HP1 2015

HYDROLOGICAL PROCEDURE NO.
1
(REVISED AND UPDATED 2015)
Estimation of Design Rainstorm in Peninsular

Malaysia
DEPARTMENT OF IRRIGATION AND DRAINAGE

HYDROLOGICAL PROCEDURE NO. 1
(REVISED AND UPDATED 2015)
ESTIMATION OF DESIGN RAINSTORM IN PENINSULAR

MALAYSIA
2015
Water Resources Management and Hydrology Division

Department of irrigation and Drainage
Ministry of Natural Environment Malaysia
Hydrological Procedure No. 1 (Revised and Updated 2015)
DISCLAIMER
Although every effort and care has been taken in selecting the methods and proposing the
recommendations that are appropriate to Malaysian condition, the user is wholly responsible
to make use of this hydrological procedure. The use of this hydrological procedure requires
professional interpretation and judgment to suit the particular circumstances under
consideration.
The department or government shall have no liability or responsibility to the user or any
other person or entity with respect to any liability, loss or damage caused or alleged to be
caused, directly or indirectly, by the adaptation and use of the methods and
recommendations of this publication, including but not limited to, any interruption of service,
loss of business or anticipatory profits or consequential damages resulting from the use of
this publication.
Copyright ©2015 by Department of Irrigation and Drainage (DID) Malaysia

Kuala Lumpur, Malaysia
All rights reserved. No part of this manual may be reproduced, in any form or by any means,
without permission in writing from the publisher.
i
ACKNOWLEDGMENTS
The Water Resources Management and Hydrology Division of the Department of Irrigation
and Drainage (DID), Ministry of Natural Environment, Malaysia would like to express sincere
appreciation to the National Hydraulic Research Institute (NAHRIM) especially to Ir. Mohd.
Zaki bin Mohd. Amin who had prepared and completed the study on the reviewed and
updated Hydrological Procedure No. 1 ‘ Estimation of Designstorm in Peninsular Malaysia’ in
December 2010. The appreciation also goes to ZHL Engineers Sdn Bhd, Ir Hj Zulkefli bin Hj
Mustaffa, Ir. Hj Azmi bin Md Jafri, Ir. Chong Sun Fat, Fairus binti Ahmad, Adnan Bin Ab.
Latif, Ahmad Shahrir Bin Md. Naziri, Dr. Norlida binti Mohd Dom, Hjh. Zainab binti Hashim,
Intan Shafilah Binti Abdul Azia, and Amirah Hanim Binti Mohd Puad who had carried out and
completed additional study on the revision on the areal reduction factors (ARFs) and the
inclusion of the consideration of climate change scenario to the design rainstorm recently.
Hence this Hydrological Procedure is the result of this earlier study by NAHRIM and a recent
study by ZHL Engineers Sdn Bhd. Valuable contribution and feedbacks from DID personnel
especially to the Director of Water Resources Management and Hydrology, Dato’ Ir. Hj.
Hanapi bin Mohd Noor, and his staff namely Hj Juhaimi bin Jusoh, Ir. Gapar bin Asan,
Mohd. Hazri bin Moh Khambali, Hazalizah binti Hamzah, Sazali bin Osman, Rosli bin Aman,
Nor Asiken binti Alias, Ernie Haryana binti Hamzah and Md. Ezaire Bin Md. Eusofe are also
greatly acknowledged.
ii
SUMMARY
This procedure contains estimation of Areal Rainfall Factor (ARF) and Climate Change
Factor (CCF) and it is a revised and updated version for Hydrological Procedure No 1
entitled ‘Estimation of the Design Rainstorm in Peninsular Malaysia ’, which is published in
2010 by National Hydraulic Research Institute of Malaysia (NAHRIM). The updates are
based on more recent and extended data sets, and currently accepted approaches.
The new procedure updates ARF estimates for 1, 3, 6, 12 and 24 hours for 50, 100, 150 and
200 km2 catchment areas. The estimates are based on recent data and using the fixed area
method by United States Weather Bureau (USWB). For CCF, the value produced by
NAHRIM is a quite conservative due to some uncertainties hence an additional review on
CCF is deemed necessary. The reviewed and updated CCF values are based on clustering
analysis approach by dividing the rainfall stations in five (5) regions.
Tables, graphs and formulas are prepared to facilitate the estimation of the design rainstorm
in Peninsular Malaysia.
iii
T ABLE OF CONTENTS
1 INTRODUCTION 1
1.1 Background and General Review 1
1.2 Objective 3
1.3 Scope of Revision and Update 3
1.4 Concerned Issues and Statements in the Proposed Revision and Update 3
1.4.1 Reviews on the Choice of Frequency Distribution 3
1.4.2 Short Duration Analysis 4
1.4.3 Formulation of Regional IDF Relationship for Gauged and
Ungauged Sites
4
2 ORGANIZATION OF TASK 6
2.1 Brief Overview of the Task 6
2.2 Objectives of the Designated Tasks 7
2.2.1 T1: Task 1 – Data Mining Assembly 7
2.2.2 T2: Task 2 – Choice of Rainfall Frequency Models 7
2.2.3 T3: Task 3 – Choice of Distribution to be used in the Chosen
Model PD/POT) 7
2.2.4 T4: Task 4 – Method of Parameter Estimation 8
2.2.5 T5: Task 5 – Estimation of Design Storm for Low and High Return
Period 8
2.2.6 T6: Task 6 – Construction and Formulation of at-site IDF Curve 8
2.2.7 T7: Task 7 – Design Storm Profile (Temporal Pattern) 9
2.2.8 T8: Task 8 – Areal Reduction Factor 10
3 APPROACH AND METHODOLOGY 16

3.1 Data Mining and Assembly 16
3.2 Choice of Rainfall Frequency Model 16
3.3 Choice of Distribution to be Used in the Chosen Model (AM or PD/POT) 16
3.3.1 Candidates of the AM model – the Generalized Extreme Value
Distribution (GEV) 16
3.3.2 Candidate Distribution of the PD/POT Model – the Generalized
Pareto Distribution (GPA) 18
3.4 Methods of Parameter Estimation Using L-Moments 19
3.5 Estimation of Design Storm/Rainfall Intensity of Low and High Return
Period 20
iv
3.6 Construction and Formulation of At-Site IDF Curve 21

3.6.1 An Overview on the Mathematical Expression of an IDF relationship 21
3.6.2 One-Step Least Square Method of the IDF Relationships 24
4 ASSESSMENT OF THE PROPOSED METHODOLOGY 25

4.1 Assessment Procedure 25
4.2 Choice of Rainfall Frequency Model 25
4.3 Robustness Study and Efficiency Procedure 26
4.4 Results of the Assessment for the Choice of Rainfall Model, Parent
Distribution and Parameter Estimation 29
5 DEVELOPING THE INTENSITY-DURATION FREQUENCY (IDF) RELATIONSHIP 31

– GAUGED SITES
5.1 Choice of Mathematical Formulation for IDF Relationship 31
5.2 Comparison of New Polynomial Equation and MSMA (2000) 31
5.3 Comparison of New Polynomial and Empirical Equation 33
6 DEVELOPING THE INTENSITY-DURATION FREQUENCY (IDF) RELATIONSHIP 35

– UNGAUGED SITES
6.1 Brief Description 35
6.2 Mathematical Formulation for the IDF Relationship of Ungauged Sites 36
6.3 Summary of Findings 36
7 DEVELOPING THE REGION OF TEMPORAL STORM PROFILES BY MEANS

OF CLUSTERING ANALYSIS 44
7.1 Introduction 44
7.2 Data Availability and Acquisition 45
7.3 Data Screening 45
7.4 Formation of Region by Clustering Analysis 45
7.5 Results of Clustering Analysis 53
8 DEVELOPING THE DESIGN STORM PROFILES (TEMPORAL STORM) 55

8.1 Introduction 55
8.2 Derivation of Storm Profiles (Temporal Pattern) 56
8.3 Summary of Results 57
v
9 DEVELOPING THE AREAL REDUCTION FACTOR 68

9.1 Introduction 68
9.2 Derivation Procedure of Areal Reduction Factor (ARF) 69
9.3 Summary of Results 70
10 ESTIMATION OF DESIGN STORM UNDER CLIMATE CHANGE SCENARIO 73

10.1 Effects of Climate Change on Design Storms 73
10.2 Review of Climate Change Factors by NAHRIM 74
11 WORKED EXAMPLE 83
12 APPENDIX 1 - Isopleths Map of IDF Parameter 87

vi
LIST OF TABLES
Table 5.1: Polynomial Equation Parameters of Site 3117070 32

Table 5.2: Rainfall Intensity of Site 3117070 32
Table 5.3: Design Rainfall Intensity for Site 3117070 at D ID Ampang 33
Table 7.1: Summary of selected 56 automatic rainfall stations for Peninsular
Malaysia 47
Table 7.2: Available Data of Site Characteristics 49
Table 7.3: Transformation of Site Characteristics 50
Table 7.4: Site Characteristics Combinations of Cluster Analysis 50
Table 7.5: Summary of Clustering Analysis of A5 Combination 51
Table 7.6: Summary of Cluster Centres of A5 Combination 51
Table 8.1: Derived Temporal Pattern for Region 1 – Terengganu, Kelantan and
Northern Pahang 55
Table 8.2: Derived Temporal Pattern for Region 2 - Johor, Negeri Sembilan,
Melaka, Selangor and Pahang 56
Table 8.3: Derived Temporal for Region 3 - Perak, Kedah, P Pinang and Perlis 57
Table 8.4: Derived Temporal Pattern for Region 4 - Mountainous Area 58
Table 8.5: Derived Temporal Pattern for Region 5 - Urban Area (Kuala Lumpur) 59
Table 8.6: Normalized Temporal Pattern For Region 1 - Terengganu & Kelantan 60
Table 8.7: Normalized Temporal Pattern for Region 2 - Johor, Negeri Sembilan,
Melaka, Selangor and Pahang 61
Table 8.8: Normalized Temporal Pattern for Region 3 - Perak, Kedah, P Pinang &
Perlis 62
Table 8.9: Normalized Temporal Pattern for Region 4 - Mountainous Area 63
Table 8.10: Normalized Temporal Pattern for Region 5 - Urban Area (Kuala
Lumpur) 64
Table 9.1: The ARFs Estimated for Kuala Lumpur 69
Table 9.2: The ARFs Estimated for Kelantan 69
Table 10.1: Summary of Minimum, Mean, Median and Maximum Values of Climate
Change Factors for Region 1 (Kelantan, Terengganu and Northern
Pahang) 76
Change Factors for Region 2 (Pahang, Johor, Melaka, N.Sembilan and
Selangor) 78
Change Factors for Region 3 (Perak, Kedah, Pulau Pinang and Perlis) 80
vii
Change Factors for Region 4 (Mountainous). 81
Change Factors for Region 5 (Wilayah Persekutuan) 82
viii
LIST OF FIGURES
Figure ‎2:1: F low chart of the designated tasks for the review and update process of
Hydrological Procedure No.1 (HP1) 6
Figure ‎4:1: Flow C hart in Assessment Procedure of the Proposed Methodology 27
Figure ‎5:1: Site 3117070 IDF curve fitted by Polynomial Equation 34
Figure ‎5:2: Site 3117070 IDF curve fitted by Empirical Equation 34
Figure ‎7:1: The “region” created by means of the clustering analysis approach 52
Figure ‎8:1: Block diagrams of temporal storm profile corresponding with storm
duration (0.25 to 12-hr) for Region 1 65
Figure ‎8:2: Block diagrams of temporal storm profile corresponding with storm
duration (24, 48 and 72-hr) for Region 1 66
Figure 9.1: Derivation ARFs Steps 68
Figure 9.2: The Relationship Graph of ARF values derived and rainfall duration for
Kuala Lumpur 70
Figure 9.3: The Relationship Graph of ARF values derived and rainfall duration for
Kelantan 71
Figure 10.1: Expected Changes in Project Design Return Periods due Climate Change 72
Figure 10.2: Boundary of East Coast and West Coast Rainfall Stations for Peninsular Malaysia 74
Figure 10.3: The Region Created using Clustering Approach 75
Figure 12.1: IDF Parameter f 86
Figure 12.2: IDF Parameter of  87
Figure 12.3: IDF Parameter of  88
Figure 12.4: IDF Parameter of  89
ix
CHAPTER 1
INTRODUCTION
1.1 BACKGROUND AND GENERAL REVIEW
First edition of HP1 by Heiler (1973) was developed using 80 rainfall stations with
available record length up to 1970. Second edition of HP1 authored b y Mahmood, et al.,
(D ID, 1982), on the other hand, use approximately 210 rainfall stations with data recorded
to year 1979/80. It was affirmed that only 4 rainfall stations has data recorded for more
than 20 years, 59 rainfall Stations have less than 10 years and the remaining ranging
from 10 to 20 years.
Due to restrictions o f records length, the estimation of design rainstorm/rainfall intensity is

only able to give an estimation utmost to 50 years return period. Adversely, there is a
common practice to use 100 - years return period as a level of protection for designing a
major water resources or hydraulic structure in Malaysia. As for the methodology adopted,
the reviewed and updated HP1 (1982) was still using similar methodology as per first
edition (1972). This is purportedly acceptable while the annual maximum series of
rainfall was considered as a model of the data series in frequency y analysis. The Gumbel
distribution maintained as the frequency distribution type and Gumbel paper has been
used to estimate the 2-parameters Gumbel distribution. Cunnane (1989) expressed that
the error of estimate increases with return period (T), population Coefficient of Variation
(Cv) and Coefficient of Skewness (Cs ) and is inversely proportional to sample size. This
signifies larger error of estimate could occur from small sample size that produces large
Cv and Cs , at a high return period and it could also be contributed by the choice of parent
distribution and method of estimation.
Effect of rainfall spatial variability particularly for long-duration of rainfall (i.e. longer
time of concentration) and large catchments, however, US Area Reduction Factor (ARF )
as shown in Table 6 - Value of Areal Average Rainfall - Point Rainfall in existing HP1
(1982,Pg12) has been adopted. Since then, this spatial correction factor has been widely
applied without notice of accuracy assurance.
As for the effect of rainfall temporal variability, it has optimized local data from historical
rainfall records by means of the standardize storm profiles technique. The temporal storm
1
profiles were sub-divided into two regions, which were recognizing as the West Coast
Region and the East Coast Region of Peninsular Malaysia.
Despite the disparity mentioned, HP1 (1982) has been widely used by the government
agencies and the public sectors for determining the design rainstorm or rainfall intensity
in water related project. This procedure was particularly used i n conjunction with other
DID procedures or associated with other approaches such as rainfall-runoff model with
respect to water resources engineering either for planning, designing and operating of
water related projects.
The estimation of design rainfall intensity based on the rainfall Intensity - Duration
Frequency - relationship (IDF relationship) has been used as standard practice for many
decades for the design of water resources and hydraulic structures. The IDF-relationship
gave an idea about the frequency or return period of a mean rainfall intensity or rainfall
volume that can be expected within a certain period of storm duration.
For the past 30 years, the numbers of rainfall stations have tremendously increased. To
date, there are about 294 and 952 of automatic and daily rainfall gauging stations
respectively which has been registered and managed by DID throughout Peninsular
Malaysia. The utilization of larger volume and longer record of available rainfall data could
assure accurate quantiles estimation.
Therefore, the major aims of reviewing and updating this procedure are mainly to
overcome the following issues:
 To enhance and improve the accuracy of quantiles estimation particularly at high
return period;
 To improve the estimation of design rainstorm/rainfall intensity with respect to the
temporal storm variability;
 To improve the estimation of design rainstorm/rainfall intensity with respect to the
spatial storm variability;
 To facilitate the Urban Storm water Management Manual (MSMA) with respect to
the estimation of design rainstorm at low return period and to provide more at-site
IDF relationship; and
To provide the estimation of design rainstorm/rainfall intensity and IDF relationship at

ungauged site.
2
1.2 OBJECTIVE
The project objective is primarily to revise and update HP1 (1982) based on data
available in the custodian of DID with extended data record. For IDF relationship, climate
change and temporal pattern analysis, the recent data up to 2004 was used. While for
ARF analysis, the data used was up to 2014. In view of the users` ease of use, it is
necessary to maintain the arrangement and presentation as per existing edition. An
effort to apply the current, most appropriate and relevant techniques associated with the
methodology was used. It is a guide to improve quantiles accuracy for the reviewed and
updated edition.
1.3 SCOPE OF REVISION AND UPDATE
Key subjects in the proposed revision and updating of HP 1 (1982) can be summarized as
follows:
 Review existing techniques used in HP1;

 Review the method of estimation using Method of Moments (MOM) and L-
Moments (LMOM));
 Review the frequency distribution by means of the Gumbel or Extreme Value Type
1 (EV1), Generalized Extreme Value (GEV), Generalized Logistic (GLO) and
Generalized Pareto (GPA) distribution;
 Derive quantiles estimate for high and low return period for long and short
duration;
 Develop Intensity-Duration-Frequency (IDF) curves and relationship for gauged
sites;
 Formulation of regional IDF relationship for ungauged sites;
 Develop new Areal reduction factor (ARF) for catchment rainfall;
 Develop temporal pattern or storm profiles.
3
1.4 CONCERNED ISSUES AND STATEMENTS IN THE PROPOSED REVISION AND

UPDATE
1.4.1 Reviews on the Choice of Frequency Distribution
The choice of frequency distribution or accurately determination of parent distribution is

subject to the type of data series used either Annual Maximum Series (AM) or Partial
Duration Series (PD)/Plot over Threshold (POT). If AM series is chosen, the most
appropriate parent distribution is likely to be either the Gumbel distribution/Extreme Value
Type 1(EV1) or Generalized Extreme Value (GEV). As for the PD series, the Generalized
Pareto (GPA) or Exponential distribution would be the most appropriate frequency type.
Therefore, the review of parent distribution will involve AM and PD data series.
1.4.2 Short Duration Analysis
To facilitate shorter time of concentration particularly in urban areas, it was suggested that
the derivation of design rainstorm or rainfall intensity should accommodate a one-minute
temporal resolution. Nevertheless, due to errors in digitizing and processing of rainfall data;
the minimum 15 -minutes temporal resolution was adopted. Therefore, for short duration
storm the data interval of 15min, 30min, 60min, 3 -hour and 6-hour are selected for analysis,
while 12-hour, 24-hour, 3-day, 5-day and 7-day were considered long -duration storm.
Design rainstorm or rainfall intensity for the duration less than 15-minutes can however be
estimated from the IDF relationship derivations.
1.4.3 Formulation of Regional IDF Relationship for Gauged and Ungauged Sites
An appropriate regional IDF relationship can be established if method of regional frequency

analysis is chosen. It will produce a dimensionless regional growth curve (RGC) of the
recognized homogeneous region. In this context, we can assume a few homogeneous
regions within the entire Peninsular Malaysia can be produced, which is possibly dominated
by the geographical factors and hydrologic characteristics such as location, altitude, average
annual rainfall and annual maximum rainfall.
These factors will produce more than one regional growth curve of the IDF relationships. The
analysis of regional growth curve can be conducted according to the index flood approach
(Dalrymple, 1956) where it is representing the ratio of extreme rainfall of the return period
4
concerned to an index rainfall (RT / RD). The development of a regional index-flood type
approach to frequency analysis based on L-moments (Hosking and Wallis; 1993, 1997),
termed the regional L-moments algorithm (RLMA) has many reported benefits, and has the
potential of unifying current practices of regional design rainfall analysis as conducted by
Smithers et al. (2000). Basically, regional rainfall frequency analysis with the index rainfall
approach consists of two major components, namely the development of a dimensionless
frequency curve or growth curve and the estimation of the value of the index rainfall. Further
detailed description and showcase of the applicability and workability using the mentioned
methodology can be explored in Amin (2002 & 2003). Second option is to utilize the
proposed procedure that will allow the constructed IDF relationships and the derived
parameters at gauged sites possibly to be extended for the formulation of regional or
ungauged IDF relationship. Under these circumstances, the parameters of the rigorous IDF
relationship in the form of
𝜆𝛵𝜅
𝑖 = (𝑑+𝜃)𝜂
can be generalized for the entire specified area of interest. Koutsoyiannis (1998) has first
motivated the idea of this approach, which explains deliberately on the mathematical
expression of IDF relationship with respect to the probability distributions of annual maxima.
As expected to remain in the presentation of HP1 (1982), and to minimize the error of
estimates and its simplicity in developing the IDF relationship for gauged and ungauged
sites, the second approach was adopted. This means, Component II – Rainfall Depth-
Duration Plotting Diagram and Component III – Rainfall Depth – Frequency Plotting Diagram
is excluded from the analysis.
5
CHAPTER 2
ORGANIZATION OF T ASK
2.1 BRIEF OVERVIEW OF THE TASK
The required revision and update of the procedure has been organized based on the
designated tasks and can be simplified as per Figure 2.1 below.
Figure ‎2.1: Flow chart of the designated tasks for the review and update process of
Hydrological Procedure No.1 (HP1)
6
2.2 OBJECTIVES OFTHE DESIGNATED TASKS
2.2.1 T1: Task 1 – Data Mining and Assembly
To collect, collate and screen the identified rainfall data provided by DID. Insufficient data set
(quantity and quality) will trigger inaccuracy of estimation. Two types of possible data sets
are identified as Annual Maximum series (AM) and Partial Duration series/Peak over
Threshold (PD/POT). Assembly of data sets is much depending on the choice of estimation
method. List of automatic rainfall stations used are summarized and shown in Figure 2.2.
Linkages: Provide information for the components of T2, T3, T4, T5, T6, T7, and T8.
2.2.2 T2: Task 2 – Choice of Rainfall Frequency Models
To determine the best type of data series that can be used in the analysis. Insufficient
records length and missing records of data series will produce inaccuracy of estimation
particularly at high return period. The AM and POT model has been selected for the rainfall
frequency models. Choice of the PD/POT data series will definitely lengthened the data sets
and can assure and gain accuracy estimates. The series of AM rainfall can be extracted
without difficulty from hydrometric records and it has been applied onto short and long
duration storms. However, the extraction of the PD/POT series of rainfall is less
straightforward because of the occasional occurrence of rainfall events. The PD/POT model
has been applied onto automatic recorded rainfall data for determining the design
rainstorm/rainfall intensity of low (1 year and below) and high (2 years and above) return
period.
Linkages: Provide information for the components of T3 and T4.
2.2.3 T 3 : Task 3 – Choice of Distribution to be Used in the Chosen Model (AM or

