HB 212—2002

Design Wind Speeds for the

Asia–Pacific Region
This page is intentionally left blank.
 HB 212—2002

Design Wind Speeds

for the

Asia–Pacific Region

by
John Holmes
and Richard Weller
HB 212—2002 2

PREFACE
The majority of this Handbook was prepared by JDH Consulting for Standards
Australia. The document is intended for use by all persons engaged in the structural
design of buildings and other structures where wind loads are to be considered.
In preparing this document the aim has been to provide a consistent set of wind
speeds for a large portion of the earth so that wind speeds can be compared and a
single loading code used over a wide area. The wind speeds in this Handbook are
presented in the spirit of scientific investigation and are not intended in any way to
substitute for the requirements of any authorities. The authors regard this Handbook
as a living document subject to review and any new information leading to revision
would be gratefully accepted.
Using the wind speeds in this Handbook together with the AS/NZS 1170 series of
loading Standards (particularly AS/NZS 1170.0: Structural design actions—General
principles and AS/NZS 1170.2: Wind actions—see Chapter 4), should enable
designs for different countries across the region to be directly compared. For
example, a design performed for a building in Singapore can be directly compared
with one in Melbourne.
Standards Australia would like to request that any information that may be used to
update the maps in this Handbook be sent to our Sydney address.

ACKNOWLEDGEMENTS
Helpful advice was received from members of the international wind design
community, especially the APEC Informal Loading Network, from members of the
Standards Committee BD-006 General Design Requirements and Loading on
Structures and from others who were consulted during the preparation of this
Handbook.

COPYRIGHT
© Standards Australia International
All rights are reserved. No part of this work may be reproduced or copied in any form or by any means,
electronic or mechanical, including photocopying, without the written permission of the publisher.
Published by Standards Australia International Ltd
GPO Box 5420, Sydney, NSW 2001, Australia
Published on 11 December 2002.
ISBN 0 7337 4817 1
 3 HB 212—2002

CONTENTS

Page

Chapter 1 INTRODUCTION
1.1 BACKGROUND ........................................................................................................... 5
1.2 EXTREME WINDS IN THE REGION .......................................................................... 6

Chapter 2 ZONAL WIND SYSTEM FOR THE ASIA–PACIFIC REGION ............................... 7

Chapter 3 WIND SPEED VERSUS RETURN PERIOD

3.1 LEVEL I (equatorial regions—strong thunderstorms and monsoon winds)................... 9
3.2 LEVEL II (moderately severe thunderstorms and extra-tropical gales)......................... 9
3.3 LEVEL III (severe thunderstorms and moderate or weakening typhoons/tropical
cyclones)...................................................................................................................... 9
3.4 LEVEL IV (strong typhoons/tropical cyclones) ............................................................. 9
3.5 LEVEL V (very strong typhoons/tropical cyclones)..................................................... 10
3.6 SUMMARY................................................................................................................. 10

Chapter 4 WIND SPEEDS AND AS/NZS 1170 ................................................................... 11

Chapter 5 WIND SPEEDS FOR SPECIFIC REGIONS

5.1 GENERAL .................................................................................................................. 13
5.2 AUSTRALIA AND AUSTRALIAN TERRITORIES....................................................... 13
5.3 NEW ZEALAND ......................................................................................................... 16
5.4 MELANESIA, MICRONESIA AND POLYNESIA ........................................................ 18
5.5 EQUATORIAL SOUTH-EAST ASIAN ARCHIPELAGO .............................................. 20
5.6 PHILIPPINES ............................................................................................................. 22
5.7 INDOCHINA ............................................................................................................... 24
5.8 THAILAND ................................................................................................................. 26
5.9 MYANMAR................................................................................................................. 26
5.10 CHINA REGION......................................................................................................... 27
5.11 JAPAN ....................................................................................................................... 30
5.12 KOREA ...................................................................................................................... 32
5.13 SRI LANKA ................................................................................................................ 33
5.14 INDIA ......................................................................................................................... 34

APPENDICES
A EXTREME WIND ANALYSIS ..................................................................................... 36
B BASIC STRONG WIND TYPES ................................................................................. 40
C WIND LOADING CODES AND STANDARDS............................................................ 43
D ANALYSIS OF WIND SPEEDS FOR LEVEL I AND HONG KONG ............................ 45
E REFERENCES AND BIBLIOGRAPHY ....................................................................... 47
HB 212—2002 4

NOTES
 5 HB 212—2002

CHAPTER 1 INTRODUCTION

1.1 BACKGROUND

There are many design Standards and codes in the east Asia, south-east
Asia and Oceania regions that give wind speeds. However, there is no
generally accepted method of comparison between these Standards. Each
one has been developed for use in particular countries and legislative
jurisdictions.

