A Module in

Hydrometeorology

Prepared By:

NATHANIEL R. ALIBUYOG

Mariano Marcos State University

College of Engineering
City of Batac, Ilocos Norte

August 2020

Module 3
Precipitation
Introduction
Precipitation replenishes surface water bodies, renews soil moisture for plants, and recharges aquifers. Its
principal forms are rain and snow. The relative importance of these forms is determined by the climate of
the area under consideration. Hydrologic modeling and water resources assessments depend upon a
knowledge of the form and amount of precipitation occurring in a region of concern over a time period of
interest.

Learning Outcomes

At the end of the lesson, you should be able to:

• Compare the different types and forms of precipitation
• Discuss the variability of precipitation over time and space and the factors affecting them
• Analyze rainfall data and present results in different type of graphs
• Compute the mean precipitation using different methods and interpret the results
• Apply different techniques of adjusting precipitation data
• Develop and interpret intensity-duration-frequency curves

Learning
Input

3.1 Formation of Precipitation

The conditions for precipitation to take place may be summarized stepwise as follows:

1. Supply of moisture
2. Cooling to below point condensation
3. Condensation
4. Growth of particles

The supply of moisture is obtained through evaporation from wet surfaces, transpiration from
vegetation or transport from elsewhere. The cooling of moist air may be through contact with a cold
earth surface causing dew, white frost, mist or fog, and loss of heat through long wave radiation (fog
patches). However, much more important is the lifting of air masses under adiabatic conditions
(dynamic cooling) causing a fall of temperature to near its dew point.

Five lifting mechanism can be distinguished:

1. Convection, due to vertical instability of the air. The air is said to be unstable if the temperature
gradient is larger than the adiabatic lapse rate. Consequently, a parcel moving up obtains a
temperature higher than its immediate surroundings. Since the pressure on both is the same
the density of the parcel becomes less than the environment and buoyancy causes the parcel
to ascend rapidly. Instability of the atmosphere usually results from the heating of the lower air
layers by a hot earth surface and the cooling of the upper layers by outgoing radiation.
 Convective rainfall is common in tropical regions and it usually appears as a thunderstorm in
temperate climates during the summer period. Rainfall intensities of convective storms can be
very high locally; the duration, however is generally short.
2. Orographic lifting. When air passes over a mountain it is forced to rise which may cause
rainfall of the windward slope. As a result of orographic lifting rainfall amounts are usually
highest in the mountainous part of the river basin.
3. Frontal Lifting. The existence of an area with low pressure causes surrounding air to move
into the depression, displacing low pressure air upwards, which may then be cooled to dew
point. If cold air is replaced by warm air (warm front) the frontal zone is usually large and the
rainfall of low intensity and long duration. A cold shows a much steeper slope of the interface
of warm and cold air usually resulting in rainfall of shorter duration and higher intensity.
4. Cyclones, tropical depressions or hurricanes. These are active depressions, which gain
energy while moving over warm ocean water, and which dissipate energy while moving over
land or cold water. They may cause torrential rains and heavy storms. Typical characteristics
of these tropical depressions are high intensity rainfall of long duration (several days).
5. Convergence. The Inter-tropical Convergence Zone (ITCZ) is the tropical region where the
air masse originating from the Tropic of Cancer and Capricon converge and lift. In the tropics,
the position of the ITCZ governs the occurrence of wet and dry seasons. In July, the ITCZ lies
to the north of the equator and in January it lies to the south. In the tropics the position of the
ITCZ determines the main rain-bringing mechanism which is also called monsoon. Hence, the
ITCZ is also called the Monsoon Trough.

3.2 Amount of Precipitable Water

Estimate of the amount of precipitation that might occur over a given region with favorable conditions
are often useful. These may be obtained by calculating the amount of water contained in a column of
atmosphere extending up from the earth’s surface. This quantity is known as the precipitable water,
W, although it cannot all be removed from the atmosphere by natural processes.

The amount of precipitable water in the atmosphere is equal to the water mass contained in a column,
having a square base of 1 cm, between elevation zero and some height z. In equation form, it is given
as

"
𝑊 = ∫# 𝜌! 𝑑𝑧

where rw = absolute humidity, W is the depth of precipitable water in cm.

3.3 Formation of Precipitation

1. Convective Precipitation. Convective precipitation is typical of the tropics and is brought

about by heating of the air at the interface with the ground. This heated are expands with a
resultant reduction in weight. During this period, increasing quantities of water vapor are taken
up; the warm moisture-laden air become unstable; and pronounced vertical current developed.
Dynamic cooling takes place, causing condensation and precipitation. Convective precipitation
may be in the form of light showers or storms of extremely high intensity.
2. Orographic Precipitation. Orographic precipitation results from the mechanical lifting of moist
horizontal air currents over natural barriers such as mountain ranges. Factors that are
important in this process include land elevation, local slope, orientation of land slope, and
distance from the moisture source.

3. Cyclonic Precipitation. Cyclonic precipitation is associated with the movement of air masses
from high-pressure regions to low-pressure regions. These pressure differences are created
by the annual heating of the earth’s surface. Cyclonic precipitation may be classified as frontal
and nonfrontal.
 a. Nonfrontal precipitation - is produce when air is lifted through horizontal convergence
of the inflow into a low-pressure area.
b. Frontal precipitation - results from the lifting of warm air over cold air at the contact zone
between air masses having different characteristics.
c. Warm front – if the air masses are moving so that warm air replaces colder air
d. Cold front - if the cold air displaces warm air
e. Stationary front - if the front is not in motion.