PD/POT)
To identify the most appropriate parent distribution that can be analyzed using local data of
AM series or PD/POT series. Apparently, the most appropriate parent distribution for the
PD/POT model is most likely the Generalized Pareto distribution (GPA) or Exponential
Distribution. The most likely parent distribution for am model is either the gumbel/extreme
value type 1(EV1) or generalized extreme value distribution (GEV). The task will be
explained in detail in Chapter 3 – Approach and Methodology.
7
Linkages: Provide information for the components of T4, T5, T7, and T8.
2.2.4 T4: Task 4 – Method of Parameter Estimation
The most flexible, practical, robust and recent technique is the L-moments method, which
has been flexibly used and plugged for the AM and PD/POT model. Its superior method that
can be used is the at-site frequency analysis or regional frequency analysis whether by the
2-parameter or more parameter distribution. The application of L-moments approach
(Hosking and Wallis, 1987 & 1997) has received widespread attention from researchers from
all over the world. Maidment (1993) has expressed the advantage of L-moments as due to
the sample estimators of L-moments which is in linear combination of the ranked
observations, thus do not involve squaring or cubing the observations as the product-
moment estimators. These resulting L-moment estimators of the dimensionless coefficients
of variation and skewness are almost unbiased. In a wide range of hydrologic applications,
L-moments provide simple and reasonably efficient estimators of the characteristics of
hydrologic‎data‎and‎of‎a‎distribution’s‎parameters.
Linkages: Provide information for the components of T5, T6, T7, and T8.
2.2.5 T5: Task 5 – Estimation of Design Storm for Low and High Return Period
The objective is to determine the magnitude of design rainstorm/rainfall intensity at gauged

sites and ungauged sites. Direct estimation can be obtained for gauged sites; however, it is
much complicated to estimate an ungauged site which is dominated by the choice of
estimation techniques. If the choice is to maintain the existing technique currently in HP1, the
depth-duration-plotting diagram and rainfall depth-frequency diagram shall be re-determined
using new data sets. The mentioned technique is most likely inappropriate to be used onto
the PD/POT data series. The only option is to follow what has been described in Chapter
1.4.3 which is based on the rigorous formulation of suggested IDF.
Linkages: Provide information for the components of T6, T7 and T8.
2.2.6 T6: Task 6 – Construction and Formulation of at-site IDF Curve
To formulate a mathematical relationship (duration, magnitude of design rainstorm/rainfall

intensity and return period) of the established IDF curve particularly from gauged sites. This
8
will make IDF relationships easier to use, and they are often estimated by regression curve.
The polynomial formula and the modified Bernard and Koutsoyianis equation of IDF
relationship has been constructed for low and high return period.
Linkages: Provide information to component T9 and the existing polynomial equation curves
in MSMA, and possible to provide more information on other cities or identified urban areas
that were not listed in the manual.
2.2.7 T7: Task 7 – Design Storm Profile (Temporal Pattern)
To derive temporal storm variability which is often times in hydrologic modelling require
design rainfall/rainstorm hyetographs. Design rainstorm/rainfall intensity that coupled with
temporal storm variability (profile) provides input to hydrologic models, whereas the resulting
flows and flow rates of the system are calculated using rainfall-runoff and flow routing
procedure.
Linkages: Provide information to the MSMA procedure and the reviewed and updated HP1
particularly for updating existing storm profiles.
2.2.8 T8: Task 8 – Areal Reduction Factor
Areal Reduction Factor (ARF) is defined as the ratio between the design values of areal
average rainfall and point rainfall that is calculated for the same average recurrence interval
(ARI). However, information from the IDF relationship is generally in the form of point design
rainstorm/rainfall intensity. But the fact that larger catchments are less likely than smaller
catchments to experience high intensity storms over the entire catchments area, the ARF is
needed to reduce/convert point design rainfall to catchments design rainfall in order to
estimate the areal average design rainfall intensity over the catchments. Due to the lack of
adequate researches carried out in Malaysia that is probably due to data availability and
station density, the ARF obtained from a study of a part in the United States were
recommended for use in existing HP1 (1982).
Linkages: Provides information for the preparation of final report and the proposed
procedure.
9
Table 2.1: List of Automatic Rainfall Gauges Stations throughout

Peninsular Malaysia
Station Location
State No. Station Name
ID
Long(o) Lat (o)
Perak 1 4010001 JPS Teluk Intan 101.036 4.017
2 4207048 JPS Setiawan 100.700 4.218
3 4311001 Pejabat Daerah Kampar 101.156 4.306
4 4409091 Rumah Pam Kubang Haji 100.901 4.461
5 4511111 Politeknik Ungku Umar 101.125 4.589
6 4807016 Bukit Larut Taiping 100.793 4.863
7 4811075 Rancangan Belia Perlop 101.175 4.893
8 5005003 Jln. Mtg. Buloh Bgn Serai 100.546 5.014
9 5207001 Kolam Air JKR Selama 100.701 5.217
10 5210069 Stesen Pem. Hutan Lawin 101.058 5.299
11 5411066 Kuala Kenderong 101.154 5.417
12 5710061 Dispensari Keroh 101.000 5.708
Selangor 13 2815001 Dispensari
JPS SungaiKeroh
Manggis 101.542 2.826
14 2913001 Pusat Kwln. JPS T Gong 101.393 2.931
15 2917001 Setor JPS Kajang 101.797 2.992
16 3117070 JPS Ampang 101.750 3.156
17 3118102 SK Sungai Lui 101.872 3.174
18 3314001 Rumah Pam JPS P Setia 101.413 3.369
19 3411017 Setor JPS Tj. Karang 101.174 3.424
20 3416002 Kg Kalong Tengah 101.664 3.436
21 3516022 Loji Air Kuala Kubu Baru 101.668 3.576
22 3710006 Rmh Pam Bagan Terap 101.082 3.729
Pahang 23 2630001 Sungai Pukim 103.057 2.603
24 2634193 Sungai Anak Endau 103.458 2.617
25 2828173 Kg Gambir 102.938 2.813
26 3026156 Pos Iskandar 102.658 3.028
27 3121143 Simpang Pelangai 102.197 3.175
28 3134165 Dispensari Nenasi 103.442 3.138
29 3231163 Kg Unchang 103.189 3.288
30 3424081 JPS Temerloh 102.426 3.439
31 3533102 Rumah Pam Pahang Tua 103.357 3.561
32 3628001 Pintu Kaw. Pulau Kertam 102.856 3.633
33 3818054 Setor JPS Raub 101.847 3.806
34 3924072 Rmh Pam Paya Kangsar 102.433 3.904
35 3930012 Sungai Lembing PCC Mill 103.036 3.917
36 4023001 Kg Sungai Yap 102.325 4.032
37 4127001 Hulu Tekai Kwsn.”B” 102.753 4.106
38 4219001 Bukit Bentong 101.940 4.233
39 4223115 Kg Merting 102.383 4.243
40 4513033 Gunung Brinchang 101.383 4.517
10

Peninsular Malaysia (Cont’d)
Station Location
ID
Long(o) Lat (o)
Terengganu 41 3933001 Hulu Jabor, Kemaman 103.308 3.918
42 4131001 Kg, Ban Ho, Kemaman 103.175 4.133
43 4234109 JPS Kemaman 103.422 4.232
44 4332001 Jambatan Tebak, Kem. 103.263 4.378
45 4529001 Rmh Pam Paya Kempian 102.979 4.561
46 4529071 SK Pasir Raja 102.974 4.564
47 4631001 Almuktafibillah Shah 103.199 4.139
48 4734079 SM Sultan Omar, Dungun 103.419 4.763
49 4832077 SK Jerangau 103.200 4.844
50 4930038 Kg Menerong, Hulu Trg 103.061 4.939
51 5029034 Kg Dura. Hulu Trg 102.942 5.067
52 5128001 Sungai Gawi, Hulu Trg 102.844 5.143
53 5226001 Sg Petualang, Hulu Trg 102.663 5.208
54 5328044 Sungai Tong, Setiu 102.886 5.356
55 5331048 Setor JPS K Terengganu 103.133 5.318
56 5426001 Kg Seladang, Hulu Setiu 102.675 5.476
57 5428001 Kg Bt. Hampar, Setiu 102.815 5.447
58 5524002 SK Panchor, Setiu 102.489 5.540
59 5725006 Klinik Kg Raja, Besut 102.565 5.797
Kelantan 60 4614001 Brook 101.485 4.676
61 4726001 Gunung Gagau 102.656 4.757
62 4819027 Gua Musang 101.969 4.879
63 4915001 Chabai 101.579 5.000
64 4923001 Kg Aring 102.353 4.938
65 5120025 Balai Polis Bertam 102.049 5.146
66 5216001 Gob 101.663 5.251
67 5320038 Dabong 102.015 5.378
68 5322044 Kg Lalok 102.275 5.308
69 5522047 JPS Kuala Krai 102.203 5.532
70 5718033 Kg Jeli, Tanah Merah 101.839 5.701
71 5719001 Kg Durian Daun Lawang 101.867 5.701
72 5722057 JPS Machang 102.219 5.788
73 5824079 Sg Rasau Pasir Putih 102.417 5.871
74 6019004 Rumah Kastam R Pjg 101.979 6.024
75 6122064 Setor JPS Kota Bharu 102.257 6.217
N Sembilan 76 2719001 Setor JPS Sikamat 101.872 2.738
77 2722202 Kg Sawah Lebar K Pilah 102.264 2.756
78 2723002 Sungai Kepis 102.315 2.701
79 2725083 Ladang New Rompin 102.513 2.719
80 2920012 Petaling K Kelawang 102.065 2.944
11

Station Location
ID
Long(o) Lat (o)
Melaka 81 2222001 Bukit Sebukor 102.268 2.232
82 2224038 Chin Chin Tepi Jalan 102.492 2.289
83 2321006 Ladang Lendu 102.193 2.364
Pulau 84 5204048 Sg Simpang Ampat 100.544 5.295
Pinang & 85 5302001 Tangki Air Besar Sg Png 100.200 5.383
Perlis 86 5302003 100.250 5.383
Kolam Tkgn Air Hitam
Melaka 87 5303001 100.304 5.392
Rmh Kebajikan P Png
88 5303053 Komplek Prai 100.392 6.382
89 5402001 Klinik Bkt Bendera P Png 100.383 5.567
90 5402002 Kolam Bersih P Pinang 100.383 5.500
91 5404043 Ibu Bekalan Sg Kulim 100.481 5.433
92 5504035 Lahar Ikan Mati K Batas 100.431 5.535
93 6401002 Padang Katong, Kangar 100.188 6.446
Kedah 94 5507076 Bt. 27, Jalan Baling 100.736 5.583
95 5704055 Kedah Peak 100.439 5.796
96 5806066 Klinik Jeniang 101.067 3.717
97 5808001 Bt. 61, Jalan Baling 100.894 5.881
98 6103047 Setor JPS Alor Setar 100.361 6.113
99 6108001 Komppleks Rumah Muda 100.847 6.106
100 6206035 Kuala Nerang 100.613 6.254
101 6207032 Ampang Padu 100.772 6.240
102 6306031 Padang Sanai 100.690 6.343
Johor 103 1437116 Stor JPS Johor Baharu 103.458 1.471
104 1534002 Pusat Kem. Pekan Nenas 103.494 1.515
105 1541139 Johor Silica 104.185 1.526
106 1636001 Balai Polis Kg Seelong 103.697 1.631
107 1737001 SM Bukit Besar 103.719 1.764
108 1829002 Setor JPS B Pahat 102.925 1.840
109 1834124 Ladang Ulu Remis 103.468 1.849
110 1839196 Simpang Masai K. Sedili 103.965 1.850
111 1931003 Emp. Semberong 103.179 1.974
112 2025001 Pintu Kaw. Tg. Agas 102.578 2.051
113 2033001 JPS Kluang 103.319 2.022
114 2231001 Ladang Chan Wing 103.147 2.250
115 2232001 Ladang Kekayaan 103.422 2.251
116 2235163 Ibu Bekalan Kahang 103.599 2.229
117 2237164 Jalan Kluang-Mersing 103.736 2.257
118 2330009 Ladang Labis 103.017 2.584
119 2528012 Rmh. Tapis Segamat 102.814 2.517
120 2534160 Kg Peta Hulu Sg Endau 103.419 2.539
121 2636170 Setor JPS Endau 103.621 2.650
12

Station Location
ID
Long(o) Lat (o)
W. 122 3015001 Puchong Drop,K Lumpur 101.597 3.019
Persekutuan 123 3116003 Ibu Pejabat JPS 102.358 6.006
124 3116004 Ibu Pejabat JPS1 101.682 3.156
125 3116005 SK Taman Maluri 101.636 3.197
126 3116006 Ladang Edinburgh 102.417 2.133
127 3216001 Kg. Sungai Tua 101.686 3.272
128 3216004 SK Jenis Keb. Kepong 102.217 2.683
129 3217001 Ibu Bek. KM16, Gombak 101.729 3.268
130 3217002 Emp. Genting Kelang 101.753 3.236
131 3217003 Ibu Bek. KM11, Gombak 101.714 3.236
132 3217004 Kg. Kuala Seleh, H. Klg 101.768 3.258
133 3217005 Kg. Kerdas, Gombak 101.713 3.238
134 3317001 Air Terjun Sg. Batu 101.704 3.335
135 3317004 Genting Sempah 101.771 3.368
13
[a] [b]
Figure 2.2: Location Map of [a] Automatic and [b] Daily Rain Gauges Station throughout Peninsular Malaysia
14
CHAPTER 3
APPROACH AND METHODOLOGY
3.1 DATA MINING AND ASSEMBLY
Error in rainfall data can be introduced at several stages: [1] at the rain gauge; problems
can be caused by a poorly sites gauge, splashing of rainfall in and out, or losses due to
high winds and vandalism, [2] human error or technical failure is always possible, both in
reading the gauge and in archiving the results. Data mining that focuses on data checking
and screening aimed to identify and investigate suspicious annual maximum series (AM) or
partial duration series (PD) of rainfall. AM or PD series abstracted from continuously hourly
data will be checked against nearby daily totals. The hourly data will be compared to the
totals for the day on which the maximum was recorded, from the nearest daily gauges. Any
suspicious large hourly totals will be investigated further by inspecting the continuous data
from which the AM or PD is abstracted. The most suspicious data either from the AM or PD
will be statistically tested for the outlier. Thus, the identified outlier (low or high outlier) will
be excluded from the analysis. The PD series will focus on independency of the data
retrieved or abstracted, in order to ensure no overlapping of each maxima data.
3.2 CHOICE OF RAINFALL FREQUENCY MODEL
Two general approaches are available for modelling flood, rainfall, and many other
hydrologic series. One option is recognized as an annual maximum series (AM) that
considers the largest event in each year; and second option is using a partial duration
series (PD) or peak-over threshold (POT) approach that performs analysis on all peaks
above a truncation or threshold level. An objection to using AM series is that it employs only
the largest e vents in each year, regardless of whether the second largest event in a year
exceeds the largest events of other years. Moreover, the largest annual maxima in a dry
year and calling them storms are misleading. Furthermore, if hydrometric records are of
insufficient records length, it will reflect the accuracy of estimation particularly at high return
period. As reported by Cunnane (1989), the AM series has received widespread attention
not due to objective manner but argued in general manner such as widely accepted, simple
and convenient to apply. The PD series analysis avoids such problems by considering all
dependent peaks, which exceed a specified threshold. Stedinger et. al.,
15
(1993) cited that arguments in favour of PDS are that relatively long and reliable PDS
records are often available, and if the arrival rate for peaks over threshold is large enough
(1.65 events/year for the Poisson arrival with exponential exceedance model), PDS
analyses should yield more accurate estimates of extreme quantiles than the corresponding
annual maximum frequency analysis. Still, the drawback of PDS analyses is that one must
have criteria to identify only independent peaks (and not multiple peaks corresponding to
the same event). However, to avoid counting any multiple peaks in the same event, an
independency criterion has to be incorporated to distinguish dependent rainfall events that
lead to the same effect. Vaes (2000) has specified that a rainfall volume is independent if in
a certain period antecedent and posterior to the considered rainfall volume no larger than or
equal rainfall volume occurs. For this period the maximum between 12-hours and the
aggregation period is assumed.
Statistically if we denote the estimate of RT obtained by the AM series as RT and that

*
obtained from the same hydrometric record by the PD method as RT, it is usually observed
that these two estimates are unequal. Furthermore the sampling variance of RT is not equal
to that of R*T, i.e. var (RT) var (R*T). From a statistical point of view that method which has
the smallest sampling variance enjoys an advantage. Cunnane (1973) examined the
* *
relative values of var (RT) and var (RT) and found that var (RT) and var (RT) provided 1.65
where  is the mean number of peaks per year included in the PD series. If  1.65 the
opposite was true. This to show that the AM method is statistically efficient when  is small
and is less efficient when  is large. These results have been re-examined by Yevjevich
and Taesombut (1978) that suggested a value of   1.8 or 1.9 may be required to ensure
greater efficiency of PD estimates of RT.
3.3 CHOICE OF DISTRIBUTION TO BE USED IN THE CHOSEN MODEL (AM OR

PD/POT)
3.3.1 Candidates of the AM model – the Generalized Extreme Value Distribution

(GEV)
This is a general mathematical form which incorporates the Gumbel’s type I, II and III of
extreme value distributions for maxima. The GEV distribution’s cdf can be written as:
 


F (x) = exp   x  }  
for  
[1]
16
The Gumbel distribution is obtained when = 0For, the general shape of the
GEV distribution is similar to the Gumbel distribution, though the right-hand tail is thicker for
= 0and thinner for < 0. Here  is a location parameter,  is a scale parameter, and is
the important shape parameter. For the distribution has a finite upper bound at 
and corresponds to the EV type III distribution for maxima that are bounded above; for < 0
the distribution has a thicker right-hand tail and corresponds to the EV type II distribution for
maxima from thick-tailed distribution like the Generalized Pareto distribution with < 0. The
parameters of the GEV distribution in term L-moments are:
= 7.8590c + 2.9554c2 [2]
=   [3]

 
=  [4]


Where
c=
ln (2)  21  o ln (2) [5]
ln32  o ln
The quantiles of the GEV distribution can be calculated from:
XT = 

   1   ln F   [6]
Where F = 1 – 1/T is the cumulative probability of interest. When data are drawn from a
Gumbel distribution (= 0), using the biased estimator b*r in equation [16] to calculate the L-
moments estimators in variance Var(0.563/n. Comparison of the statistic
Z = √𝑛/0.563 with standard normal quantiles allows construction of a powerful test of
whether = 0 or not when fitting with a GEV distribution.
17
3.3.2 Candidate Distribution of the PD/POT Model – the Generalized Pareto

Distribution (GPA)
The GPA distribution’s cdf is given by:

F (x) = 1   x Xo 


 }  for 
[7]
where Xo is the threshold value,  and are scale and shape parameter respectively. For
positive this cdf has upper bound = Xo + 
x max ; for < 0, an unbounded and thick-tailed
distribution results; = 0 yields a two-parameter exponential distribution in the form of

F (x) = 1 – exp [-1 (x X0)] . The parameters of the GEV distribution in term L -moments

are:
(a) The threshold (Xo) is know
 4 1 - 30 + Xo    

[8]
0 - 21
 ( 0 - Xo) (1 + )         [9]
(b) The threshold (Xo) is know
 92 - 10 1 + 20        [10]

(21 - 32)
 (21 - 0) (1 + )(2

 + )       [11]
0 - 
Xo 
(1 + )        [12]
The quantiles of the GPA distribution can be calculated from:
X o  1 - exp Y T or

XT  [13]
 
X o  1 - 1 - F  or
XT  [14]
 
where YT = - ln(1 - F) and F = 1 – 1/whileis the average number of events per year
larger than a threshold Xo.
18
3.4 METHODS OF PARAMETER ESTIMATION USING L-MOMENTS
Just as the variance, or coefficient of skewness, of a random variable are functions of the
moments E(X), E(X2), and E(X3), L-moments can be written as functions of probability-
weighted moments (PWMs), which can be defined as:
r E { X [F(X)]r } [15]
where F(X) is the cdf for X. Probability-weighted moments are the expectation of X times
powers of F(X). For r=0, 0 is the population mean  x. Estimators of L-moments are
mostly simply written as linear function of estimators of PWMs. The first PWM estimator b0 of
0 is the sample mean X. To estimate other PWMs, one employs the ordered observations,
or the order statistics X(n) .....  X(1), corresponding to the sorted or ranked observation in a
sample (Xi i = 1, ….., n). A simple estimator of r for r  1 is:
n

1  j=1X(j) 1 – (j – 0.35) 
* r
br        [16]
n n
1– (j – 0.35)
where are estimators of F(X(j)). b* r is suggested for use when estimating
n
quantiles and fitting a distribution at a single site. Although it is biased, it generally yields
smaller mean square error quantiles estimators than the unbiased estimators as in equation
below. When unbiasedness is important, one can employ unbiased PWM estimators as:
bo  X [17]
XX j(j )
n1 n(n–j j)
b1   nn 1 [18]
j 1
n2 n(n
 –j   1 X (jj)
j)n(n –j j-1)
b2   [19]
j 1 nn  1n  2
n3 n(n –j j)
n(n–jj-1)
 1(nn  j  2X X( j )j 
– j-2)
b3   [20]
j 1 nn  1n  2n  3
19
These are examples of the general formula:
𝑛−𝑗 𝑛−𝑗
1 ( )(j) 1 ( )(𝑗)
̂ r = r = ∑𝑛−𝑟 𝑟 ∑𝑛−𝑟 𝑟
b
𝑛 𝑗=1 𝑛−𝑟 =
(𝑟+1) 𝑗=1 𝑛 [21]
( ) ( )
𝑟 𝑟+1
for (which defines PWMs in terms of powers of (1-F); this r  1r formula can be derived
using the fact that is the expected value r  1 of the largest observation in a sample of size.
The unbiased estimators are recommended for calculating L-moments diagrams and for use
with regionalization procedures where unbiasedness is important. For any distribution, L-
moments are easily calculated in term of PWMs from:
1   o [22]
2  21   o [23]
3  6 2  61   o [24]
4  20 3  30 2  121   o [25]
3.5 ESTIMATION OF DESIGN STORM/RAINFALL INTENSITY OF LOW AND HIGH

