RAINFALL ANALYSIS

Of the many variables of rainfall that might be of interest, we are concerned primarily with four:
1. Space: the average rainfall over the area
2. Intensity: how hard it rains
3. Duration: how long it rains at any given time
4. Frequency: how often it rains at any given intensity and duration
Point Precipitation Analysis
Data from a single nearby rain gage are often sufficiently representative to allow their use in the design
of small projects. The analysis of data from a single gage is called point precipitation analysis. Spatial
analysis is much more complex and is left for advanced courses.
Intensity and Duration
Intensity is a measure of how hard it rains. The units of measure are mm/h. It is computed from rainfall
records by determining the accumulation of rain for a selected time interval:

∆p = p2 – p1
∆t t2 – t1
where p1 and p2 are accumulated precipitation in mm at times t1 and t2. Because the time intervals is
typically less than an hour, the conversion factor is typically 60 min/h. The duration of precipitation is equal
to t2 –t1.
Frequency from Probability Analysis
The relative frequency of an event such as coin toss is a probability. A rainfall of a given intensity for a
duration is such event. The probability of a single rainfall event, say E 1, is defined as the relative number of
occurrences of the event in a long period of record of rainfall events. Thus, P(E 1), the probability of rainfall
event E1, is n1/N for n1 occurrences of the same event in a record of N events if N is sufficiently large. The
number of occurrences of n1 is the frequency, and n1/N is the relative frequency.
P(E1) = 0.10 implies a 10 percent chance each year that a rainfall event will occur. Because the probability
of any single, exact value of a continuous variable is zero, “occur” also means the rainfall event will be reached
or exceeded.
Hydrologists and engineers often use reciprocal of annual average probability because it has some
temporal significance. The reciprocal is called the average return period or average recurrence interval (T):

T = 1
Annual average probability

Weibull’s formula for calculating return period (Weibull, 1939):

T = n + 1
m

where T = average return period in years

n = number of years of record
m = rank of storm, with most intense storm given a rank of 1
Weibull’s formula allows for a small correction when the number of years of record is small. At larger
values of n it closely approximates T = n / m.

page 9
Rainfall record for the Dismal Swamp (1 Oct. 1954 – 30 Set. 1999)

The above Table is the compilation of a partial series of rainfall events. Rather than a record of all rainfalls,
it is a record of rainfall intensities above some practical minimum. It gives the frequency or number of times
that a rainfall of given intensity and duration will be equaled or exceeded for the period of record. For
example, looking at the first row of the Table, one would expect seven rainfall events with an intensity of
160 mm/h or more and a duration of five minutes to occur in any 45-year period (1954 – 1999) in the Dismal
Swamp.

Example:
Prepare a table of plotting points for an IDF (Intensity-Duration-Frequency) curve for a 5 – year storm at
the Dismal Swamp. Compute points for each duration given in Table above.
Given: n = 1999 – 1954 = 45 yr
T = 5 yr
Required: The points for each duration.
Figure. Refer to the Table.
Equation:
T = n + 1
m
Solution: Solving the rank of the 5-yr storm.
for the m using the Weibull’s formula:
T = n + 1
m
m = n + 1
T

page 10
= 45 + 1
 5
= 46 yr
5 yr
m = 9.2
Starting with the 5-minute duration, we note that the 9.2-ranked storm lies between the 16 th and 7th-
ranked; that is form the Table,
Intensity (mm/h)
140 0 160.0
16 9.2 7

It is noted that the ranks increase form right to left while the intensities increase from left to right.
Assuming a linear relationship between intensity and rank, we can apply interpolation by simple proportion:
16 140.0
9.2 X
7 160.0
So, 16 – 9.2 = 140.0 – X
16 – 7 140.0 – 160
6.8 = 140.0 – X
9 20.0
9 (140.0 – X) = - 20(6.8)
1,260 – 9X = - 136
- 9X = - 136 – 1,260
- 9X = - 1,396
X = - 1,396
- 9
X = 155.1 mm/h
Thus, the rank 9.2 storm has rainfall intensity of 155.1 mm/h in a duration of 5 minutes.

