460Ha-m, determine the Φ index for the
basin. (1 Ha-m represents the volume of water required to cover 1Ha to a depth of 1m i.e. 1Ha-m = 100,00m3)

Solution
Total runoff volume = 2640*10,000 = 26.4 *106 m3
Area of basin = 800 km2 = 800 * 106 m2
Depth of runoff = 264*105/800*106 = 0.033m = 33mm

The given rainfall hytegraph is as shown above

By trial and error if the Φ index is chosen as 8 mm/hr., the rainfall volume above the Φ index = 0 + (18-8) +
(25-8) + (12-8) + (10-8) + 0 = 0 + 10 + 17 + 4 + 2 + 0 = 33mm, which is equal to observed runoff: Therefore Φ =
8mm/hr. Repeat the process taking Φ index as 6mm/hr. and 9mm / hr. respectively.

Example 2:
A storm with a 10.00cm rainfall produced a direct runoff depth of 5.8cm. Given time distribution of the storm
is as shown in the table below, estimate the Φ index.
Time from start (hours) 1 2 3 4 5 6 7 8
Incremental rainfall per hour (cm) 0.4 0.9 1.5 2.3 1.8 1.6 1.0 0.5

Solution:
Total Infiltration = Rainfall depth – Runoff depth = 10.00-5.8 = 4.2 cm
Assume t = time of rainfall excess = 8 hours for the first trial
Then Φ index = 4.2/8 = 0.525cm / hr.
This value makes rainfalls for the first and eighth hour ineffective as they are less than 0.525 cm/hr.
For the second trial assume t = 6 hours
Infiltration = 10.00-0.4-0.5 -5.8 = 3.3cm which gives Φ index = 3.3/6 = 0.55 cm/hr.

This value is satisfactory as it gives t = 6 and by calculating rainfall excess

Time from start (h) 1 2 3 4 5 6 7 8
Incremental rainfall per h (cm) 0.0 0.35 0.95 1.75 1.25 1.05 0.45 0..
Total rainfall excess = 0.55cm = total runoff

7
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
INFILTRATION
Infiltration is the process through which water enters the earth’s surface strata. It replenishes soil moisture
deficiency after which excess water moves downwards by force of gravity (deep seepage or percolation) to
build up the groundwater table. The maximum rate at which the soil in any given condition is capable of
absorbing water is called its infiltration capacity. Infiltration (f) often begins at a high rate and decreases to a
fairly steady state rate (fc) as the rain continues. When a soil is dry, infiltration rate is high but decreases, as
the soil gets wet until reaches a uniform rate. For an initially dry soil subjected to rain, infiltration capacity
curve therefore shows an exponentially decaying trend.

Infiltration rate decreases as more water infiltrates into the soil

For consistency in hydrological calculations, a constant value of infiltration rate for the entire storm duration
called Infiltration Index is normally adopted. Information on infiltration rates is useful in the estimation of
surface runoff and ground water recharge. Infiltration rate is not constant but varies according to storm
duration and season of the year. Infiltration rate at any time is given by Horton’s equation who through field
experiments observed that infiltration begins at some rate fo and decreases exponentially until it reaches a
constant rate fc. f (t) = (fo-fc) e-kt where k is a decaying constant with dimension, T-1

Infiltration takes place at capacity rates only when rainfall intensity equals or exceeds infiltration capacity, fp;
i.e., f = fp when i ≥ fp; but when i < fp, f < fp and the actual infiltration rates are approximately equal to rainfall
rates. Infiltration depends upon intensity and duration of rainfall, weather, soil characteristics, vegetation
cover, land use, initial soil moisture content (wetness), entrapped air and depth of the ground water table.
Vegetation cover provides protection against rain drop impact and thus helps to increase infiltration.

Methods of determining infiltration include: Infiltrometers, observation in pits and ponds, artificial rain
simulators and hydrograph analysis. The most common is use of Infiltrometers (double or single ring). Rainfall
simulators involves allowing water to fall on an area by artificial showers at a uniform rate, measuring the
resulting runoff and estimating infiltration using hydrograph analysis.