RETURN PERIOD
The estimated parameters of the chosen probability distributions as to be carried out in Task
5 (T5), will lead to the possibility of calculating quantile estimation of design storm/rainfall
intensity for low and high return period. It can be calculated from the proposed equations of
[11], [12] and [13] associated with return period, T; and duration, D. The calculated quantiles
estimation at low return period of T=1 -month, 2-month, 3-month and 6- month (less than
one-year) at specified durations is intentionally calculated to accommodate the construction
of IDF relationship at specified urban/city areas in the urban stormwater/sewer design. It is
also purposely carried out to supplement the existing discrepancies in MASMA (JPS, 2000).
The calculated quantiles estimation at high return period (with respect to T=2, 5, 10, 20, 50
and 100-year return period) is definitely to enhance and improve the rainfall intensity design
values of the existing HP1 and integral for the construction of IDF relationship/curves for the
entire gauged and ungauged sites of Peninsular Malaysia.
20
3.6 CONSTRUCTION AND FORMULATION OF AT-SITE IDF CURVE
The formulation of a mathematical expression on the at-site IDF relationships is definitely for
the benefit of the users and it will assist them to calculate the quantiles estimation easily and
quickly. The polynomial equations have been introduced in the Urban Storm Water
Management Manual, MSMA (JPS, 2000), however, the equations is limited to the duration
of an hour to 1000 minutes. Possible reasons are due to the proposed polynomial equation
that has failed to fit the small storm duration (less than 1 -hour) and larger storm duration for
more than 24-hours. For duration less than one hour, a relationship of the required duration
and the factor of 2-years return period 24-hours rainfall that explicitly showed in the manual
as in Chapter 13- equation [13.3] has been introduced. But no explanation has been
proposed or introduced on how to perform estimation for more than 1000 minutes duration in
particular.
Consequently, as quoted in the procedure, the error of estimation is likely to be up  20%

particularly for the shorter duration of 30-minutes and longer duration of 15-hours. To give a
more precise estimation and for minimizing the error of estimates due to the chosen
mathematical expression, we proposed general equation [26] and the identical equation[27]
to be adopted as general mathematical formulation of the IDF relationship. Under these
circumstances, for the specified formulation of the GEV distribution, the Gumbel distribution
and the GPA distribution can be explicitly performed using equation [35], [36] and [37]
respectively.
3.6.1 An Overview on the Mathematical Expression of an IDF Relationship
IDF relationship is a mathematical relationship between the rainfall intensity i, the duration d,
and the return period T (or, equivalently, the annual frequency of exceedance, typically
referred to as ‘frequency’ only) (Koutsoyiannis, Kozonis and Manetas; 1998).
The typical IDF relationship for a specific return period is a special case of the
generalized formula as given in equation [25] where ,,  and  are non-negative
coefficients with   1. This expression is an empirical formula that encapsulates the

experience from several studies. A numerical study shows if assumed =1, the
corresponding error are much less than the typical estimation errors which results equation
[26].
21

i= [26]
(𝑑𝑣 + )

i= [27]
(𝑑+ )
For any two return periods T1 and T2 where T2<T1 yields the set of restriction in
equation [26] which 1   0, 0  1   2    1, and 1  2  0. With these
restrictions,  is considered as a (increasing) function of the return period T. This leads

to a general IDF relationship shows in equation [27], which has the advantage of a
separable functional dependence of i on T and d. The function of
bd is bd  d   is where  and  is parameter to be estimated (>0, 0<<1).
i = a (T) [28]
bd 
The function of a(T); however, completely could be determined from the probability
distribution function of the maximum rainfall intensities I(d). Therefore, if the intensity I(d) of
a certain duration d has a particular distribution FId  i; d, yields the distribution of variable ,
which is no more than the intensity rescaled by X  I(d)b(d), which is no more than the
intensity rescaled by b(d). Mathematically, this can be expressed by
FId  I; d  FX xT   1  1/ T
(non-exceedance probability), which can be shown in the form of equation [28]; therefore
proved that a(T) can completely be determined from the distribution function of maximum
intensity.
XT = a (T) = Fy—1 (1-1/T) [29]
The distribution function of the proposed GEV, the Gumbel and the GPA distribution
respectively can be written in the form of equation [29], [30] and [31] where >0, >0, and
 are shape, scale and location parameters respectively. Subsequently, for the GEV, the
Gumbel and the GPA distribution can be directly obtained from equation [29], [30] and
[31], which in turns into equation [32], [33] and [34] respectively. Finally, general formula
for idf relationship is shown in equation [27] can be written in specific form of the GEV, the
Gumbel and the GPA distribution respectively in the form of equation [35], [36] and [37].
22
x
F(x)=exp{-[1+( − )]−1/ [30]

F(x) = exp {-exp(− x⁄ + )} [31]
[32]
𝑥 −1/
F(x)={-[1+( − )]

[− ln(1−1⁄𝑇 )]− −1
XT = a(T) =  +

 [33]
XT = a(T) =  { - ln[-ln(1-1⁄𝑇)]} [34]
XT  aT    T   1 ] [35]


−
− ln(1−1⁄𝑇)]
 {  +[ 
i = (𝑑+)
} [36]
XT   ln ln1  1 T

[37]
(d+ )
 + 𝑇𝑘 −1
[ 𝑘
] [38]
i = (𝑑+ )
For the case of the GEV, the Gumbel and the GPA distribution, the parameters of the
function of a(T) (i.e. ,  and  ) and b(d) (i.e  and ) could be separately determined
either function a(T) or b(d), or simultaneously solving for function a(T) and b(d) .
The function of a(T), however, as for simplicity used, can be expressed in Bernard
equation (1932) i n the form of:
aT   T k [39]
and finally equation [37](13) can be transformed in general term as follow:
𝜆𝛵𝜅
𝑖= [40]
(𝑑+𝜃)𝜂
23
Equation [39] has been used to formulate the gauged IDF relationship and the derived
parameters of , ,  and  h as been generalized for the construction of ungauged IDF
relationship. As for the MSMA polynomial equation, it has been reviewed and updated
using new quantile estimation derivations.
3.6.2 One-Step Least Square Method of the IDF Relationships
For solving equation [39], one-step least square method is chosen

b a  T d due to its ability solving function, and simultaneously. To this aim, an empirical
return period can be assigned using the Gringorten potting formula
𝑛𝑗 +0.12
Tjl =
𝑙−0.44
to each data value i jl (j refer to a particular duration d, j=1, …. k; l denoting the rank, l 
1,…nj is the length of the group j). Each data will have a triplet of numbers ilj , Tij , dj) and
resulted in the intensity model as
𝑎(𝑇 )
iˆ jl =  𝑗𝑙
𝑏(𝑑𝑗 )
The corresponding error could be measure model as
e jl  ln i jl ln iˆ jl = ln (i jl / iˆ jl )
The overall mean square error is
e 2 1 ∑𝑘𝑗=1 𝑛1 ∑𝑛𝑙=1 𝑒𝑗𝑙2

𝑗
k
which leads into an optimization procedure defined as e  f 2  , , ,  . Simultaneous solution
to perform the optimization as defined can be executed using the embedded solver tools of
common spreadsheet package.
24
CHAPTER 4
ASSESSMENT OF THE PROPOSED METHODOLOGY
4.1 ASSESSMENT PROCEDURE
Assessment procedure of the proposed methodology has been conducted as per Figure
4.1. The objectives of this procedure are:
1. Data mining and assembly which are among others to identify and investigate
suspicious annual maximum series (AM) or partial duration series (PD) of rainfall
data; identification data independency for PD/POT data series in order to avoid any
overlapping each of maxima data; and to ensure clean data set (quantity and quality)
for the AM and PD/POT model analysis;
2. To determine the best type of data series that can be used in analysis. Two models
are identified as Annual Maximum model (AM) and Partial Duration series/Peaks
over Threshold model (PD/POT);
3. To identify the most appropriate parent distribution that can be used in analysis of
AM series or PD/POT data series;
4. To determine the best method of parameter estimator between the Method of
Moment (MOM ) and L-Moments (LMOM) approach;
5. To determine the best fit or appropriate distribution-estimates (D/E) model; which
can be carried out by robustness study in which includes determination of good
performance (bias) and accuracy of estimation (RMSE) of the model;
6. To estimate the magnitude of design rainstorm in corresponds with return period (low
and high) which includes developing design rain depth-duration and rainfall intensity-
duration relationship;
7. To construct and formulate the Intensity-Duration-Frequency relationship for gauged
sites.
In order to perform assessment of the proposed methodology, annual maximum data series
(AM) and partial duration data series are collected from eight (8) selected rainfall stations as
listed below:
1. Site 2033001 at Pekan Nenas, Johor;
2. Site 3428081 at Temerloh, Pahang;
3. Site 3613004 at Ibu Bekalan Sg Bernam, Selangor;
4. Site 5005003 at Bagan Serai, Perak;
5. Site 5328044 at Sungai Tong, Terengganu;
6. Site 6019004 at Kastam Rantau Panjang, Kelantan;
7. Site 6103047 at Hospital Alor Setar, Kedah; and
8. Site 6401002 at Padang Katong, Perlis
25
4.2 CHOICE OF RAINFALL FREQUENCY MODEL
There are two choices for the rainfall frequency model; Annual Maximum Series (AM) model
and the Partial Duration Series/Peak over Threshold (PD/POT) model. One-hour duration
historical data records have been extracted from eight (8) rainfall stations as listed above.
All selected rainfall stations has been assumed to have similar statistical characteristics and
has been tested using the models proposed.
For the record, PD/POT model was tested for high and low return period while AM model
was only tested for high return period. Quantile estimates of low return period calculates for
T= 0.5, 1, 3 and 6-months meanwhile high return period refers to T= 2, 5, 10, 20, 25, 50 and
100 years. Comparatively, the PD/POT model has advantage against the AM model as the
later could not derived quantile estimates for low return period.
Comparatively, this analysis yields quantiles estimation of the PD/POT constantly greater
than the AM model. In addition, the analysis using the PD/POT model subsequently
produced the quantile estimation of low return period with respect to T=0.5, 1, 3 and 6-
months, which definitely could not derived from the AM model. Therefore, based on this
findings, the PD/POT model quite certain can be the most appropriate rainfall model, which
it has capability and ability to derive the quantiles estimation of low and high return period
simultaneously.
4.3 ROBUSTNESS STUDY AND EFFICIENCY PROCEDURE
Objective quantile estimation is based on methods developed for use with random samples
from stationary populations. Such random samples have the characteristics that different
samples, when treated in the same way, generally yield numerically different values of
quantile estimates.
A procedure for estimating RT is robust if it yields estimates of which are good estimations
(low bias, high efficiency) even if the procedure is based on an assumptions which is not
true. A procedure is not robust if it yields poor estimates of RT when the procedure’s
assumption departs even slightly from what is true.
26
Figure ‎4.1: Flow C hart in Assessment Procedure of the Proposed Methodology
27
Since we do not know how the distribution of AM series or PD series behaves naturally, we
have to seek out and find a distribution and an estimation procedure which are robust and
able to be used with distributions that gives random samples of a storm-like behaviour. It
should be emphasized that split samples test based on historical rainstorm records are
inadequate for testing the robustness of any distribution and estimation (D/E) procedure
(Cunnane, 1989).
A suitable method of testing a D/E procedure involves simulating random samples from a
parent distribution in which the R-T relationships is exactly known (Hosking et. al., 1985a).
To be authentic, in this context, the parent distribution must produce random samples
which are rainstorms-like in their behaviour. Such a parent distribution would be a GEV
and EV1 of the AM model and a GPA and EXP of the PDS/POT model. Then the D/E
under test is applied to each sample and ȒT is obtained from each sample for a selection of
T values. This is repeated for M samples (M large) and the equations [40] to [44] are used
to calculate Bias and RMSE from the M values of ȒT:
Ȓ𝑇
mean = ȒT = ∑𝑀
𝑖=1 ( ) [40]
𝑀
2 1/2
St. Dev = SȒT = ∑ [[(Ȓ𝑇)𝑖− Ȓ𝑇] ] [41]
𝑀
Bias = bT = ȒT - RT [42]
1/2
2
RMSE = rT =∑ [[(Ȓ𝑇)𝑖− Ȓ𝑇] ] [43]
𝑀
In these expressions ȒT is known population value. The sampling distribution of ȒT

is also examined and frequently this can be approximated by a Normal distribution so that
5% and 95% quantiles of the sampling distribution, denoted lower and upper confidence
levels, LCL and UCL, can be obtained as:
LCL  ȒT  1.645SȒT [44]
UCL  ȒT  1.645SȒT [45]
28
Hydrological Procedure No. 1
All these quantiles can be made dimensionless by division of population value RT. This
practice is usually done to enable inter-comparison of D/E procedures. Based on the
procedures mentioned, the D/E was tested by means of the following combinations (1) 3P-
GPA/LMOM, (2) 3PGEV/LMOM, (3) 2P-GPA/EXP/LMOM; (4) 2P-EV1/LMOM, and (5)
2pEV1/MOM. The D/E technique as explained above is referred to as predictive ability
procedure, but it is also guided with descriptive ability which is based on visual inspection of
the probability plot of R-T relationship.
4.4 RESULTS OF THE ASSESSMENT FOR THE CHOICE OF RAINFALL MODEL,

PARENT DISTRIBUTION AND PARAMETER ESTIMATION
The AM and PD/POT model has been tested for determining quantile estimation at high
return period (T) which are corresponding with T=2, 5, 10, 20, 25, 50 and 100 years.
Meanwhile, the quantile estimation of PD/POT model was tested for low return period (less
than T=1 year) that corresponds with T=0.5, 1, 3, and 6-month return period. The
assessment of PD/POT model was highly motivated due to insufficient at-site information in
MSMA (2000) particularly for quantiles estimation of low return period.
The assessment have been carried out to obtain the most efficient model of the PD/POT
model that represented by 3P-Generalized Pareto (GPA) and 2P-GPA/Exponential
distribution (EXP) to the AM model of 3PGeneralized Extreme Value (GEV) and 2P-Extreme
Value Type 1 (EV1/Gumbel) distribution.
Parameters of probable distribution of the proposed model were estimated by a robust

approach of the L-Moment (LMOM) and conventional technique of the Method of Moments
(MOM). The analysis results the following conclusions:
a. For less than 6-hr rainfall duration, the D/E test showed that the best options are
represented by the 2P-EV1/LMOM and 2P-GPAEXP/LMOM. However, for 6-hr
rainfall duration and greater, the 3PGPA/LMOM and 3P-GEV/LMOM is pretty well
fitted particularly in Johor, Kelantan and Terengganu;
b. Robustness study shows the 2P-EV1/LMOM and 2P-GPAEXP/LMOM produced
small Root Mean Square Error (RMSE); however, the 2P-GPA-EXP/LMOM has been
chosen due to the major advantage of this model which is its ability for determining
quantile estimates at high and low return period;
29
c. Method of parameter estimation study showed that L-Moments was selected instead
MOM where the former has advantages as follows; (1) the method was accepted
worldwide; (2) flexible and easy to use with other types of distribution; and (3)
recommended method for the regionalization approach as it will accommodate
important tool in Task 8 (T8);
d. Hypothesis for determining k=0 or not when fitting with GEV has been carried out for
the AM model of 3P-GEV/LMOM by means of comparing the statistic
Z = √𝑛/0.563with standard normal quantiles level which is found that for all
stations-duration shows not significantly large at 5% significant level. Hence the
hypothesis that k=0 is not rejected;
e. This conclude that the 2P-EV1/LMOM distribution/estimation is accepted for
representing the AM model of daily rainfall data series;
f. The 2P-GPA/EXP distribution is considered the best option for the PDS/POT model
as 2P-EV1 and 2P-GPA/EXP is special case of the 3P-GEV and 3P-GPA distribution
when the shape parameter k=0;
In summary, the quantile estimate of design rainstorm throughout Peninsular Malaysia was
derived based on [1] 188 nos. of automatic rain gauged stations throughout Peninsular
Malaysia analysed using PDS/POT model of 2P-GPA/EXP distribution; [2] 827 nos. of daily
rain gauged stations in the entire of Peninsular Malaysia were modelled with the AM model
of 2P-EV1/LMOM; and [3] 135 nos. of IDF curves have been produced for high and low
return period. As for the location of automatic and daily raingauges station in Peninsular
Malaysia, it can be seen at Figure 2.2 in Chapter 2.
30
CHAPTER 5
DEVELOPING THE INTENSITY – DURATION FREQUENCY (IDF) RELATIONSHIP
– GAUGED SITES
5.1 CHOICE OF MATHEMATICAL FORMULATION FOR IDF RELATIONSHIP
As explained in Chapter 3.6.1, the formulation of IDF relationship was constructed based on
equation [39]. This equation has been formulated based on formula derived by
Koutsoyiannis (1998) and Bernard (1932) as shown in equation [26] and [38] respectively.
General term of the IDF relationship or recognized as an empirical formula is finally in the
form of
𝜆𝛵 𝜅
𝑖=
(𝑑 + 𝜃)𝜂
The required IDF model parameters of 𝜆, 𝜅 , 𝜃 and  were derived using simultaneous
solution of the embedded MS Excel SOLVER by means of One -Step Least Square (OSLS)
method.
As for accommodating the MSMA polynomial equation (2000) as stated in Table 13.A1
(Volume 4, Chapter 13), new polynomial parameters of a, b, c and d were reviewed and
updated using new quantile estimates derived. The new polynomial formula was derived
particularly for accommodating longer time period for the duration of 15 to 4320-minutes (72-
hrs) which is in contrast to the current MSMA polynomial formula that is valid only for the
duration of 30 to 1000 minutes.
The formulated equations of empirical and polynomial formula has been established onto
135 nos. of selected rainfall gauging stations throughout Peninsular Malaysia and it has been
applied to quantiles estimates of high (more than or equal to 2-year) and low (less than or
equal to 1 -year) return period.
5.2 COMPARISON OF NEW POLYNOMIAL EQUATION AND MSMA (2000)
For comparison purposes, Site 3117070 at DID Ampang is selected where the site IDF
curve was regular and widely used for determining design rainstorm/intensity in Kuala
Lumpur area. The polynomial parameters of a, b, c and d that derived from the recent
exercise and based on current MSMA are summarized in Table 5.1 while Table 5.2 shows
31
quantiles estimate from the two fitted equations. As was mentioned previously, the new
formula has an advantage and ability to accommodate longer period of time; 15 to 4320
minutes. This makes its unnecessary to have additional tool for quantiles estimate for the
duration of less than 30 minutes and beyond 1000 minutes. According to Table 5.2,
significant different in the estimated design rainstorm can be seen. For instance, say
quantile estimate for short duration of one-hour corresponding with 100-year ARI is found
to be 114.2mm and 110.2mm which represents new fitted parameters and current
parameters respectively or about 3.6% increase.
Table ‎5.1: Polynomial Equation Parameters of Site 3117070

Value of derived parameters (new) associated with return
period (ARI)
Parameter
2 5 10 20 50 100
a 4.1889 4.3678 4.4705 4.5603 4.6658 4.7382
b -0.7113 -0.7153 -0.7174 -0.7190 -0.7207 -0.7217
c -0.0929 -0.0817 -0.0763 -0.0721 -0.0676 -0.0648
d 0.0165 0.0142 0.0131 0.0122 0.0113 0.0108
Value of present parameters(MSMA, 2000) associated with
Parameter return period (ARI)
a 5.3255 5.1086 4.9696 4.9781 4.8047 5.0064
b 0.1806 0.5037 0.6796 0.7533 0.9399 0.8709
c -0.1322 -0.2155 -0.2584 -0.2796 -0.3218 -0.3070
d 0.0047 0.0112 0.0147 0.0166 0.0197 0.0186
Table ‎5.2: Rainfall Intensity of Site 3117070

Duration Quantiles estimate associated with new parameters (mm/hr)
(hr.)
2 5 10 20 50 100
0.25 141.5 175.0 197.1 218.3 245.9 266.5
0.5 102.7 123.9 137.9 151.4 168.9 182.0
1 65.9 78.9 87.4 95.6 106.3 114.2
3 27.6 33.2 36.9 40.4 45.0 48.5
6 15.0 18.3 20.4 22.4 25.1 27.1
12 8.2 10.0 11.2 12.4 13.9 15.0
24 4.6 5.6 6.3 7.0 7.8 8.5
Duration Quantiles estimate associated with current parameters (mm/hr)
(hr.) (MSMA, 2000)
0.5 99.0 117.9 130.4 142.4 156.6 172.2
1 64.8 75.7 83.9 91.3 100.5 110.2
3 28.7 32.5 36.2 39.4 43.2 47.2
6 15.9 18.0 20.4 22.4 24.7 26.8
12 8.4 9.8 11.5 12.9 14.4 15.6
32
5.3 COMPARISON OF NEW POLYNOMIAL AND EMPIRICAL EQUATION
As for assessing the variation of quantiles estimates from the new fitted polynomial equation
and new derived empirical equation, previous site which is Site 3117070 has been adopted.
Figure 5:1 and Figure 5:2 depicts the IDF curves that were fitted by means of polynomial and
empirical equation respectively. Table 5.3 shows quantiles estimate of the former and latter,
respectively.
Table ‎5.3: Design Rainfall Intensity for Site 3117070 at D ID Ampang

Duration Quantiles estimate of rainfall i intensity by Polynomial (mm/hr)
(hr.)
2 5 10 20 50 100
0.25 141.5 175.0 197.1 218.3 245.9 266.5
0.5 102.7 123.9 137.9 151.4 168.9 182.0
1 65.9 78.9 87.4 95.6 106.3 114.2
3 27.6 33.2 36.9 40.4 45.0 48.5
6 15.0 18.3 20.4 22.4 25.1 27.1
12 8.2 10.0 11.2 12.4 13.9 15.0
24 4.6 5.6 6.3 7.0 7.8 8.5
48 2.7 3.3 3.7 4.1 4.6 4.9
72 2.1 2.5 2.8 3.1 3.4 3.7
Duration
(hr.) Quantiles estimate of rainfall intensity by Empirical (mm/hr)
0.25 155.1 177.7 196.9 218.2 249.9 276.9