For the duration of 10 minutes rainfall at rank 9.2, using the Table,
15 120.0
9.2 X
7 140.0
So, 15 – 9.2 = 120.0 – X
15 – 7 120.0 – 140.0
5.8 = 120.0 – X
8 - 20.0
8(120.0 – X) = 5.8 (- 20.0)
960 – 8X = - 116
- 8X = - 116 – 960
- 8X = - 1,076
X = - 1,076
-8
X = 134.5
Thus, the rank 9.2 has rainfall intensity of 134.5 mm/h in a duration of 10 minutes.

page 10
Completed table for the IDF curve for a 5-yr storm at the Dismal Swamp:
Duration ( min) Intensity (mm/h)
5 155.1
10 134.5
15 114.7
20 82.7
30 59.5
40 39.5
50 33.1
60 ---

Figure. Intensity-duration-frequency curve for the Dismal Swamp

RUNOFF ANALYSIS
Three runoff questions are of interest:
1. How much of the precipitation that falls on a watershed reaches the stream or storm sewer draining it?
2. How long does it take for the runoff to reach the stream or storm sewer?
3. How often does the runoff cause a flood?
Stream Gages
Streamflow measurements are made by recording the height of the surface of the water above a reference
datum at a location with a fixed geometry (known as a control section). The control section allows the use of
hydraulic equations to estimate the flowrate based on the depth of water above the reference datum.
The elevation (stage) readings are calibrated in terms of streamflow (discharge). At manual stations,
readings are made from a graduated roof (staff gage) placed in the stream.
At automatic recording stations, three components are required: a device to sense the water stage, a
method of recording the stage, and a method of storing the data. One of the oldest systems that is still in use
is the float and cable system.

page 11
Figure. A float and cable system of measuring streamflow and a stilling well.
 strip chart

FLOAT

counter
weight

Estimation of Amount of Runoff

Four categories use to estimate runoff:
1. Hydrograph analysis: This is a comparison of precipitation and the resulting streamflow record.
2. Correlation with meteorological data: Statistical techniques such as cross-correlation, regression
analysis, and frequency analysis are used in this approach.
3. Correlation with hydrological data at another site: Stream gage data at one site may have transferable
value at a nearby site on the same stream. This may involve extrapolation or interpolation of information
gathered at two stations.

page 12

4. Sequential data generation: Synthetic data are generated based on a time series that includes a random
component in this technique.
Rational Method.
This method is used to determine the diameter of a storm sewer to carry the runoff from a small
watershed.
QR = kQP
where QR = runoff
k = coefficient
QP = precipitation
The coefficient must account for all other terms in the hydrologic mass balance: antecedent moisture,
ground slope, ground cover, depression storage, shape of the drainage area, and so forth. For instance, if the
rainfall on the lot continues for a long period at a constant intensity, at some time the system will reach steady
state. At steady state each drop of water that falls on the watershed conceptually displaces a drop through the
storm sewer. Thus, further rainfall at the same intensity does not increase the discharge at the storm sewer.
The time that it takes for steady state to be achieved is called the time of concentration (t C). The time of
concentration is primarily a function of the basin geometry, surface conditions, and slope
The input (QP) is a volume-per-unit time that may easily be shown to be equal to the product of the rainfall
intensity (i) and the area of the watershed (A):

QP = (i) A

The output volume per unit time (QR) is direct runoff. It is equal to the discharge. So the ratio between the
QP and QR is called runoff coefficient and can expressed as

C = Q R / QP
Substituting the value of QP = (i)A and QR = kQP, then k = QR / QP,

C = QR / (i) A
QR = C (i)A

Q = C(i)A

where Q = peak runoff rate (m3/s)

C = runoff coefficient
i = rainfall intensity (mm/h)
A = area (m2)

Table for selected runoff coefficients.

