Department of Civil Engineering
Faculty of Engineering, Zarqa University
Hydrographs
Chapter 5
HYDROGRAPH COMPONENTS
A hydrograph is a graph
showing discharge (i.e.,
stream flow at the
concentration point) versus
time. The various components
of a natural hydrograph are
shown in Fig. At the beginning,
there is only base flow (i.e., the
ground water contribution to
the stream) gradually depleting
in an exponential form. After
the storm commences, the
initial losses like interception
and infiltration are met and
then the surface flow begins.
HYDROGRAPH COMPONENTS
HYDROGRAPH COMPONENTS
The hydrograph gradually rises and reaches its
peak value after a time tp (called lag time or
basin lag) measured from the controid of the
hyetograph of net rain. Thereafter it declines and
there is a change of slope at the inflection point,
i.e., there has been, inflow of the rain up to this
point and after this there is gradual withdrawal of
catchment storage.
By this time the ground water table has been built up by the infiltrating and percolating
water, and now the ground water contributes more into the stream flow than at the
beginning of storm, but thereafter the GWT declines and the hydrograph again goes on
depleting in the exponential form called the ground water depletion curve or the
recession curve. If a second storm occurs now, again the hydrograph starts rising till it
reaches the new peak and then falls and the ground water recession begins, Fig.
HYDROGRAPH COMPONENTS
Thus, in actual streams gauged, the
hydrograph may have a single peak or multiple
peaks according to the complexity of storms. For
flood analysis and derivation of unit hydrograph, a
single peaked hydrograph is preferred. A complex
hydrograph, however, can be resolved into simple
hydrographs by drawing hypothetical recession
lines as shown in Fig.
It has been found from many hydrographs
that the ground water depletion curves for a
given drainage basin are nearly the same and As Qt is the derivation of storage with
hence it is termed as the normal ground water respect to time, dS Q dt
t t
depletion curve. It has been found that such Qt
curves, or at least their segments, follow a St
log e K r
simple inverse exponential function of the Hence, the discharge at any time is
elapsed time of the form, t proportional to the water remaining in
Q t Q0 K r
storage i.e.,
Q0 = discharge at start of period Qo S
Qt = discharge at end of time t o
Qt St
Kr = recession constant
HYDROGRAPH COMPONENTS
If a continuous stream flow record is
available for a number of years, the
hydrograph can be plotted on a semi-log
paper, i.e., log Q vs. t. (Fig.). Starting with
the lowest recession flow line, a line is
drawn tangential to the lower portion on
a tracing paper. This tangent line is
progressively extended by moving the
tracing-paper towards the origin with the
abscissae coincident, such that the line is
tangential to the lower portion of the
successive depletion curves of increasing
magnitude. This common line is the log
plot of the master depletion curve, which
is then converted to linear vertical scale
and is called the composite ground
water depletion curve.
HYDROGRAPH COMPONENTS
A composite ground water depletion curve
can be constructed from the recession
graphs resulting from a number of storms.
The various segments of the recession
graphs are shifted with respect to the time
axis until they appear to match and then,
an average or composite curve is drawn
through them as shown in Fig.
SEPARATION OF STREAMFLOW
COMPONENTS
Barnes (1940) proposed that the stream flow
components may be separated by plotting the
hydrograph on a semi-log paper (Fig.). The tail end of the
hydrograph plots as a straight line, i.e., ground water
recession (CD). If this straight line plot is extended
backwards up to the point E directly under the inflection
point I and line BE drawn, the area under BEC represents
the ground water contribution to the stream flow. If the
ordinates of this area are deducted from the ordinates
of the total hydrograph and re-plotted, the hydrograph
of surface runoff and interflow (subsurface flow) is
obtained, which plots as a straight line (HG) at the tail
end. By extending this line backwards up to the point L
directly under I and drawing the line FL, the area under
FLG gives the interflow component. By deducting the
ordinates of this from the ordinates of the hydrograph of
surface runoff and interflow, the hydrograph of surface
runoff is replotted whose tail end again, may plot as a
straight line representing the surface recession or
channel storage . The slopes of the straight line plots at
the tail ends of the separated hydrographs give the
recession constants.