8
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Two rings are driven into the ground to a depth of 15cm using a plate and hammer to penetrate the soil without
tilt or undue disturbance of the soil surface. Point gauges are placed in centre of the rings and in the annular
space between the two rings. Water is poured in the two rings to maintain the desired depth (2.5 to 15cm,
minimum 5cm) and water added to maintain the original constant depth at regular intervals of 5, 10, 15, 20,
40 etc. up to at least 6 hours noted. A curve of infiltration against time is then plotted. The purpose of the
outer tube is to eliminate the edge effect of the surrounding drier soil and prevent the water within the inner
space from spreading over a larger area after penetrating below the bottom of the ring. The rate at which
water is required to be added to the inner ring so as to maintain constant level determines the infiltration
capacity and its rate of variation.

STREAM FLOW
Streamflow constitutes historical data and is measured in m3/s. It relates directly to water supply, flood control,
reservoir design, irrigation, drainage etc. and serves as basis for many water resources engineering designs.
It’s most important characteristics is volume (monthly, annual etc.). When overland flow starts, some water is
held in puddles, pits and ponds (depression storage). The volume of water in transit as overland flow that has
not reached the stream channel is called detention storage. The portion of runoff in a rising flood in a stream,
absorbed by permeable stream boundaries above the normal phreatic surface is called bank storage, Figure 2.

Figure 2: Bank storage

Classification of streams: A stream is defined as a flow channel into which surface runoff from a specified
basin drains. Streams vary in width and length and may contain water perennially/intermittently. Streams
cannot be defined by length, size or shape, because they vary from storm to storm and season to season.
Descriptive definitions are therefore based on permanency of flows:

9
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
1. Ephemeral streams: Carry surface runoff only and flow only during and immediately after periods of
rainfall. They have no permanent or well defined channels but follow depressions in the natural
contour of the land. The drainage basin is either impervious or groundwater table is always below
stream bed throughout its length.

2. Intermittent streams: Carry flow during wet seasons and are dry during dry seasons. Groundwater
table lies above stream bed during the wet season but drops below the bed during the dry season.
Flow is derived principally from surface runoff but it may receive contributions from groundwater
during the wet season.

3. Perennial streams: Flow at all times even during the most severe drought. Groundwater table never
drops below the stream bed, thus maintaining a continuous supply. These streams are also called
effluent streams. In contrast in influent streams groundwater table is below river bed level and river
flow continuously recharges the groundwater reservoir at deeper depths.

Influent stream

Effluent stream
Streamflow generation: Mechanisms of stream flow generation as verified by field experiments can be
summarized as dominated by overland flow mechanisms (Hortonian and Saturation overland flow) and sub-
surface flow mechanism:
I. Hortonian overland flow: If the soil has an infiltration capacity (f), expressed as depth (mm)
absorbed per hour, then when rainfall intensity (i) is less than (f) then all the rain will be absorbed and
there will be no surface runoff. If (i) is greater than (f) then surface runoff will occur at the rate of (i-f).
Horton termed this difference (i-f) “rainfall excess” and considered surface runoff to take the form of
sheet flow whose depth can be measured in mm. As flow accumulates going down slope, depth of this
sheet flow increases until it discharge into a stream channel. Along with overland flow there is
depression storage in surface hollows and detention storage proportional to the depth of overland
flow itself. The soil stores infiltrated water which it releases gradually as sub-surface flow to enter the
stream as base flow during dry periods. Hortonian overland flow is applicable to impervious surfaces

10
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
in urban areas and to natural surfaces with thin soil layers and low infiltration capacities as in ASALS.
It rarely occurs on vegetated surfaces in humid regions, under which conditions, soil infiltration
capacity exceeds observed rainfall intensities for all except the most extreme rainfalls and sub-surface
flow becomes the primary mechanism for transporting storm water to streams.
II. Saturated overland flow: is produced when sub-surface flow saturates the soil near the bottom of
a slope and overland flow then occurs as rain falls onto the saturated soil. It differs from Hortonian
overland flow in that while in Hortonian overland flow the soil is saturated by infiltration from above,
in saturated overland flow it is saturated from below by subsurface flow. Saturated overland flow
occurs mostly at the bottom of hill slopes and near river banks. Velocity of subsurface flow is so low
that the whole watershed cannot contribute subsurface flow or saturated overland flow to a stream
during a storm.