Description of area or Runoff Description of area or Runoff
character of surface coefficient character or surface coefficient
Business Railroad yard 0.20 to 0.35
Downtown 0.70 to 0.95 Unimproved 0.10 to 0.30
Neighborhood 0.50 to 0.70 Pavement
Residential Asphaltic and concrete 0.70 to 0.95
Single-family 0.30 to 0.50 Brick 0.70 to 0.85
Multi-units, detached 0.40 to 0.60 Roofs 0.75 to 0.95
Multi-units, attached 0.60 to 0.75 Lawns, sandy soil
Residential (suburban) 0.25 to 0.40 Flat, 2 percent 0.05 to 0.10
Apartment 0.50 to 0.70 Average, 2 to 7 percent 0.10 to 0.15
Industrial Steep, 7 percent 0.15 to 0.20
Light 0.50 to 0.80 Lawns, heavy soil
Heavy 0.60 to 0.90 Flat, 2 percent 0.13 to 0.17
Parks, cemeteries 0.10 to 0.25 Average, 2 to 7 percent 0. 18 to 0.22
Playgrounds 0.20 to 0.35 Steep, 7 percent 0.25 to 0.35

page 13
Example:
What is the peak discharge from the grounds of the Beauregard Long Ashby High School during a 5-year
storm? The school grounds encompasses a 16.2 ha plot that is 1.3 km east of the Dismal Swamp rain gage.
Assume that the time of concentration of the grounds is 41 minutes. The composition of the grounds is as
 follows:

Character of surface Area (m 2) Runoff coefficient

Building 10,800 0.75
Parking lot, asphaltic 11,150 0.85
Lawns, heavy soil
2.0% slope 35,000 0.17
6.0% slope 105,050 0.20
Total = 162,000

Required: The peak discharge from the grounds.

Solution: Solving the peak discharge of the grounds
for the Q:
Q = C(i)A
where
CA = [ (10,800 m2)(0.75) + (11,150 m2)(0.85) + (35,000 m2)(0.17) + (105,050 m2)(0.20) ]
= [ 8,100 m2 + 9,477.5 m2 + 5,950 m2 + 21,010 m2 ]
CA = 44,537.5 m2
because the Dismal Swamp rain gage is only 1.3 km from the school, we can use the IDF curve from
the table in a 5-year storm curve considering a 41-minute duration.
Duration (min) Intensity (mm/h)
40 39.5
41 X
50 33.1
applying interpolation,
40 ---------- 39.5
41 ---------- X
50 ---------- 33.1

40 – 41 = 39.5 – X
40 – 50 39.5 – 33.1
- 1 = 39.5 – X
- 10 6.4
- 10 (39.5 – X) = - 1(6.4)
- 395 + 10 X = - 6.4
10 X = - 6.4 + 395
X = 388.6 mm/h
10
X = 38.86 mm/h (the intensity of rainfall at Dismal Swamp in a duration of 41 min)
Q = C(i)A = CA (i)
= 44,537.5 m2 ( 38.86 mm/h)
= 44,537.5 m2 ( 38.86 mm/h x 1 m/1000 mm x 1h/3600 s)
= 44,537.5 m2 ( 38.86 m / 3,600,000 s )
= 44,537.5 m2 (0.000108 m/s)
Q = 0.48 m3/s (peak discharge runoff of the entire school grounds)

Thus, a storm sewer large enough to handle 0.48 m3/s of flow is required to carry storm water away from
the BLAHS grounds.

page 14

Hydrographs
A graphical representation of the discharge of a stream at a single gaging station is called a hydrograph.
During the period between storms the base flow is a result of exfiltration of groundwater from the banks of
the stream. Discharge from precipitation excess, that is, that which remains after abstractions, causes a hump
in the hydrograph. This hump is called the direct runoff hydrograph (DRH).

Estimation of Time of Arrival

In addition to the quantity of discharge, it is often desirable to know when the peak flow will arrive at the
watershed outlet or at some point along the discharge channel. This is particularly important when analyzing a
series of watersheds that contribute to a river or sewer at various distance downstream from the headwater.
The coincident arrival of two peaks would influence the design dramatically.
Time of Concentration
The time of concentration (tC) is the time required for direct runoff to flow from the hydraulically most
remote part of the drainage area to the watershed outlet. Thus it is represented by the formula,
tC = 1.8 (1.1 – C) √ 3.28D
3 S

where tC = time of concentration (min)

C = runoff coefficient
D = overland flow distance (m)
S = slope (%)

Example:
Estimate the tC for the Beauregard Long Ashby High School 6-percent-slope lawn. Assume that the
overland flow distance was 300.0 m. (Refer the data on the table)

