PRECIPITATION

What is Precipitation?
• Precipitation is that part of atmospheric moisture that reaches the
earth’s surface in different forms.
The essential requirement of for
precipitation to occur:
*Moisture in the atmosphere
*Presence of nuclei around which condensation vapor takes place
*Dynamic cooling responsible for condensation of water vapor
*Precipitation product must reach the ground in some form
TYPES OF PRECIPITATION
• Precipitation formation is classified according to
the factors responsible for lifting the air mass.

• --Convective
• --Orographic
• --Cyclonic
ESTIMATING MISSING
PRECIPITATION DATA
MISSING DATA
• ABSENCES OF THE OBSERVER
• INSTRUMENTAL FAILURES
METHODS:
1. STATION-AVERAGE METHOD (also ARITHMETIC MEAN
METHOD/LOCAL MEAN METHOD)
2. NORMAL-RATIO METHOD
3. QUADRANT METHOD (also DISTANCE POWER
METHOD/RECIPROCAL-DISTANCE APPROACH/INVERSE-DISTANCE
METHOD)
4. REGRESSION METHOD
1. STATION-AVERAGE/ARITHMETIC 2. NORMAL-RATIO METHOD
MEAN/LOCAL MEAN METHOD
• USED WHEN THE NORMAL • USED WHEN THE NORMAL
ANNUAL PRECIPITATION IS WITHIN ANNUAL PRECIPITATION AT ANY OF
10% OF THE GAUGE FOR WHICH THE INDEX STATION DIFFERS FROM
DATA ARE BEING RECONSTRUCTED THAT OF THE INTERPOLATION
STATION BY MORE THAN 10%
• ASSUMES EQUAL WEIGHTS FROM • WEIGHS THE EFFECT OF EACH
ALL NEARBY RAIN GAUGE SURROUNDING STATION
STATIONS
1. STATION-AVERAGE/ARITHMETIC 2. NORMAL-RATIO METHOD
MEAN/LOCAL MEAN METHOD
• •
WHERE: WHERE:
Px = PRECIPITATION AT THE MISSING LOCATION Px = PRECIPITATION AT THE MISSING LOCATION
Pi = PRECIPITATION AT INDEX STATION Pi = PRECIPITATION AT INDEX STATION
n = NUMBER OF RAIN GAUGES Nx = AVERAGE ANNUAL RAIN AT “MISSING DATA” GAUGE
Ni = AVERAGE ANNUAL RAIN AT GAUGE
n = NUMBER OF RAIN GAUGES
EXAMPLE:
GAGE > A B C X
NORMAL RAINFALL 420 410 390 410
(mm)
EVENT RAINFALL 26 31 23 ?
(mm)

SOL’N:
Px
USE
Px 26.667 mm ARITHMETIC
MEAN
EXAMPLE:
GAGE > A B C X
NORMAL RAINFALL 420 410 390 410
(mm)
EVENT RAINFALL 26 31 23 26.667
(mm)

SOL’N:
Px
USE
Px 26.667 mm ARITHMETIC
MEAN
EXAMPLE:
GAGE > A B C X
NORMAL RAINFALL 410 370 460 400
(mm)
EVENT RAINFALL 24 23 31 ?
(mm)

SOL’N:
Px
USE
Px 25.079 mm NORMAL-
RATIO
EXAMPLE:
GAGE > A B C X
NORMAL RAINFALL 410 370 460 400
(mm)
EVENT RAINFALL 24 23 31 25.079
(mm)

SOL’N:
Px
USE
Px 25.079 mm NORMAL-
RATIO
3. QUADRANT/DISTANCE POWER/RECIPROCAL-
DISTANCE/INVERSE-DISTANCE METHOD
• RAINFALL AT A STATION IS ESTIMATED AS A WEIGHTED AVERAGE OF
THE OBSERVED RAINFALL AT THE NEIGHBOURING STATIONS
• WEIGHTS FOR EACH SAMPLE ARE INVERSELY PROPORTIONATE TO ITS
DISTANCE FROM THE POINT BEING ESTIMATED
3. QUADRANT/DISTANCE POWER/RECIPROCAL-
DISTANCE/INVERSE-DISTANCE METHOD
•