4. Thunderstorms. Thunderstorm cells develop from vertical air movements associated with
intense surface heating or orographic effects.

3.4 Types of Precipitation

1. Rain - Water drops have a diameter of at least 0.5 mm. It can be classified based on intensity
as
• Light rain – up to 2.5 mm/h
• Moderate – 2.5 mm/h to 7.5 mm/h
• Heavy rain – greater than 7.5 mm/h

2. Snow - Precipitation in the form of ice crystals which usually combine to form flakes, with an
intensity density of 0.1 g/cm3

3. Drizzle - Rain droplets of size less than 0.5 mm and rain intensity of less than 1 mm/h
4. Glaze - When rain or drizzle touches ground at 0oC, glaze or freezing rain is formed

5. Sleet - It is frozen raindrops of transparent grains which form when rain falls through air at
subfreezing temperature

6. Hail - It is a showery precipitation in the form of irregular pellets or lumps of ice of size more
than 8 mm

3.4 Measurement of Precipitation

The instrument used to collect and measure the precipitation is called rain gauge.

Types of rain gauge

1. Non-recording type – The standard gage has a collector of 8-in diameter. Rain passes from the
collector into a cylindrical measuring tube inside the overflow can. The measuring tube has a
 cross-sectional area one-tent that of the collector so that 0.1 in rainfall in the tube will have a
depth of 1 in.
2. Recording type – The instrument records the graphical variation of the rainfall, the collected
quantity in a certain time interval and the intensity of the rainfall (mm/hr). It allows continuous
measurement of the rainfall.
a. Tipping bucket type - These buckets are so balanced that when 0.25 mm of rain falls into
one bucket, it tips bringing the other bucket in position.
b. Weighing -bucket type – The catch empties into a bucket mounted on a weighing scale.
The weight of the bucket and its contents are recorded on a clock work driven chart. The
instrument gives a plot of cumulative rainfall against time (mass curve of rainfall).

3.5 Presentation of Rainfall Data

Hyetograph – plot of rainfall intensity against time. The hyetograph is derived from the mass curve and is
usually represented by a bar chart. It is a very convenient way of representing the characteristics of a storm
and is particularly important in the development of design storms to predict extreme floods.

Mass Curve of Rainfall -plot of accumulated precipitation against time, plotted in chronological order.
Mass curves of rainfall are very useful in extracting the information on the duration and magnitude of a
storm. Also, intensities at various time intervals in a storm can be obtained by the slope of the curve.

Point Rainfall – It is also known as station rainfall. It refers to the rainfall data of a station.

Moving average. Moving average is a technique for smoothening out the high frequency fluctuations of a
time series and to enable the trend, if any, to be noticed. The basic principle is that a window of time range
m years is selected. Starting from the first set of m years of data, the average of the data of m years is
calculated and placed in the middle year of the range m. The window is next moved sequentially one time
unit (year) at a time and he mean of m terms in the window is determined at each window location. The
value of m can be 3 or more years; usually ad odd value.
 Hyetograph Mass Curve of Rainfall

3.6 Mean Precipitation Over an Area

For most hydrologic analyses, it is important to know the areal distribution of precipitation. The reliability
of rainfall measured at one gauge in representing the average depth over a surrounding area is a function
of:
a. The distance from the gauge to the center of the representative area
b. The size of the area
c. Topography
d. The nature of the rainfall of concern
e. Local storm pattern characteristics

The following methods are used to measure the average precipitation over an area:
1. Arithmetic mean method
2. Theissen polygon method
3. Isohyetal method
4. Inverse distance weighting

Arithmetic Mean Method – Simplest method for determining areal average. In this method, only stations
within the watershed boundary are considered.
!
𝑃" = " ∑"
#$! 𝑃#

where: Pi = rainfall at the ith raingauge station

N = total number of raingauge stations

Thiessen Polygon Method - In this method the rainfall recorded at each station is given a weightage on
the basis of an area closes to the station. The procedure of determining the weighing area is as follows:
Consider a catchment area as shown below containing three raingauge stations. There are three stations
outside the catchment but in its neighborhood. The catchment area is drawn to scale and the positions of
the six stations marked on it. Stations 1 to 6 are joined to form a network of triangles. Perpendicular
bisectors for each of the sides of the triangle are drawn. These bisectors form a polygon around each station.
The boundary of the catchment, if it cuts the bisector is taken as the outer limit of the polygon. Thus, for
station 1, the bounding polygon is abcd. For station 2, kade is taken as the bounding polygon. These
bounding polygons are called Thiessen polygons. The areas of the polygons are determined either with a
planimeter or by using an overlay grid.

The average precipitation over the watershed area is computed as:

∑ "
&! '!
𝑃" = !#$
∑'

where: Pi = rainfall at the ith raingauge station

Ai = area of the polygon enclosing the ith raingauge station
N = total number of raingauge stations

The Thiessen polygon method is superior to the arithmetic average method as some weightage is given to
the various stations on a rational basis. Further, the raingauge stations outside the catchment are also used
effectively.