This Handbook gives a system of five wind speed classifications for

countries in the Asia–Pacific region (Levels I to V). The wind speeds given
enable comparison of extreme wind hazard in different countries across
the region. Designs based on these wind speeds could easily be verified
for any of the countries covered.

These classifications are related to the regional wind speed system given
in AS/NZS 1170.2. This enables the design methods in AS/NZS 1170.2 to
be used for design purposes in other countries in the region.

The wind speed classifications given in this Handbook have been based on
analyses of wind speed recordings at large numbers of stations throughout
the region and data published in national codes and Standards. Published
analyses are used in conjunction with raw data to establish the appropriate
wind speeds for each area.

The basis for presentation of the wind speeds is given in Appendix A,

including extreme value analysis, use of return period and errors in
estimation.

Other appendices give information on types of wind, wind loading codes

and Standards, and an analysis of wind speeds for the equatorial Level I
and for Hong Kong.

COPYRIGHT
HB 212—2002 6

1.2 EXTREME WINDS IN THE REGION

The region, defined approximately as the land bounded by latitude ±50°, and
longitude 90° to 180° east, has been a region of high economic growth, and
rapidly expanding trade. The influences in the region producing strong winds
are, in general terms, the same as those influencing atmospheric motion in
other parts of the world—namely solar heating (producing pressure
differences), the earth’s rotation (Coriolis forces), and frictional effects,
particularly near the surface. A feature of greater importance in this region is
the large expanse of ocean, mainly the Pacific Ocean itself. This affects the
weather in two related ways: as a source of latent heat and moisture, and by
the occurrence of ocean currents.

Other important influences are the Equatorial Trough that is particularly

strong over the Pacific Ocean, and the Asian Monsoon, which produces a
reversal in the general wind regime over large parts of east and south-east
Asia.

Severe tropical cyclones (usually called 'typhoons' in the north-west Pacific)

are the source of the most severe winds in the Region. The north-west and
south-west Pacific combined, in fact, has produced about 55% of the world
total of these events. The 'seasons' for these events are July to October in the
north, and December to April in the south.

The equatorial region with low design wind speeds covers a number of
countries, and around 300 million population. The whole of Indonesia,
Malaysia, Singapore and most of Papua New Guinea are included. It is
known that tropical cyclones, or typhoons, cannot form within about 5° of
the Equator, so that these countries are fortunate in not receiving any
influence from the world’s worst extreme wind events. Although there are
likely to be local effects that might justify some micro-zonation within the
classification, the reduced importance of wind load, relative to other loads,
means that this is probably not justifiable or necessary for the majority of
structures.

COPYRIGHT
 7 HB 212—2002

CHAPTER 2 ZONAL WIND SYSTEM FOR THE ASIA–

PACIFIC REGION
Table 1 gives a simple classification system of 5 Levels, which can be
used to 'grade' any country or region in terms of its general level of wind
speed (see also Table 2). Nothing is stated in this Table with regard to the
type of windstorm that is dominant in a country. A dominant storm type in
one country can produce similar extreme value statistics to another storm
type in a different country. The classification system is based on maximum
3 s gust, which is a more relevant indicator than the 10 min mean for some
storm types, such as downbursts generated by thunderstorms.

TABLE 1
CLASSIFICATION SYSTEM FOR DESIGN WIND SPEEDS
(3 s gust, 10 m height, open country terrain*)

Nominal 500 year

Handbook Nominal 50 year Related region in
return period
Level return period (m/s) AS/NZS 1170.2
(m/s)
I 32 40 —
II 39 45 A
III 44 57 B
IV 52 66 C
V 60 80 D
* Open country terrain is terrain with a roughness length of 0.02.

