College of Engineering – Civil Eng.

‫دنية‬+‫كلية الهندسة – الهندسة ا‬

CE 331 | HYDROLOGY

Lecture 5: Introduction to Hydrology and

Climate

“Precipitation: Checking Data Consistency”

Dr. Abdullah Alodah 1

Precipitation

Test for Consistency of hydrological data

-

Like any other hydrologic timeseries*, precipitation time series often exhibit spurious (nonclimatic) jumps

z
and/or gradual shifts due to changes in station location, environment (exposure), instrumentation, or
observing practices. Homogeneity means that all elements of the data series originate from a single
population.

*Timeseries: is a series of data points

indexed in time order.
~
Moving Average method (MA)
Graphical methods for detecting nonhomogeneity include moving average plots. A moving average is a
technique to get an overall idea of the trends in a data set; it is an average of any subset of numbers. The
-

moving average is extremely useful for forecasting long-term trends. You can calculate it for any period of
time. For example, if you have data for a twenty-year period, you can calculate a five-year moving average,
a four-year moving average, a three-year moving average and so on.
2
Precipitation

Test for homogeneity of hydrological data

Moving Average method (MA)

~ 650 Actual
Moving Average, w=3
Smooth
Simple Moving Average 600
Moving average, w=5a
550

The simplest form of a moving average, known as a simple 500

Q (cms)
moving average (MA), is calculated by taking the
450
arithmetic mean of a given set of values. In other words, a
set of numbers–or prices in the case of financial 400
instruments–are added together and then divided by the 350
number of prices in the set. The formula for calculating
300
the simple moving average of a security is as follows:

1941
1943
1945
1947
1949
1951
1953
1955
1957
1959
1961
1963
1965
1967
1969
1971
1973
1975
Year

where Ai is the data point in the ith period.

3
Precipitation

Test for homogeneity of hydrological data

Moving Average method (MA)

~
Example 1. Find the moving average using a window of 3 and 5.

#
Year PCP MA (3) MA (5)
2003 75 T

2004 81
2005 64 - 73 .
2

2006 91
2007 55
2008 61
2009 72
2010 59
2011 73
4
2012 81
Precipitation

Test for homogeneity of hydrological data

~
Moving Average method (MA)

Example 1. Find the moving average using a window of 3 and 5.

100

Year PCP MA (3) MA (5)

90

2003 75
80
2004 81 73.3
2005 64 78.7 73.2 70

PCP (mm)
2006 91 70 70.4 60

2007 55 69 68.6
50
2008 61 62.7 67.6
2009 72 64 64 40

2010 59 68 69.2 30
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
2011 73 71 Year

2012 81 PCP M(3) M(5)

5
Precipitation
Test for homogeneity of hydrological data

Moving Average method (MA) in Excel

Excel: Steps
Step 1: Click the “Data” tab and then click “Data Analysis.”
Step 2: Click “Moving average” and then click “OK.”
Step 3: Click the “Input Range” box and then select your data. If you include column headers,
make sure you check the Labels in first Row box.
Step 4:Type an interval into the box. An interval is how many prior points you want Excel to
use to calculate the moving average. For example, “5” would use the previous 5 data points to
calculate the average for each subsequent point. The lower the interval, the closer your
moving average is to your original data set.
Step 5: Click in the “Output Range” box and select an area on the worksheet where you
want the result to appear. Or click the “New worksheet” radio button.
Step 6: Check the “Chart Output” box if you want to see a chart of your data set (if you
forget to do this, you can always go back and add it or choose a chart from the “Insert” tab.”
Step 7: Press “OK.” Excel will return the results in the area you specified in Step 6.
6
Precipitation

Test for Consistency of hydrological data ~

-
Inconsistency in records may occur for many reasons such as:
• Shifting of rain gauge to a new location
-

• Change in the vicinity of station, e.g. Construction, Landslide, frost, Flood, Earthquake, etc.
- - -

• Occurrence of observational error from a certain date

-

• Observational procedure might have changed

- -
Many studies (and project designs) require long term rainfall data; therefore, a test must be conducted to
check homogeneity or self consistency of the rainfall record. The inconsistency of rainfall record can be
checked by- graphical or statistical methods including double mass curve, the von Neumann ratio test,
cumulative deviation, run test. Double mass curve method described here is one of the most common
and widely accepted methods for checking the consistency of rainfall record.