Isohyetal Method. An isohyets is a line joining points of equal rainfall magnitude. In the isohyetal method.
The catchment area is drawn to scale and the raingauge stations are marked. The recorded values for which
areal average 𝑃" is to be determined are then marked on the plot at appropriate stations. Neighboring stations
outside the catchment are also considered. The isohyets of various values are then drawn by considering
point rainfalls as guide and interpolating between them by the eye. The procedure is similar to the drawing
of elevation contours based on spot levels.
The area between two isohyets are then determined with a planimeter. If the isohyets go out of the
catchment, the catchment boundary is used as the bounding line. The average value of the rainfall indicated
by two isohyets is assumed to be acting over the inter-isohyet area. Thus, P1, P2, …, Pn are the values of
isohyets and if A1, A2, …, An-1 are the inter-isohyets areas respectively, then the mean precipitation over
the catchment of area A is given by
% &% % &% % &%
'$ ( $ ' )*'' ( ' ()*⋯*')*$ ( )*$ ))
𝑃" = ' '
'
'

The Isohyet method is superior to other two methods especially when stations are large in number.

Inverse DIstance Weighing (IDW) Method. Prediction at a point is more influenced by nearby
measurements than that by distant measurements. The prediction at an ungauged point is inversely
proportional to the distance to the measurement points.

Steps:
a. Computed distance (di) from ungauged point to all measurement points

𝑑!, = '(𝑋! − 𝑋, ), + (𝑌! − 𝑌, ),

b. Compute the precipitation at the ungauged point using the

formula

𝑃
∑-#$! . #, /
𝑑#
𝑃" =
1
∑-#$! 0 2
𝑑# ,

where: P = precipitation at station i

d12 = distance between point station 1 and 2
n = number of gauged points
Example 1. In a catchment area, approximated by a circle of diameter 100 km, four rainfall stations are
situated inside the catchment and one station is outside in its neighborhood. The coordinate of the center
of the catchment and of the five station are given below. Also given are the annual precipitation recorded
by the five stations in 2019 Determine the average annual precipitation by the Thiessen polygon method.

Center: (100,100) Diameter: 100 km

Distance are in km

Station 1 2 3 4 5
Coordinates (30,80) (70,100) (100,140) (130,100) (100,70)
Precipitation 85.0 135.2 95.3 146.4 102.2
(cm)

Thiessen polygon

Solution:

The catchment area is drawn to scale and the stations are marked on it. The stations are joined to form a set
of triangles and the perpendicular bisector of each side is then drawn. The Thiessen-polygon area enclosing
each station is then identified. It may be noted that Station 1 in this problem does not have any area of
influence in the catchment. The areas of various Thiessen polygons are determined either by a planimeter
or by placing an overlay grid.

Station Boundary of Area (km2) Fraction of Rainfall Weighted P

area #Grids total area (cm) (cm) (Col 4 x
Col 5)
1 - - - 85.0 -
2 abcd 2141 0.2726 135.2 36.86
3 dce 1609 0.2049 95.3 19.53
4 ecbf 2141 0.2726 146.4 39.91
5 fba 1963 0.2499 102.2 25.54
Total 7854 1.000 121.84
Mean Precipitation = 121.84 cm
Example 2. The Isohyets due to a storm in a catchment were drawn as shown below and the area of the
catchment bounded by isohyets were tabulated as below.

Isohyets (cm) Area (km2)

Station – 12.0 30
12.0 – 10.0 140
10.0 – 8.0 80
8.0 – 6.0 180
6.0 – 4.0 20

Estimate the mean precipitation due to the storm.

Solution:

For the first area consisting of a station surrounded by a closed isohyet, a precipitation value of 12.0 cm is
taken. For all other areas, the mean of two bounding isohyets are taken.

Isohyets Average value Area (km2) Fraction of Weighted P

of P (cm) total area (cm)
(Col 3 /450) (Col 2 x Col
4)
1 2 3 4 5
12.0 12.0 30 0.0667 0.800
12.0 – 10.0 11.0 140 0.3111 3.422
10.0 – 8.0 9.0 80 0.1778 1.600
8.0 – 6.0 7.0 180 0.4000 2.800
6.0 – 4.0 5.0 20 0.0444 0.222
Total 450 1.0000 8.844
Mean Precipitation 𝑃"= 8.84 cm

3.6 Adjustments of Precipitation Data

Before using the rainfall records of a station, it is necessary to first check the data for continuity and
consistency. The continuity of a record may be broken with missing data due to many reasons such as
damage or fault in a raingauge during a period. The missing data can be estimated by using the data of the
neighboring stations. In these calculations the Normal Rainfall is used as standard comparison. The normal
rainfall is the average value of rainfall at a particular date, month, or year over a specified 30-year period.
The 30-year normal are recomputed every decade. Thus, the term Normal annual precipitation at Station
A means the average annual precipitation at A based on a specified 30-year of record.

Estimation of Missing Data

Given the annual precipitation values, P1, P2, P3, …, Pm at neighboring M stations 1, 2, 3, …, M respectively,
it is required to find the missing annual precipitation Px at a station X not included in the above M stations.
Further, the normal annual precipitations N1, N2, …, Ni at each of the above (M+1) stations including
station X are known.