Figure 1 shows a general indication of the zoning system as applied to the

whole Asia–Pacific region. The maps given in Chapter 5 should be used
for classification as they give more detail for individual countries and
territories.

The classification system given in Table 1 is based on the Australia–New

Zealand regional system with an additional Level I added to accommodate
the equatorial countries of Indonesia, Malaysia, Papua New Guinea, and
Singapore. Normal sampling errors in the analysis of historical wind
speeds (Paragraph A3, Appendix A) often do not justify more closely
spaced values for design wind speeds.

Changing from one level to a higher in the above Table gives a factor on
wind load (assumed proportional to wind speed squared) of about 1.5.

NOTE: The classification system in Table 1 should not be confused with the Saffir–
Simpson Scale for hurricanes used in the United States, and sometimes used for
tropical cyclones and typhoons in the Asia–Pacific region.

COPYRIGHT
HB 212—2002 8

Note: This map gives a general indication only, as many of the areas are too small to
show. Refer to the text and maps in Section 5 (Figures 2 to 11) for classifications of
particular locations.

FIGURE 1 GENERAL MAP OF LEVELS FOR THE REGION

COPYRIGHT
 9 HB 212—2002

CHAPTER 3 WIND SPEED VERSUS RETURN PERIOD

In order to make use of the Australia/New Zealand loading Standards,
which use different return periods (more correctly known as an annual
probability of exceedance) depending on the importance of the structure, it
is necessary to have relationships between the basic wind speed, and the
return period.
The basic wind speed, VR , defined in the following is a 3 s gust at 10 m
height in flat open country terrain, corresponding to an R-year return
period. Relationships are given for the five wind speed Levels defined in
Table 1 of this Handbook. Wind speeds calculated from these equations
should be rounded to the nearest whole number.

3.1 LEVEL I (equatorial regions—strong thunderstorms and

monsoon winds)
A suitable relationship for the equatorial countries of Singapore, Malaysia,
Indonesia, and Papua New Guinea is the following, which is derived from a
composite analysis of 11 stations (Paragraph C1, Appendix C):
− 0.1
V R = 70 − 56R ...1

This gives a value of V 50 (50 year return period gust) of 32.1 m/s. The
value for V 500 is 39.9 m/s.

3.2 LEVEL II (moderately severe thunderstorms and extra-tropical

gales)
The following relationship is appropriate for areas classified as Level II:
− 0.1
V R = 67 − 41R ...2

This gives a value of V 50 of 39.3 m/s for a 50 year return period, and a
value for V 500 of 45.0 m/s.

3.3 LEVEL III (severe thunderstorms and moderate or weakening

typhoons/tropical cyclones)
The following relationship is appropriate for areas classified as Level III:
− 0.1
V R = 106 − 92R ...3

This gives a value of V 50 of 43.8 m/s for a 50 year return period, and a
value for V 500 of 56.6 m/s.

3.4 LEVEL IV (strong typhoons/tropical cyclones)

The following relationship is appropriate for areas classified as Level IV:
− 0.1
V R = 122 − 104R ...4

This gives a value of V 50 of 51.7 m/s for a 50 year return period, and a
value for V 500 of 66.1 m/s.

COPYRIGHT
HB 212—2002 10

3.5 LEVEL V (very strong typhoons/tropical cyclones)

The following relationship is appropriate for areas classified as Level V:
− 0.1
V R = 156 − 142R ...5

This gives a value of V 50 of 60.0 m/s for a 50 year return period, and a
value for V 500 of 79.7 m/s

3.6 SUMMARY
Table 2 summarises the proposed relationships between 3 s gust wind
speed and return period for the five Levels in this Handbook. The values
for 50 years and 500 years return periods correspond to those in Table 1.

TABLE 2
WIND SPEED VERSUS RETURN PERIOD
(3 s gust, 10 m height, open country terrain)

Handbook Equation for

Description V 50 V 500
Level VR
Strong thunderstorms and − 0.1
I 70 − 56R 32 40
monsoon winds
Moderately severe
− 0.1
II thunderstorms and extra- 67 − 41R 39 45
tropical gales
Severe thunderstorms and
− 0.1
III moderate or weakening 106 − 92R 44 57
typhoons/tropical cyclones
Strong typhoons/tropical − 0.1
IV 122 − 104R 52 66
cyclones
Very strong typhoons/tropical − 0.1
V 156 − 142R 60 80
cyclones
NOTE: Calculated wind speeds are rounded to the nearest m/s as more accuracy
cannot be justified.