Solution: Solving the tC

for the tC:
tC = 1.8 (1.1 – C) √ 3.28D
3
S
= 1.8 (1.1 – 0.20) √ 3.28(300.0 )
3
6.0
= 50.82
1.82
tC = 28 min

Storage of Reservoirs
Reservoirs are classified either by size or by use. The size of the reservoir is used to establish the degree of
safety to be incorporated into the design of dam and spillway. The use or uses of the reservoir are a basis for
evaluating the benefit-cost ratio.
Major dams (reservoir capacity greater than 6 x 107 m3) are designed to withstand the maximum probable
flood. Intermediate-sized dams (1 x 106 to 1 x 107 m3) are designed to handle the discharge from the most
severe storm considered to be reasonably characteristic of the watershed. For minor reservoirs
(less than 1 x 106 m3) the dams are designed to handle floods with return periods of 50 to 100 years.
Benefits derived from the reservoirs:
1. flood control
2. hydroelectric power
3. irrigation
4. water supply
5. navigation
6. preservation of aquatic life
7. recreation

page 15
Volume of Reservoirs.
The techniques for determining the storage volume required for a reservoir are dependent both on the size
and use of the reservoir. The simplest procedure which is satisfactory for small water-supply impoundments,
storm-water retention ponds, and wastewater equalization basins is called the mass diagram or Rippl method
(Rippl, 1883). The main disadvantage of the Rippl method is that it assumes that the sequence of events
leading to a drought or flood will be the same in the future as it was in the past.
The Rippl procedure for determining the storage volume is an application of the mass balance approach.
dS = d(Input) – d(output)
dt dt dt
dS = Qin - Qout
dt
dS = Qin(dt) – Qout(dt)

then, ∆S = (Qin)(∆t) - (Qout)(∆t)

where ∆S = change of volume storage
Table shown below is the summarized computations for the storage required to meet demand of 2.0 m 3/s of
the Wash River at Watapitae, Michigan from August 1976 to December 1978.
Month Qin (m3/s) Qin (∆t) x 106 m3 Qout ( m3/s) Qout (∆t) x 106 m3 ∆S (106 m3) ∑ (∆S) x 106 m3
1976
Aug 1.70 4.553 2.0 5.357 - 0.8035 - 0.8035
Sep 1.56 4.043 2.0 5.184 - 1.140 - 1.944
Oct 1.56 4.178 2.0 5.357 - 1.178 - 3.122
Nov 2.04 5.287 2.0 5.184 0.1036 - 3.019
Dec 2.35 6.294 2.0 5.357 0.9374 - 2.081
1977
Jan 2.89 7.741 2.0 5.357 2.384
Feb 9.57
Mar 17.7
Apr 16.4
May 6.83
Jun 3.74
Jul 1.60 4.285 2.0 5.357 - 1.071 - 1.071
Aug 1.13 3.027 2.0 5.357 - 2.330 - 3.402
Sep 1.13 2.929 2.0 5.184 - 2.255 - 5.657
Oct 1.42 3.803 2.0 5.357 - 1.553 - 7.210
Nov 1.98 5.132 2.0 5.184 - 0.052 - 7.262
Dec 2.12 5.678 2.0 5.357 0.3214 - 6.941
1978
Jan 1.78 4.768 2.0 5.357 - 0.5892 - 7.530
Feb 1.95 4.717 2.0 4.838 - 0.121 - 7.651
Mar 7.25 19.418 2.0 5.357 14.061
Apr 24.7
May 6.26
Jun 8.92
Jul 3.57
Aug 1.98 5.303 2.0 5.357 - 0.0536 - 0.0537
Sep 1.95 5.054 2.0 5.184 - 0.1296 - 0.1832
Oct 3.09 8.276 2.0 5.357 2.919
Nov 3.94
Dec 12.7