If the normal annual precipitations at various stations are within about 10% of the normal annual
precipitation at station X, then a simple arithmetic average procedure is followed to estimate Px. Thus
!
𝑃. = / [𝑃! + 𝑃, + ⋯ + 𝑃0 ]

If the normal precipitation vary considerably, then Px is estimated by weighing the precipitation at the
various stations by the ratios of normal annual precipitations. This method, known as the Normal Ratio
Method, gives Px as

"+ &$ & &

𝑃. = 6 + "' + ⋯ + ", 7
/ "$ ' ,

Example 3. The normal annual rainfall at station A, B, C, and D in a basin are 80.97, 67.59, 76.28, and
92.01 cm respectively. In the year 2019, the station D was inoperative and the stations A, B, and C recorded
annual precipitation of 91.11, 72.23, and 79.89 cm, respectively. Estimate the annual rainfall at station D
in that year.

Solution:

As the normal rainfall values vary more than 10% , the normal ration method is adopted. Hence,

"+ &$ & &

𝑃. = / "$
6 + "' + ⋯ + ", 7
' ,

1,.3! 1!.!! 6,.,4 61.51

𝑃. = 4
653.61 + 76.81 + 67.,57

𝑃. = 99.48 𝑐𝑚

Test for Consistency of Record

If the conditions relevant to the recording of a raingauge station have undergone a significant change during
the period of record, inconsistency would arise in the rainfall data of that station. This inconsistency would
be felt from the time the significant change took place.

Some of the cause of inconsistence of record are:

• Shifting of a raingauge station to a new location
• The neighborhood of the station undergoing a marked change
• Change in the ecosystem due to calamities, such as forest fires, landslides
• Occurrence of observational error from a certain date

The checking for inconsistency of a record is done by the Double Mass Curve Technique. This technique
is based on the principle that when each recorded data comes from the same parent population, they are
consistent.

Steps:

1. A group of 5 to 10 base stations in the neighborhood of the problem station X is selected.

2. The data of the annual (or monthly or seasonal mean) rainfall of the station X and also the average
rainfall of the group of bases stations covering a long period is arranged in the reverse
chronological order (i.e., the latest record as the first entry and the oldest record as the last entry
in the list)
3. The accumulated precipitation of the station X (i.e., ∑ 𝑃. ) and the accumulated values of the
average of the group of base stations (i.e., ∑ 𝑃9:; ) are calculated starting from the latest record.
4. Values of ∑ 𝑃. are plotted against ∑ 𝑃9:; for various consecutive time periods.
5. A decided break in the slope of the resulting plot indicates a change in the precipitation regime of
station X.
6. The precipitation values at station X beyond the period of change of regime is corrected by using
the relation

𝑀<
𝑃<. = 𝑃. > @
𝑀9

where: Pcx = corrected precipitation at any time period t1 at station X

Px = original recorded precipitation at time period t1 at station X
Mc = corrected slope of the double mass curve
Ma = original slope of the double mass curve

Double Mass Curve

Example 4. Annual rainfall data for station M as well as the average annual rainfall values for a
group of ten neighboring stations located in a meteorologically homogenous region are given
below.

Annual Average Annual

Average
Rainfall of Annual Rainfall of
Year Year Annual
Station M Rainfall of the Station M
Rainfall of
(mm) Group (mm) (mm)
 the Group
(mm)
1950 676 780 c 1244 1400
1951 578 660 1966 999 1140
1952 95 110 1967 573 650
1953 462 520 1968 596 646
1954 472 540 1969 375 350
1955 699 800 1970 635 590
1956 479 540 1971 497 490
1957 431 490 1972 386 400
1958 493 560 1973 438 390
1959 503 575 1974 568 570
1960 415 480 1975 356 377
1961 531 600 1976 685 653
1962 504 580 1977 825 787
1963 828 950 1978 426 410
1964 679 770 1979 612 588

Test the consistency of the annual rainfall data of station M and correct the record if there is any
discrepancy. Estimate the mean annual precipitation at station M.

Solution:

a. The data is sorted in descending order of the year, starting from the latest year 1979.
b. Cumulative value of station M rainfall (∑ 𝑃0 ) and the ten station average rainfall values
(∑ 𝑃9:; ) are calculated as shown in Table below.
c. The data is then plotted with ∑ 𝑃0 on the y-axis and ∑ 𝑃9:; on the x-axis to obtain a double
mass curve plot as shown below
d. The value of the year corresponding to the plotted points is also noted on the plot.
e. It is that the data plots as two straight lines with a break of grade at the year 1969. This
represent a change in the regime of the station M after the year 1968.
f. The slope of the best straight line for the period 1979 – 1968 is Mc = 1.029 and the slope of
the best straight line for the period 1968 – 1950 is Ma = 0.8779
g. The correction ratio to bring the old record (1950 – 1968) to the current (post 1968) regimes
is = Mc/Ma = 1.029/0.8779 = 1.173
h. Each of the pre 1960 annual rainfall value is multiplied by the correction ration of 1.173 to get
the adjusted value.
i. The adjusted values at station M are shown in Col. 5 of Table. The finalized values of Pm
(rounded off to nearest mm) for all the 30 years of record are shown in Col. 7
j. The mean annual precipitation at station M (based on the corrected time series) = (19004/30)
= 633.5 mm

