LECTURE 5

PRECIPITATION
GAUGE NETWORK
  To have an idea of areal distribution of precipitation there should
PRECIPITATION be adequately planned rain gauge network

GAUGE  There are two reasons to plan a network:

1. Catch area of a precipitation gauge is very small
NETWORK 2. Amount of precipitation varies from place to place
 PRECIPITATIO
N GAUGE
NETWORK IN
PAKISTAN
AREA=796095km2
No of Gauges = 56
Gauge Density = 14215
km2/gauge

(Faisal, Ghafar, 2012)

CURRENT MET OBSERVATORIES IN PAKISTAN:=97-with GD of 8,000 km 2/Gauge

http://www.pmd.gov.pk/Observatories/
  On average basis the area served by one rain gauge is called gauge
density.
G.D =
GAUGE
DENSITY  Lesser value of G.D means more no. of rain gauges in the area
which are closely spaced and vice-versa
(G.D)  World Meteorological Organization (W.M.O) has recommended
certain gauge densities
 1. For Flat regions 600-900 Km2 /station
W.M.O 2. For mountainous regions 100-250 Km2 /station
RECOMMENDED 3. For small mountainous islands 25 Km2 /station
GAUGE DENSITIES 4. For arid and polar zones 1500-10,000 Km2 /station
  Topography
 For hilly terrains G.D should be lesser because generally more rains and
more variations
 Hydro-meteorology
 Areas with more precipitation should have less G.D

FACTORS  Stream network

 Denser stream network lesser will be G.D
AFFECTING  Population density
GAUGE  More population lesser should be G.D

DENSITY  Economy
 Lesser G.D.P more G.D

 Interest
 Climatological station is must on airports and water control reservoirs

 Research
 Lesser G.D results in better output
  Interpretation of rain data coming from a rain gauge network
INTERPRETATION includes
OF PRECIPITATION 1. Estimation of missing precipitation data
DATA FROM A 2. Checking consistency in precipitation data record

GAUGE NETWORK 3. Determination of average areal precipitation

  According to US Environmental Data Services, for estimation of
missing precipitation record, there should be at least three
surrounding stations (Index Stations)
 Index stations should be located more or less at same effective
1.ESTIMATION radii from station “X” (station for which precipitation data is
OF MISSING missing)
 Index stations should be as close to the station, with missing
PRECIPITATION record, as possible.
DATA  Index stations should be as evenly spaced around the station
with missing record, as possible.

C
A
X

B
  There are two methods for estimation of missing precipitation
based on Normal Annual Precipitation (N.A.P.)
1.ESTIMATION 1. Arithmetic Mean method

OF MISSING 2. Normal Ratio method

Normal Annual Precipitation : mean of annual precipitation of 30
PRECIPITATION years record is called Normal Annual Precipitation.
DATA
NAP=
C
A
X

B
  When NAP of the index stations are within 10% NAP
of the station with missing data, then:
Missing data can be estimated by simple Arithmetic Mean
a) ARITHMETIC of the precipitation of that day or the year at index stations
MEAN C

METHOD Px= A
X
Px : Missing precipitation at station X
B
PA , PB ,PC =Precipitations at index stations at same time
  When normal annual precipitation (NAP) of index stations differs
more than 10% of NAP of the stations with missing data then
missing precipitation is assessed by NORMAL RATIO METHOD.
 In this method amount of precipitation at index stations is weighted
by the ratios of normal annual precipitation values, i.e.

b)NORMAL
RATIO Px = P A + P B + PC ] C

METHOD Px = + + ] A
X

B
Px : Missing precipitation at station X
PA , PB ,PC : Precipitations at index stations at same time
NX : normal annual precipitation at station X
NA , NB , NC : normal annual precipitations at index stations
  Precipitation station X was inoperative on 24 august, 2010. The
NUMERICAL precipitation records for the same date at three surrounding
stations A,B and C were 82,87 and 97mm respectively. Normal
PROBLEM Annual Precipitations at station X, A,B and C are 920,800,810 and
950mm respectively
  Solution:
Which method is to be used? Based on NAP
10% of NAP at Station X (NX) = 92
NX +10% = 1012mm and NX -10% = 828mm
For arithmetic method NA , NB and NC should lie between (1012-828)
NUMERICAL Which is untrue for this case.
PROBLEM Apply Normal Ratio Method and determine missing precipitation for
the date mentioned