WHERE:
Px = PRECIPITATION AT THE MISSING LOCATION
Pi = PRECIPITATION AT INDEX STATION
di = DISTANCE FROM EACH INDEX STATION FROM THE “MISSING DATA” GAUGE
n = NUMBER OF RAIN GAUGES
EXAMPLE:
IN A RIVER BASIN A STATION A WAS IN
OPERATIVE DURING A STORM, WHILE
STATIONS B, C, D, AND E,
SURROUNDING A WERE IN
OPERATION, RECORDING 74 mm, 88
mm, 71 mm, and 80 mm OF MONTHLY
RAINFALL. THE COORDINATES OF THE
STATION IS GIVEN IN THE FIGURE
BELOW. ESTIMATE THE MONTHLY
RAINFALL OF STATION A BY THE
DISTANCE POWER METHOD.
EXAMPLE:

GAGE > X2 Y2 D2
B 92 62 117
C 122 (-9)2 225
D (-11)2 (-6)2 157
E (-7)2 72 98
EXAMPLE:

GAGE > X2 Y2 D2 Pi
B 92 62 117 1/117 74
B 92 2 62 2 117 1/117 74
C 12 (-9) 225 1/225 88
C 122 2 (-9)22 225 1/225 88
D (-11) (-6) 157 1/157 71
D (-11) 2
(-6) 2
157 1/157 71
E (-7) 2
7 2
98 1/98 80
E (-7)2 72 98 1/98 80

SOL’N:
Px
Px 77.529 mm
4. REGRESSION METHOD
• A PROCEDURE FOR FITTING AN EQUATION TO A SET OF DATA
• PROVIDES A MEANS FOR FINDING THE VALUES OF THE COEFFICIENTS
(a and b) FOR THE STRAIGHT LINE (y = a + bx) THAT BEST FITS THE
DATA

LEAST SQUARE METHOD:

na +b =
a+b=
EXAMPLE:
THE RAINFALL DATA OF THE TABLE YEAR ANNUAL RF MONTHLY RF
1990 1100 200
BELOW ARE THE ANNUAL RAINFALL
1992 1200 220
(mm) AND MONTHLY RAINFALL
1992 1000 180
(mm) OF A METEOROLOGICAL 950 950 170
STATION FROM 1990-1995. 1994 1150 210
ESTIMATE THE MONTHLY RAINFALL 1995 1250 230
VALUE OF THE STATION FOR THE 1996 1300 X

YEAR 1996 USING THE REGRESSION

METHOD.
EXAMPLE:
YEAR ANNUAL RF (x) MONTHLY RF (y) x2 y2
1990 1100 200 1210000 220000
1992 1200 220 1440000 264000
1992 1000 180 1000000 180000
950 950 170 902500 161500
1994 1150 210 1322500 241500
1995 1250 230 1562500 287500
1996 1300 ?
SUM 6650 1210 7437500 1354500

SOL’N:
6 a + 6650 b = 1210 y = a + bx
6650 a + 7437500 b = 6650 ( 1210) y = -20 + 0.20 x
na +b =
a = -20 @ x = 1300
b = 0.2 y = 240 mm a+b=
METHODS:
1. STATION-AVERAGE METHOD (also ARITHMETIC MEAN
METHOD/LOCAL MEAN METHOD) FLAT
2. NORMAL-RATIO METHOD
3. QUADRANT METHOD (also DISTANCE POWER
METHOD/RECIPROCAL-DISTANCE APPROACH/INVERSE-DISTANCE
METHOD)
4. REGRESSION METHOD
METHOD USES
ARITHMETIC MEAN UPCOUNTRY WET ZONE
NORMAL-RATIO MID COUNTRY AND UPCOUNTRY
INTERMEDIATE ZONE STATION
DISTANCE POWER LOW COUNTRY STATIONS
REGRESSION
METHOD USES
ARITHMETIC MEAN FLAT AREAS
NORMAL-RATIO WHERE AVERAGE ANNUAL RAINFALL IS
LOW BUT HAS HIGH ANNUAL VARIABILITY
DISTANCE POWER ---
REGRESSION ---