From the table above, the third column is the product of the second column and the time interval for the
given month. Example, for August (31 days) and Sep (30 days), in 1976;
(1.70 m3/s)(31 d)(86,400 s/d) = 4,553,280 m3
(1.56 m3/d)(30 d)(86,400 s/d) = 4,043,420 m3
and the fourth column is the demand given in the problem. The fifth column is the product of the demand and
time interval for the month.
page 16
(2.0 m3/s)(31 d)(86,400 s/d) = 5,356,800 m3
(2.0 m3/s)(30 d)(86,400 s/d) = 5,184,000 m3
The sixth column (∆S) is the difference between the third and fifth columns.
4,553,280 m3 – 5,356,800 m3 = - 803,520 m3
 4,043,420 m3 – 5,184,000 m3 = - 1,140,480 m3
The last (7th) column is the (∑ ∆S) sum of the last value in that column and the value in the sixth column. For
the month of August 1976, it is – 803,520 m3 since it is the first value. For September 1976, it is
(- 803,520 m3) + (- 1,140,480 m3) = - 1,944,000 m3

Based from the table, it can be interpreted that from the month of August through December 1976, the
demand exceeds the flow, and the storage must be provided. The maximum storage required for this interval
is 3.122 x 106 m3. In January 1977, the storage (∆S) exceeds the deficit (∑ ∆S) from December 1976. If the
deficit is viewed as the volume of water in a virtual reservoir with a total capacity of 3.122 x 10 6 m3, then in
December 1976, the volume of water in the reservoir is 1.041 x 10 6 m3 (3.122 x 106 m3 – 2.081 x 106 m3). The
January 1977 inflow exceeds the demand and fills the reservoir deficit of 2.081 x 10 6 m3.
Because the inflow (Qin) exceeds the demand (2.0 m3/s) for the months of February through June 1977, no
storage is required during this period. Hence there is no computations performed.
From July 1977 through February 1978, the demand exceeds inflow, and storage is required. The maximum
storage required is 7.651 x 106 m3. Note that the computations for storage did not stop in December 1977,
even though the inflow exceeded the demand. This is because the storage was not sufficient to fill the
reservoir deficit. The storage was sufficient to fill the reservoir deficit in March 1978.

Groundwater and Wells

Although the portion of the population of the United States supplied by surface water is 58 percent than
that supplied by groundwater, the number of communities supplied by groundwater is almost 12 times that
supplied by surface water. The reason for this pattern is that larger cities are supplied by surface water while
many small communities use groundwater.
Fig. (a) Percentage of the population served by drinking-water system source.

Groundwater Surface
39% water
61%

Fig. (b) Percentage of drinking-water systems by supply source.

Surface water
8%

Groundwater
92%

page 17
Groundwater has several characteristics that make it desirable as a water supply source. First, the
groundwater system provides natural storage, which eliminates the cost of impoundment works. Second,
since the supply frequently is available at the point of demand, the cost of transmission is reduced
significantly. Third, because groundwater is filtered by the natural geologic strata, groundwater is clearer to
the eye than surface water.
 Groundwater is not without its drawbacks. It dissolves naturally occurring minerals which may give the
water undesirable characteristics such as hardness, red color from iron oxides, and toxic contaminants such as
arsenic.

Construction of Wells
Modern wells consist of more than a simple hole in the ground. A steel pipe called a casing is placed in the
well to maintain the integrity of the hole. The casing is sealed to the surrounding soil with a cement grout, and
a screen is placed at the bottom of the casing to allow water in and to keep soil material out. Two types of
pump may be used. A pump motor is at the ground surface and the pump itself is placed down in the well
above the well screen. The alternative is a submersible pump. The pump and the motor are lowered into the
casing; water is pumped out of the well through a discharge pipe or drop pipe.
Figure. A Pumphouse.

Basic sanitary aspects to be considered and followed in the construction of wells:

1. The annular space outside the casing should be filled with watertight cement grout or puddled clay from the
surface to the depth necessary to prevent entry of contaminated water. A minimum of 6 m is recommended.
2. For artesian aquifers, the casing should be sealed into the overlying impermeable formations so as to retain
the artesian pressure.
3. When a water-bearing formation containing water of poor quality is penetrated, the formation should be
sealed off to prevent the infiltration of water into the well and aquifer.

page 18

4. A sanitary well-seal with an approved vent should be installed at the top of the well casing to prevent the
entrance of contaminated water or other objectionable material.

Figure. A Sanitary considerations in well construction.