COPYRIGHT
 11 HB 212—2002

CHAPTER 4 WIND SPEEDS AND AS/NZS 1170

The AS/NZS 1170 series of Standards includes a number of parts, each

accompanied by a Commentary in a separate volume.
AS/NZS 1170, Structural design actions —
Part 0: General principles
Part 1: Permanent, Imposed and other actions
Part 2: Wind actions
Part 3: Snow and ice actions (to be published)
Part 4: Earthquake actions (to be published)
Part 0 is the 'head' Standard and covers the general process of structural
design. It links the Standards for loads to the materials design Standards
and gives a simple format for specifying structural performance in terms of
a design event.
It provides for the selection of annual probability of exceedance (either by
reference to the Building Code of Australia, the appropriate requirements
for New Zealand (in AS/NZS 1170.0) or by setting a value for a particular
structure). It also covers combinations of actions (e.g., wind, permanent
and imposed loads), and verification of design by limit states methods.
The annual probability of exceedance of the design event is equivalent to
the inverse of the 'return period' (P = 1/R) used in AS/NZS 1170.2 and in
this Handbook.
NOTE: Return period expressed in years (sometimes called the average
recurrence interval) should not be taken to imply that the wind will occur once in
each period of years (see Paragraph A2, Appendix A). It is a statistical device
only. In practice, the wind speed may exceed the value more than once, or
approach the value a number of times within the period of years due to the
statistical scatter.

Once a return period is chosen, as appropriate for the design of the

structure, AS/NZS 1170.2 may be used to give a wind speed (for Australia
and New Zealand) and to calculate the forces on the structure.

For locations outside Australia and New Zealand, the wind speeds given in
Chapters 3 and 5 of this Handbook may be substituted for those given in
AS/NZS 1170.2. This means that the designer does the following:
(i) Finds the appropriate design event (expressed as the return period)
for the structure to be designed—by reference to AS/NZS 1170.0, a
building code or other source for the appropriate degree of risk.
NOTE: National codes may give requirements for specific risk levels. These
may be different to those given in AS/NZS 1170.0. For example, there may
be no consideration given for post-disaster structures (with all structures
designed for the same loads), different actions may be treated differently or
additional specific factors may be given for different types of structures.
Other risk-related variations may be hidden in the factors used in
combinations of actions.

COPYRIGHT
HB 212—2002 12

NOTE: If the method of specifying the design event is used, the need to
specify such additional factors is removed. The AS/NZS 1170 series is set
up to use this method. The use of design event as the measure of structural
performance has the potential to improve trade in building products and
services.

(ii) Looks up the location in this Handbook (Chapter 5) and finds the
appropriate wind classification, and hence wind speed for that
degree of risk (Chapter 3).
(iii) Calculates the forces due to wind using AS/NZS 1170.2.
(iv) Combines the loads using AS/NZS 1170.0;
(v) Implements the design process (using materials design Standards).

For all users, this Handbook provides a means of comparing the wind
speeds in various countries across a large portion of the earth. It provides
a means of comparison for design methods, outcomes and products
consistent across the region.

For Australian and New Zealand users, this Handbook enables design of
structures in a large number of countries to be carried out using methods
already familiar to the designer.

The use of a combination factor of 1.0 for wind, coupled with adoption of
the structure 'importance level' scheme used in AS/NZS 1170.0, leads to a
design event for 'normal' structures of 1/500. This is similar to using a
factor of 1.5 for combinations and an annual probability of exceedance of
1/50. Thus, the annual probability of exceedance for 'normal' structures
used in Australia and New Zealand (1/R = 1/500) is nominally the same as
that used by many other countries and also that used in many earthquake
Standards and codes.

For the loading side of the 'load vs resistance' equation, the design
performance is set for a structure simply by specifying the annual
probability of exceedance of the design event (1/R). The value of 1/500
represents a chance of the load being exceeded of approximately 10% in
50 years (assuming 50 years is the reference period for the design life of
'normal' structures). Other performance levels are then defined through
changing the probability of exceedance (i.e., to 1/200 or 1/1000, see
AS/NZS 1170.0 and its Commentary).