1 2 3 4 5 6 7 8
YEA Pm Pave SPm SPave Adjusted Finalized Adjusted
R Values of Pm values of Cumulative
Pm value of Pm
1979 612 588 612 588 612 612
1978 426 410 1038 998 426 1038
1977 825 787 1863 1785 825 1863
1976 685 653 2548 2438 685 2548
1975 356 377 2904 2815 356 2904
1974 568 570 3472 3385 568 3472
1973 438 390 3910 3775 438 3910
1972 386 400 4296 4175 386 4296
1971 497 490 4793 4665 497 4793
1970 635 590 5428 5255 635 5428
1969 375 350 5803 5605 375 5803
1968 596 646 6399 6251 698.92 699 6502
1967 573 650 6972 6901 671.95 672 7174
1966 999 1140 7971 8041 1171.51 1172 8345
1965 1244 1400 9215 9441 1458.82 1459 9804
1964 679 770 9894 10211 796.25 796 10600
1963 828 950 10722 11161 970.98 971 11571
1962 504 580 11226 11741 591.03 591 12162
1961 531 600 11757 12341 622.70 623 12785
1960 415 480 12172 12821 486.66 487 13272
1959 503 575 12675 13396 589.86 590 13862
1958 493 560 13168 13956 578.13 578 14440
1957 431 490 13599 14446 505.43 505 14945
1956 479 540 14078 14986 561.72 562 15507
1955 699 800 14777 15786 819.71 820 16327
1954 472 540 15249 16326 553.51 554 16880
1953 462 520 15711 16846 541.78 542 17422
1952 95 110 15806 16956 111.41 111 17533
1951 578 660 16384 17616 677.81 678 18211
1950 676 780 17060 18396 792.73 793 19004
Total of Pm = 19004 mm
Mean of Pm = 633.5 mm
 30000
Cumulative annual rainfall at station M (mm)

Old
25000
Adjusted

20000 New
Linear (Old)
15000
Linear (New)
10000 y = 0.8779x + 917.93

5000
y = 1.0295x + 12.48
0
0 5000 10000 15000 20000
Cumulative ten station average (mm)

3.7 Frequency of Point Rainfall

In many hydraulic engineering applications such as those concerned with floods, the probability of
occurrence of a particular extreme rainfall, e.g., 24-h maximum rainfall, will be of importance. Such
information is obtained by the frequency analysis of the point-rainfall data.

The probability of occurrence of an event of a random variable (e.g. rainfall) whose magnitude is equal to
or in excess of a specified magnitude X is denoted by P. The recurrence interval (also known as return
period) is defined as

T = 1/P
This represent the average interval between the occurrence of a rainfall of magnitude equal to or greater
than X. Thus, if it is stated that the return period of rainfall of 20 cm in 24 hours is 10 years at a certain
station A, it implies that on an average rainfall magnitude equal to or greater than 20 cm in 24 hour occur
once in 10 year, i.e., in a long period of say 100 years. However, it does not mean that every 10 years one
such event is likely, i.e., periodicity is not implied. The probability of a rainfall of 20 cm in 24 hour
occurring in anyone year at station A is 1/T = 1/10 = 0.10.

If the probability of an event occurring is P, the probability of the event not occurring in a given year is

q=1–p

The binomial distribution can be used to find the probability of occurrence of the event r times in n
successive years. Thus

𝑃=,- = -𝐶= 𝑃= 𝑞-?=

𝑛!
𝑃=,- = 𝑃= 𝑞-?=
(𝑛 − 𝑟)! 𝑟!

Where: Pr, n = probability of a random hydrologic event (rainfall) of given magnitude and exceedance
probability P occurring r times in n successive years. Thus, for example,

a) The probability of an event of exceedance probability P occurring 2 times in n successive years

is

𝑛!
𝑃=,- = 𝑃= 𝑞-?=
(𝑛 − 𝑟)! 𝑟!

𝑛!
𝑃,,- = 𝑃, 𝑞-?,
(𝑛 − 2)! 2!

b) The probability of the event occurring at all in n successive years is

𝑛!
𝑃@,- = 𝑃3 𝑞-?3
(𝑛 − 0)! 0!
𝑃@,- = 𝑞- = (1 − 𝑃- )
 c) The probability of the event occurring at least once in n successive years

𝑃! = 1 − 𝑞- = 1 − (1 − 𝑃- )

Example 6. Analysis of data on maximum one-day rainfall depth at Ilocos Norte indicates that a depth of
280 mm had a return period of 50 years. Determine the probability of a one-day rainfall depth equal to
or greater than 280 mm at Ilocos Norte occurring (a) once in 20 successive years, (b) two times in 15
successive years, and (c) at least once in 20 successive years.

Solution:

1
𝑃 = = 0.02
50

Using equation

𝑛!
𝑃=,- = 𝑃= 𝑞-?=
(𝑛 − 𝑟)! 𝑟!

a) Once in 20 successive years: n = 20, r = 1

𝑛!
𝑃=,- = 𝑃= 𝑞-?=
(𝑛 − 𝑟)! 𝑟!
20!
𝑃!,,3 = (0.02)! (1 − 0.02),3?!
(20 − 1)! 1!