HYDROGRAPH SEPARATION
For the derivation of unit hydrograph,
the base flow has to be separated from the
total runoff hydrograph (i.e., from the
hydrograph of the gauged stream flow).
Some of the well-known base flow
separation procedures are given in (Fig.)
(1) Simply by drawing a line AC
tangential to both the limbs at their lower
portion This method is very simple but is
approximate and can be used only for
preliminary estimates
(2) Extending the recession curve
existing prior to the occurrence of the
storm up to the point D directly under the
peak of the hydrograph and then drawing a
straight line DE, when E is a point on the
hydrograph N days after the peak, and N (in
N 0.83 A 0.2
days) is given by
HYDROGRAPH SEPARATION
(3) Simply by drawing a straight line AE,
from the point of rise to the point E, on the
hydrograph, N days after the peak.
(4) Construct a line AFG by projecting
backwards the ground water recession curve
after the storm, to a point F directly under the
inflection point of the falling limb and sketch
an arbitrary rising line from the point of rise of
the hydrograph to connect with the projected
base flow recession. This type of separation is
preferred where the ground water storage is
relatively large and reaches the stream fairly
rapidly, as in lime-stone terrains.
Many a time a straight line AE meets the requirements for practical purposes. Location of
the point E is where the slope of the recession curve changes abruptly, and as a rough guide E
is N days after the peak. In all the above four separation procedures, the area below the line
constructed represents the base flow, i.e., the ground water contribution to stream flow.
HYDROGRAPH SEPARATION
In Fig.(5.7), ADE is the assumed base flow.
Actually, when the water level in the stream rises
due to floods, the stream feeds the ground water
and the permeable boundaries (called bank storage
Fig.(5.8) and this is termed as negative base flow.
flow
But actually the variation in base flow may be much less than indicated in Fig.(5.7)
depending on the permeability of the boundaries and the gradient of the phreatic surface.
The resulting variation will be a slight dip from the extrapolated depletion curve, followed
by a gradual rise to a higher level than the initial value as the flood recedes as shown in Fig.
(5.9).
UNIT HYDROGRAPH
The unit hydrograph is defined as the
hydrograph of storm runoff resulting
from an isolated rainfall of some unit
duration occurring uniformly over the
entire area of the catchments, produces
a unit volume (i.e., 1 cm) of runoff.
Derivation of the unit hydrographs.
The following steps are adopted to
derive a unit hydrograph from an
observed flood hydrograph (Fig.).
UNIT HYDROGRAPH
1)Select from the records isolated (single-
peaked) intense storms, which occurring
uniformly over the catchments have produced
flood hydrographs with appreciable runoff (>1
cm, say, 8 to 16 cm). The unit period selected
should be such that the excess rainfall (i.e.,
Pnet) occurs fairly uniformly over the entire
drainage basin. Larger unit periods are
required for larger basins. The unit periods
may be in the range of 15-30% of the 'peak
time' period, i.e., the ne from the beginning of
surface runoff to the peak, and the typical
unit periods may be 3, 6, 8, 12 hours. (The
time of concentration may be a little longer
than the peak time).
The unit storm is a storm of such
duration that the period of surface runoff is
not much less for any other storm of shorter
duration.
UNIT HYDROGRAPH
(2) Select a flood hydrograph, which has
resulted from a unit storm chosen in item (i) above.
(3) Separate the base flow from the total runoff
by the well-known base flow separation
procedures).
(4) From the ordinates of the total runoff
hydrograph (at regular time intervals) deduct the
corresponding ordinates of base flow, to obtain the
ordinates of direct runoff.
(5) Divide the volume of direct runoff by the
area of the drainage basin to obtain the net
precipitation depth over the basin.
(6) Divide each of the ordinates of direct runoff
by the net precipitation depth to obtain the
ordinates of the unit hydrograph.
(7) Plot the ordinates of the unit hydrograph
against time since the beginning of direct runoff.
This will give the unit hydrograph for the basin, for
the duration of the unit storm (producing the flood
hydrograph) selected in item (i) above.