0.5 103.8 118.9 131.8 146.0 167.2 185.3
1 64.6 74.0 82.0 90.8 104.1 115.3
3 27.9 31.9 35.4 39.2 44.9 49.7
6 15.9 18.2 20.2 22.4 25.7 28.4
12 9.0 10.3 11.4 12.7 14.5 16.1
24 5.1 5.8 6.4 7.1 8.2 9.0
48 2.8 3.3 3.6 4.0 4.6 5.1
72 2.0 2.3 2.6 2.9 3.3 3.6
Duration
(hr) Difference (%) of quantiles estimate
0.25 9.61 1.55 -0.08 -0.06 1.65 3.91
0.5 1.08 -4.06 -4.48 -3.57 -0.98 1.84
1 -2.05 -6.20 -6.20 -4.99 -2.07 0.94
3 1.00 -3.84 -4.12 -3.10 -0.36 2.56
6 5.85 -0.18 -0.93 -0.22 2.21 4.96
12 10.26 3.10 1.91 2.30 4.45 7.03
24 11.02 3.51 2.17 2.45 4.47 6.97
48 4.85 -1.55 -2.49 -1.97 0.24 2.82
72 -2.81 -7.76 -8.18 -7.31 -4.82 -2.13
33
Figure 5.1: Site 3117070 IDF curve fitted by Polynomial Equation
Figure 5.2: Site 3117070 IDF curve fitted by Empirical Equation
34
CHAPTER 6
DEVELOPING THE INTENSITY-DURATION FREQUENCY (IDF) RELATIONSHIP –
UNGAUGED SITES
6.1 BRIEF DESCRIPTION
As for determining quantiles estimation at ungauged sites from the current HP1 (1982), the
so called Component II – Rainfall Depth-Duration Plotting Diagram and Component III –
Rainfall Depth – Frequency Plotting Diagram has been used in association with the
isopleths maps of 0.5hr, 3hr, 24hr and 72hr which is in correspond with 2 and 20 years
return period.
The required quantiles estimation in correspond with return period acquires information to be
retrieved from the isopleths map mentioned and it has to be transformed onto the rainfall
depth–duration plotting diagram and rainfall depth–frequency (return period) plotting
diagram. As shown in Appendix C of the HP1 (1982), the error of estimates contributed by
this approach for 2 and 20 years return period are ranging from -30% to +18% and -58% and
+53% respectively. Apparently, it clearly demonstrates that the worst performances are
contributed at shorter duration of 0.25hr and higher return period while also demonstrating
good performance for longer duration.
Large error of estimates could be contributed particularly from [1] the isopleths map
developed using less and shorter rainfall data, and [2] flaws from the rainfall depth-duration
and frequency plotting diagram developed.
As the analysis was performed and derived at 2 and 20 years return period, the required
quantiles estimate particularly at higher return period which was produced by means of
extrapolation, in turn could lead to larger error. Eventually, the method described only has
the ability for determining quantiles estimate but it would not be able to establish the IDF
curve and IDF relationship of ungauged sites required.
As to anticipate and minimize the error of estimates and its simplicity in developing the IDF
curve and IDF relationship at ungauged sites, eventually the constructed IDF relationship of
gauged sites can be extended in the formulation of ungauged IDF relationship. In turn, the
component II and III of rainfall depth-duration and rainfall depth-frequency plotting diagrams
were excluded in the analysis.
35
6.2 MATHEMATICAL FORMULATION FOR THE IDF RELATIONSHIP OF

UNGAUGED SITES
As described in Chapter ‎3.6.1 and it has also discussed in Chapter ‎1, the formulation of IDF
curve and IDF relationship at ungauged site was extended from the rigorous general term of
IDF relationship used for gauged site in the form of
𝜆𝛵 𝜅
𝑖=
(𝑑 + 𝜃)𝜂
The four parameters or coefficients derived from gauged sites which are 𝜆, 𝜅 , 𝜃 and  can
be separately generalized in order to produce the isopleths map of each parameter.
Advantages for using this approach are gained from [1] the ungauged parameters are
directly transformed from gauged sites, [2] ungauged IDF relationship can directly be
formulated at any point from the four parameters isopleths maps, [3] IDF curve can easily be
generated at any point of interest, and [4] the required design rainstorm can easily be
derived in correspond with any return period (low and high return period) and duration
(15minutes to 72hrs).
6.3 SUMMARY OF FINDINGS
The four parameters derived from 135 nos. of raingauge stations are tabulated in Table 6.1a-
6.1d and Table 6.2a - 6.2d for the IDF relationship with corresponding to high return period
and low return period respectively. The high and low return periods are associated with T=2,
5, 10, 20, 50, 100-years and T=1, 2, 3, 6 and 12-month respectively. Figure 11.1 to 11.4 in
Appendix 1, depicts the generalized isopleths map of 𝜆, 𝜅 , 𝜃 and .
36
Table 6.1a: Derived IDF Parameters of High A RI for Peninsular Malaysia
Station Derived Parameters

ID
   
Perak 1 4010001 JPS Teluk Intan 54.017 0.198 0.084 0.790
2 4207048 JPS Setiawan 56.121 0.174 0.211 0.854
3 4311001 Pejabat Daerah Kampar 69.926 0.148 0.149 0.813
4 4409091 Rumah Pam Kubang Haji 52.343 0.164 0.177 0.840
5 4511111 Politeknik Ungku Umar 70.238 0.164 0.288 0.872
6 4807016 Bukit Larut Taiping 87.236 0.165 0.258 0.842
7 4811075 Rancangan Belia Perlop 58.234 0.198 0.247 0.856
8 5005003 Jln. Mtg. Buloh Bgn Serai 52.752 0.163 0.179 0.795
9 5207001 Kolam Air JKR Selama 59.567 0.176 0.062 0.807
10 5210069 Stesen Pem. Hutan Lawin 52.803 0.169 0.219 0.838
11 5411066 Kuala Kenderong 85.943 0.223 0.248 0.909
12 5710061 Dispensari Keroh 53.116 0.168 0.112 0.820
Selangor JPS
Selangor 1 2815001 JPS Sungai
Sungai Manggis
Manggis 56.052 0.152 0.194 0.857
2 2913001 Pusat
Pusat Kwln. JPS
Kwln. JPS TT Gong
Gong 63.493 0.170 0.254 0.872
3 2917001 Setor JPS Kajang 59.153 0.161 0.118 0.812
4 3117070 JPS Ampang 65.809 0.148 0.156 0.837
5 3118102 SK Sungai Lui 63.155 0.177 0.122 0.842
6 3314001 Rumah Pam JPS P Setia 62.273 0.175 0.205 0.841
7 3411017 Setor JPS Tj. Karang 68.290 0.175 0.243 0.894
8 3416002 Kg Kalong Tengah 61.811 0.161 0.188 0.816
9 3516022 Loji Air Kuala Kubu Baru 67.793 0.176 0.278 0.854
10 3710006 Rmh Pam Bagan Terap 60.793 0.173 0.185 0.884
Pahang 1 2630001 Sungai
Sungai Pukim
Pukim 46.577 0.232
0.232 0.169
0.169 0.687
0.687
2 2634193 Sungai Anak
Sungai Anak Endau
Endau 66.179 0.182
0.182 0.081
0.081 0.589
0.589
3 2828173 Kg Gambir
Kg Gambir Pos 47.701 0.182
0.182 0.096
0.096 0.715
0.715
4 3026156 Iskandar Simpang
Pos Iskandar 47.452 0.184
0.184 0.071
0.071 0.780
0.780
5 3121143 Pelangai
Simpang Pelangai 57.109 0.165
0.165 0.190
0.190 0.867
0.867
6 3134165 Dispensari Nenasi 61.697 0.152
0.152 0.120
0.120 0.593
0.593
7 3231163 Kg Unchang 55.568 0.179
0.179 0.096
0.096 0.649
0.649
8 3424081 JPS Temerloh 73.141 0.173
0.173 0.577
0.577 0.896
0.896
9 3533102 Pahang Tua
Rumah Pam Pahang Tua 58.483 0.212
0.212 0.197
0.197 0.586
0.586
10 3628001 Kaw. Pulau
Pintu Kaw. PulauKertam
Kertam 50.024 0.211
0.211 0.089
0.089 0.716
0.716
11 3818054 Setor JPS Raub 53.115 0.168
0.168 0.191
0.191 0.833
0.833
12 3924072 Rmh Pam Paya Kangsar 62.301 0.167
0.167 0.363
0.363 0.868
0.868
13 3930012 Sungai Lembing PCC Mill 45.999 0.210
0.210 0.074
0.074 0.590
0.590
14 4023001 Kg Sungai Yap 65.914 0.195
0.195 0.252
0.252 0.817
0.817
15 4127001 Hulu Tekai Kwsn.”B” 59.861 0.226
0.226 0.213
0.213 0.762
0.762
16 4219001 Bukit Bentong 73.676 0.165
0.165 0.384
0.384 0.879
0.879
17 4223115 Kg Merting 52.731 0.184
0.184 0.096
0.096 0.805
0.805
18 4513033 Gunung Brinchang 42.004
42.004 0.164
0.164 0.046
0.046 0.802
0.802
37
Table 6.1b: Derived IDF Parameters of High ARI for Peninsular Malaysia (cont’d)

ID    

Tereng- 1 3933001 Hulu Jabor, Kemaman 103.519 0.228 0.756 0.707
ganu 2 4131001 Kg, Ban Ho, Kemaman 65.158 0.164 0.092 0.660
3 4234109 JPS Kemaman 55.899 0.201 0.000 0.580
4 4332001 Jambatan Tebak, Kem. 61.703 0.185 0.088 0.637
5 4529001 Rmh Pam Paya Kempian 53.693 0.194 0.000 0.607
6 4631001 Almuktafibillah Shah 66.029 0.199 0.165 0.629
7 4734079 SM Sultan Omar, Dungun 51.935 0.213 0.020 0.587
8 4832077 SK Jerangau 54.947 0.212 0.026 0.555
9 4930038 Kg Menerong, Hulu Trg 60.436 0.204 0.063 0.588
10 5029034 Kg Dura. Hulu Trg 60.510 0.220 0.087 0.617
11 5128001 Sungai Gawi, Hulu Trg 48.101 0.215 0.027 0.566
12 5226001 Sg Petualang, Hulu Trg 48.527 0.228 0.000 0.547
13 5328044 Sungai Tong, Setiu 52.377 0.188 0.003 0.558
14 5331048 Setor JPS K Terengganu 58.307 0.210 0.123 0.555
15 5426001 Kg Seladang, Hulu Setiu 57.695 0.197 0.000 0.544
16 5428001 Kg Bt. Hampar, Setiu 55.452 0.186 0.000 0.545
17 5524002 SK Panchor, Setiu 53.430 0.206 0.000 0.524
18 5725006 Klinik Kg Raja, Besut 52.521 0.225 0.041 0.560
Kelantan 1 4614001 Brook 49.623 0.159 0.242 0.795
1 4614001 Brook 49.623 0.159 0.242 0.795
2 4726001 Gunung Gagau 43.024 0.220 0.004 0.527
2 4726001 Gunung Gagau 43.024 0.220 0.004 0.527
3 4819027 Gua Musang 57.132 0.155 0.119 0.795
3 4819027 Gua Musang 57.132 0.155 0.119 0.795
4 4915001 Chabai 47.932 0.169 0.108 0.794
4 4915001 Chabai 47.932 0.169 0.108 0.794
5 4923001 Kg Aring 47.620 0.187 0.020 0.637
Kelantan 5 4923001 Kg Aring 47.620 0.187 0.020 0.637
6 5120025 Balai Polis Bertam 61.338 0.168 0.193 0.811
7 5216001 Gob 41.783 0.175 0.122 0.720
7 5216001 Gob
Dabong 41.783 0.175 0.122 0.720
8 5320038 51.442 0.189 0.077 0.710
8 5320038 Dabong 51.442 0.189 0.077 0.710
9 5322044 Kg Lalok 53.766 0.197 0.121 0.705
9 5322044 Kg Lalok 53.766 0.197 0.121 0.705
10 5522047 JPS Kuala Krai 39.669 0.231 0.000 0.563
10 5522047 JPS Kuala Krai 39.669 0.231 0.000 0.563
11 5718033 Kg Jeli, Tanah Merah 72.173 0.196 0.360 0.703
12 5719001 Kg Durian Daun Lawang 51.161 0.193 0.063 0.745
13 5722057 JPS Machang 48.433 0.219 0.000 0.601
13 5722057 JPS Machang 48.433 0.219 0.000 0.601
14 5824079 Sg Rasau Pasir Putih 51.919 0.216 0.062 0.560
15 6019004 Rumah
Rumah Kastam
Kastam RR Pjg
Pjg 49.315 0.228 0.000 0.609
15 6019004 Setor JPS Kota Bharu 49.315 0.228 0.000 0.609
16 6122064 Setor JPS Kota Bharu 60.988 0.214 0.148 0.616
16 6122064 60.988 0.214 0.148 0.616
Negeri 1 2719001 Setor
Setor JPS
JPS Sikamat
Sikamat 52.823 0.167 0.159 0.811
1 2719001 52.823 0.167 0.159 0.811
Sembilan 2 2722202 Kg
Kg Sawah Lebar
Sawah Lebar K
K Pilah
Pilah 44.811 0.181 0.137 0.811
2 2722202 44.811 0.181 0.137 0.811
3 2723002 Sungai Kepis
Sungai Kepis 54.400 0.176 0.134 0.842
3 2723002 54.400 0.176 0.134 0.842
4 2725083 Ladang
Ladang New
New Rompin
Rompin 57.616 0.191 0.224 0.817
4 2725083 57.616 0.191 0.224 0.817
5 2920012 Petaling
Petaling K
K Kelawang
Kelawang 50.749 0.173 0.235 0.854
5 2920012 50.749 0.173 0.235 0.854
Negeri
Sembilan 38
Table 6.1c: Derived IDF Parameters of High ARI for Peninsular Malaysia (cont’d)

ID    

Melaka 1 2222001 Bukit Sebukor 95.823 0.169 0.660 0.947
2 2224038 Chin Chin Tepi Jalan 54.241 0.161 0.114 0.846
3 2321006 Ladang Lendu 72.163 0.184 0.376 0.900
Pulau 1 5204048 Sg Simpang Ampat 62.089 0.220 0.402 0.785
Pinang & 2 5302001 Tangki Air Besar Sg Png 67.949 0.181 0.299 0.736
Perlis 3 5302003 Kolam Tkgn Air Hitam 52.459 0.191 0.106 0.729
4 5303001 Rmh Kebajikan P Png 57.326 0.203 0.325 0.791
5 5303053 Komplek Prai 52.771 0.203 0.095 0.717
6 5402001 Klinik Bkt Bendera P Png 64.504 0.196 0.149 0.723
7 5402002 Kolam Bersih P Pinang 53.785 0.181 0.125 0.706
8 5404043 Ibu Bekalan Sg Kulim 57.832 0.188 0.245 0.751
9 5504035 Lahar Ikan Mati K Batas 48.415 0.221 0.068 0.692
10 6401002 Padang Katong, Kangar 57.645 0.179 0.254 0.826
Kedah 1 5507076 Bt. 27, Jalan Baling 52.398 0.172 0.104 0.788
2 5704055 Kedah Peak 81.579 0.200 0.437 0.719
3 5806066 Klinik Jeniang 59.786 0.165 0.203 0.791
4 5808001 Bt. 61, Jalan Baling 47.496 0.183 0.079 0.752
5 6103047 Setor JPS Alor Setar 64.832 0.168 0.346 0.800
6 6108001 Komppleks Rumah Muda 52.341 0.173 0.120 0.792
7 6206035 Kuala Nerang 54.849 0.174 0.250 0.810
8 6207032 Ampang Padu 66.103 0.177 0.284 0.842
9 6306031 Padang Sanai 60.331 0.193 0.249 0.829
Johor 1 1437116 Stor JPS Johor Baharu 59.972 0.163 0.121 0.793
2 1534002 Pusat Kem. Pekan Nenas 54.265 0.179 0.100 0.756
3 1541139 Johor Silica 59.060 0.202 0.128 0.660
Kelantan 4 1636001 Balai Polis Kg Seelong 50.115 0.191 0.099 0.763
5 1737001 SM Bukit Besar Setor 50.554 0.193 0.117 0.722
6 1829002 JPS B Pahat 64.099 0.174 0.201 0.826
7 1834124 Ladang Ulu Remis 55.864 0.166 0.174 0.810
8 1839196 Simpang Masai K. Sedili 61.562 0.191 0.103 0.701
9 1931003 Emp. Semberong 60.568 0.163 0.159 0.821
10 2025001 Pintu Kaw. Tg. Agas 80.936 0.187 0.258 0.890
11 2033001 JPS Kluang 54.428 0.192 0.108 0.740
12 2231001 Ladang Chan Wing 57.188 0.186 0.093 0.777
13 2232001 Ladang Kekayaan 53.457 0.180 0.094 0.735
14 2235163 Ibu Bekalan Kahang 52.177 0.186 0.055 0.652
15 2237164 Jalan Kluang- Mersing 56.966 0.190 0.144 0.637
16 2330009 Ladang Labis 45.808 0.222 0.012 0.713
17 2528012 Rmh. Tapis Segamat 45.212 0.224 0.039 0.711
18 2534160 Kg Peta Hulu Sg Endau 59.500 0.185 0.129 0.623
19 2636170 Setor JPS Endau 62.040 0.215 0.103 0.592
39
Table 6.1d: Derived IDF Parameters of High ARI for Peninsular Malaysia (cont’d)
Derived Parameters
State No.Station Station Name
ID
   

W. 1 3933001 Puchong Drop,K Lumpur 69.650 0.151 0.223 0.880
Perseku- 2 4131001 Ibu Pejabat JPS 61.976 0.145 0.122 0.818
tuan 3 4234109 Ibu Pejabat PS1 64.689 0.149 0.174 0.837
4 4332001 SK Taman Maluri 62.765 0.132 0.147 0.820
5 4529001 Ladang Edinburgh 63.483 0.146 0.210 0.830
6 4631001 Kg. Sungai Tua 64.203 0.152 0.250 0.844
7 4734079 SK Jenis Keb. Kepong 73.602 0.164 0.330 0.874
8 4832077 Ibu Bek. KM16, G ombak 66.328 0.144 0.230 0.859
9 4930038 Emp. Genting Kelang 70.200 0.165 0.290 0.854
10 5029034 Ibu Bek. KM11, Gombak 62.609 0.152 0.221 0.804
11 5128001 Kg. Kuala Seleh, H. Klg 61.516 0.139 0.183 0.837
12 5226001 Kg. Kerdas, Gombak 63.241 0.162 0.137 0.856
13 5328044 Air Terjun Sg. Batu 72.992 0.162 0.171 0.871
14 5331048 Genting Sempah 61.335 0.157 0.292 0.868
1 5426001
2 5428001
Table
3 6.2a: Derived IDF Parameters of Low ARI for Peninsular Malaysia
5524002
4 5725006
Derived Parameters
5 Station Name
State No. Station
6 4614001
ID
   

Perak 7 1 4726001
4010001 JPS Teluk Intan 65.185 0.368 0.255 0.846
8 2 4819027
4207048 JPS Setiawan 56.270 0.343 0.206 0.847
9 3 4915001
4311001 Pejabat Daerah Kampar 79.271 0.183 0.305 0.853
104 4923001
4409091 Rumah Pam Kubang Haji 47.832 0.353 0.104 0.802
115 5120025
4511111 Politeknik Ungku Umar 62.932 0.344 0.170 0.823
126 5216001
4807016 Bukit Larut Taiping 83.396 0.319 0.177 0.817
137 5320038
4811075 Rancangan Belia Perlop 57.491 0.320 0.203 0.870
148 5322044
5005003 Jln. Mtg. Buloh Bgn Serai 63.236 0.318 0.333 0.846
159 5522047
5207001 Kolam Air JKR Selama 67.050 0.316 0.226 0.808
Kelantan 16 5718033
10 5210069 Stesen Pem. Hutan 53.731 0.337 0.224 0.835
111 5719001
5411066 Lawin Kuala Kenderong 68.536 0.420 0.156 0.838
212 5722057
5710061 Dispensari Keroh 59.220 0.327 0.162 0.852
3 5824079
Selangor 4 1 2815001
6019004 JPS Sungai Manggis 57.350 0.276 0.169 0.867
5 2 2913001
6122064 Pusat Kwln. JPS T Gong 65.356 0.328 0.345 0.863
3 2917001
2719001 Setor JPS Kajang 62.956 0.329 0.130 0.827
4 3117070
2722202 JPS Ampang 69.173 0.249 0.192 0.837
5 3118102
2723002 SK Sungai Lui 68.459 0.304 0.204 0.873
6 3314001
2725083 Rumah Pam JPS P Setia 65.186 0.282 0.218 0.870
7 3411017
2920012 Setor JPS Tj. Karang 70.991 0.300 0.293 0.906
8 3416002 Kg Kalong Tengah 59.975 0.244 0.164 0.807
9 3516022 Loji Air Kuala Kubu Baru 66.888 0.280 0.349 0.833
10 3710006 Rmh Pam Bagan Terap 62.264 0.317 0.280 0.867
40
Table 6.2b: Derived IDF Parameters of Low ARI for Peninsular Malaysia