III. Sub-surface flow mechanism: Subsurface flow (interflow) delivers water directly to a permanent
stream or expanding intermittent channel network in a variable source area. Control of sub-surface
flow is by hydrogeology. Recharge to groundwater is controlled by the overlying unsaturated zone.
Subsurface flow velocity is normally so low that alone it does not contribute significant amounts of
storm precipitation directly through flow except under special circumstances e.g. where the hydraulic
conductivity of the soil is very high.

IV Ground water (Base flow): Is the portion of precipitation that percolates downwards until it reaches
the ground water table. This water accretion eventually discharges into the stream if the ground water
table intersects the stream channel. However its contribution to stream flow can fluctuate rapidly
because of its very low velocity.

Streamflow measurement
Streamflow provides water for domestic, commercial and industrial use; irrigation; HEP generation; transport
channels for commerce and a medium for recreation. Stream flow records provide information on its
availability and variability in both time and space. These records are used in the planning and design of surface-
water related projects and in their management and operation. When stream flow occurs in excess it causes
hazards such as floods that result in damage to life and property. Records of flood events serve as the basis for
the design of bridges, culverts, dams, flood-control reservoirs and development of flood-warning systems.

Measurement of stream flow

Indirect methods:
Slope-area method using flow measuring devices - flumes, weirs etc. where there are no RG stations.

Direct methods:
Area-Velocity method (using current meters and floats)
Electromagnetic method
Dilution (tracer or chemical) method
Echo sounder method

i. Area-Velocity method involves (i) Selection of a gauging site (ii) Measurement of parameters of
discharge equation (Q = AV) or measurement of stage and development of a rating curve / table.
ii. Selection criteria for a regular gauging station (RGS):
 A way from confluence with tributary streams, to avoid backwater effect.
 Where stage is not affected by operations of H-E power stations
 Immediately upstream of natural riffles / falls / artificial control structures
 Where channel bed is stable to ensure stability of the rating curve / table.

11
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
 Where a stream does not overflow its banks to avoid loss of records during high stage.
 Straight and unobstructed reach with no aquatic growths e.g. algae etc.
 Easily accessible, especially for manual stations.

Although such sites may not be available in natural streams, good results can be obtained through proper
judgment and taking adequate precautions during actual flow and stage measurements.

iii. Measurement of discharge parameters (Q = AV)

 Stream cross sectional area (A) involves determining width and depth of the river and using area
formulae for regular sections. The river cross section is divided into a number of panels, each
containing approx. 10% of flow whose widths and depths are then measured. Panel areas are treated
as rectangular / trapezoidal depending on method of analysis.
 Velocity is measured along verticals bounding each panel at an appropriate point along the vertical.
One or more measurements may be taken along a vertical to obtain mean velocity.
 The measured data (panel width, depth of verticals and point velocities along verticals) are plotted
on a stream cross sectional sketch.
 Mathematical methods (mid- or mean-section methods) are then used to compute area and mean
velocity for each panel. The product of mean velocity and panel area is the panel discharge. Total
stream discharge is the sum of all panel discharges.

Measurement of width:
Stream width is measured from a fixed reference point (initial point) using a measuring tape/tag line stretched
across the stream. Accuracy of current-meter measurements depends on the number of verticals used. Should
be as many as possible and closely spaced in order to reflect the true shape of stream bed.

Measurement of depth:
Depth of the vertical and position of the current meter in the vertical are measured using a wading rod or
sounding line. In order to obtain accurate depth measurements using a sounding line, the sounding weight is
equipped with an electrical bottom-contact which gives a signal when the weight touches the stream bed. If
the sounding weight cannot maintain the line within 5° of the perpendicular to the water surface, the angle
between the sounding line and the vertical is measured to facilitate depth adjustment (should be <30°).
Accuracy of measurement improves if the suspension line is maintained in an almost vertical position.
iv. Air and wet line correction in depth measurement when using a current meter suspension cable.
Current meter cable, though designed to carry the meter can also be used to determine the appropriate
position at which it should be lowered below the water surface. The cable is lowered from a cable car
and does not remain vertical once in water due to water wave swipe (figure below).

Cableway with cable car

12
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
 When the vertical angle of departure is below 5º no correction is required.
 When the vertical angle is 5º< Ө < 30º, correction may be applied or sounding weight increased
in order to reduce Ө to between 0 and 5º.
 If vertical angle > 30º increase weight to bring it <30° or discard readings altogether.