The appropriate factors from AS/NZS 1170.2 for use with the Handbook
classifications are as follows:
(a) Handbook levels I, II and III—use Region A and B values.
(b) Handbook levels IV and V—use Region C and D values.
This applies to terrain/height multipliers and turbulence intensity. The
factors FC and FD should only be applied for Australia.

COPYRIGHT
 13 HB 212—2002

CHAPTER 5 WIND SPEEDS FOR SPECIFIC REGIONS

5.1 GENERAL

This Chapter gives maps that show wind speed classifications for various
countries in the Asia–Pacific region, which can be used in conjunction with
AS/NZS 1170.2, by means of the conversion in Table 1.

Chapter 4 indicates how this information may be used with the

AS/NZS 1170 series of Standards.

5.2 AUSTRALIA AND AUSTRALIAN TERRITORIES

This large continental country has a variety of severe wind types, with
large extra-tropical gales along the south coast and Tasmania moving from
west to east, and 'east coast lows' in the Tasman Sea affecting the eastern
coastline. Thunderstorm-generated downbursts originating from local
convection are the dominant windstorms in the interior. The strongest
recorded winds, at 10 m height, in the four major capitals of Sydney,
Melbourne, Adelaide and Brisbane are also caused by local downbursts
from thunderstorms. Severe tropical cyclones can affect the coastline
within about 100 km from the sea between 25° south and 10° south
latitudes. The most common and most severe occur on the west coast
between 20° south and 25° south.

COPYRIGHT
HB 212—2002 14

Figure 2 is a copy of the map given in AS/NZS 1170.2 with the equivalent
classifications noted and directional wind information removed.

In the Australian/New Zealand Standard (AS/NZS 1170.2), four regions are

defined; these are labelled from A to D with increasing basic design wind
speeds. Regions C and D are considered dominated by severe to very
severe tropical cyclones. In region B, generally covering the tropical
coastal strip between 50 km and 100 km inland (including Brisbane),
weaker tropical cyclones can potentially occur. An analysis of historical
recorded extreme wind speeds was carried out recently by Holmes (2002).

Table 3 shows the main sources of strong winds in the various defined
regions. The correspondence with the Handbook classifications is also
shown in Table 3.

TABLE 3
AUSTRALIA

AS/NZS
Description Handbook Level
Region
Thunderstorm downbursts and synoptic winds
A II
(gales)
B Weakening tropical cyclones III
C Moderately severe to severe tropical cyclones IV
D Severe tropical cyclones V
Off-shore Australian territories:
A Lord Howe Island II
Torres Strait Islands, Christmas Island, Norfolk
B III
Island
C Cocos Islands IV

COPYRIGHT
15 HB 212—2002

FIGURE 2 AUSTRALIA

COPYRIGHT
HB 212—2002 16

5.3 NEW ZEALAND

The main extreme winds in New Zealand are temperate synoptic systems,
although the north of the country can experience the effects of decaying
tropical cyclones (extra-tropical transitions). Most of New Zealand has
similar all-directional basic wind speeds but differ in their directional
multipliers. There are a number of mountain areas, especially in the South
Island, where downslope winds occur—for these the wind speed is
increased by a 'Lee Multiplier' of up to 1.35 (see AS/NZS 1170.2 for
details).

Table 4 and Figure 3 show the suggested new classifications.

In addition to the lee multiplier, much of New Zealand is hilly and subject to
hill shape and elevation multipliers as defined in AS/NZS 1170.2.
Therefore, resulting site wind speeds may approach values for flat level
ground in tropical areas.

Extreme wind classification: II, III (some mountain areas)

TABLE 4
NEW ZEALAND

Region Description Handbook Level

Synoptic winds (gales) and decaying
A II
tropical storms
W Synoptic winds in Cook Strait III

COPYRIGHT
 17 HB 212—2002

FIGURE 3 NEW ZEALAND

COPYRIGHT
HB 212—2002 18

5.4 MELANESIA, MICRONESIA AND POLYNESIA

Guam
This Pacific Island has experienced some of the strongest recorded
tropical cyclones. The ASCE Standard, Minimum Design Loads for
Buildings and Other Structures (ASCE-7-98) specifies a 50 year return
period gust speed of 76 m/s.