ID
   
Pahang 1 2630001 Sungai Pukim 63.978 0.391 0.256 0.872
2 2634193 Sungai Anak Endau 79.431 0.364 0.143 0.705
3 2828173 Kg Gambir 61.193 0.386 0.188 0.824
4 3026156 Pos Iskandar 59.990 0.349 0.226 0.877
5 3121143 Simpang Pelangai 64.965 0.323 0.300 0.900
6 3134165 Dispensari Nenasi 88.648 0.383 0.404 0.761
7 3231163 Kg Unchang 71.647 0.352 0.181 0.789
8 3424081 JPS Temerloh 62.208 0.353 0.351 0.837
9 3533102 Rumah Pam Pahang Tua 80.889 0.361 0.480 0.758
10 3628001 Pintu Kaw. Pulau Kertam 63.507 0.383 0.288 0.820
11 3818054 Setor JPS Raub 61.343 0.369 0.393 0.845
12 3924072 Rmh Pam Paya Kangsar 58.376 0.333 0.242 0.843
13 3930012 Sungai Lembing PCC Mill 77.000 0.453 0.570 0.813
14 4023001 Kg Sungai Yap 77.149 0.373 0.344 0.881
15 4127001 Hulu Tekai Kwsn.”B” 60.224 0.465 0.124 0.802
16 4219001 Bukit Bentong 67.613 0.271 0.246 0.866
17 4223115 Kg Merting 62.751 0.284 0.363 0.902
18 4513033 Gunung Brinchang 42.176 0.283 0.147 0.785
Terengganu
Terengganu 1
1 3933001
3933001 Hulu Jabor,
Hulu Jabor, Kemaman
Kemaman 74.805
74.805 0.217
0.217 0.253
0.253 0.728
0.728
2
2 4131001
4131001 Kg, Ban Ho, Kemaman
Kg, Ban Ho, Kemaman 68.666
68.666 0.316
0.316 0.116
0.116 0.697
0.697
3
3 4234109
4234109 JPS Kemaman
JPS Kemaman 75.826
75.826 0.239
0.239 0.381
0.381 0.730
0.730
4 4332001 Jambatan Tebak,
Jambatan Tebak, Kem.
Kem. 77.283 0.346 0.304 0.730
4 4332001 77.283 0.346 0.304 0.730
8 4832077 SK Jerangau 81.498 0.374 0.423 0.759
8 4832077 SK Jerangau 81.498 0.374 0.423 0.759
9 4930038 Kg Menerong, Hulu Trg 80.965 0.378 0.256 0.716
9
10 4930038
5029034 Kg
Kg Menerong,
Dura. Hulu Hulu
Trg Trg 80.965
62.786 0.378
0.350 0.256
0.110 0.716
0.664
10
11 5029034
5128001 Kg Dura.Gawi,
Sungai Hulu Hulu
Trg Trg 62.786
59.306 0.350
0.400 0.110
0.131 0.664
0.680
11
12 5128001
5226001 Sungai Gawi, Hulu
Sg Petualang, HuluTrg
Trg 59.306
51.786 0.400
0.297 0.131
0.070 0.680
0.659
12
13 5226001
5328044 Sg Petualang, Hulu
Sungai Tong, Setiu Trg 51.786
63.414 0.297
0.386 0.070
0.100 0.659
0.654
14
13 5331048
5328044 Sungai Tong,
Setor JPS Setiu
K Terengganu 67.027
63.414 0.284
0.386 0.263
0.100 0.669
0.654
15
14 5426001
5331048 Kg Seladang,
Setor Hulu Setiu
JPS K Terengganu 76.909
67.027 0.451
0.284 0.164
0.263 0.683
0.669
16
15 5428001
5426001 Kg Bt. Hampar,Hulu
Seladang, SetiuSetiu 57.946
76.909 0.249
0.451 0.038
0.164 0.600
0.683
17
16 5524002
5428001 SK Panchor, Setiu
Kg Bt. Hampar, Setiu 75.149
57.946 0.415
0.249 0.258
0.038 0.676
0.600
17 5524002 SK Panchor, Setiu 75.149 0.415 0.258 0.676
41
Table 6.2c: Derived IDF Parameters of Low ARI for Peninsular Malaysia (cont’d)

ID
   
Kelantan 1 4614001 Brook 49.731 0.316 0.198 0.792
2 4915001 Chabai 56.296 0.299 0.197 0.838
3 4923001 Kg Aring 70.265 0.381 0.242 0.819
5 5216001 Gob 47.465 0.283 0.153 0.785
6 5320038 Dabong 67.791 0.378 0.274 0.812
7 5322044 Kg Lalok 67.766 0.329 0.237 0.819
8 5522047 JPS Kuala Krai 63.069 0.468 0.310 0.783
11 5722057 JPS Machang 57.376 0.344 0.174 0.709
13 6019004 Rumah Kastam R Pjg 65.365 0.443 0.158 0.753
Negeri 14
1 4023001
2719001 Kg Sungai
Setor Yap
JPS Sikamat 77.149
60.423 0.279 0.269 0.881
0.854
Sembilan 15
2 4127001
2722202 Hulu
Kg TekaiLebar
Sawah Kwsn.”B”
K Pilah 60.224
49.323 0.373
0.272 0.344
0.216 0.802
0.850
16
3 4219001
2723002 Bukit
SungaiBentong
Kepis 67.613
61.334 0.465
0.254 0.124
0.329 0.866
0.872
17
4 4223115
2725083 Kg Merting
Ladang New Rompin 62.751
65.025 0.271
0.358 0.246
0.355 0.902
0.875
18
5 4513033
2920012 Gunung
Petaling Brinchang
K Kelawang 42.176
51.734 0.284
0.292 0.363
0.264 0.785
0.863
0.283 0.147
Melaka 1 2222001
3933001 Bukit Jabor,
Hulu Sebukor Kemaman 78.148
74.805 0.269 0.368 0.897
0.728
2 2224038
4131001 Kg, Ban Ho,
Chin Chin Tepi Kemaman
Jalan 66.059
68.666 0.336
0.217 0.330
0.253 0.891
0.697
3 2321006
4234109 JPS
LadangKemaman
Lendu 64.759
75.826 0.298
0.316 0.290
0.116 0.879
0.730
Jambatan Tebak, Kem.
Pulau Pinang 4
1 4332001
5204048 Sg Simpang
Rmh Pam Paya Ampat
Kempian
77.283
59.312 0.239
0.339 0.381
0.335 0.730
0.809
& Perlis 5
2 4529001
5302001 Tangki Air Besar Sg Png 65.279
71.748 0.346
0.293 0.304
0.293 0.667
0.778
Almuktafibillah Shah
6
3 4631001
5302003 Kolam Tkgn Air Hitam 81.886
56.115 0.364
0.298 0.148
0.178 0.746
0.763
SM Sultan Omar, Dungun
7
4 4734079
5303001 66.426 0.340 0.260 0.702
SK Jerangau P Png
Rmh Kebajikan 60.108 0.358 0.275 0.830
8
5 4832077
5303053 Kompleks PraiHulu
P Pinang 81.498
49.486 0.329
0.331 0.215
0.052 0.759
0.712
Kg Menerong, Trg
9
6 4930038
5402001 Klinik Bkt Hulu
Bendera 80.965 0.374 0.423 0.716
Kg Dura. Trg P Png 68.100 0.311 0.190 0.766
10
7 5029034
5402002 Sungai
Kolam Bersih
Gawi, Hulu
P Pinang
Trg 62.786
62.753 0.378
0.269 0.256
0.249 0.664
0.776
11
8 5128001
5504035 Sg Petualang,
Lahar Ikan MatiHulu Trg
K Batas 59.306
60.860 0.350
0.337 0.110
0.232 0.680
0.798
12
9 5226001
6401002 Sungai Tong, Setiu
Padang Katong, Kangar 51.786
52.151 0.400
0.357 0.131
0.158 0.659
0.786
Kedah 1 5507076 Bt. 27,
Kg Jalan Baling
Seladang, Hulu Setiu 62.761 0.258 0.304 0.835
14 5331048 67.027 0.386 0.100 0.669
2 5704055 Kedah
Kg Peak
Bt. Hampar, Setiu 58.596 0.339 0.064 0.661
15 5426001 76.909 0.284 0.263 0.683
3 5806066 Klinik
SK JeniangSetiu
Panchor, 67.120 0.382 0.238 0.823
16 5428001 57.946 0.451 0.164 0.600
4 5808001 Bt. 61, Jalan Baling 56.399 0.388 0.252 0.803
17 5524002 75.149 0.249 0.038 0.676
5 6103047 Setor JPS Alor Setar 67.641 0.334 0.274 0.828
0.415 0.258
6 6108001 Komppleks Rumah Muda 58.404 0.278 0.234 0.829
7 6206035 Kuala Nerang 62.960 0.308 0.359 0.859
8 6207032 Ampang Padu 70.997 0.293 0.382 0.863
9 6306031 Padang Sanai 63.615 0.313 0.309 0.852
42
Table 6.2d: Derived IDF Parameters of Low ARI for Peninsular Malaysia (cont’d)

ID
   
Pulau Pinang 1 5204048 Sg Simpang Ampat 59.312 0.339 0.335 0.809
& Perlis 2 5302001 Tangki Air Besar Sg Png 71.748 0.293 0.293 0.778
3 5302003 Kolam Tkgn Air Hitam 56.115 0.298 0.178 0.763
4 5303001 Rmh Kebajikan P Png 60.108 0.358 0.275 0.830
5 5303053 Kompleks Prai P Pinang 49.486 0.331 0.052 0.712
6 5402001 Klinik Bkt Bendera P Png 68.100 0.311 0.190 0.766
7 5402002 Kolam Bersih P Pinang 62.753 0.269 0.249 0.776
8 5504035 Lahar Ikan Mati K Batas 60.860 0.337 0.232 0.798
9 6401002 Padang Katong, Kangar 52.151 0.357 0.158 0.786
Johor 10
1 5719001
1437116 Stor JPS Johor Baharu 67.240
73.679 0.277 0.182
0.293 0.862
11
2 5722057
1534002 Pusat Kem. Pekan Nenas 57.376
62.651 65
0.323 0.174
0.156 0.753
0.821
12
3 5824079
1541139 Kg Durian
Johor SilicaDaun Lawang 68.508
79.536 0.344
0.336 0.202
0.295 0.709
0.810
13
4 6019004
1636001 JPS Machang
Balai Polis Kg Seelong 65.365
61.212 0.408
0.337 0.158
0.238 0.700
0.843
14
5 4023001
1737001 Sg
SMRasau Pasir Setor
Bukit Besar Putih 77.149
61.351 0.443
0.303 0.203 0.753
0.824
15
6 4127001
1829002 Rumah Kastam R Pjg
JPS B Pahat 60.224
62.158 0.306 0.344
0.142 0.881
0.825
16
7 4219001
1834124 Kg
Ladang UluYap
Sungai Remis 67.613
59.171 0.373
0.294 0.124
0.185 0.802
0.838
17
8 4223115
1839196 Hulu Tekai Kwsn.”B”
Simpang Masai K. Sedili 62.751
71.795 0.465
0.268 0.246
0.186 0.866
0.807
18
9 4513033
1931003 Bukit
Emp. Bentong
Semberong 42.176
66.885 0.271
0.355 0.363
0.211 0.902
0.838
10 2025001 Kg Merting
Pintu Kaw. Tg. Agas 77.772 0.284
0.310 0.147
0.281 0.785
0.879
1
11 3933001
2033001 Gunung
JPS KluangBrinchang 74.805
- 0.283
- - -
2
12 4131001
2231001 Ladang Chan Wing 68.666
66.144 0.324 0.253
0.178 0.728
0.849
3
13 4234109
2232001 Hulu
LadangJabor, Kemaman
Kekayaan 75.826
66.754 0.217
0.308 0.116
0.227 0.697
0.838
4 4332001 Kg, Ban Ho, Kemaman 77.283 0.316 0.381 0.730
14 2235163 Ibu Bekalan Kahang 62.339 0.279 0.163 0.739
5 4529001 JPS Kemaman 65.279 0.239 0.304 0.730
15 2237164 Jalan Kluang-
Jambatan Mersing
Tebak, Kem. 73.236 0.343 0.220 0.773
6
16 4631001
2330009 Ladang Labis 81.886
65.222 0.346
0.395 0.148
0.235 0.667
Rmh Pam Paya Kempian 0.846
7
17 4734079
2528012 Rmh. Tapis Segamat 66.426
63.689 0.364
0.382 0.260
0.259 0.746
Almuktafibillah Shah 0.871
8
18 4832077
2534160 Kg 81.498 0.340 0.215 0.702
SM Sultan Omar,Endau
Peta Hulu Sg Dungun 69.958 0.350 0.181 0.706
9
19 4930038
2636170 Setor JPS Endau 80.965
77.630 0.329
0.399 0.423
0.250 0.759
SK Jerangau 0.693
10 5029034 62.786 0.374 0.256 0.716
W. 1 3015001 Puchong Drop,K
Kg Menerong, Lumpur
Hulu Trg 68.587 0.352 0.170 0.849
11 5128001 59.306 0.378 0.110 0.664
Perseku- 2 3116004 Ibu Pejabat
Kg Dura. JPS
Hulu SK
Trg 65.992 0.286 0.160 0.834
12 5226001 Sungai 51.786 0.350 0.131 0.680
tuan 3 3116005 Taman Gawi,
Maluri Hulu
LadangTrg 74.451 0.266 0.312 0.861
13 5328044 Sg Petualang, Hulu Trg 63.414 0.400 0.070 0.659
4 3116006 Edinburgh Kg. 64.503 0.275 0.181 0.833
14 5331048 Sungai 67.027 0.297 0.100 0.654
5 3216001 Sungai Tong,
Tua Setiu 62.940 0.258 0.199 0.837
6 3216004 SK Jenis Keb. Kepong Ibu 69.788 0.296 0.167 0.851
16 5428001 Kg Seladang, Hulu Setiu 57.946 0.284 0.164 0.683
7 3217001 Bek.
Kg Bt.KM16, Gombak
Hampar, Setiu Emp. 66.069 0.257 0.229 0.840
17 5524002 75.149 0.451 0.038 0.600
8 3217002 Genting Kelang
SK Panchor, Setiu 66.258 0.262 0.242 0.845
0.249 0.258 0.676
9 3217003 Ibu Bek. KM11, Gombak 73.954 0.298 0.324 0.824
0.415
10 3217004 Kg. Kuala Seleh, H. Klg 64.318 0.234 0.182 0.865
11 3217005 Kg. Kerdas, Gombak 68.853 0.298 0.202 0.882
12 3317001 Air Terjun Sg. Batu 75.935 0.248 0.266 0.867
13 3317004 Genting Sempah 55.393 0.282 0.184 0.835
43
CHAPTER 7
DEVELOPING THE REGION OF TEMPORAL STORM PROFILES BY MEANS OF
CLUSTERING ANALYSIS
7.1 INTRODUCTION
Cluster analysis is a multivariate analysis technique or procedure in order to organize

information of variables to form relatively homogeneous groups, or “cluster”. There are
several types of cluster analysis such as K- Means Cluster Analysis and Hierarchical Cluster
Analysis.
In this study, regions were formed by K-Means Cluster Analysis method to identify
homogeneous groups of cases that based on selected of site characteristics by using an
algorithm that can handle large numbers of cases. A data vector is associated with each site,
and sites are partitioned into groups according to the similarity of their data vectors that can
include at-site statistics, site characteristics or combi nation of two. But, in this clustering
analysis, site characteristics only selected, and did not involve any at-site statistics
measuring the shape of the frequency distribution of rainfall. When cluster analysis is based
on site characteristics, the at-site statistics are available for use as the basis of an
independent test of the homogeneity of the final regions.
Most clustering algorithms measure similarity by the reciprocal of Euclidean distance in a

space of site characteristics. This distance measure is affected by the scale of measurement
or rescale of the site characteristics in order to have same amount of variability, as
measured by their range or standard deviation across all of the sites in the data set. In
determining clusters, it may not be appropriate when the rescaling gives equal weight to
each site characteristics that have greater influence on the form of the frequency distribution
and it should be given greater weight in the clustering. There is no assumption that there are
distinct clusters of sites that satisfy the homogeneity condition and no ‘correct’ number of
clusters, instead a balance must be sought between using regions that are too small or too
large. The output from the cluster analysis need not be final because some subjective
adjustment can be done in order to improve the physical coherence of the regions and to
reduce the heterogeneity of the regions that measured by the heterogeneity test, H.
44
The clustering analysis is aimed to form relatively homogenous ‘groups’ or ‘regions’ that are
able to accommodate and creates new regions for the storm profiles or storm temporal
pattern as in existence HP1 (1982) divided into the region of East Coast and West Coast.
7.2 DATA AVAILABILITY AND ACQUISITION
The numerical analysis was performed using 1-day duration rainfall of 56 selected automatic
recording rainfall stations maintained by DID. Pertinent details on the rainfall station ID and
length of records for each 56 automatic rainfall stations throughout Peninsular Malaysia is
tabulated in Table 7.1. While in the environmental application study, five variables of site
characteristics were chosen such as latitude, longitude, elevation, mean annual rainfall and
the ratio of the minimum average two-month rainfall to maximum average two-month rainfall.
The available data for the site characteristics that used for clustering analysis is tabulated in
Table 7.2.
‎
7.3 DATA SCREENING
Data screening represents an important step in all statistical computations. The first
important step of any statistical data analysis is to check that the data are appropriate for the
analysis. Before carrying out the frequency analysis, the data integrity check was carried out
where there should not be too long gaps in the data records in each year.
In this study, we stated that more than 10% yearly gaps are discarded from the analysis.
Perhaps, a check of each site’s data separately is needed in order to identify outlying values
and repeated value, which may be due to error of recording data.
7.4 FORMATION OF REGION BY CLUSTERING ANALYSIS
Identifying clusters in a space of site characteristics formed regions. At-site statistics are
used to assess the homogeneity of the regions that are formed in the clustering procedure,
and the validity of this assessment is compromised if the same data are used both to form
regions and to test their homogeneity.
In this study, five (5) variables of site characteristics were chosen such as site latitude, site
longitude, site elevation, mean annual rainfall and the ratio of the minimum average two-
month rainfall to maximum average two-month rainfall. The variables need to be transformed
45
in order to get comparable ranges because the standard methods of cluster analysis are
very sensitive to such scale differences. All the variables were rescaled so that their values
lay between 0 and 1. Table 7.3 shows the transformations from the five site characteristics
to the variables used in cluster analysis. For this study, some combi nations of this site
characteristics or variables as shown in Table 7.4 would be done in order to see the impact
through the result of clustering process.
46
Table 7.1: Summary of Selected 56 Automatic Rainfall Stations for Peninsular Malaysia
Station Data Period No. of Station Data Period No. of
No Station Name No Station Name
ID Record Selected Years ID Record Selected Years
1 6401002 Padang Katong at Kangar 741103- 750101- 25 15 4409091 Rumah Pam Kubang Haji, Perak 700627- 710101- 29
Perlis 1010104 1001231 1010414 1001231
2 6402008 Ngolang at Perlis 830220- 840101- 16 16 4209093 JPS Telok Sena, Perak 700703- 710101- 29
1010103 1001231 1010414 1001231
3 6306031 Padang Sanai, Kedah 700701- 710101- 29 17 4010001 JPS Telok Intan, Perak 700701- 710101- 29
1010107 1001231 1010417 1001231
4 6207032 Ampang Pedu, Kedah 700629- 710101- 29 18 3516022 Logi Air Kuala Kubu Baru, 700629- 710101- 29
1010107 1001231 Selangor 1010102 1001231
5 6206035 Kuala Nerang at Kedah 700627- 710101- 29 19 3416002 Kg. Kalong Tengah (AB), 780830- 790101- 21
1010107 1001231 Selangor 1010102 1001231
6 6108001 Kompleks Rumah Muda, 741215- 710101- 29 20 3411017 Stor JPS Tanjung Karang, 700629- 710101- 29
Kedah 1010102 1001231 Selangor 1010103 1001231
7 5808001 Bt 61 Jalan Baling, Kedah 740929- 750101- 25 21 3317004 Genting Sempah, Wilayah 741001- 750101- 25
1010103 1001231 Persekutuan 1010116 1001231
8 5704055 Kedah Peak, Kedah 750102- 750101- 25 22 3314001 Rumah Pam Paya Setia, 740102- 740101- 26
1010101 1001231 Selangor 1010103 1001231
9 5504035 Lahar Ikan Mati at Pulau 700701- 710101- 29 23 3118102 Sek. Keb. Lui at Selangor 700723- 710101- 29
Pinang 1010115 1001231 1010404 1001231
10 5710061 Dispensari Kroh, Perak 400101- 700101- 30 24 2917001 Stor JPS Kajang, Selangor 750402- 760101- 24
1010503 1001231 1010102 1001231
11 5210069 Stesen Pemeriksaan HUtan 700629- 710101- 29 25 2723002 Sg. Kepis at Masjid site 2, Negeri 770529- 780101- 22
Lawin, Perak 1010619 1001231 Sembilan 1010605 1001231
12 5005003 Jalan Matang Buloh Bagan 740401- 750101- 25 26 2719001 Stor JPS Sikamat Seremban, 700626- 710101- 29
Serai, Perak 1010601 1001231 Negeri Sembilan 1010606 1001231
13 4708084 Ibu Bekalan Talang, Kuala 700704- 710101- 29 27 2321006 Ladang Lendu, Melaka 740511- 750101- 25
Kangsar, Perak 1010619 1001231 1010507 1001231
14 4511111 Politeknik Ungku Omar, Ipoh, 720501- 730101- 27 28 2224038 Chin Chin (Tepi Jalan), Melaka 700702- 710101- 29
Perak 1010418 1001231 1010419 1001231
47
Table 7 .1: Summary of Selected 56 Automatic Rainfall Stations for Peninsular Malaysia (cont’d)
Station Data Period No. of Station Data Period No. of