Px = + + ]
PX = 95.68 mm
  Double mass analysis tests the consistency of the record at a
station by comparing its accumulated annual or seasonal
precipitation with concurrent accumulated values of mean
precipitation for a group of surrounding stations
2. CONSISTENCY
OF PRECIPITATION  A change in the slope indicates a change in precipitation regime at
RECORD station.
 Any change due to meteorological reasons will not cause change
in slope because all the base stations and the station whose
consistency is in question will be affected equally
DOUBLE MASS
CURVE
  If change in slope occurs the year of change is determined using
double mass curve and station history is seen for any important
event in that year.
DOUBLE MASS  Conditions before and after change year are assessed and data is
adjusted according to reliable data.
CURVE  Adjustment is made by using ratio of the slopes of the lines before
and after change
Adjustment factor =k=
  Number of surrounding stations should be at least 10.
 If surrounding stations are less than 10, then inconsistency of each
station should be checked, and any inconsistent station should not
be used for checking consistency of the station.
 Considerable caution must be exercised in applying this technique.
DOUBLE MASS Because the plotted points deviate about a mean line and changes
in slope should be only accepted if there is a marked difference
CURVE substantiated by other evidences too.
 If the slope change remains for a period less than 5 years this
change is ignored. Data is considered consistent
 Depending upon conditions any of the data before or after the
change year may be reliable but generally latest data is taken as
reliable.
  Change in the location e.g., gauge near a river site
CAUSES OF  Change in the instrument e.g., modern, better performance
INCONSISTENCY  Change in observational procedures
  Average depth of precipitation over a specific area on a storm,
seasonal or annual basis is required in many types of hydrologic
problems.
 This average depth when multiplied with the area under
3. AVERAGING consideration provides total volume of water generated in a storm,
PRECIPITATION or seasonally or annually.
 There are mainly three methods to compute average depth
OVER AN AREA 1. Arithmetic average
2. Thissen’s method
3. Isohyetal method
  Simplest method of obtaining average depth over an area is
arithmetic mean of the gauges present in the area
 This technique yields good estimate where
 the gauges are more or less uniformly spaced OR
 individual gauge readings (catches) do not vary much from the mean
a) ARITHMETIC value.

AVERAGE  This technique yields good estimate in flat countries where

 the gauges are more or less uniformly spaced And individual gauge
METHOD catches do not vary much from the mean value.
 These limitations can be partially overcome if topographical
influences and areal representation are considered while installing
gauges
  Applicability:
1. Uniform precipitation almost at all stations
2. Stations are equally spaced in the area
3. For very small and plane areas results may be satisfactory
 Merits:
a) ARITHMETIC  This is a quick and simple method

AVERAGE  Demerits:
METHOD  When there is variation in precipitation almost 10% from the mean
value outcome is not very true. Also not applicable to areas with
unequal spacing between stations. These are generally most
encountered conditions, so arithmetic average method is not used
quite frequently.
  This method attempts to allow for non-uniform distribution of
gauges by providing a weighing factor for each gauge
 Stations are plotted on a scale like a map and connecting lines are
drawn
 Perpendicular bisectors are marked and by their intersection
b) THE polygons are made. Each polygon is the assumed effective area
for that gauge.
THISSEN’S  Effective Area for each gauge is determined by Planimetry (or
METHOD square measurement) and is presented as percentage of total area.
 Weighted average rainfall for the total area is computed by:
 multiplying the precipitation over each gauge and its effective area
 Then summing the product of areas and precipitations and dividing
the sum by total area
b) THE
THISSEN’S
METHOD
Pakistan average ppt based on
1960-2010 data) is 297mm
(Faisal, Ghafar, 2012)

Ref: Faisal, N.1, 2, A. Gaffar2 (2012), Development of Pakistan’s New Area Weighted Rainfall Using Thiessen Polygon Method,
Technical Note In Pakistan Journal of Meteorology ,Vol. 9, Issue 17: July 2012 , pp 107 -116.
  Applicability:
 Applicable for large and variable precipitation areas.
Effective areas are computed to reduce variations
which was a hindrance in calculations of simple
arithmetic method.
 Merits:
b) THE  Applicable to large and variable areas
 Better results as compared to arithmetic mean method
THISSEN’S  Demerits:
METHOD  This is a time taking and hectic technique
 Changes are rigid and abrupt in precipitation values
for any polygon which is un-natural
 And there is a requirement for replotting the map and
do all the work again if there is a change in the
location of a gauge or there is a new instrument
installed
  This is the most accurate method when used by
experienced analyst
 Station locations and amounts of precipitations are
plotted on the map, on a certain scale
c) THE  Lines of equal precipitation (isohyets) are drawn,
ISOHYETAL by interpolating the ppt in-between the gauges
METHOD  Average is computed by weighing the average
precipitation between successive isohyets (usually
taken as average of the two Isohyetal values)
 And as done in Thissen's methods the sum of these
products is divided by total area
  Applicability:
 Applicable to large areas with non uniform precipitation
patterns
 Merits
c) THE  Applicable to areas with non-uniform precipitation
behavior and unequal spacing between stations which
ISOHYETAL is closer to natural conditions.
 Most accurate method if analyst is expert because this
METHOD may involve the personal judgment of the analyst
 Demerits
 This is a hectic and time consuming method
 It require a total change in the map and plotting if there
is any change in the gauges in that area