The correction applied depends on the relative lengths of the cable line above the water surface (air-line)
and below the water surface (wet-line). Percentage air-line correction is then given by:

ka = (sec   1) * 100 where  is the angle made by the cable line with the vertical. Similarly, percentage
wet-line correction depends on  .

Thus if y is the vertical distance between the index point and the water surface and la is the air-line length of
the cable, then:
ka
la = y (1 + ). Similarly if d is the true vertical depth of the current meter below the water surface and lw is
100
the wet line length of the cable, then
kw
lw = y (1 + ) where kw is the percentage wet-line correction.
100

Position of current meter in SWIFT water

Vertical angle % air- line % wet-line Vertical angle Percentage air Percentage wet-
(Degrees) correction correction (Degrees) line correction line correction
4 0.24 0.06 18 5.15 1.64
6 0.55 0.16 20 6.42 2.04
8 0.98 0.32 22 7.85 2.48
10 1.54 0.50 24 9.46 2.96
12 2.23 0.72 26 11.26 3.50
14 3.06 0.98 28 13.36 4.08
16 4.03 1.28 30 15.47 4.72

13
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Example:
If the vertical distance between the index point and the water surface is 3.5m and we want to reach a point at
a depth of 4.5m below the water surface, then the total length of cable to be lowered when the cable line
makes an angle of 20° with the vertical is given by:
6.42 2.04
L = la + lw = (1 + ) * 3.5 + (1 + ) * 4.5 = 8.32m
100 100
Example:
The following observations were made using a current meter suspension cable from a cable cord. Vertical angle
ө = 10º, free depth = 3.8m, total observed depth = 10.3m. Calculate the true depth of the vertical. Take airline
correction as 1.54% of free depth and wet line correction as 0.50% of free depth. (Solution = 6.409m)

v. Determination of mean velocity in a vertical: Mean velocity can be calculated from point readings at
current meter positions in a vertical. The readings (rev/per time) are converted into velocity using current
meter equation (V= a+bN), where N is in rev/sec and a and b are current meter constants. The number of
point velocities necessary to determine mean velocity in a vertical depends on the accuracy required.
However, the following methods are common.

 Surface velocity: Vmean = 0.85Vs

 One-point method; Vmean = V0.6D (for shallow streams)

V  V0.8 D
 Two-point method: Vmean  0.2 D
2
V0.15 D  V0.5 D  V0.85 D
 Three-point method: Vmean  (for deep streams)
3

Current meter position D is measured from the water surface

Vertical velocity distribution and average velocity of flow for a river

14
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I

Vertical axis (cup type) current meter

Horizontal axis (Propeller type) current meter

Exercise
Using data given in table 1, plot the velocity distribution curve along a vertical and use the graph to: (i) Estimate
mean velocity by 1- point, 2- point and 3- point methods. (ii) Calculate the ratio of mean velocity by graphical
integration method and velocity at the water surface.
Table 1
Depth below surface (m) Point velocity (m/s)
0.0 2.1
0.2 2.2
0.4 2.25
0.6 2.22
0.8 1.98
1.0 2.0
1.2 1.7
1.4 1.22
1.6 0.56
1.7 0.0 river bed

vi. Computation of stream discharges: From the computed panel widths (b) depths of verticals (d) mean
velocities for verticals, panel discharges are computed using mean or mid-section methods:

15
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Mean section method:
Assumptions (1) end panels are triangular in shape (2) mid-panels are trapezoidal in shape (3) mean panel
velocity is the average of the means of the bounding verticals.

Mid-section method involves introducing mid ordinates in the middle of the panels to divide the widths into
two equal parts. Panels are then re-defined as per the new vertical ordinates.

Assumptions (1) areas bounded by mid-ordinates are rectangular in shape (2) end areas outside extreme
ordinates are neglected (3) velocity in each panel bounded by the new ordinates is taken as the velocity in
mid- vertical.
Mean Section Method

 V  Vx 1  d x  d x 1 
qx   x  Wx
 2  2 
[

Mid-Section Method:

[  W  Wx 1 
qx  Vx  x 1 d x
 2 

Depth-Velocity Integration Method:

 Draw depth-velocity curve for each vertical by plotting velocity observations against their
corresponding depths and joining with a smooth curve.
 Measure the area enclosed by each curve and its vertical by means of a planimeter.
 Plot the areas obtained above over the water-surface line of the measuring cross-section and draw a
smooth curve through the points. The area enclosed between this curve and the water- surface line
represents discharge through the cross section.