𝑃!,,3 = 0.272

b) Two times in 15 successive years: n = 15, r =2

𝑛!
𝑃=,- = 𝑃= 𝑞-?=
(𝑛 − 𝑟)! 𝑟!
15!
𝑃,,!8 = (0.02), (1 − 0.02)!8?,
(15 − 2)! 2!

𝑃,,!8 = 0.323

c) At least once in 20 successive years

𝑃! = 1 − 𝑞- = 1 − (1 − 𝑃- )

𝑃! = 1 − (1 − 0.02),3 )

𝑃! = 0.332
Plotting Position

The purpose of the frequency of an annual series is to obtain a relation between the magnitude of the event
and its probability of exceedance. The probability analysis may be made either by empirical or by analytical
methods.

A simple empirical technique is to arrange the given annual extreme series in descending order of magnitude
and to assign an order number m. Thus for the first entry, m = 1, for the second entry, m= 2, and so one,
till the last event for which m = N = number of years of record. The probability P of an event equaled or
exceeded is given by the Weibull formula

0
𝑃 = K"*!L

The recurrence interval, T

! "*!
𝑇 = & = 0

Other formulas use are given in the table below:

Plotting position formulae

Method P
California m/N
Hazen (m – 0.5)/N
Weibull m/(N+1)
Chegodayev (m-0.3)/(N+0.4)
Blom (m-0.44)/(N+0.12)
Gringorten (m-3/8)/(N+1/4)

Having calculated P (and hence T) for all events in the series, the variation of the rainfall magnitude is
plotted against the corresponding T on a semi-log paper or log-log paper. By suitable extrapolation on
this plot, the rainfall magnitude of specific duration for any recurrence interval can be estimated.

Return periods of annual rainfall at Station A

Example 7. The record of annual rainfall at Station A covering a period of 22 years is given below.
(a) Estimate the annual rainfall with return periods of 10 years and 50 years.
(b) What would be the probability of an annual rainfall of magnitude equal to or exceeding 100 cm
occurring at Station A?
(c) What is the 75% dependable annual rainfall at station A?

Year Annual rainfall (cm) Year Annual rainfall (cm)

1960 130 1971 90
1961 84 1972 102
1962 76 1973 108
1963 89 1974 60
1964 112 1975 75
1965 96 1976 120
1966 80 1977 160
1967 125 1978 85
1968 143 1979 106
1969 89 1980 83
1970 78 1981 95

Solution:

The data are arranged in descending order and the rank number assigned to the recorded events. The
probability of P of the event being equaled to or exceed ed is calculated by using Weibull formula. It may
be noted that when two or more events have the same magnitude (as for m=13 and 14) the probability P
is calculated for the largest m value of the set. The return period T is calculated as T= 1/P.

250

200
Annual rainfall (cm)

150

100

50

0
1 10 100
Retrun period T in years
 Return
Annual
Probability Period
Year rainfall m
P= m/(N+1) T = 1/P
(cm)
(years)
1977 160 1 0.043 23.000
1968 143 2 0.087 11.500
1960 130 3 0.130 7.667
1967 125 4 0.174 5.750
1976 120 5 0.217 4.600
1964 112 6 0.261 3.833
1973 108 7 0.304 3.286
1979 106 8 0.348 2.875
1972 102 9 0.391 2.556
1965 96 10 0.435 2.300
1981 95 11 0.478 2.091
1971 90 12 0.522 1.917
1963 89 13 0.565
1969 89 14 0.609 1.643
1978 85 15 0.652 1.533
1961 84 16 0.696 1.438
1980 83 17 0.739 1.353
1966 80 18 0.783 1.278
1970 78 19 0.826 1.211
1962 76 20 0.870 1.150
1975 75 21 0.913 1.095
1974 60 22 0.957 1.045

a. For T = 10 years, the corresponding rainfall magnitude is obtained by interpolation between two
appropriate successive values in the Table, viz, those having T = 11.5 and 7.667 years
respectively, as 137.9 cm

For T = 50 years, the corresponding rainfall magnitude, by extrapolation of the best fit straight
line, is 187 cm.

b. Return period of an annual rainfall of magnitude equal to or exceeding 100 cm, by interpolation,
is 2.4 years. As such, the exceedance probability 𝑃 = 1/2.4=0.417

c. 75% dependable annual rainfall at Station A = annual rainfall with probability P = 0.75, i.e., T =
1/0.75 = 1.333 years.
 By interpolation between two successive values in the Table having T = 1.28 and 1.35
respectively, the 75% dependable annual rainfall at Station A is 83.27 cm

83 cm - 1.353
R - 1.333
80 - 1.278

A?54 !.444?!.484
= !.484?!.,65
54?54

(!.444?!.484)
𝑅 = (!.484?!.,65) (82 − 83) + 83

𝑅 = 83.27 𝑐𝑚

3.8 Depth-Area-Duration Relationship

The total amount of rain falling at a point is the usual basic precipitation figure available. For many
purposes, however, this is not adequate and information may be required which include the intensity,
duration frequency and areal extent.