Extreme wind classification: V

Vanuatu, Fiji, Tonga, Samoa, Western Samoa

These islands are subject to frequent visits from tropical cyclones. ASCE-
7-98 specifies for American Samoa a 50 year return period gust speed of
56 m/s.

Extreme wind classification: IV

New Caledonia

Extreme wind classification: III

Islands within ±5° of the Equator

Extreme wind classification: I

COPYRIGHT
 19 HB 212—2002

FIGURE 4 SOUTH WEST PACIFIC

COPYRIGHT
HB 212—2002 20

5.5 EQUATORIAL SOUTH-EAST ASIAN ARCHIPELAGO

The Equatorial area, with low design wind speeds covers the whole of
Indonesia, Malaysia and Singapore and most of Papua New Guinea, with
around 300 million population. It is known that tropical cyclones/typhoons
cannot form within about 5° to 6° of the Equator, so that these countries are
fortunate in not receiving any significant influence from the world’s worst
extreme wind events.

Although there are likely to be local effects that might justify some micro-
zonation within the region, the reduced importance of wind load relative to
other loads, means that this is probably not justifiable or necessary for the
majority of structures. To this end, historical wind gust data from several
stations in Malaysia, Indonesia and stations in Singapore have been
composited and analysed in Appendix C.

The resulting classifications are shown in Figure 1.

Indonesia
Similar to Malaysia and Singapore, Indonesia is entirely in the equatorial
area, does not experience typhoons, and design wind speeds from weak
thunderstorms and monsoonal winds are low (see Figure 1).

Extreme wind classification: I

Malaysia
Malaysia is entirely in the Equatorial area, does not experience typhoons,
and has very low extreme winds from weak thunderstorms and monsoonal
winds. Monthly maximum wind data is available from more than 30 stations
in the country, including Miri and Kuching in East Malaysia (Sarawak).

Analysis of this data for 50 year return period gust values for 20 stations
by the Malaysian Meteorological Service gave values between 24 m/s and
32 m/s. There is some evidence of higher wind speeds in the highland
stations away from the coastal plains (see Figure 1).

Extreme wind classification: I

COPYRIGHT
 21 HB 212—2002

Singapore
Good quality corrected monthly maximum extreme gust data is available
from Tengah and Changi airfields. Extreme value analysis by Choi and
Tanurdjaja (2001) gives 50 year return period gusts of 30 m/s and 25 m/s,
respectively (a single gust of 40.1 m/s was recorded at Tengah in 1984).
These values are quite compatible with the values obtained for Malaysia.

Extreme wind classification: I

Papua New Guinea

The majority of Papua New Guinea (including Port Moresby) is in the
equatorial Level I, and the design winds, originating from thunderstorms
produced by local convective activity, are quite low. An extreme value
(Gumbel) analysis for Port Moresby by Whittingham (1964) using only 11
years of data, gives a 50 year return period gust of 31 m/s. The addition of
some extra years gives even lower values. The Papua New Guinea loading
code gives a contour map with 50 year return period gust wind speeds
ranging from 24 m/s to 32 m/s. For the south-west tip 40 m/s is specified.
Values for major centres are: Port Moresby 28 m/s, Lae 23 m/s, Rabaul 26
m/s. 34 m/s is specified for Honiara (Solomons)

The islands on the south west are occasionally exposed to developing

Coral Sea cyclones, and should have higher design wind speeds as shown
in Figure 4.

Extreme wind classification: I (most of country), II (south west tip)

COPYRIGHT
HB 212—2002 22

5.6 PHILIPPINES

The Philippines experiences typhoons from the Pacific Ocean, which often
cross the northern Philippines (Luzon) and reform in the south China sea.
On the other hand, the southern island of Mindanao has little or no
influence from typhoons, and effectively has an equatorial extreme wind
climate. An extreme value analysis of 1 min average extreme wind speeds
in the Philippines was carried out in the early 1970s by Kintenar (1971).
This gave widely ranging 50 year return period values, and probably
suffered from sampling errors due to short records.

The National Structural Code of the Philippines specifies three extreme

wind zones with 1 min sustained wind speeds of 200 kph (55.5 m/s),
175 kph (48.6 m/s) and 150 kph (41.7 m/s) respectively. These values
have been used for Table 4 below, but are probably conservative for
Mindanao.