16
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Example
Using the following data compute the river discharge by both mid and mean section methods.

Distance from initial point (m) 2 4 5 8 10 12 14 16 17 18

Depth (m) 0 0.3 0.5 1.5 1.0 2.0 1.2 0.8 0.2 0
Mean velocity m/s - 0.1 0.14 0.5 0.16 1.20 0.90 0.92 0.1 -
Answers: Mid-Section = 10.464m3/s and Mean section 9.328 m3/s

Depth-velocity integration method of computing current-meter measurement

Example
Using data given in Table 2, compute streamflow by the mid-section method. Assume current meter
equation as given by V = 0.04+0.65N where N = Rev/sec

Table 2:
Distance from Depth (m) Meter depth (m) Rev Time (sec)
Bank (m)
0.6 0.3 0.18 10 50
1.2 1.07 0.85 22 55
0.21 35 52
1.8 1.58 1.28 28 53
0.30 40 58
2.7 1.92 1.52 32 58
0.40 45 60
3.4 1.34 1.07 28 45
0.27 33 46
4.0 0.67 0.40 22 50
4.6 0.24 0.15 12 49
5.2 0

17
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Solution
i. Calculate point velocities at each current meter position using the current meter equation (N- rev/sec)
ii. Determine mean velocity in each vertical using 1 or 2-point method after verifying CM position.
iii. Sketch the river cross-section, indicating on each vertical its depth and mean velocity.
iv. Draw mid-ordinates in each panel bounded by verticals indicated in (iii) above
v. Calculate the new panel widths as defined by the mid-ordinates. Mean velocities are now at the center
of these new panels.
vi. Calculate areas of these panels by multiplying the widths calculated in (v) with centre depths.
vii. Calculate panel discharge by multiplying areas calculated in (VI) with mean velocities as in (v).
viii. Sum up all panel discharges in step (vii) to obtain total discharge for the stream.
Repeat using the mean section method and compare the results

Computation of stream discharge using stage - discharge relationship (Rating curve):

River stage is the elevation of the water surface relative to a datum. By plotting discharge (Q) against
corresponding river stage (H), the following relationships can be developed: (1) rating curve (2) rating table (3)
rating equation Q = aHb. Rating curve is taken as the average curve between stage and discharge (H-Q) during
rising and falling stages. Although it is a normal curve, individual Q measurements may deviate from values
read against a corresponding H due to many factors for example:
 Change in river roughness
 Change in channel bed elevation due to scouring / silting
 Vegetal growth (aquatic growth)
 Variation in backwater from either main or tributary stream

To obtain accurate H-Q relationship it is necessary to check the cross sectional elevation with respect to the
TBM and to measure H and Q and compare with Q for same H on the rating curve. River stage is usually
measured using manual staff gauges or automatic level recorders. Staff gauges can be segmental for easy
installation. Extreme stages are captured by automatic recorders or special methods such as flood marks on
permanent objects, bottle gauges buried away from banks or flood sticks coated with a soluble paint.

Once a relationship has been established between measured H and Q, then stage information which is easily
obtained can be used to obtain discharge. This relationship must however be monitored for any changes. The
rating curve represents the integrated effects of a wide range of channel and flow parameters termed control.
A permanent control is where H-Q relationship for a station does not change with time. If it changes with time

18
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
it is a shifting control. Changes in relationship may arise from changes due to weed growth, aggradation /
degradation or unsteady flow effects of a rapidly changing stage. There are no permanent corective measures
for these problems except frequent current meter gauges to update rating curves.