Intensity – This is a measure of the quantity of rain falling in a given time; for example, mm per hour.
Duration – This the period of time during which rain falls.
Frequency – This refers to the expectation that a given depth of rainfall will fall in a given time. Such an
amount may be equaled or exceeded in a given number of days or years.
Areal extent – This concerns the area over which a point’s rainfall can be held to apply.

Maximum Intensity – Duration Relationship

In any storm, the actual intensity as reflected by the slope of the mass curve of rainfall varies over a wide
range during the course of the rainfall. If the mass curve is considered divided into N segments of time
interval ∆t such that the total duration of the storm is

D = N ∆t

Then, the intensity of the storm for various sub-duration tj are:

Tj = (1,∆t), (2,∆t), (3,∆t), …, (j,∆t)… and (N,∆t)

could be calculated. It will be found that for each duration (say, tj), the intensity will have a maximum
value and this could be analyzed to obtain a relationship for the variation of the maximum intensity with
duration for the storm. This process is basic to the development of maximum intensity duration
frequency relationship for the station.

The procedure for analysis of mass curve of rainfall for developing maximum intensity-duration
relationship of the storm is as follows:
 1. Select a convenient time step ∆t such that duration of the storm D = N ∆t
2. For each duration (say tj = j ∆t) the mass curve of rainfall is considered to be divided into
consecutive segments of duration tj. For each segment the incremental rainfall dj in duration tj is
noted and intensity Ij = dj/tj obtained.
3. Maximum value of the intensity (Imj) for the chosen tj is noted.
4. The procedure is repeated for all values of j = 1 to N to obtain a data set of Imj as a function of
duration tj.
5. Plot the maximum intensity Imj as function of duration t.
6. It is common to express the variation of Im with t as

𝑐
𝐼0 =
(𝑡 + 𝑎)D

Where a, b, c are coefficients obtained through regression analysis.

Example 8. The mass curve of rainfall in a storm of total duration 270 minutes is given below. (a) Draw
the hyetograph of the storm at 30 minutes time step. (b) Plot the maximum intensity-duration curve for
this storm. (c) Plot the maximum depth-duration curve for the storm.

Times since start in min 0 30 60 90 120 150 180 210 240 270
Cumulative rainfall (mm) 0 6 18 21 36 43 49 52 53 54

Solution:
a. Hyetograph: The intensity of rainfall at various time duration is calculated as shown below:

Times since start in min 30 60 90 120 150 180 210 240 270
Cumulative rainfall (mm) 6 18 21 36 43 49 52 53 54
Incremental depth of rainfall in 6 12 3 15 7 6 3 1 1
the interval (mm)
Intensity (mm/h) 12 24 6 30 14 12 6 2 2

35
30
30
Rainfall Intensity (mm/h)

24
25

20

15 14
12 12

10
6 6
5
2 2
0
30 60 90 120 150 180 210 240 270
Time since start of rainf (h)
 Hyetograph of storm
Maximum
Intensity 30 22 3 15 7 6 3 1 1
(mm/h)
Duration
30 60 90 120 150 180 210 240 270
(min)
Maximum
Depth 15 22 30 37 43 49 52 53 54
(mm)

b) Various duration ∆t = 30, 60, 90, …, 240, 270 minutes are chosen. For each duration ∆t a series of
running totals of rainfall depth is obtained by starting from various points of the mass curve. This can be
done systematically as shown in the Table below. By inspection the maximum depth for each tj is identified
and corresponding maximum intensity is calculated. The data obtained from the above analysis is plotted
as maximum depth vs duration and maximum intensity vs duration.

Table. Maximum intensity-duration relation

Times Cumulative 30 60 90 120 150 180 210 240 270

since start rainfall (mm)
in min
0 0
30 6 6
60 18 12 18
90 21 3 15 21
120 36 15 18 30 36
150 43 7 22 25 37 43
180 49 6 13 28 31 43 49
210 52 3 9 16 31 34 46 52
240 53 1 4 10 17 32 35 47 53
270 54 1 2 5 11 18 33 36 48 54

Table. Maximum intensity-maximum depth-duration relation

Maximum Intensity
30.0 22.0 20.0 18.5 17.2 16.3 14.9 13.3 12.0
(mm/h)
Duration (min) 30 60 90 120 150 180 210 240 270
Maximum Depth (mm) 15 22 30 37 43 49 52 53 54
 60.0

Max depth (mm) and Max intensity (mm/h)

Max. Intensity-duration
50.0
Max. Depth-duration
40.0

30.0

20.0

10.0

0.0
30 60 90 120 150 180 210 240
Duration (min)

Maximum intensity-duration and maximum depth-duration curves for the storm

The world’s highest recorded intensities are of the order of 40 mm in a minute, 200 mm in 20 minutes
and 26 m in a year. The table below shows the World’s greatest recorded point rainfalls.