Figure 5 shows a converted map based on a recent re-analysis with

suggested new classifications and Levels II and III used for Mindanao.

TABLE 5
PHILIPPINES

Zone in Philippines
Description Handbook Level
Structural Code
I Eastern Luzon V
II Remainder of Philippines IV
II Eastern Mindanao III
III Western Mindanao II
III Palawan II

COPYRIGHT
 23 HB 212—2002

FIGURE 5 PHILLIPINES

COPYRIGHT
HB 212—2002 24

5.7 INDOCHINA

Vietnam
Vietnam is influenced by typhoons over most of its coastline, although the
influence is weaker on the southern provinces. For design wind speeds,
Vietnam is divided into five zones with 20 year return period gust speeds
ranging from 33 m/s to 55 m/s, in the national loading code TCVN-2737
(values of dynamic pressure are given in the code). The zones of higher
wind speeds occur close to the coast and reflect different degrees of
influence from typhoons. Table 6 and Figure 6 detail the new
classifications based on the contours given in TCVN-2737.

TABLE 6
VIETNAM

Zone in Vietnamese
Description Handbook Level
Loading Code
I Inland north and south I
II Inland southern delta II
III, IV Central and northern coastline III
V Offshore islands in north IV

Cambodia
No data is available. Assume Level I (as for Vietnam Level I) for the
eastern part; Level II for the western part (as for Thailand).

Laos
No data is available. Assume Level I (as for Vietnam Level I) for the
eastern part; Level II for the western part (as for Thailand).

COPYRIGHT
 25 HB 212—2002

FIGURE 6 VIETNAM, CAMBODIA, LAOS

COPYRIGHT
HB 212—2002 26

5.8 THAILAND

Thailand has a particularly mixed wind climate. Most of the country

appears to be dominated by extreme winds from thunderstorms and
monsoons. Typhoons have occasionally impacted on southern Thailand,
as did Typhoon 'Gay' in 1989, inflicting considerable damage. Post-landfall
typhoons can also affect north-west Thailand.

An analysis of historical gust data for 60 meteorological stations is

described by Davenport et al (1995), using Type I (Gumbel) Extreme Value
Distributions. There were apparently siting problems for many of the
anemometers, and although extreme winds caused by typhoons were
separated, those from thunderstorms apparently were not.

The analysis by Davenport et al (1995) proposed two design wind speeds

based on 50 year return period 10 min means of 26.5 m/s and 30 m/s. The
latter value applies to small areas on the east and north-west of Thailand.
In recent proposals for a new Thailand loading code (Lukkunaprasit, 1997),
these values were converted to nominal mean hourly speeds of 24.9 m/s
and 28.2 m/s respectively.

Extreme wind classification: II

5.9 MYANMAR

No data or analysis is available, but design wind speeds can be assumed

to be the same as for Thailand.

Extreme wind classification: II

COPYRIGHT
 27 HB 212—2002

5.10 CHINA REGION

China (including Taiwan)

China is a large country with a range of extreme wind types ranging from
severe gales arising from synoptic systems in Siberia in the north-west, to
typhoons along the southern coastline. There is also a region with
downslope winds.

The GBJ-9–87 loading code published by the Department of Standards

and Norms, Ministry of Construction, includes a wind loading section. This
is also available as an English translation (GBJ-9 – Department of
Standards and Norms, 1994). A map is included with this Standard, which
2
gives contours of dynamic pressure in kN/m (kPa). The Standard states
that the 'wind reference pressure' is calculated from the 10 min mean wind
2
speed at 10 m height by the formula, wo = v o /1600. These values have a
50 year return period (recently revised from 30 year return period). Values
of dynamic pressure on these contours range from 0.30 kPa to 0.90 kPa.
For most of the country, the values are in the range 0.30 to 0.50 kPa.
A revision was published in early 2002 with the same map of wind speeds
(GB 50009—2001).

A converted map is given in Figure 7 and the relationships in Table 7.