Stage –Discharge plot:Arithmetic plot

Stream discharges decrease as stream stage and cross sectional area decrease. The relationship between stage
and discharge is unique for a gauging station. A graph of measured stream stage and discharge at a gauging
station defines this relationship. A curve that best represents the position of data points is called a best fitting
curve and is referred to as the stream rating curve. The constants a and b in the equation Q = aHb can be found
by a log-log plot of H-Q or by the method of least squares.
Example: Using the stage discharge values given below, establish a rating equation.
H (m) 0.5 0.8 1.2 1.32 1.4 1.56
Q(m3/s) 0.9 4 6.4 7.1 8.3 8.95

Solution
Plot Q vs H on a log – log paper, a is y-intercept and b is the gradient
Or write two equations (1) log Q = log (a) + b log H and (2) multiply equation 1 by log H to give the form log
HlogQ = log (a) log H +b (log H)2. Substitute each row of data in the two equations and sum the results for
equation 1 and 2. Then solve simultaneously for a and b. Approximate answer Q = 4.4 H1.7meaning a = 4.4 and
b = 1.7 and equation is the form Q = 4.4H1.7

Extension of Rating Curves

The stage discharge curve is established using discharge measurements taken around the mean stage in a river
incorporating both rising and falling stages. During high floods, the river may be inaccessible, making it
necessary to extend the stage discharge curve beyond the highest recorded discharges. The magnitude of a
design flood may also have been estimated using other methods but it is desired to get the stage corresponding
to this design flood and this requires a rating curve. Extension of rating curves is also required in river flow
forecasting and in solving flood control problems. These extensions may be subject to errors in alluvial rivers
due to aggradation / degradation and should therefore be counter checked with other methods and a field
survey made to establish the probable rating trend. Some of the methods used to extend rating curves include:

Logarithmic method: Where an equation of the form Q  bH  a  is assumed to be an appropriate fit
n

between discharge Q and river stage H. The value of a is assumed to be the stage corresponding to zero
discharge in the river. Taking logarithms on both sides Log Q = Log b + n log (H-a) suggesting that a plot of Q vs
(H-a) would be a straight line on a log paper which can be extrapolated to determine discharges corresponding
high stages. n is the slope and log b is the y intercept. A trial value of a is assumed and a plot of Q and (y-a)

19
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
made on a log-log plot. If s straight line is obtained then the assumed value of a is correct. If not the procedure
is repeated with different values until a nearly straight line is obtained.

Adjusting of flow rating curve

Analytical approaches using the least squares regression analysis may also be used to obtain values of a, b and
n. The straight line obtained may then be used in extrapolation to obtain H values for Q values outside the
range. Reliability of extra-polated values depends on stability of the gauging section control.

Example of Stage –Discharge curve: logarithmic plot

Using the data given below, establish the rating equation by both analytical and graphical methods
Stage (m) 0.7 1.2 1.7 2.0 2.5 2.7 3.1 3.3 3.7
Discahrge (m3/s) 1.6 2.4 3.4 4.5 5.4 6.2 7.7 9.0 10.3
Solution a = 2.02 and b = 1.19

20
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
ELECTRONIC INDUCTION METHOD
This involves burying a long conductor at the bottom of the channel to cover the entire width of the river bed
and which carries a current I to produce a controlled vertical magnetic field. As the flowing water in the
stream cuts the magnetic field, it produces an electromagnetic force (emf) which is related to stream
discharge, flow depth and the current in the conductor at the bottom of the river. Electrodes placed at the
sides of the channel section measure the emf produced by flow in the stream.
n
 Ed 
Q  k1   k 2  Where Q is the rate of flow in the river, d is the depth of flow, E is the emf induced, I is the
 I 
current in the conductor and k1, k 2 and n are system constants to be obtained on calibration. The method is
suitable for rivers with varying cross sections due to sedimentation, weed growth etc. and for tidal rivers where
the channel changes rapidly both in magnitude and direction.

ECHO SOUNDER METHOD

This uses an electro acoustic instrument and is suitable for rivers that are deep compared to their widths with
very high velocities and where conventional measurement of depth is not possible due to steep topography.
It is based on electrical principle of transmitting a pressure wave from water surface level to the river bed by a
transmitter for a short duration and receiving back the reflected pressure wave in the form of an echo on a
receiver. The time of transmission and reception is plotted automatically and from this knowing the velocity of
sound in a fluid medium (i.e. 1470m/s in water) the depth of flow can be determined.

DILUTION TECHNIQUES OF STREAMFLOW MEASUREMENT

The dilution method of flow measurement depends on the continuity principle applied to a tracer which is
allowed to mix completely with flow. It involves adding a non-reactive, dissolved chemical tracer to a flowing
stream and monitoring how the tracer is diluted after mixing thoroughly with water as it travels downstream.
The greater the flow, the greater will be level of tracer dilution. The technique is useful in streams that are
difficult to gauge with current meters because they are too shallow / slow and those with uneven, rocky
bottoms and or irregular wavy surfaces. A commonly used tracer is common salt (NaCl) which is cheap, easily
available and easily measurable in the stream using an electrical conductivity meter. At concentrations used
during dilution gauging, it is also not harmful to the stream ecosystem. Other tracers include fluorescent dyes,
such as rhodamine, which can be measured at very low concentrations and are thus useful for high stream
flows. Detecting low concentrations of such dyes however, requires expensive instrumentation.