Depth
Duration Location Year
in mm
1 min 1.5 38 Guadeloupe 1970
8 min 5.0 126 Bavaria 1920
20 min 8.1 206 Romania 1889
42 min 12.0 305 Missouri, USA 1947
2h 45 min 22.0 559 Texas, USA 1935
12 h 52.8 1340 Reunion (Indian Ocean Island) 1964
24 h 73.6 1870 Reunion 1952
2 days 98.4 2500 Reunion 1952
4 days 146.5 3721 Cherrapunji, India 1974
8 days 162.6 4130 Reunion 1952
1 month 366.1 9300 Cherrapunji 1861
6 months 884.0 22454 Cherrapunji 1861
1 year 1041.8 26461 Cherrapunji 1861

Paulhus suggests that if rainfall is plotted against duration, both scales being logarithmic, the world’s
greatest recorded rainfalls lie on or just under a straight line whose equation is

𝑅 = 16.6 𝐷3.E68

where: R is rainfall (in) and D is duration (h) or

 𝑅 = 422 𝐷3.E68

where: R is in mm

100000

10000
Rainfall (mm)

1000

R = 345.68D0.4967
100 R² = 0.9974

10

1
0.01 0.1 1 10 100 1000 10000
Duration (h)

Rainfall (mm) as a function of duration (h)

Intensity-Duration-Frequency Relationship

In 1935 Bilham published his well-known article on these relationships in the United Kingdom, which
contained a graph that is reproduced here as shown in Figure below. This graph used the subjective
phraseology of “very rare”, “remarkable” and “noteworthy” rather than frequency of occurrence.
However, the frequencies were calculated from the formula

𝑛 = 1.25𝑡 (𝑟 + 0.1)3.,5,

where: n = number of occurrences in 10 years

r = depth of rain in inches
t = duration of rain in hours
 Bilham’s rainfall classification

The SI form of the formula is

𝑛 = 0.5𝑡 (𝑃 + 2.54)3.,5,

where: P = depth of rain in mm

n = number of occurrences in 10 years
t = duration of rain in hours

Depth-Area Relationship

For a rainfall of a given duration, the average depth decreases with the area in an exponential fashion given
by

𝑃" = 𝑃@ exp(−𝑘𝐴- )

where: 𝑃"= average depth in cm over and area A, km2

Po = highest amount of rainfall in cm at the storm center
K and n are constants for a give region

Value of K and n for storms of different duration (Dhar and Bhattacharya (1975)
Duration K n
1 days 0.0008526 0.6614
2 days 0.0009877 0.6306
3 days 0.001745 0.5961

Since it is very unlikely that the storm center coincides over a raingauge station, the exact determination of
Po is not possible. Hence, in the analysis of large areas storms the highest station rainfall is taken as the
average depth over an area of 25 km2.

Maximum Depth -Area-Duration Curves

Storm precipitation is analyzed with respect to time and area. The development of maximum depth-area-
duration relationship is known as DAD analysis. DAD analysis gives the maximum amounts of
precipitation within various durations over areas of various sizes.

Steps:
1. Develop the isohyets of a storm (with single major center) over a basin
 2. Plot the mean precipitation within each isohyet with respect to area

3. To obtain the maximum amount of precipitation for a certain duration, it is necessary to take as
many storms as possible with that duration and plot an enveloping curve to all the points obtained

4. Repeat the analysis for different duration storms to get family of curves.

The figure below shows a typical DAD curves. For the figure, it could be shown that:
• For the same area, as the duration increases the depth increases
• For the same duration, as the area increases the depth decreases
 Typical DAD Curves

Depth-Area-Time Relationship

Precipitation rarely occurs uniformly over an area. Variations in intensity and total depth of fall occur from
the centers to the peripheries of storms. It is useful to quantify the ratio between point and areal rainfall
over areas up to 10 km2 and for storms lasting from 2 to 120 minutes.

Holland approximated this relationship by the formula:

&F 3.4√'
&
=1− H∗

where: 𝑃" = average rain depth over the area

P = point rain depth measured at the center of the area
A = the area in km2
t* = an inverse gamma function of storm time obtained from the correlation shown in the Figure
below

Correlations between storm t and t*

Example 5. What is the average rainfall intensity over an area of 5 km2 in Ilocos Norte during a 60 minute
storm with a frequency of once in 10 years? Use the Rain intensity-frequency graph.
Solution:
From the figure, the frequency line of once in 10 years just cuts the 25 mm depth at about 1 hour, hence P
= 25 mm.

Given t = 60 minutes, then correlation between storm t and t*, t* = 5.6 so,

&F 3.4√'
=1−
& H∗

&F 3.4√8
=1− = 0.88
& 8.7

Hence, 𝑃" = 0.88 P = 0.88 (25 mm) = 22 mm in 1 hour

Rain intensity-frequency graph

Intensity-Duration-Frequency (IDF) Curves

• It is necessary to know the rainfall intensities of different durations and different return periods, in
case of many design problems such as runoff disposal erosion control, highway construction, culvert
design etc.
• The curve that shows the inter-dependency between I (cm/hr), D (hour) and T (Year) is called IDF curve
• The relation can eb expressed in general form as:

𝑘 𝑇 .
𝑖 =
(𝐷 + 𝑎)-

where: I = intensity (cm/hr)

D = duration (hours)
 k, x, a, n = area constant for a given catchment

Typical IDF Curves

• In general, the higher the intensity of the rainfall the shorter the duration of it will be.
• Intensity-duration-frequency (IDF) relationship is important for engineers in designing hydraulic
structure
• Each curve is drawn for a certain frequency, and indicates the chance of intensity with respect to the
time interval called the reference time interval, (duration of the storm).
• For the same duration, as the frequency increases the intensity increases

• For the same frequency, as the duration increases the intensity decreases