No Station Name No Station Name
ID Record Selected Years ID Record Selected Years
29 230009 Ladang Sg. Labis at Labis 700629- 710101- 29 43 5331048 Stor JPS Kuala Terengganu 700629- 710101- 29
Johor 1010101 1001231 1010528 1001231
30 2033001 Stor Baru JPS Kluang, Johor 761205- 770101- 22 44 5029034 Kg Dura Terengganu 710704- 720101- 28
1000703 991231 1010528 1001231
31 2023001 Pintu Kawalan Tg. Agas, Muar 740810- 750101- 25 45 4930038 Kg Menerong Terengganu 710811- 700101- 30
Johor 1010101 1001231 1010527 1001231
32 1839196 Simpang Mawai, Kuala Sedeli, 700630- 710101- 29 46 4929001 Kg Embong Sekayu Ulu 750411- 760101- 24
Johor 1010102 1001231 Terengganu 1010526 1001231
33 1737001 Sek. Men. Bukit Besar at Kota 740727- 750101- 25 47 4234109 JPS Kemaman, Terengganu 700628- 710101- 29
Tinggi, Johor 1010101 1001231 1010606 1001231
34 1732004 Parit Madironon at Site 4, Johor 781011- 790101- 21 48 4513033 Gunung Berinchang, 750701- 760101- 24
1010101 1001231 Cameron Highlands, Phg 1010202 1001231
35 1534002 Pusat Kemajuan Perikanan, 781030- 790101- 21 49 4023001 Kg. Sg Yap, Pahang 731108- 740101- 26
Pkn Nanas, Jhr 1010101 1001231 1010119 1001231
36 5824079 Sg. Rasau Pasir Puteh, 700629- 710101- 25 50 4019001 JKR Benta, Benta, Pahang 770103- 780101- 22
Kelantan 970225 961231 1010207 1001231
37 5718002 Air Lanas, Kelantan 800714- 810101- 19 51 3924072 Rumah Pam Paya Kangsar, 700629- 710101- 29
1010101 1001231 Pahang 1010104 1001231
38 5320038 Dabong at Kelantan 7190913- 720101- 27 52 3818054 Stor JPS Raub, Pahang 700701- 710101- 29
1010109 9912311 1010109 1001231
39 4923001 Kg Aring at Kelantan 741116- 750101- 24 53 3717001 Bukit Peninjau at Pahang 751014- 760101- 24
1000901 991231 1010104 1001231
40 5725006 Klinik Kg Raja, Besut, 700704- 720101- 28 54 3533102 Rumah Pam Pahang Tua, 700704- 730101- 27
Terengganu 1010524 1001231 Pekan Pahang 1010402 1001231
41 5428002 Klinik Chalok Barat S1 780202- 790101- 21 55 3519125 Kuala Marong, benta, Pahang 700629- 710101- 29
Terengganu 1010527 1001231 1010109 1001231
42 5428001 Kg Batu Hampar at Chalok Site 780202- 790101- 21 56 3231163 Kg. Unchang at Pahang 740306- 750101- 25
1 Terengganu 1000529 1001231 1010207 1001231
48
7.2: Available Data of Site Characteristics
No
No Station ID
Station ID Long (Deg)
Long (Deg) Lat (Deg)
Lat (Deg) Elev (M
Elev (M)) Mean
Mean (Mm)
(Mm) Ratio
Ratio No
No Station ID
Station ID Long
Long (Deg)
(Deg) Lat
Lat (Deg)
(Deg) Elev
Elev(M
(M)) Mean
Mean (Mm)
(Mm) Ratio
1 6401002 100.19 6.45 2.6 2012 0.1238 29 2330009 103.02 2.38 32.0 1975 0.6000
2 6402008 100.25 6.48 7.0 1405 0.1885 30 2033001 103.33 2.01 40.0 2050 0.5723
3 6306031 100.77 6.24 34.8 1614 0.119 31 2025001 102.58 2.02 3.0 1991 0.4033
4 6207032 100.69 6.34 61.0 1946 0.1416 32 1839196 103.97 1.85 14.0 2612 0.5253
5 6206035 100.61 6.25 78.3 1701 0.1397 33 1737001 103.72 1.76 45.1 2127 0.5539
6 6108001 100.85 6.11 152.4 2084 0.1313 34 1732004 103.27 1.71 40.0 2167 0.6253
7 5808001 100.89 5.88 128.9 2406 0.1406 35 1534002 103.49 1.52 40.0 2376 0.7473
8 5704055 100.44 5.8 1063.8 3193 0.1347 36 5824079 102.42 5.83 3.0 2694 0.1324
9 5504035 100.43 5.53 3.7 1973 0.2344 37 5718002 101.89 5.85 74.1 3857 0.2653
10 5710061 101.00 5.71 313.0 2168 0.2162 38 5320038 102.02 5.38 76.2 2182 0.2482
11 5210069 101.06 5.3 103.0 1686 0.2811 39 4923001 102.31 5.83 91.1 2714 0.2655
12 5005003 100.55 5.01 2.0 2037 0.5159 40 5725006 102.57 5.8 5.1 2705 0.1242
13 4708084 100.89 4.78 50.1 1491 0.5861 41 5428002 102.82 5.41 33.0 3682 0.2095
14 4511111 101.13 4.59 61.0 2327 0.4813 42 5428001 102.82 5.45 10.0 3211 0.1745
15 4409091 100.90 4.46 23.2 1731 0.5267 43 5331048 103.13 5.32 87.0 2834 0.1541
16 4209093 100.9 4.26 12.8 2098 0.59 44 5029034 102.94 5.07 55.0 3187 0.2228
17 4010001 101.04 4.02 14.9 2442 0.4466 45 4930038 103.06 4.94 15.0 3509 0.2268
18 3516022 101.45 3.58 143.9 2488 0.4450 46 4929001 102.97 4.95 70.0 4646 0.2743
19 3416002 101.66 3.44 70.1 2595 0.3621 47 4234109 103.42 4.23 5.5 2783 0.245
20 3411017 101.17 3.42 2.4 1690 0.5105 48 4513033 101.38 4.52 2031.2 2398 0.4636
21 3317004 101.77 3.37 818.1 2242 0.4016 49 4023001 101.33 4.03 76.2 1636 0.4903
22 3314001 101.41 3.37 17.1 2029 0.5603 50 4019001 102.00 4.03 121.9 2033 0.5597
23 3118102 101.94 3.16 85.0 2492 0.4684 51 3924072 102.43 3.90 45.7 1656 0.4074
24 2917001 101.80 2.99 39.0 2353 0.5313 52 3818054 101.85 3.81 228.6 1942 0.5718
25 2723002 102.32 2.70 121.9 1709 0.5468 53 3717001 101.80 3.72 1323.1 2243 0.4213
26 2719001 101.96 2.74 121.9 1933 0.4754 54 3533102 103.36 3.57 7.0 2519 0.2585
27 2321006 102.19 2.36 33.0 1762 0.458 55 3519125 101.92 3.51 91.5 1876 0.4833
28 2224038 102.49 2. 29 8.6 1628 0.4807 56 3231163 103.20 3.30 40.0 2114 0.3669
49
Table 7.3: Transformation of Site Characteristics
Site Characteristic, X Cluster Variable, Y

Longitude (deg) Y= X / 90
Longitude (deg) Y= X / 150
Elevation (deg) Y= X / 10000
Mean Annual Rainfall (mm) Y = X / 100
Ratio of minimum average two -month rainfall to maximum Y= X
average two-month rainfall
Table 7.4: Site Characteristics Combinations of Cluster Analysis
Site Characteristics Combinations Combination Code

Latitude + Longitude + Elevation A1
Latitude + Longitude + Mean of Rain A2
Latitude + Longitude + Ratio A3
Latitude + Longitude + Elevation + Mean of Rain A4
Latitude + Longitude + Elevation + Mean of Rain +
Ratio A5
Clustering analysis was performed by Ward’s method where the distance between two
clusters is the sum of squares between the two clusters summed over all the variables. This
is an “agglomerative hierarchical” clustering procedure.
The method tends to join clusters that contains a small number of sites and strongly biased
in favour of producing clusters containing approximately equal number of sites.
This method is based on the Euclidean distances and also sensitive to redundant information
that may be contained in the variables as well as to the scale of the variables being clustered
(Fovell and Fovell, 1993). Initially each site is a cluster by itself, and clusters are then
merged one by one until all sites belong to a single cluster.
The assignment of sites to clusters can be determined for any number of clusters and there
is no formal measure of an “optimal” number of clusters where the choice is subjective.
50
7.5 RESULTS OF CLUSTERING ANALYSIS
In this study, for Peninsular Malaysia that consists of 56 selected automatic rainfall stations,
it is decided that four clusters would be an appropriate number.
The clusters obtained by Ward’s method were adjusted by K –means algorithm of Hartigan
and Wong (1979), which yield clusters that were little more compact in the space of cluster
variables. The result of heterogeneity measures showed that the best combination of site
characteristics is found to be group A5 where cluster no.1, 2, 3 and 4 were classified as
acceptably homogeneous (H=0.72), possibly heterogeneous (H=2.13b), acceptably
homogeneous (H=-1.23a) and possibly heterogeneous (H=1.48b) respectively.
Summary of cluster membership for group A5 is given in Table 7.5 and summary of cluster
centre is tabulated in Table 7.6. Figure 7.1 shows final region created and region no.4 was a
distinct region as it is located and represents mountainous area; meanwhile Region No.5
was specifically created for accommodating an urban area.
Table ‎7.5: Summary of Clustering Analysis of A5 Combination
Total
Cluster Station No.
Members
1 11 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 38
2 29 12, 13, 14, 15, 16, 17, 18, 19, 20, 22, 23, 24, 25, 26,
27, 28, 29, 30, 31, 32, 33, 34, 35, 49, 50, 51, 52, 55,
56
3 12 36, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 54
4 4 8, 21, 48, 53
Table ‎7.6: Summary of Cluster Centers of A5 Combination
Cluster Latitude Longitude Elevation Mean Ratio
1 100.76 5.97 65.47 1914.66 0.18
2 102.07 3.21 45.3 2035.24 0.51
3 102.81 5.19 29.36 3168.66 0.21
4 101.35 4.35 1272.02 2504.65 0.36
51
Figure ‎7.1: The “Region” Created By Means of the Clustering Analysis Approach
52
CHAPTER 8
DEVELOPING THE DESIGN STORM PROFILES (TEMPORAL STORM)
8.1 INTRODUCTION
A variety of methods to generate design storm hyetograph exist in the literature, but as cited
by Veneziano and Villani (1999) suggested that the most practical methods can be divided
into three categories:
1. Specification of simple geometrical shapes anchored to a single point of the IDF

curve/relationships – the traditional approach uses a rectangular design hyetograph
with duration equal to the concentration time of the basin and rainfall rate derived
from the IDF relationship (i.e. frequently used in a combination of the rational method
as shown in Hydrological Procedure No. 5)
2. Use of the entire IDF curves to specify a rainfall profile that reflects the entire IDF
relationship and not only the IDF value at a single duration.
3. Use of standardized profiles obtained directly from historic rainfall records which is
able to reduce a rainfall event to a dimensionless curve by dividing time by the total
duration of the event and cumulative rainfall by the total rainfall volume (i.e. as
appeared in the existing procedure and has directly been adopted in the MSMA).
Based on the categories mentioned, the last two methods are recognized as the best choice
to adopt, but to continue as in the existing HP1 (1982), the method of standardized profiles
is selected
Use of standardized rainfall profiles is quite common in the hydrology literature. Prodanovic
and Simonovic (2004) cited that the most popular are those of Huff (1967) and SCS (1986).
Standardized profiles, also known as mass curve, transform a precipitation event to a
dimensionless curve with cumulative fraction of storm time on the horizontal and cumulative
fraction of total rainfall on the vertical axis. Since rainfall records are highly variable
because of the uncertainty of what actually constitutes a rainfall event, as well as
randomness of the rainfall phenomena itself, the standardized profiles method must use
some sort of temporal smoothing, or assemble averaging. In the Soil Conservation System
(SCS) hypothetical storm method uses standardized rainfall intensities arranged to
maximize the peak runoff at a given storm depth. Although primarily has been used for the
design of small dams, it has been applied in many rural and urban areas. The required
input parameters are distribution type and total storm depth.
53
The Huff method has features similar to the SCS method, except that it gives the user more
flexibility – restrictions are not placed on storm duration. The required input parameters are
quantile distribution, storm duration, d and total storm depth, D. The main appeal of this
category of methods of design rainstorm/rainfall intensity hyetographs is that the resulting
output is based on the actual data of intense regional rainfall. Furthermore, as the methods
do not rely on IDF data, rainfall exceeding return period of 100-years can be easily used, if
available. In the context of available records of rainfall data managed by DID in Peninsular
Malaysia, however, it apparently shows that the maximum length of historic rainfall records
are mostly found to be about 30-40 years. Under these circumstances, the mentioned
methodology probably has limited ability for producing design hyetograph at high return
period for more than 50 year. This method also requires large sample data sets for the
construction of regional profiles, which in turn generates large uncertainties. Therefore,
temporal smoothing needs to be performed and this might overlook some of the important
features of rainfall at the locality interest.
8.2 DERIVATION OF STORM PROFILES (TEMPORAL PATTERN)
About 441 number of storms was considered in the analysis, with durations ranging from
0.25-hr to 72-hrs. Generally, the storms were selected and identified from 5 nos. of annual
maximum rainfall intensity at each state. However, due to lack of station density, Melaka and
Negeri Sembilan, and Pulau Pinang and Perlis were grouped as two distinct areas. The
required input parameters are storm duration and total storm depth where the mass curves
of selected duration were constructed and temporal smoothing has been carried out by
means of mass curve averaging. As reported in Chapter 7, the clustering analysis has
produced 4 distinct regions throughout Peninsular Malaysia and in addition, Federal Territory
of Kuala Lumpur region was specifically created. Therefore, the regional storm profiles
basically refer to:
1. Northeast Region – Kelantan, Terengganu and Northen Pahang

2. Central and Southern Region – Pahang (except Northern Pahang), Selangor, Negeri
Sembilan, Melaka and Johor;
3. Northwest Region – Perak, Kedah, Pulau Pinang and Perlis;
4. Mountainous Region – covers an area of high altitude which is no longer recognized
by administrative boundary;
5. Urban Region – specifically for Federal Territory of Kuala Lumpur
54
Thus, final regional storm profiles were obtained by means of averaging the mass curves
from the stated states in each derived region. With the newly created regions as stated
above, the current East and West Coast region of HP1 (1982) is no longer usable and
appropriate. Figure 7:1 depicts the derived region.
8.3 SUMMARY OF RESULTS
Based on the final regions created, actual storm profiles for each region are summarized in
Table 8.1 - Table 8.5. However, the normalization (standardization) of actual storm profile is
produced by generating accurate peak discharge or runoff volumes estimation. Table 8.6 –
Table 8.10 show normalized temporal storm profile for the region of 1 to 5. Example of storm
profile block diagrams is illustrated in Figure 8:1 - Figure 8:2 associated with storm duration.
Table ‎8.1: Derived Temporal Pattern for Region 1 – Terengganu, Kelantan and
Northern Pahang
No. of Duration
Block
15-min 30-min 60-min 180-min 6-hr 12-hr 24-hr 48-hr 72-hr
1 0.316 0.202 0.091 0.071 0.057 0.064 0.025 0.029 0.022
2 0.368 0.193 0.060 0.060 0.063 0.070 0.027 0.046 0.020
3 0.316 0.161 0.062 0.059 0.071 0.073 0.050 0.049 0.021
4 0.100 0.054 0.060 0.069 0.084 0.048 0.058 0.029
5 0.133 0.061 0.061 0.059 0.084 0.058 0.054 0.030
6 0.211 0.115 0.080 0.073 0.097 0.058 0.028 0.033
7 0.082 0.078 0.086 0.086 0.036 0.019 0.052
8 0.087 0.100 0.067 0.070 0.046 0.029 0.053
9 0.087 0.120 0.082 0.099 0.044 0.028 0.048
10 0.097 0.110 0.119 0.083 0.039 0.060 0.038
11 0.120 0.132 0.130 0.106 0.057 0.053 0.036
12 0.084 0.069 0.123 0.083 0.049 0.055 0.041
13 0.056 0.038 0.042
14 0.050 0.037 0.047
15 0.043 0.040 0.059
16 0.068 0.044 0.053
17 0.048 0.027 0.038
18 0.050 0.033 0.037
19 0.042 0.030 0.033
20 0.028 0.046 0.067
21 0.019 0.048 0.056
22 0.016 0.065 0.058
23 0.022 0.048 0.055
24 0.022 0.034 0.030
55
Table ‎8.2: Derived Temporal Pattern for Region 2 - Johor, Negeri Sembilan, Melaka,
Selangor and Pahang
No. of Duration
Block
1 0.255 0.103 0.103 0.042 0.044 0.041 0.024 0.026 0.023
2 0.376 0.124 0.110 0.080 0.090 0.045 0.040 0.022 0.035
3 0.370 0.126 0.046 0.097 0.081 0.048 0.031 0.013 0.016
4 0.130 0.063 0.129 0.083 0.056 0.032 0.012 0.016
5 0.152 0.059 0.151 0.090 0.046 0.022 0.025 0.033
6 0.365 0.088 0.128 0.081 0.106 0.020 0.045 0.024
7 0.069 0.079 0.115 0.146 0.024 0.036 0.022
8 0.053 0.062 0.114 0.124 0.039 0.041 0.049
9 0.087 0.061 0.106 0.116 0.033 0.059 0.038
10 0.057 0.053 0.085 0.127 0.054 0.058 0.027
11 0.060 0.054 0.074 0.081 0.050 0.066 0.047
12 0.153 0.063 0.037 0.064 0.047 0.068 0.067
13 0.031 0.062 0.057
14 0.029 0.059 0.051
15 0.029 0.051 0.036
16 0.039 0.022 0.049
17 0.042 0.026 0.048
18 0.093 0.022 0.049
19 0.052 0.026 0.068
20 0.035 0.056 0.043
21 0.083 0.040 0.079
22 0.065 0.093 0.050
23 0.057 0.039 0.043
24 0.028 0.032 0.030
56
Table ‎8.3: Derived Temporal for Region 3 - Perak, Kedah, P Pinang and Perlis
No. of Duration
Block
1 0.215 0.141 0.077 0.085 0.047 0.040 0.048 0.021 0.044
2 0.395 0.173 0.064 0.100 0.041 0.046 0.033 0.045 0.026
3 0.390 0.158 0.098 0.086 0.070 0.036 0.034 0.060 0.063
4 0.161 0.087 0.087 0.099 0.066 0.033 0.086 0.074
5 0.210 0.068 0.087 0.081 0.066 0.034 0.039 0.021
6 0.158 0.074 0.088 0.113 0.060 0.036 0.028 0.050
7 0.078 0.100 0.121 0.081 0.031 0.020 0.058
8 0.072 0.100 0.099 0.092 0.044 0.026 0.049
9 0.075 0.085 0.078 0.119 0.036 0.015 0.008
10 0.104 0.063 0.076 0.114 0.027 0.014 0.031
11 0.106 0.060 0.129 0.113 0.023 0.028 0.030
12 0.099 0.059 0.045 0.166 0.035 0.017 0.044
13 0.041 0.057 0.025
14 0.053 0.039 0.022
15 0.039 0.044 0.044
16 0.055 0.035 0.024
17 0.032 0.038 0.024
18 0.031 0.052 0.025
19 0.039 0.069 0.023
20 0.080 0.046 0.070
21 0.076 0.056 0.078
22 0.044 0.046 0.081
23 0.042 0.045 0.028
24 0.056 0.073 0.058
57
Table ‎8.4: Derived Temporal Pattern for Region 4 - Mountainous Area
No. of Duration
Block
1 0.146 0.117 0.028 0.055 0.054 0.120 0.026 0.018 0.116
2 0.177 0.121 0.028 0.098 0.040 0.041 0.007 0.057 0.011
3 0.677 0.374 0.066 0.132 0.041 0.065 0.023 0.037 0.005
4 0.107 0.079 0.164 0.062 0.052 0.050 0.033 0.006
5 0.130 0.073 0.197 0.020 0.056 0.055 0.047 0.011
6 0.152 0.064 0.169 0.019 0.048 0.048 0.081 0.000
7 0.106 0.095 0.045 0.052 0.023 0.018 0.014
8 0.058 0.027 0.016 0.157 0.142 0.027 0.018
9 0.280 0.019 0.060 0.058 0.049 0.024 0.096
10 0.042 0.019 0.171 0.059 0.060 0.007 0.035
11 0.052 0.019 0.390 0.038 0.009 0.003 0.060
12 0.119 0.006 0.082 0.253 0.112 0.000 0.039
13 0.034 0.002 0.028
14 0.040 0.080 0.016
15 0.001 0.066 0.005
16 0.002 0.007 0.009
17 0.000 0.031 0.065
18 0.026 0.036 0.028
19 0.008 0.026 0.023
20 0.007 0.204 0.034
21 0.000 0.037 0.127
22 0.027 0.062 0.027
23 0.227 0.053 0.056
24 0.027 0.043 0.171
58
Table ‎8.5: Derived Temporal Pattern for Region 5 - Urban Area (Kuala Lumpur)
No. of Duration
Block
1 0.184 0.072 0.058 0.095 0.023 0.007 0.080 0.017 0.047
2 0.448 0.097 0.050 0.175 0.161 0.003 0.054 0.012 0.031
3 0.368 0.106 0.061 0.116 0.118 0.003 0.011 0.001 0.006
4 0.161 0.108 0.096 0.096 0.051 0.023 0.001 0.027
5 0.164 0.096 0.093 0.107 0.074 0.025 0.033 0.060
6 0.400 0.103 0.097 0.102 0.086 0.017 0.026 0.049
7 0.106 0.078 0.092 0.206 0.015 0.020 0.022
8 0.065 0.050 0.096 0.081 0.047 0.027 0.009
9 0.065 0.060 0.091 0.140 0.021 0.053 0.067
10 0.056 0.048 0.045 0.180 0.012 0.041 0.023
11 0.068 0.062 0.037 0.107 0.035 0.068 0.019
12 0.164 0.030 0.033 0.064 0.032 0.096 0.014
13 0.009 0.132 0.050
14 0.002 0.015 0.040
15 0.003 0.018 0.014
16 0.075 0.011 0.025
17 0.055 0.031 0.003
18 0.087 0.030 0.072
19 0.076 0.004 0.110
20 0.052 0.024 0.054
21 0.103 0.036 0.087
22 0.048 0.142 0.052
23 0.027 0.033 0.050
24 0.091 0.129 0.070
59
Table ‎8.6: Normalized Temporal Pattern for Region 1 - Terengganu, Kelantan and
Northern Pahang
No. of Duration
Block
1 0.316 0.133 0.060 0.060 0.059 0.070 0.019 0.027 0.021
2 0.368 0.193 0.062 0.061 0.067 0.073 0.022 0.028 0.029
3 0.316 0.211 0.084 0.071 0.071 0.083 0.027 0.029 0.030
4 0.202 0.087 0.080 0.082 0.084 0.036 0.033 0.033
5 0.161 0.097 0.110 0.119 0.097 0.042 0.037 0.037
6 0.100 0.120 0.132 0.130 0.106 0.044 0.040 0.038
7 0.115 0.120 0.123 0.099 0.048 0.046 0.042
8 0.091 0.100 0.086 0.086 0.049 0.048 0.048
9 0.087 0.078 0.073 0.084 0.050 0.049 0.053
10 0.082 0.069 0.069 0.083 0.056 0.054 0.055
11 0.061 0.060 0.063 0.070 0.058 0.058 0.058
12 0.054 0.059 0.057 0.064 0.068 0.065 0.067
13 0.058 0.060 0.059
14 0.057 0.055 0.056
15 0.050 0.053 0.053
16 0.050 0.048 0.052
17 0.048 0.046 0.047
18 0.046 0.044 0.041
19 0.043 0.038 0.038
20 0.039 0.034 0.036
21 0.028 0.030 0.033
22 0.025 0.029 0.030
23 0.022 0.028 0.022
24 0.016 0.019 0.020
60
Table ‎8.7: Normalized Temporal Pattern for Region 2 - Johor, Negeri Sembilan, Melaka,
Selangor and Pahang
No. of Duration
Block
1 0.255 0.124 0.053 0.053 0.044 0.045 0.022 0.027 0.016
2 0.376 0.130 0.059 0.061 0.081 0.048 0.024 0.028 0.023
3 0.370 0.365 0.063 0.063 0.083 0.064 0.029 0.029 0.027
4 0.152 0.087 0.080 0.090 0.106 0.031 0.033 0.033
5 0.126 0.103 0.128 0.106 0.124 0.032 0.037 0.036
6 0.103 0.153 0.151 0.115 0.146 0.035 0.040 0.043
7 0.110 0.129 0.114 0.127 0.039 0.046 0.047
8 0.088 0.097 0.090 0.116 0.042 0.048 0.049
9 0.069 0.079 0.085 0.081 0.050 0.049 0.049
10 0.060 0.062 0.081 0.056 0.054 0.054 0.051
11 0.057 0.054 0.074 0.046 0.065 0.058 0.067
12 0.046 0.042 0.037 0.041 0.093 0.065 0.079
13 0.083 0.060 0.068
14 0.057 0.055 0.057
15 0.052 0.053 0.050
16 0.047 0.048 0.049
17 0.040 0.046 0.048
18 0.039 0.044 0.043
19 0.033 0.038 0.038
20 0.031 0.034 0.035
21 0.029 0.030 0.030
22 0.028 0.029 0.024
23 0.024 0.028 0.022
24 0.020 0.019 0.016
61
Table ‎8.8: Normalized Temporal Pattern for Region 3 - Perak, Kedah, P Pinang & Perlis
No. of Duration
Block
1 0.215 0.158 0.068 0.060 0.045 0.040 0.027 0.015 0.021
2 0.395 0.161 0.074 0.085 0.070 0.060 0.031 0.020 0.023
3 0.390 0.210 0.077 0.086 0.078 0.066 0.033 0.026 0.024
4 0.173 0.087 0.087 0.099 0.092 0.034 0.028 0.025
5 0.158 0.099 0.100 0.113 0.114 0.035 0.038 0.028
6 0.141 0.106 0.100 0.129 0.166 0.036 0.039 0.031
7 0.104 0.100 0.121 0.119 0.039 0.045 0.044
8 0.098 0.088 0.099 0.113 0.042 0.046 0.049
9 0.078 0.087 0.081 0.081 0.044 0.052 0.058
10 0.075 0.085 0.076 0.066 0.053 0.057 0.063
11 0.072 0.063 0.047 0.046 0.056 0.069 0.074
12 0.064 0.059 0.041 0.036 0.080 0.086 0.081
13 0.076 0.073 0.078
14 0.055 0.060 0.070
15 0.048 0.056 0.058
16 0.044 0.046 0.050
17 0.041 0.045 0.044
18 0.039 0.044 0.044
19 0.036 0.039 0.030
20 0.034 0.035 0.026
21 0.033 0.028 0.025
22 0.032 0.021 0.024
23 0.031 0.017 0.022
24 0.023 0.014 0.008
62
Table ‎8.9: Normalized Temporal Pattern for Region 4 - Mountainous Area
No. of Duration
Block
1 0.146 0.117 0.028 0.019 0.019 0.041 0.000 0.002 0.005
2 0.677 0.130 0.052 0.019 0.040 0.052 0.002 0.007 0.006
3 0.177 0.374 0.064 0.055 0.045 0.056 0.007 0.018 0.011
4 0.152 0.073 0.098 0.060 0.059 0.009 0.024 0.014
5 0.121 0.106 0.164 0.082 0.120 0.023 0.027 0.018
6 0.107 0.280 0.197 0.390 0.253 0.026 0.033 0.027
7 0.119 0.169 0.171 0.157 0.027 0.037 0.028
8 0.079 0.132 0.062 0.065 0.040 0.043 0.035
9 0.066 0.095 0.054 0.058 0.049 0.053 0.056
10 0.058 0.027 0.041 0.052 0.055 0.062 0.065
11 0.042 0.019 0.020 0.048 0.112 0.080 0.116
12 0.028 0.006 0.016 0.038 0.227 0.204 0.171
13 0.142 0.081 0.127
14 0.060 0.066 0.096
15 0.050 0.057 0.060
16 0.048 0.047 0.039
17 0.034 0.037 0.034
18 0.027 0.036 0.028
19 0.026 0.031 0.023
20 0.023 0.026 0.016
21 0.008 0.018 0.011
22 0.007 0.007 0.009
23 0.001 0.003 0.005
24 0.000 0.000 0.000
63
Table ‎8.10: Normalized Temporal Pattern for Region 5 - Urban Area (Kuala Lumpur)
No. of Duration
Block
1 0.184 0.097 0.056 0.048 0.033 0.003 0.003 0.001 0.006
2 0.448 0.161 0.061 0.060 0.045 0.051 0.011 0.011 0.014
3 0.368 0.400 0.065 0.078 0.092 0.074 0.015 0.015 0.019
4 0.164 0.096 0.095 0.096 0.086 0.021 0.018 0.023
5 0.106 0.106 0.097 0.107 0.140 0.025 0.024 0.027
6 0.072 0.164 0.175 0.161 0.206 0.032 0.027 0.040
7 0.108 0.116 0.118 0.180 0.047 0.031 0.049
8 0.103 0.096 0.102 0.107 0.052 0.033 0.050
9 0.068 0.093 0.096 0.081 0.055 0.041 0.054
10 0.065 0.062 0.091 0.064 0.076 0.068 0.067
11 0.058 0.050 0.037 0.007 0.087 0.129 0.072
12 0.050 0.030 0.023 0.003 0.103 0.142 0.110
13 0.091 0.132 0.087
14 0.080 0.096 0.070
15 0.075 0.053 0.060
16 0.054 0.036 0.052
17 0.048 0.033 0.050
18 0.035 0.030 0.047
19 0.027 0.026 0.031
20 0.023 0.020 0.025
21 0.017 0.017 0.022
22 0.012 0.012 0.014
23 0.009 0.004 0.009
24 0.002 0.001 0.003
64
Figure 8:1: Block diagrams of temporal storm profile corresponding with storm duration (0.25 to 12-hr) for Region 1
0.38 0.25hr 0.25 0.50 0.25 1-hr