Two approaches used in dilution gauging are (1) constant rate injection method which involves injecting a
tracer at a constant rate and measuring downstream dilution once concentration reaches a steady state (2)
Gulp injection method which involves injecting the tracer as an instantaneous "slug," and monitoring the
passage of rising-and-falling "tracer wave" downstream. Constant-rate method gives better results, especially
for low flows, but requires equipment for controlling tracer injection. It is impractical for high-volume flows.

Consider a tracer which does not react with the fluid or boundary.
Let C0 be the small initial concentration of the tracer in stream flow.
At section 1, a small quantity of (Volume V1) of high concentration C1 of this tracer is added (Fig. 1). Section 2
is sufficiently downstream of section 1 such that the tracer mixes thoroughly with the fluid due to the turbulent
mixing process while passing through the reach. The concentration profile taken at section 2 is schematically
shown in fig. 1.

21
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I

Figure: 1 Sudden- injection method

The concentration will have a base value of C0, increases from time t1 to a peak value and gradually reaches
the base value of C0 at time t2. The stream flow is assumed to be steady.

By continuity of the tracer material

M1 = mass of tracer added at section 1= V1*C1

Neglecting the second term on the RHS as insignificantly small,

Equation
1

Thus discharge Q in the stream can be estimated if for a known M1 the variation of C2 with time at section 2
and C0 are determined. This method is known as the sudden injection, gulp or integration method.

Another way of using the dilution principle is to inject the tracer of concentration C1 at a constant rate Qt at
section 1. At section 2, the concentration gradually rises from the background value of C0 at time t1 to a
constant value C2 (fig.2). At steady state, the continuity equation for the tracer is

Equation 2

This technique in which Q is estimated by knowing C1, C2, C0 and Qt is known as constant rate injection
method or plateau gauging.

22
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I

Figure 2: Constant rate injection method

Dilution gauging is based on the assumption of steady flow. If the flow is unsteady and the flow rate changes
appreciably during gauging, there will be a change in storage volume in the reach and the steady-state
continuity equation used to develop equations 1 and 2 will not be valid. Systematic errors can be expected in
such cases.

The diagram above illustrates how stream discharge Q is calculated using the constant-rate- injection method.
A solution with tracer concentration Cis is injected at a rate of q into a stream with a flow Q and background
tracer concentration Cbg. After adequate mixing downstream, the tracer reaches a constant diluted
concentration Cs.
From the law of conservation of mass, discharge-concentration products of stream flow ([Link]) and the
injected solution ([Link]) equals the product of cumulative discharge and concentration downstream ([Q+q]Cs).
The equation when re-arranged can be used to solve for the unknown stream discharge
Q = [(Cis - Cs) / (Cs - Cbg)] q
Tracers
A good tracer should have the following properties:
 Be non-toxic.
 Capable of being detected in a distinctive manner in small concentrations.
 Be cheap.

23
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
Three common types of tracers are:
1. Chemicals (common salt and sodium dichromate)
2. Fluorescent dyes (Rhoda mine-WT and Sulpho-Rhodamine B Extra)
3. Radioactive materials (Bromine-82, Sodium-24 and Iodine-132).

Common salt can be detected within an error of ± 1% up to a concentration of 10ppm while sodium dichromate
can be detected up to concentrations of 0.2 ppm. Fluorescent dyes can be detected at levels of tens of
nanograms per litre and hence require small amounts of solution. Radioactive tracers are detectable up to
accuracies of tens of picocuries per litre and permit large-scale dilutions. They require sophisticated equipment
and handling by trained personnel. Availability of detection instruments, environmental effects of tracer and
operational costs are factors that determine the tracer to be used.