UNIT HYDROGRAPH
In unit hydrograph derivation, such storms
should be selected for which reliable rainfall and
runoff data are available. The net rain graph P = total rainfall
(hyetograph of excess rain) should be determined by
Pnet = net precipitation (from
deducting the storm loss and adjusting such that the
total volume of net storm rain is equal to the total hyetograph) or direct runoff as
volume of direct surface runoff. The unit hydrograph equivalent depth over the
derived, which, when applied to the known net rain basin.
data, should yield the known direct runoff Losses = due to infiltration (Fp ),
hydrograph. etc.
The steps given above for the derivation of unit
hydrograph can be formulated as follows (exemplified
A=area
A= of the drainage basin
in Fig.). Qd, DRO = direct runoff
Pnet P Losses ordinate
TRO = total runoff ordinate
Pnet
Q d t
,.........Q d DRO t = time interval between
A
successive direct runoff
TRO BFO DRO
ordinates
DRO
UGO BFO = base flow ordinate
Pnet
UNIT HYDROGRAPH
Elements of unit hydrograph.
The various elements of a unit hydrograph are
shown in Fig.
Base width (T)—The period of direct surface
runoff (due to a unit storm) of the unit hydrograph
is called the time base or the base width.
Unit storm—The storm of unit duration (.i.e.,
duration of the unit hydrograph) regardless of its
intensity is called unit storm.
Unit period—The time duration of the unit
storm (i.e., the duration of the unit hydrograph) is
called unit period.
Lag time (tp)—The time from the centre of a
unit storm to the peak discharge of the
corresponding unit hydrograph is called lag time.
Recession time (Tr)—The duration of the
direct surface runoff after the end of the excess or
net rainfall, is called recession time in hydrograph
analysis.
UNIT HYDROGRAPH
Propositions of the Unit Hydrograph
The following are the basic propositions of the unit
hydrograph:
(1) Same runoff duration. For all unit storms of
different intensities, the period of surface runoff (i.e.,
time base, base width or base period) is approximately
the same, although they produce different runoff
volumes (Fig. (a)).
(2) Proportional ordinates. For unit storms of
different intensities, the ordinates of the hydrograph at
any given time are in the same proportion as the
rainfall intensities (Fig.(a)).
(3) Principle of superposition. If there is a
continuous storm and/or isolated storms of uniform
intensity net rain, they may be divided into unit storms
and hydrographs of runoff for each storm obtained,
and the ordinates added with the appropriate time lag
to get the combined hydrograph (Fig. (b)).
UNIT HYDROGRAPH
(4) Same distribution percentages.
If the total period of surface runoff (i.e.,
time base or base width) is divided into
equal time intervals the percentage of
surface runoff that occurs during each of
these periods will be same for all unit
storms of different intensities (Fig. ).
UNIT HYDROGRAPH FROM
COMPLEX STORMS
Unit hydrographs from complex storms, involving
varying intensities of rain can be obtained by considering
the complex storm as successive unit storms of different
intensities and the runoff hydrograph (due to complex
storm) as the result of superposition of the successive
storm hydrographs. The ordinates of each storm
hydrograph are obtained as 'the storm intensity times the
corresponding ordinate of the unit hydrograph' as shown
in Fig. The unit hydrograph ordinates U 1, U2 ... are thus
obtained by writing a series of equations for each of the
ordinates Q1 ,Q2 ...., of the runoff hydrograph (due to
complex storm) and successively solving them. In Fig. ,
Q1=xu1 .: u1=?
Q2=xu2+yu1 .: u2=?
Q3=xu3 +yu2 + zu1 :. U3= ?
Thus, the tr-hour unit graph ordinates can be determined. Although the method is straight
forward, errors will creep in due to the assumptions on the intensity and duration of rainfall
and deduction of an assumed base flow; many trials are required to get a reasonable unit
graph.
UNIT HYDROGRAPH FROM
COMPLEX STORMS
Average Unit Hydrograph.
It is better if several unit graphs are
derived for different isolated (single
peaked) uniform intensity storms. If the
durations of storms are different, the
unit hydrographs may be altered to the
same duration (sect.... or art.....). From
several unit hydrographs for the same
duration, so obtained, an average unit
hydrograph can be sketched by
computing the average of the peak flows
and times to peak and sketching a
median line, (Fig.), so that the area under
the graph is equal to a runoff volume of 1
cm.