The assumption is that the mean accumulated precipitation for a large group of stations is not
significantly affected by a change or changes in individual stations (i.e., reliable data set). If we
plot the mean accumulated precipitation for several stations against the accumulated precipitation of the
record for the station that needs to be adjusted, any change in slope will indicate a “break” in the station
record. Such break needs to be justified or corrected. 7
Precipitation

Test for Consistency of hydrological data

Suppose we want to check the consistency of timeseries S(y) with S(x):

Fig. (a) reflects that series S(y) has a pattern consistent with that of series S(x) while Fig. (b) reflects
the inconsistency in the data. Referring to Fig. (b) the point where the slope changes from s1 to s2
marks the time where the inconsistency occurs. Therefore, part of the record of y must be adjusted.
Which part of the record must be adjusted (before the break point or after) depends on the
particular case. For example, if the inconsistency in y has occurred because the gage was moved from
the original location to another location (the current location), then it is logical to adjust the record
before the break point.
(a) (b)

8
 Precipitation

Test for Consistency of Record: # Break

&
Double Mass curve technique

r
te
af
!!
The double mass curve is used to check the consistency of
many kinds of hydrologic data by comparing data for a single
station (X) with that of a pattern composed of the data from o re onlay
f
several other stations in the area. The double-mass be
curve can be used to adjust inconsistent precipitation data. !"

slope
: **
+

• If the slope of the period before break is (!! ) and of the slope of the period after break is (!" ), the
adjustment of the inconsistent data is made by the ration of the slope of two line-segments.

• The data adjusted to reflect the conditions that existed prior (before) to the indicated break. This is
done by multiplying each recent precipitation value (after break point) of station X (being tested) by
the ratio !! /!" .
9
Precipitation

Test for Consistency of hydrological data

Double Mass curve technique

Ø The double mass curve obtained by plotting the accumulated precipitation at the station in
question along Y-axis and the average accumulated precipitation of a number of other
nearby stations which are situated under the same meteorological conditions along X-axis.

o If the curve has a constant slope, the record of station “X” is consistent.
o However, if there is any break in the slope of the curve, the record of the station is

X
inconsistent and has to be adjusted by the formula:
!" (-%.#$% -$%!/)
""#$$%"&%' = "#$()(*!+ >> All values after break are to be adjusted.
!! (!.&%$ -$%!/)
where .

r
""#$$%"&%' = Adjusted precipitation

te
af
"#$()(*!+ = Observed (original) precipitation !!
!" (-%.#$% -$%!/) = Slope prior (before) to the break in the curve

&
fo re
!! (-%.#$% -$%!/) = Slope after the break in the curve be
!"
10
 i
Precipitation

Test for Consistency of hydrological data

↑

Double Mass curve in detailed steps:

1. Select a group of stations (5 to10 stations) in the neighborhood of the problem station "
2. Take the average of the neighboring stations (#avg)
3. Arrange the data of station " rainfall (#1) and the average of the neighboring stations [the most-
trusted data are first; usually, but not always, in reverse chronological order from recent to old
record].
4. Accumulate the precipitation of station " (Σ#1) and the average values of the group base stations
(Σ#avg) starting from the most-trusted record.
5. The double mass curve is obtained by plotting the (%#1) in the y-axis against (%#avg) in the x-axis
6. A decided break in the slope of the resulting plot is observed that indicates a change in precipitation
regime of station ", i.e., inconsistency in data (a change in slope is taken as significant only if it
persists for more than 5 years).
7. Therefore, apply correction to the points lying after the breaking point (away from the straight line)
by the factor:
!" (-%.#$% -$%!/)
""#$$%"&%' = "#$()(*!+
!! (!.&%$ -$%!/) 11
‫‪Precipitation‬‬ ‫عشان تشوف إذا فيه بريك او ال … مثال فيه عشر نقاط ‪،‬‬
‫أخذ اول خمس نقاط وارسم خط مستقيم باملسطرة ثم‬
‫راح يظهر لك البريك‬
‫‪Different ways to find the correction factor:‬‬
‫‪Remember:‬‬

‫‪>> Also remember:‬‬ ‫‪y= m x + constant‬‬ ‫‪12‬‬

 2! (#$%&'$ #'$())
Precipitation Remember: ""#$$%"&%' = "#$()(*!+
i
2( ((%+$' #'$())

Test for Consistency of Record: Double Mass curve technique

Example 2. Check for Consistency of data at station X using Double Mass curve method, given that the recent
data are more accurate.