Length of Reach
Length of reach between dosing and sampling sections should be adequate to achieve complete mixing. Length
depends dimensions of channel cross-section, discharge and turbulence levels. It varies from about 1 km for
mountain streams (about 1.0 m3/s) to about 100km for streams in plains (300m3/s). Mixing length becomes
very long for large rivers and is a constraint of this method. Artificial tracer mixing at the dosing station may
prove beneficial for small streams in reducing the length of the reach.

Application
The method estimates discharge directly and in an absolute way. Suitable for small turbulent mountain
streams. Can be used to check calibration, stage discharge curves etc. obtained through other methods.

DISCHARGE MEASUREMENT BY OTHER METHODS

Moving Boat method
Applicable to rapid measurement of Q in large streams e.g. during floods when sites become inudated and
inaccessible or at locations where unsteady flow conditions require rapid measurementof discharge. Requires
no fixed facilities and lends itself to use of alternative sites if conditions allow. Uses the velocity area-method
the major difference being the method of data collection which is dynamic as opposed to static. Current meter
position is fixed at a constant depth from the free surface which is a departure from the conventional area-
velocity method. Data is collected while the observer is on a boat rapidly traversing the cross section. Since
the current meter moves with the same velocity as that of the boat, it records the resultant of boat and stream
velocities which is the vectoral sum of the two. As the boat trverses the stream, a sonic sounder records the
depth of flow at different points which helps define the geometry of the cross section. Limitation is the high
velocity of water and safety considerations.

Slope area method

Channel flow is computed using either Chezy or Manning’s open channel formulae.
1
Q  AC RS and Q  A R 2 / 3 S 1 / 2 where A is the cross sectional area, S is the slope, R is the
n
hydraulic radius, C is Chezy’s constant and n manning’s coefficient. A straight reach of the river free from
disturbances is selected. Gives approximate discharge values because of the difficulty of choosing the right n,
cross sectional areas used being different from those prior to the flood and slope being an estimate as the
water surface during a flood can be warped.

Hydraulic structures
Applicable to small streams using flow measuring devices e.g. rectangular, triangular or trapezoidal weirs or
venturi-meters or Parshall flumes. Flow is expressed as a function of structure geometry and some reference
head. The discharge equation contains an empirical coefficient. An existing dam across a stream, or a bridge

24
 Lecture notes Dr. J.K. Mwangi ECE 2402: Hydrology I
opening may be used to measure discharge indirectly. In case of a dam, discharge is expressed as a function of
the length of dam and the head of flow above the dam. In the case of a bridge discharge is a function of the
area of flow at the constriction and the drop in water surface near the bridge.

Floats
Used to measure velocity at various verticals across the width of a stream. Discharge is then computed by the
velocity area method. A float is allowed to travel on the water surface and the time it takes to travel a known
distance is noted. Mean velocity at a vertical is taken as 0.85 times the velocity of the float at the surface. The
distance travelled by the float must be large enough (min 180m) to give accurate velocity. Floats can be
influenced by prevailing winds and hence not travel along straight paths between sections. Several floats
should therefore be introduced into the stream at various points to give mean velocities for several verticals.

Velocity measurement by floats

Examples 1
A 25g/l solution of a fluorescent tracer was discharged into a stream at a constant rate of 10m3/s. The
background concentration of the dye in the stream water was found to be zero. At a downstream section
sufficiently far away, the dye was found to reach an equilibrium concentration of 5 ppb. Estimate the stream
discharge.

Solution:
This is a constant rate injection method.
Q1 C1  C 2 
Q= , Q1 = 10 cm3/s = 10*10-6 m3/s
C2  C0
C1 = 0.025, C2 = 5x10-9, C0 = 0
10 x10 6
Q= 9
(0.025  5 x10 9  50m 3 / s
5 x10
Example 2
Into a stream with no initial trace of salt, a salt solution of concentration 20mg/cc is introduced at a constant
rate of 2 l/min. Samples collected at a point sufficiently downstream indicated an equilibrium salt
concentration of 0.05 ppm. Use the above information to determine the discharge in the stream.
Solution:
C0 = 0, C1 = 20 mg/c.c. = 20/1000 = 0.02 mg/l
C2 = 0.05 ppm = 0.05/1,000,000 = 5*10-8
q = 2 l/min = 0.002/60 = 3.33*10-5 = m3/s
C   0.02 
Q  q 1  1 = 3.33*10-5  8
 1 = 13.33 m3/s
 C2   5 *10 

25