0.37
0.368
0.211 0.211
0.36 0.20 0.193 0.20 0.193
0.35
0.161 0.161
0.34 0.15 0.15

0.133 0.133
0.33
0.32 0.316 0.316 0.10 0.10
0.31
0.30 0.05
0.05
0.29
0.28 0.00
0.00
1 2 3 1 2 3 4 5 6
1 2 3 4 5 6
0.14 3-hr 0.14 6-hr 0.14

12-hr
0.132 0.130
0.123
0.120 0.120 0.119
0.12 0.115 0.12 0.12
0.110
0.100
0.10 0.097 0.10
0.10
0.091
0.087 0.087 0.086
0.082 0.084 0.082
0.080 0.078
0.08 0.08 0.08
0.071 0.071 0.069 0.073
0.069 0.067
0.063
0.060 0.062 0.061 0.060 0.059 0.060 0.061 0.059
0.06 0.06 0.057
0.054 0.06
0.04 0.04 0.04
0.02 0.02 0.02
0.00 0.00 0.00

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
65
Figure 8:2: Block diagrams of temporal storm profile corresponding with storm duration (24, 48 and 72-hr) for Region 1
66
CHAPTER 9
DEVELOPING THE AREAL REDUCTION FACTOR
9.1 INTRODUCTION
The fixed area method used by the United State Weather Bureau (USWB) was adopted for
estimating the ARF for Kuala Lumpur and North Kelantan.
The maximum point rainfalls from each station and maximum areal rainfall shall be
extracted for each study area and each duration considered. The ARF was then
estimated by dividing the maximum areal rainfall by the average of the maximum point
rainfalls:
ARF = maximum areal rainfall

average of maximum point rainfalls
ARF estimates obtained in this manner for each season were averaged to give the final
estimated ARF for the area and storm duration considered. The estimates of ARF
obtained in all available years will be average to give the final rainfall for a specific area and
storm duration.
The following data was extracted for the derivation of ARF for the study area:
i. The maximum 1-hour, 3-hour, 6-hour, 12-hour and 24-hour point rainfall recorded by
the rainfall stations compounded within the study area.
ii. The maximum 1-hour, 3-hour, 6-hour, 12-hour and 24-hour areal rainfall for the study
area. (Areal rainfall was determined by creating Thiessen Polygons of rainfall stations
compounded within the study area).
9.2 DERIVATION PROCEDURE OF AREAL REDUCTION FACTOR (ARF)
Figure ‎9:1 shows the basic steps in the derivation of ARFs for each
sample/hypothetical catchment.
67
Determine the maximum areal rainfall (Ra) for each year
Determine the Maximum point rainfall recorded by each

station in the study eg. R1, R2, R3....
(Note that the storms from which maximum point
rainfalls were extracted need not be from the storm
where maximum areal rain was recorded
Calculate Rp, the weighted average (using Thiessen

weights for instance) of maximum point rainfall obtained;
Rp = W 1.R1 + w2.R2+W 3R3+…
where W 1, W 2, W 3….are‎Thiessen‎weights
Compute ARF = Ra / Rp
Obtain ARFs from other years and average to get ARF
Figure 9.1: Derivation ARFs Steps
68
9.3 SUMMARY OF RESULTS
Main outcomes of these tasks are as follows:
1. The result of this analysis can be tabularized by the relationship of areal

reduction factor (ARF) against catchment area a specific duration;
2. The derived ARF for both studies (Kuala Lumpur and Kelantan) for different storm
duration are tabulated in Table 9.1 and 9.2 respectively.
3. The ARF relationship mentioned, for example, can be seen in the respective
Figure 9.2 and 9.3 that shows the plot of ARF and storm duration (hr.) for both
studies;
4. It is recommended that the adopted ARF values of National Hydraulic Research of

Malaysia, NAHRIM (2010) should be replaced by the derived ARF values from this
present study for the rainfall duration of 1 hour to 24 hours.
Table 9.1: The ARF`s Estimated for Kuala Lumpur
ARFs for various storm duration (hours)

Area (km2)
1 3 6 12 24
50 0.76 0.78 0.80 0.82 0.83
100 0.68 0.70 0.72 0.75 0.76
150 0.63 0.66 0.68 0.70 0.71
200 0.60 0.63 0.65 0.67 0.68
Table 9.2: The ARF`s Estimated for Kelantan
ARFs for various storm duration (hours)

Area (km2)
1 3 6 12 24
50 0.86 0.89 0.96 0.97 0.99
100 0.78 0.82 0.86 0.9 0.93
150 0.74 0.78 0.83 0.85 0.89
200 0.70 0.74 0.78 0.81 0.84
69
ARF Curve for Kuala Lumpur
0.90
Areal Reduction Factor (ARF)
0.80
1hr
0.70 3hr
6hr
12hr
0.60 24hr
0.50
0.40
0 50 100 150 200 250
Catchment Area (km2)
Figure 9.2: The Relationship Graph of ARF Values Derived and Rainfall Duration for Kuala Lumpur
70
ARF Curve for Kelantan

1
0.9
Areal Reduction Factor (ARF)
0.8
0.7
0.6
0.5
0.4
0 50 100 150 200 250
Catchment Area (km2)
Figure 9.3: The Relationship Graph of ARF Values Derived and Rainfall duration for Kelantan
71
CHAPTER 10
ESTIMATION OF DESIGN STORM UNDER CLIMATE CHANGE SCENARIO
10.1 EFFECTS OF CLIMATE CHANGE ON DESIGN STORMS
Design storms which have derived based on historical data will no longer valid for the design
of hydraulic infrastructure especially with higher return periods. It is expected that future
changes in rainfall intensity due to climate change are expected to alter the level of
protection of hydraulic infrastructure. Increased rainfall intensity will result in more frequent
flooding. Therefore engineers have no choice but to adapt to climate change impact to
design storms. Hans, et al. C.B. (2006) found that two consequences would result in the
operation and design of hydraulic infrastructure. The first consequence to which a structure
is designed is no longer constant over time. Secondly the level of protection to which the
structure was designed will more frequently. Such consequences are shown in Figure 10.1
below.
Figure 10.1: Expected Changes in Project Design Return Periods due Climate Change
(Hans, A and Brian, C.B. 2006)
72
10.2 REVIEW OF CLIMATE CHANGE FACTORS BY NAHRIM
Based on the studies done by researches, the climate change factors are being
implemented as adding certain percentage to the design storm. National Hydraulic Research
Institute of Malaysia (NAHRIM) has published a technical guide on Estimation of Future
Design Rainstorm under the Climate Change Scenario in Peninsular Malaysia on January
2013. However, the study carried out by NAHRIM is a quite conservative due to some
uncertainties; hence an additional review on the climate change factors is deemed
necessary.
The review was carried out by dividing the rainfall stations in five (5) regions based on the
region created by means of the clustering analysis approach (Figure 10.3). The climate
change factors (CCFs) for each rainfall station as produced in NAHRIM’s Technical Guide
No. 1 on “Estimation of Future Design Rainstorm under the Climate Change Scenario in
Peninsular Malaysia’ (January 2013) has been summarized with respect to the five (5)
regions. The statistical values of minimum, median, mean and maximum climate change
factors for East Coast and West Coast rainfall stations (Figure 10.2) has been calculated
and summarized as in Table 10.1 to Table 10.5 respectively. Thus, the median values of
CCFs for various ARIs for each region is selected for design consideration because the
values are comparable to the study carried by Canada (Downscaled Global Climate Change
Model–CGCM2) and are considered reasonable and not too high compared to the mean of
the mean of the maximum values of CCFs.
73
Figure 10.2: Boundary of East Coast and West Coast Rainfall Stations for Peninsular Malaysia
74
Figure 10.3: The Region Created using Clustering Approach (NAHRIM, 2010)
75
Table 10.1: Summary of Minimum, Mean, Median and Maximum Values of Climate Change Factors for Region 1
(Kelantan, Terengganu and Northern Pahang)
Climate Change Factor
Station
State No Station Name Return Period, T
ID
2 5 10 20 25 50 100 200
Kelantan 1 5120025 Balai Polis Bertam 1.14 1.23 1.28 1.32 1.34 1.37 1.40 1.43
2 5320038 Dabong 1.06 1.21 1.30 1.37 1.39 1.45 1.50 1.54
3 4819027 Gua Musang 1.08 1.12 1.15 1.17 1.17 1.19 1.21 1.22
4 5522047 JPS Kuala Krai 1.17 1.24 1.27 1.30 1.30 1.32 1.34 1.35
5 5722057 JPS Machang 1.07 1.21 1.28 1.33 1.34 1.38 1.41 1.44
6 4923001 Kg. Aring 1.14 1.22 1.26 1.30 1.31 1.33 1.36 1.38
7 5718033 Kg. Jeli Tanah Merah 1.12 1.18 1.22 1.24 1.25 1.27 1.29 1.31
8 5322044 Kg. Lalok 1.11 1.20 1.25 1.28 1.30 1.32 1.35 1.37
9 6122064 Setor JPS Kota Bharu 1.10 1.28 1.37 1.43 1.45 1.50 1.55 1.58
1 4631001 Almuktafibillah Shah 1.06 1.16 1.22 1.27 1.28 1.32 1.35 1.38
2 3933001 Hulu Jabor, Kemaman 1.07 1.12 1.14 1.15 1.16 1.17 1.18 1.19
3 4332001 Jambatan Tebak, Kem 1.14 1.16 1.18 1.19 1.19 1.19 1.20 1.21
4 4234109 JPS Kemaman 1.12 1.20 1.24 1.26 1.27 1.29 1.31 1.33
5 5428001 Kg. Bt. Hampar, setiu 0.93 1.08 1.16 1.22 1.24 1.29 1.33 1.37
6 4131001 Kg. Ban Ho, Kemaman 1.08 1.11 1.13 1.14 1.14 1.15 1.16 1.17
7 4930038 Kg. Menerong, Hulu Trg 0.99 1.06 1.10 1.12 1.13 1.15 1.17 1.19
Terengganu 8 4526001 Kg. Seladang, Hulu Setiu 0.91 1.11 1.21 1.29 1.31 1.37 1.42 1.47
9 5725006 Klinik Kg. Raja, Besut 1.05 1.12 1.16 1.19 1.20 1.22 1.24 1.26
10 5331048 Setor JPS K. Terengganu 0.99 1.10 1.14 1.16 1.16 1.17 1.18 1.18
11 4832077 S K Jerangau 1.07 1.12 1.15 1.17 1.17 1.19 1.20 1.21
12 5524002 SK Panchor, Setiu 0.96 1.08 1.14 1.19 1.20 1.24 1.27 1.30
13 4734079 SM Sultan Omar, Dungun 1.09 1.18 1.23 1.26 1.27 1.30 1.32 1.34
14 5128001 Sungai Gawi, Hulu Trg 0.98 1.13 1.13 1.27 1.28 1.34 1.38 1.42
15 5328044 Sungai Tong, Setiu 1.04 1.21 1.21 1.37 1.39 1.44 1.49 1.53
76
(Kelantan, Terengganu and Northern Pahang) (cont`d)
Station
State No Station Name Return Period, T
ID
2 5 10 20 25 50 100 200
Pahang 1 4219001 Bukit Bentong 1.21 1.33 1.39 1.44 1.45 1.49 1.52 1.55
Minimum 0.91 1.06 1.1 1.12 1.13 1.15 1.16 1.17
Median 1.08 1.18 1.22 1.26 1.27 1.3 1.33 1.35
Mean 1.07 1.17 1.22 1.27 1.28 1.31 1.34 1.36
Maximum 1.23 1.37 1.44 1.5 1.52 1.57 1.61 1.64
77
(Pahang, Johor, Melaka, N. Sembilan and Selangor)
State No Station ID Station Name Return Period, T
2 5 10 20 25 50 100 200
Pahang 1 4127001 Hulu Tekai Kwsn B 1.09 1.16 1.20 1.22 1.23 1.26 1.27 1.29
2 3424081 JPS Temerloh 1.17 1.27 1.33 1.38 1.40 1.44 1.48 1.51
3 4223115 Kg. Merting 1.24 1.35 1.41 1.46 1.47 1.51 1.55 1.58
4 4023001 Kg. Sungai Yap 1.19 1.32 1.39 1.45 1.47 1.52 1.57 1.61
5 3628001 Pintu Kaw. Pulau Ketam 1. 16 1.38 1.49 1.56 1.58 1.62 1.64 1.65
6 3924072 Rumah Pam Paya Kangsar 1.2 1.33 1.40 1.46 1.47 1.52 1.56 1.59
7 3533102 Rumah Pam Pahang Tua 1.02 1.19 1.27 1.34 1.36 1.41 1.46 1.50
8 3818054 Setor JPS Raub 1.12 1.21 1.27 1.31 1.33 1.37 1.40 1.43
9 3121143 Simpang Pelangai 1.08 1.22 1.29 1.36 1.38 1.43 1.48 1.53
Johor 1 1636001 Balai Polis Kg Seelong 1.05 1.22 1.31 1.38 1.40 1.46 1.51 1.55
2 1931003 Emp. Semberong 1.09 1.14 1.16 1.18 1.18 1.20 1.21 1.23
3 2235163 Ibu Bekalan Kahang 1.21 1.32 1.37 1.41 1.43 1.46 1.49 1.52
4 2237164 Jalan Kluang Mersing 1.27 1.36 1.40 1.43 1.43 1.46 1.48 1.49
5 1541139 Johor Silica 1.25 1.27 1.28 1.29 1.29 1.29 1.30 1.30
6 2033001 JPS Kluang 0.98 0.99 0.99 0.99 0.99 0.99 0.99 0.99
7 2534160 Kg Peta Hulu Sg Endau 1.05 1.09 1.10 1.12 1.12 1.13 1.14 1.15
8 2231001 Ladang Chan Wing 1.14 1.22 1.26 1.30 1.31 1.35 1.38 1.40
9 2232001 Ladang Kekayaan 1.13 1.23 1.29 1.33 1.34 1.37 1.40 1.43
10 2330009 Ladang Labis 1.12 1.30 1.39 1.46 1.49 1.55 1.60 1.64
11 2025001 Pintu Kaw. Tg. Agas 1.13 1.29 1.37 1.45 1.47 1.53 1.58 1.63
12 1534002 Pusat Kem. Pekan Nenas 1.10 1.20 1.25 1.30 1.31 1.35 1.38 1.41
13 2528012 Rmh. Tapis Segamat 1.19 1.37 1.46 1.53 1.55 1.61 1.66 1.71
14 1829002 Setor JPS B Pahat 1.11 1.25 1.32 1.38 1.40 1.45 1.49 1.53
15 2636170 Setor JPS Endau 1.05 1.09 1.11 1.13 1.13 1.15 1.16 1.17
16 1839196 Simpang Masai K Sedili 1.27 1.48 1.58 1.65 1.67 1.73 1.78 1.82
78
(Pahang, Johor, Melaka, N. Sembilan and Selangor) (cont`d)