UNIT HYDROGRAPH FROM
COMPLEX STORMS
Alteration of Unit Hydrograph Duration. It becomes necessary, in computation
of flood hydrographs, that the duration of the unit graph available should be altered to suit
the duration of the design storm (to be used for obtaining the flood hydrograph). Two cases
arise:
Case (1) Changing a Short Duration Unit
Hydrograph to Longer Duration.
If the desired long duration of the unit graph is an
even multiple of the short, say a 3-hour unit graph is
given and a 6-hour unit graph is required. Assume two
consecutive unit storms, producing a net rain of 1 cm
each. Draw the two unit hydrographs, the second unit
graph being lagged by 3 hours. Draw now the
combined hydrograph by superposition. This combined
hydrograph will now produce 2 cm in 6 hours. To
obtain the 6-hour unit graph divide the ordinates of
the combined hydrograph by 2, (Fig.). It can be
observed that this 6-hour unit graph derived has a
longer time base by 3 hours than the 3-hour unit
graph, because of a lower intensity storm for a longer
time.
S-CURVE METHOD
Case (ii) Changing a Long Duration Unit
Hydrograph to a Shorter Duration by
S-curve Technique
S-curve or the summation curve
is the hydrograph of direct surface
discharge that would result from a
continuous succession of unit storms
producing 1 cm in tr —hr (Fig.). If the
time base of the unit hydrograph is T hr,
it reaches constant outflow (Qe) at T hr,
since 1 cm of net rain on the catchment
is being supplied and removed every tr
Qe = constant outflow (cumec)
hour and only T/tr . unit graphs are
tr = duration of the unit graph (hr)
necessary to produce an S-curve and
develop constant outflow given by, 2.78 A A = area of the basin (km2)
Q
e
tr
S-CURVE METHOD
Given a tr -hour unit graph, to derive a
tr' hour unit graph (tr' ≥ tr) —Shift the S-
curve by the required duration tr' along
the time axis. The graphical difference
between the ordinates of the two S-
curves, i.e., the shaded area in Fig.
represents the runoff due to tr' hours rain
at an intensity of 1/tr. cm/hr, i.e., runoff of
tr' / tr cm in tr' hours. To obtain a runoff of
1 cm in tr' hours (i.e., tr' -hour UG),
multiply the ordinates of the S-curve
difference by tr/ tr'. This technique may be
used to alter the duration of the given
unit hydrograph to a shorter or longer
duration. The longer duration need not
necessarily be a multiple of short
BERNARD'S DISTRIBUTION GRAPH
The distribution graph, introduced by
Bernard in 1935, shows the percentages of
total unit hydrograph, which occur during
successive, arbitrarily chosen, uniform time
increments, Fig. 5.20. It is an important
concept of the unit hydrograph theory that
all unit storms, regardless of their intensity,
produce nearly identical distribution
graphs.
The procedure of deriving the
distribution graph is first to separate the
base flow from the total runoff; the surface
runoff obtained is divided into convenient
time units, and the average rate of surface
runoff during each interval is determined.
BERNARD'S DISTRIBUTION GRAPH
If the rainfall-runoff data for a short
duration fairly uniform storm is known, the
duration of net rainfall is taken as the unit
period and the distribution percentages are
computed directly. But if there are multiple
storms of different intensities producing
different net rains. during successive unit
periods, a trial and error procedure of
applying the distribution percentages is
followed, till the direct surface runoff
during successive time intervals
corresponds to the computed values.
Once a distribution graph is derived for a drainage basin, any expected volume of
surface runoff from the basin can be converted into a discharge hydrograph. By
drawing a smooth curve along the steps of the distribution graph to give equal areas, a
unit hydrograph may be obtained as shown in Fig.
SYNTHETIC UNIT HYDROGRAPHS
There are many drainage basins (catchments)
for which no streamflow records are available and
unit hydrographs may be required for such basins.