13
 2! (#$%&'$ #'$())
Precipitation Remember: ""#$$%"&%' = "#$()(*!+
2( ((%+$' #'$())

Test for Consistency of Record: Double Mass curve technique

Example 2.

-
e
&

j Need correction 14
 2!(#$%&'$ #'$())
Precipitation Remember: ""#$$%"&%' = "#$()(*!+
2( ((%+$' #'$())

، ‫عشان تشوف إذا فيه بريك او ال … مثال فيه عشر نقاط‬

Test for Consistency of Record: Double Mass curve technique ‫أخذ اول خمس نقاط وارسم خط مستقيم باملسطرة ثم راح‬
‫يظهر لك البريك‬

·
Example 2. Check for Consistency of data at station x using Double Mass curve method, given that the recent
data are more accurate.
Add this line to check
data Consistency
Now, calculate slopes before and after deviation
by taking two points for each slope:

!" = (3410 – 1656) / (4064 – 2045)

!" = 0.868

!! = (9024 – 6923 ) / (9415 – 7665)

!! = 1.2

The correction factor = !" / Ma

= 0.868 / 1.2 = 0.72

To correct data, we multiply data after the

break by the correction factor (!" / !! ). 15
==> applicable only to data before 1950
 2! (#$%&'$ #'$())
Precipitation Remember: ""#$$%"&%' = "#$()(*!+
2( ((%+$' #'$())

Test for Consistency of Record: Double Mass curve technique

Example 2. Check for Consistency of data at station x using Double Mass curve method, given that the recent
data are more accurate.
Original Corrected
Annual Annual
Year Rainfall at Rainfall at
station X station X
(mm) (mm)

1949 284 204.48

1948 493 354.96
1947 320 230.4
1946 274 197.28
1945 322 231.84
1944 437 314.64
1943 389 280.08
1942 305 219.6
1941 320 230.4
1940 328 236.16
1939 308 221.76
1938 302 217.44 16
1937 414 298.08
Precipitation
#
Test for Consistency of Record: Double Mass curve technique
Average Annual precipitation (cm) of 25
Year Annual precipitation (cm) at Station X
surrounding stations
1981 163 135
Example 3. Check for Consistency of data at station x using Double 1982 119 111
1983 121 124
Mass curve method, given that the recent data are more accurate. 1984 129 111
1985 126 123
1986 120 90
1987 153 138
1988 172 119
1989 127 108
1990 108 107
1991 126 111
1992 190 142

Excel Application 1993

1994
112
97
112
99
1995 86 93
1996 111 131
1997 68 92
XL file on Blackboard 1998 88 142
1999 112 123
2000 95 142
2001 106 92
2002 81 91
2003 116 131
2004 112 104
2005 80 97
2006 88 111
2! (#$%&'$ #'$())
Remember: ""#$$%"&%' = "#$()(*!+
2007 85 114
2008 90 92
2( ((%+$' #'$()) 2009 120 146
2010 72 93
2011 113 138
2012 82 112
2013 116 117
2014 122 152
2015 73 90
2016 74 104

17
Precipitation

Test for Consistency of Record: Double Mass curve technique

Example 3. Check for Consistency of data at station x using Double Mass curve method

Solution:

Double Mass curve method:

1. Arrange in a Descending order (from the latest years to the recent ones) Why? b/c
the most recent data are more reliable (we ”usually” trust them more).
2. Calculate the cumulative annual PCP for station x and the nearby stations
3. Plot the double Mass curve
4. Check for a break in the slope of this plot (which indicates a change in the
precipitation) using a straight line
5. If a break exists, Precipitation values at X beyond the period of change of regime is
corrected as

!" (-%.#$% -$%!/)

""#$$%"&%' = "#$()(*!+
!! (!.&%$ -$%!/)
18
ANNOUNCEMENTS • Quiz 1 is COMING
Material: up to the end of ……….

19