2 5 10 20 25 50 100 200
17 1737001 SM Bukit Besar 0.98 1.14 1.23 1.30 1.32 1.38 1.43 1.47
18 1437116 Stor JPS Johor Bahru 1.02 1.09 1.12 1.15 1.16 1.19 1.21 1.23
Melaka 1 2222001 Bukit Sebukor 1.12 1.26 1.33 1.39 1.41 1.46 1.51 1.56
2 2224038 Chin Chin Tepi Jalan 1.12 1.29 1.39 1.47 1.50 1.57 1.63 1.68
3 2321006 Ladang Lendu 1.06 1.18 1.24 1.30 1.32 1.36 1.41 1.45
N. 1 2722202 Kg Sawah Lebar K Pilah 1.04 1.23 1.33 1.41 1.43 1.50 1.56 1.62
Sembilan 2 2725083 Ladang New Rompin 1.06 1.24 1.33 1.41 1.43 1.49 1.54 1.58
3 2920012 Petaling K Kelawang 1.10 1.18 1.22 1.25 1.26 1.28 1.31 1.32
4 2719001 Setor JPS Sikamat 1.10 1.21 1.27 1.32 1.33 1.37 1.41 1.44
5 2723002 Sungai Kepis 1.01 1.14 1.21 1.28 1.29 1.35 1.40 1.44
Selangor 1 3117070 JPS Ampang 1.08 1.14 1.17 1.20 1.21 1.23 1.25 1.27
2 2815001 JPS Sungai Manggis 1.21 1.27 1.31 1.34 1.35 1.37 1.39 1.41
3 3416002 Kg. Kalong Tengah 1.04 1.16 1.23 1.29 1.31 1.36 1.41 1.45
4 3516022 Loji Air Kuala Kubu Baru 1.10 1.13 1.14 1.16 1.16 1.17 1.18 1.19
5 2913001 Pusat Kwln. JPS T. Gong 1.19 1.26 1.30 1.33 1.34 1.36 1.39 1.41
6 3710006 Rmh Pam Bagan Terap 1.10 1.17 1.22 1.26 1.27 1.30 1.33 1.36
7 3314001 Rumah Pam JPS P Setia 1.09 1.28 1.38 1.46 1.49 1.55 1.61 1.67
8 2917001 Setor JPS Kajang 1.14 1.27 1.35 1.41 1.42 1.47 1.52 1.55
9 3411017 Setor JPS Tjg. Karang 1.08 1.14 1.17 1.19 1.20 1.22 1.24 1.26
10 3118102 SK Sungai Lui 1.19 1.24 1.27 1.29 1.30 1.31 1.33 1.34
Minimum 0.98 0.99 0.99 0.99 0.99 0.99 0.99 0.99
Median 1.10 1.23 1.29 1.33 1.34 1.37 1.41 1.45
Mean 1.12 1.23 1.29 1.34 1.35 1.39 1.42 1.45
Maximum 1.27 1.48 1.58 1.65 1.67 1.73 1.78 1.82
79
(Perak, Kedah, Pulau Pinang and Perlis)
2 5 10 20 25 50 100 200
Perak 1 4807016 Bkt Larut Taiping 1.11 1.10 1.11 1.13 1.13 1.16 1.20 1.25
2 5710061 Dispensari Kroh 1.07 1.11 1.13 1.15 1.15 1.17 1.18 1.19
3 5005003 Jln Mtg Buloh Bagan Serai 1.03 1.12 1.16 1.20 1.21 1.25 1.28 1.30
4 4207048 JPS Setiawan 1.11 1.18 1.22 1.25 1.26 1.29 1.31 1.34
5 4010001 JPS Teluk Intan 1.12 1.20 1.24 1.28 1.29 1.32 1.35 1.38
6 5207001 Kolam Air JKR Selama 1.14 1.20 1.23 1.26 1.27 1.29 1.32 1.34
7 5411066 Kuala Kenderong 1.18 1.22 1.24 1.25 1.26 1.27 1.28 1.29
8 4311001 Pej Daerah Kampar 1.05 1.14 1.19 1.24 1.25 1.29 1.32 1.36
9 4511111 Politeknik Ungku Omar 1.15 1.31 1.40 1.46 1.48 1.53 1.57 1.60
10 4811075 Rancangan Belia Perlop 1.40 1.50 1.55 1.59 1.60 1.63 1.65 1.68
11 4409091 Rumah Pam Kubang Aji 1.16 1.27 1.33 1.38 1.39 1.44 1.48 1.51
Kedah 1 6207032 Ampang Pedu 1.05 1.08 1.09 1.10 1.11 1.12 1.13 1.13
2 5507076 Bt.27, Jln Baling 1.12 1.16 1.18 1.20 1.21 1.22 1.24 1.25
3 5808001 Bt.61,Jln Baling 1.08 1.13 1.16 1.18 1.19 1.21 1.22 1.24
4 5704055 Kedah Perak 1.14 1.20 1.24 1.26 1.27 1.29 1.31 1.33
5 5806066 Klinik Jeniang 1.15 1.17 1.18 1.19 1.20 1.20 1.21 1.22
6 6108001 Komp Rmh Muda 1.15 1.24 1.29 1.33 1.34 1.38 1.41 1.44
7 6206035 Kuala Nerang 0.97 1.07 1.13 1.17 1.18 1.22 1.25 1.28
8 6306031 Padang Sanai 1.08 1.09 1.11 1.14 1.15 1.18 1.23 1.28
9 6103047 JPS Alor Setar 1.07 1.17 1.22 1.26 1.28 1.32 1.35 1.38
P. Pinang 1 5404043 Ibu Bekalan Sg Kulim 1.26 1.29 1.31 1.33 1.33 1.34 1.35 1.36
& Perlis 2 5402001 Klinik Bkt Bendera 1.20 1.23 1.25 1.27 1.27 1.29 1.30 1.31
3 5402002 Kolam Bersih PP 1.23 1.27 1.29 1.31 1.32 1.33 1.35 1.36
4 5302003 Kolam Tkgn Air Hitam 1.31 1.37 1.41 1.44 1.45 1.47 1.50 1.52
5 5303053 Komplek Prai 1.24 1.26 1.28 1.28 1.29 1.30 1.30 1.31
80
(Perak, Kedah, Pulau Pinang and Perlis) (cont`d)
2 5 10 20 25 50 100 200
6 5504035 Lahar Ikan Mati 1.23 1.31 1.35 1.39 1.40 1.43 1.46 1.49
7 6401002 Padang Katong 1.06 1.08 1.09 1.09 1.10 1.10 1.11 1.11
8 5303001 Rumah Kebajikan PP 1.24 1.27 1.28 1.29 1.30 1.31 1.32 1.33
9 5204048 Sg Simpang Ampat 1.03 1.19 1.28 1.36 1.38 1.45 1.50 1.56
10 5302001 Tangki Air Besar Sg Png 1.25 1.30 1.32 1.34 1.35 1.37 1.39 1.41
Minimum 0.97 1.07 1.09 1.09 1.10 1.10 1.11 1.11
Median 1.14 1.20 1.24 1.26 1.27 1.29 1.32 1.34
Mean 1.15 1.21 1.24 1.27 1.28 1.31 1.33 1.35
Maximum 1.4 1.5 1.55 1.59 1.6 1.63 1.65 1.68
Table 10.4: Summary of Minimum, Mean, Median and Maximum Values of Climate Change Factors for Region 4 (Mountainous)
2 5 10 20 25 50 100 200
1 4219001 Bukit Bentong 1.21 1.33 1.39 1.44 1.45 1.49 1.52 1.55
Pahang
2 4513033 Gunung Berinchang 1.43 1.46 1.48 1.49 1.49 1.50 1.51 1.51
1 4614001 Brook 1.23 1.37 1.44 1.50 1.52 1.57 1.61 1.64
Kelantan
2 5216001 Gob 1.11 1.19 1.23 1.26 1.27 1.29 1.31 1.33
Minimum 1.11 1.19 1.23 1.26 1.27 1.29 1.31 1.33
Median 1.22 1.35 1.42 1.47 1.47 1.50 1.52 1.53
Mean 1.25 1.34 1.39 1.42 1.43 1.46 1.49 1.51
Maximum 1.43 1.46 1.48 1.5 1.52 1.57 1.61 1.64
81
Table 10.5: Summary of Minimum, Mean, Median and Maximum Values of Climate Change Factors for Region 5 (Wilayah
Persekutuan)
2 5 10 20 25 50 100 200
1 3317001 Air Terjun Sg. Batu 1.04 1.14 1.21 1.26 1.28 1.33 1.37 1.42
2 3217002 Emp. Genting Kelang 1.14 1.20 1.24 1.27 1.28 1.31 1.33 1.35
3 3317004 Genting Sempah 1.12 1.20 1.25 1.30 1.31 1.34 1.38 1.41
4 3217003 Ibu Bek. KM11, Gombak 1.04 1.11 1.14 1.18 1.18 1.21 1.23 1.25
5 3217001 Ibu Bek. KM16, Gombak 1.11 1.14 1.16 1.18 1.18 1.20 1.21 1.22
Wilayah
6 3116003 Ibu Pejabat JPS 1.06 1.11 1.14 1.16 1.16 1.18 1.20 1.21
Persekutuan
7 3116004 Ibu Pejabat JPS 1 1.08 1.16 1.20 1.23 1.24 1.28 1.30 1.33
8 3217005 Kg. Kerdas, Gombak 1.09 1.19 1.25 1.30 1.31 1.35 1.39 1.42
9 3217004 Kg. Kuala Seleh, H.Klg 1.12 1.21 1.26 1.30 1.31 1.34 1.37 1.40
10 3216001 Kg. Sungai Tua 1.08 1.15 1.19 1.23 1.24 1.27 1.30 1.32
11 3116006 Ladang Edinburgh 1.12 1.23 1.29 1.34 1.35 1.39 1.43 1.46
12 3015001 Puchong Drop, K. Lumpur 1.10 1.14 1.17 1.19 1.19 1.21 1.22 1.24
13 3216004 SK Jenis Keb, Kepong 1.00 1.11 1.17 1.22 1.24 1.29 1.33 1.37
14 3116005 SK Taman Maluri 1.06 1.14 1.19 1.23 1.24 1.28 1.31 1.34
Minimum 1.09 1.15 1.20 1.23 1.24 1.29 1.32 1.35
Median 1.09 1.15 1.20 1.23 1.24 1.29 1.32 1.35
Mean 1.08 1.16 1.20 1.24 1.25 1.28 1.31 1.34
Maximum 1.14 1.23 1.29 1.34 1.35 1.39 1.43 1.46
82
CHAPTER 11
WORKED EXAMPLE
EXAMPLE 1: GAUGED LOCATION
A dam is proposed to be constructed at Station 3118102. Derive the IDF curves and future
IDF curves.
Solution:
STEP 1: Determine the derived IDF parameters from Revised HP (Table 6.1a)
From the table, the derived parameters are:
IDF Parameter Value

λ 63.155
κ 0.177
θ 0.122
𝜂 0.842
STEP 2: Calculate design storm rainfall intensity from the equation:

𝜆𝛵𝜅
Rainfall intensity, 𝑖=
(𝑑+𝜃)𝜂
Intensity (mm/hr) Return period, T (yr)

Duration, d (hr) 2 5 10 20 25 50 100
0.25 164.2 193.1 218.3 246.8 256.7 290.2 328.1
0.5 106.5 125.2 141.6 160.1 166.5 188.3 212.8
1 64.8 76.2 86.2 97.4 101.3 114.6 129.5
2 37.9 44.6 50.4 57.0 59.3 67.0 75.7
3 27.4 32.2 36.4 41.2 42.8 48.4 54.7
6 15.5 18.3 20.6 23.3 24.3 27.5 31.0
12 8.7 10.3 11.6 13.1 13.7 15.4 17.5
24 4.9 5.8 6.5 7.4 7.7 8.7 9.8
48 2.7 3.2 3.6 4.1 4.3 4.8 5.5
72 1.9 2.3 2.6 2.9 3.0 3.4 3.9
83
STEP 3: Create the graph of IDF curves
IDF Curves for SK Sungai Lui (Stn. No. 3118102)

1000.0
Intensity (mm/hr)
100.0 Series
2 yr
1
5 yr
Series
210 yr
Series
20 yr
325 yr
Series
10.0 450 yr
100 yr
Series
5
1.0
0.1 1 10 100
Duration(hr)
STEP 4: Determine CCF from Table 3.4e for the station in Selangor (Figure 3.2(b))
Return Period T (yr) CCF (Median value)

2 1.10
5 1.23
10 1.29
20 1.33
25 1.34
50 1.37
100 1.41
84
STEP 5: Calculate future design storm rainfall intensity:

Future design storm rainfall intensity, I future = I x CCF
Intensity (mm/hr) Return period, T (yr)

Duration, d (hr) 2 5 10 20 25 50 100
0.25 180.6 237.5 281.6 328.2 344.0 397.6 462.6
0.5 117.1 154.0 182.7 212.9 223.1 257.9 300.1
1 71.3 93.7 111.1 129.6 135.8 156.9 182.6
2 41.7 54.8 65.0 75.8 79.4 91.8 106.8
3 30.1 39.6 47.0 54.7 57.4 66.3 77.1
6 17.1 22.5 26.6 31.0 32.5 37.6 43.8
12 9.6 12.6 15.0 17.5 18.3 21.2 24.6
24 5.4 7.1 8.4 9.8 10.3 11.9 13.8
48 3.0 4.0 4.7 5.5 5.7 6.6 7.7
72 2.1 2.8 3.3 3.9 4.1 4.7 5.5
STEP 5: Create the graph of future IDF curves
Future IDF Curves for SK Sungai Lui (Stn. No.: 3118102

1000.0
Intensity (mm/hr)
100.0 Series
2 yr
1
5 yr
Series
2 10 yr
Series
3
20 yr
25 yr
Series
4 50 yr
10.0 Series
5 100 yr
1.0
0.1 1 10 100
Duration(hr)
85
EXAMPLE 2: A barrage is to be constructed at latitude N 50 50’ 25’’ and longitude 1010 53’
30’’ which commands a catchment area of 200 sq. km. Find the design storm of 100 year
return period for 24 hour duration and the appropriate areal reduction factor for the design
storm.
Solution:
STEP 1: Determine the derived IDF parameters from the maps in the appendices for
ungauged sites. The derived IDF parameters are:
IDF Parameter Value

λ 55.271
κ 0.207
θ 0.111
𝜂 0.684
STEP 2: Calculate design storm rainfall intensity of 100 year return period for 24 hour
duration from the equation:
𝜆𝛵𝜅
Rainfall intensity, 𝑖=
(𝑑+𝜃)𝜂
55.271 ×(100)0.207
=
(24×0.0111)0.684
= 16.3mm/hr
STEP 3: Calculate the design storm rainfall depth of 100 year return period for 24 hour
duration
Rainfall depth = I × d
= 16.3 × 24
= 390.1 mm
STEP 4: Find the appropriate ARF to be applied to the design storm
The site is located in the east coast of Peninsular Malaysia and the appropriate ARF for
design storm rainfall depth of 100 year return period for 24 hour duration is 0.84 (Table 9.2).
Therefore the areal design storm rainfall depth of 100 years return period for 24 hour
duration is 390.1 x 0.84 = 327.8 mm.
86
CHAPTER 12
APPENDIX 1 – ISOPLETHS MAP OF IDF PARAMETER
FIGURE 12:1: IDF PARAMETER OF 
87
FIGURE 12:2: IDF PARAMETER OF 
88
FIGURE 12.3: IDF PARAMETER OF 
89
FIGURE 12.4: IDF PARAMETER OF 
90
References:
Amin, M.Z.M., Shaaban, A.J.,(2004). The Rainfall Intensity-Duration-Frequency (IDF)

Relationships for Ungauged Sites in Peninsular Malaysia using a Mathematical Formulation.
Proceeding of the International Conference on the 21st Century River Management, Sept.
21-23, 2004, Penang, Malaysia.
Amin, M.Z.M., (2003). Design Rainstorm of Peninsular Malaysia: Regional Frequency

Analysis Approach. Proceeding of the International Conference on Water and Environment
(WE2003), 15-18 December 2003, Bhopal, India.
Amin, M.Z.M., (2002). Regionalization Approach in Design Rainstorm Estimation Based on

L-Moments Theory in
Peninsular Malaysia. Paper presented at International Conference on Urban Hydrology, 14-
16 October 2002, Kuala Lumpur, Malaysia
Amin, M.Z.M.,(2000) Regional flood frequency analysis in the East Coast of Peninsular
Malaysia by L-moments approach. M.Sc (Thesis), unpublished, National University of
Ireland, Galway, December 2000.
Australian Rainfall and Runoff. 2014: An Interim Guide for Considering Climate Change in
Rainfall and Runoff.
Cunnane, C. (1989). Review of statistical models for flood frequency analysis. WMO
Operational Hydrology. Rep. no. 33, WMO-no 718, World Meteorological Organization.
Chow, V.T., Maidment, D.R., and Mays , L.W. (1988). Applied hydrology. McGraw-Hill, Inc.,
New York, 571 pp.
Faulkner, D., (1999). Flood Estimation Handbook: Rainfall Frequency Analysis. Institutes of
Hydrology, Wallingford, 110pp.
Daud, Z.M., 2001. Statistical modeling of extreme rainfall processes in Malaysia. Ph.D
(Thesis), unpublished, Universiti Teknologi Malaysia.
Hans, A and Brians, C.B. 2006. Urban Drainage Infrastructure Planning and Design
Considering Climate Change. Proceedings of EIC Climate Change Technology Conference.
Ottawa, Ontario, Canada: IEEE.
Heiler, T.D., 1973. Hydrological Procedure No.1 – Estimation of design rainstorm. Ministry of
Agriculture Malaysia, Kuala Lumpur.
Hosking, J.R.M. (1990). L-moments : analysis and estimation of distribution using linear
combination of order statistics .Journal Royal Statistical Society. B, 52(1), 105-124.
Hosking, J.R.M., and Wallis, J.R (1993). Some statistics useful in regional frequency
analysis. Water Resource Research, 29(2), 271-281.
Hosking, J.R.M., and Wallis, J.R. (1997). Regional frequency analysis: an approach based
on L -moments . Cambridge University Press, New York, 224 pp.
International Practices on Climate Adaptation in Transportation.

http://www.fhwa.dot.gov/environment/climate_change/adaptation/publication. Accessed on 1
August 2015.
91
Jabatan Pengairan dan Saliran Malaysia.(1986) Variation of Rainfall with Area in Peninsular
Malaysia, Water Resources Publication No. 17.
Jenkins, G., Goonetilleke, A., 2002. Estimating Peak Runoff for Risk-Based Assessment in
Small Catchments . Australian
Journal of Water Resources, Vol .5, No.2, P177-193
Koutsoyiannis , D., Kozonis , D., Manetas, A., 1998. A Mathematical framework for studying
rainfall intensity-duration frequency relationship, Journal of Hydrology 206 (1998), 118-135.
Mahmood, M.D., Salleh, S., Leong, T.M, Teh, S.K., 1982. Hydrological Procedure No.1 -
Estimation of design rainstorm in Peninsular Malaysia. Ministry of Agriculture Malaysia,
Kuala Lumpur, 71 pp.
Maidment, D.R. (1993). Hydrology. Handbook of Hydrology, D.R. Maidment ed., McGraw-
Hill, Inc., New York, 1.1-1.5
Nguyen, V.T.V., Peyron, N., Rivard, G., 2002. Rainfall Temporal Pattern for Urban Drainage
in Southern Quebec. Proceeding of the International Conference on Urban Drainage, ACSE
2004 New York.
Norlida M.D. (2012) Review of Areal Reduction Factor.
Prodanovi c, P., Si monovic, S.P., (2004). Generation of Synthetic Design Rainstorm for
Upper Thames River Basin. Project Report V: Assessment of Water Resources Risk and
Vulnerability to Changing Climatic Conditions . University of Western Ontario, Ontario, p2-
20.
Sri wardena, L., Weinmann, P.E., (1996). Derivation of Areal Reduction Factors for Design
Rainfalls in Victoria. CRC for Catchment Hydrology, Australia, 60 pp.
Smithers , J.C., Schulze, R.E., (2000). A methodology for estimation of short duration design
rainstorm in South Africa using a regional approach based on L-moments. Journal of
Hydrology. 241, 42-52.
Town of Stratford .(2012). Impact of climate Change on Stormwater Management.

Stormwater Management Plan Update.
Veneziano, D., Villani, P., 1999. Best Linear Unbiased Design Hyetograph. Water Resources
Research, Vol . 35, No.9, P2725-2738.
Vaes, G., (2002). The Influence of Rainfall and Model Simplification on Combined Sewer
System Design. PhD (Thesis), Unpublished, University of Leuven, Belgium, April 1999.
Zaris, D., Koutsoyiannis , D., Karavokoris, G., (1996). A Simple Stochastic Rainfall
Disaggregation Scheme for Urban Drainage Modelling. Journal of Hydrology, p85-92.
92

HP1 2015

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

HP1 2015

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

HP1 2015

Uploaded by

Copyright:

Available Formats

HYDROLOGICAL PROCEDURE NO.

Estimation of Design Rainstorm in Peninsular

DEPARTMENT OF IRRIGATION AND DRAINAGE

ESTIMATION OF DESIGN RAINSTORM IN PENINSULAR

Water Resources Management and Hydrology Division

Copyright ©2015 by Department of Irrigation and Drainage (DID) Malaysia

3 APPROACH AND METHODOLOGY 16

3.6 Construction and Formulation of At-Site IDF Curve 21

4 ASSESSMENT OF THE PROPOSED METHODOLOGY 25

5 DEVELOPING THE INTENSITY-DURATION FREQUENCY (IDF) RELATIONSHIP 31

6 DEVELOPING THE INTENSITY-DURATION FREQUENCY (IDF) RELATIONSHIP 35

7 DEVELOPING THE REGION OF TEMPORAL STORM PROFILES BY MEANS

8 DEVELOPING THE DESIGN STORM PROFILES (TEMPORAL STORM) 55

9 DEVELOPING THE AREAL REDUCTION FACTOR 68

10 ESTIMATION OF DESIGN STORM UNDER CLIMATE CHANGE SCENARIO 73

12 APPENDIX 1 - Isopleths Map of IDF Parameter 87

Table 5.1: Polynomial Equation Parameters of Site 3117070 32

1.1 BACKGROUND AND GENERAL REVIEW

Due to restrictions o f records length, the estimation of design rainstorm/rainfall intensity is

To provide the estimation of design rainstorm/rainfall intensity and IDF relationship at

1.3 SCOPE OF REVISION AND UPDATE

 Review existing techniques used in HP1;

1.4 CONCERNED ISSUES AND STATEMENTS IN THE PROPOSED REVISION AND

1.4.1 Reviews on the Choice of Frequency Distribution

The choice of frequency distribution or accurately determination of parent distribution is

1.4.2 Short Duration Analysis

An appropriate regional IDF relationship can be established if method of regional frequency

2.1 BRIEF OVERVIEW OF THE TASK

2.2 OBJECTIVES OFTHE DESIGNATED TASKS

2.2.1 T1: Task 1 – Data Mining and Assembly

2.2.2 T2: Task 2 – Choice of Rainfall Frequency Models

Linkages: Provide information for the components of T3 and T4.

2.2.3 T 3 : Task 3 – Choice of Distribution to be Used in the Chosen Model (AM or

2.2.4 T4: Task 4 – Method of Parameter Estimation

The objective is to determine the magnitude of design rainstorm/rainfall intensity at gauged

Linkages: Provide information for the components of T6, T7 and T8.

2.2.6 T6: Task 6 – Construction and Formulation of at-site IDF Curve

To formulate a mathematical relationship (duration, magnitude of design rainstorm/rainfall

2.2.7 T7: Task 7 – Design Storm Profile (Temporal Pattern)

2.2.8 T8: Task 8 – Areal Reduction Factor

Table 2.1: List of Automatic Rainfall Gauges Stations throughout

Table 2.1: List of Automatic Rainfall Gauges Stations throughout

Table 2.1: List of Automatic Rainfall Gauges Stations throughout

Table 2.1: List of Automatic Rainfall Gauges Stations throughout

3.1 DATA MINING AND ASSEMBLY

3.2 CHOICE OF RAINFALL FREQUENCY MODEL

Statistically if we denote the estimate of RT obtained by the AM series as RT and that

3.3 CHOICE OF DISTRIBUTION TO BE USED IN THE CHOSEN MODEL (AM OR

3.3.1 Candidates of the AM model – the Generalized Extreme Value Distribution

parameters of the GEV distribution in term L-moments are:

= 7.8590c + 2.9554c2 [2]

=   [3]

=  [4]

The quantiles of the GEV distribution can be calculated from:

3.3.2 Candidate Distribution of the PD/POT Model – the Generalized Pareto

The GPA distribution’s cdf is given by:

distribution results; = 0 yields a two-parameter exponential distribution in the form of