In such cases, hydrographs may be synthesized
directly from other catchments, which are
hydrologically and meteorologically
homogeneous, or indirectly from other
catchments through the application of empirical
relationship. Methods for synthesizing
hydrographs for ungauged areas have been
developed Snyder (1938). Snyder analyzed a large
number of hydrographs an drainage basins ranging
in the area from 25 to 25000 km2 and selected the
three parameters for the development of unit
hydrograph, namely, base width (T), peak
discharge (Qp) and lag time (basin lag, tp),
Fig., and proposed the following empirical
formulae for the three parameters:
SYNTHETIC UNIT HYDROGRAPHS
Lag time, t p C t ( L. Lca ) 0.3
Standard duration tp
tr
of net rain 5.5
Peak flow, A
Qp C p
tp
Time base in days
tp
T 3 3
24
Peak flow per km2 of basin,
Qp
qp
tp
SYNTHETIC UNIT HYDROGRAPHS
Snyder proposed subsequently an
expression to allow for some variation in
the basin lag with variation in the net rain
duration, i.e., if the actual duration of the
storm is not equal to tr, but is tr' then:
tp
tr
5.5
t r/ t r
t pr tp
4
tpr = basin lag for a storm duration of tr' and tpr, is used instead of tp in Equations.
tp = lag time (basin lag), hr
Ct , Cp = empirical constants (Ct ≈ 0.2 to 2.2, Cp ≈ 2 to 6.5, the values depending on the basin
characteristics and units)
A = area of the catchment (km2)
L = length of the longest water course, i.e., of the mainstream from the gauging station (outlet or
measuring point) to its upstream boundary limit of the basin, (km) (Fig.).
Lca = length along the main stream from the gauging station (outlet) to a point on the stream
opposite the areal centre of gravity (centroid) of the basin.
SYNTHETIC UNIT HYDROGRAPHS
Snyder considered that the shape of the unit hydrograph is likely to be affected by the
basin characteristics like area, topography, shape of the slope, drainage density and channel
storage. He dealt with the size and shape of basin by measuring the length of the mainstream
channel. The coefficient Ct , reflects the size, shape and slope of the basin.
Linsley, Kohler and Paulhus gave an
expression for the lag time in terms of the
basin characteristics (see Fig.) as
LLca
t p Ct
S
S = basin slope, and the values of
n and Ct when L, Lca were measured
in miles are n=0.38
Ct = 1.2, for mountainous region
= 0.72, for foot hill areas
= 0.35, for valley areas
SYNTHETIC UNIT HYDROGRAPHS
Taylor and Scwarz found from an analysis of 20
drainage basins of size 50-4000 km2 in the north and
middle Atlantic States in USA that (when L and Lca were
measured in miles) t r/
0.6 T 5 t pr
Ct 2
S T 5 t peak
The usual procedure for developing a synthetic unit hydrograph for a basin
for which the streamflow records are not available is to collect the data for the 5. 6
W 50 1.08
basin like A, L, Lca and to get the coefficients, Ct and Cp from adjacent basins qp
whose streams are gauged and which are hydrometeorologically homogeneous. 3.21
From these the three parameters, i.e., the time to peak, the peak flow and the W 75
q 1p.08
time base are determined from the Snyder's empirical equations, and the unit
hydrograph can be sketched so that the area under the curve is equal to a W W 50
75
runoff volume of 1 cm. Empirical formulae have been developed by the US 1.75
Army Corps of Engineers (1959) for the widths of W50 and W75 of the
hydrograph in hours at 50% and 75% height of the peak flow ordinate,
respectively, (see Fig. ) as:
TRANSPOSING UNIT HYDROGRAPHS
LLca
t p Ct
S
LLca
log t p log C t n log
S
y c m x
LLca
Hence, a plot of tp vs S on log-log paper
from data from basins of similar hydrologic
characteristic gives a straight –line
LL
relationship
(fig.). The constant Ct = tp when S = 1, and
ca
the slope of the straight line gives n. It may be Taylor and Scwarz Eqs.
observed that for basins having different 0.6
hydrologic characteristics the straight lines Ct
S
obtained are nearly parallel, i.e., the values of
Ct depending upon the slope of the basin (as
can be seen from Eq. Taylor and Scwarz) but
the value of n is almost the same.