TABLE 7
CHINA

Handbook
Region Description
Level
Central Pressure contours from 0.30 to
II
Mainland 0.50 kPa
North west and inner Pressure contours from 0.60 to
III
southern coast 0.70 kPa
Outer southern coast and Pressure contours from 0.80 to
IV
islands, Hainan 0.90 kPa
Taiwan Severe tropical cyclones V

COPYRIGHT
HB 212—2002 28

Hong Kong and Macau

As for the rest of the south China coastline, Hong Kong and Macau are
subjected to frequent visits from moderate to severe typhoons. Hong Kong
has good quality recorded wind speed data extending more than a hundred
years from 1884 to 1957, from the Royal Hong Kong Observatory, and
since 1957, from Waglan Island. Analysis of extreme winds from typhoons
has been carried out by a number of authors including Faber and Bell
(1967), Chen (1975), Davenport et al (1984), Melbourne (1984), Holmes et
al (2001). Most of these studies have normalised the wind speeds to a
height of 50 m, rather than 10 m. Design wind speeds in Hong Kong and
Macau are set by the respective building departments.

The Hong Kong Code of Practice implies a 50 year return period 3 s gust
wind speed at 50 m height of about 68 m/s. This compares with the values
obtained by Chen (1975) of 70.5 m/s using annual maxima, and of
63.5 m/s by Melbourne (1984), and 59.0 (Observatory) and 61.8 (Waglan
Island) by Holmes et al (2001). All these studies used only recorded
typhoon data.

A relationship between the gust wind speed at 50 m height, and return

period is determined in Paragraph D2, Appendix D. This relationship (given
below) should be used for Hong Kong in preference to the wind Level
system given elsewhere in this Handbook.

Hong Kong
Equation for VR V 50 V 500
(50 m height)
− 0.1
95.5 − 54R 59 66
NOTE: 3 s gust speed at 50 m height in urban terrain.

COPYRIGHT
29 HB 212—2002

FIGURE 7 CHINA REGION

COPYRIGHT
HB 212—2002 30

5.11 JAPAN

Japan is subject to typhoons from the Pacific in Kyushu and Okinawa, and
temperate synoptic systems in the north of the country. The Architectural
Institute of Japan has a contour map of design wind speeds (10 min mean,
100 year return period) in its wind load recommendations. Values range
from 26 m/s to 44 m/s on the main islands, to 50 m/s on Okinawa, which is
subject to frequent severe typhoons. A converted map is given in Figure 8.

Extreme wind classification: II, III, IV, V

COPYRIGHT
 31 HB 212—2002

NOTE: Izu, Ogasawara, Satsunan, Okinawa, Daitou and Sakishima Islands are all
classification V.

FIGURE 8 JAPAN

COPYRIGHT
HB 212—2002 32

5.12 KOREA

The coastline of South Korea has some influence from typhoons on the
south and east coasts and the island of Cheju.

The Architectural Institute of Korea has a map of 10 min mean 100 year
return period wind speeds varying from 25 m/s in the inland centre to
40 m/s at some points on the eastern and southern coastline. Seoul is
specified as 30 m/s. A converted map is given in Figure 9.

Extreme wind classification: II, III, IV, V

FIGURE 9 KOREA

COPYRIGHT
 33 HB 212—2002

5.13 SRI LANKA

The east coast of Sri Lanka is exposed to relatively weak tropical cyclones.
A building code was prepared by an Australian consulting group in the
1970s. Three design wind zones are specified. The converted map is given
in Figure 10 and the relationships in Table 8.

TABLE 8
SRI LANKA

Zone in Sri Lanka

Description Handbook Level
Building Code
1 50 km from east coast III
2 Inland strip II
3 South and west (including Colombo) I

FIGURE 10 SRI LANKA

COPYRIGHT
HB 212—2002 34

5.14 INDIA

India, a large sub-continental tropical country, has a range of extreme wind

zones, with extreme tropical cyclones being dominant on the east (Bay of
Bengal) coast, and less frequent ones on the west coast. In inland areas,
thunderstorms and monsoon winds are prevalent.

India has a good network of meteorological stations, and there have been
a number of extreme value analyses of wind speeds summarised by
Sharma et al (1995). The Indian Standard for Wind Loads IS 875 Part 3
(1987) divides the country into six zones, giving 50 year return period gust
wind speeds ranging from 33 m/s to 55 m/s. Figure 11 shows a converted
map and Table 9 gives the relationships.

TABLE 9
INDIA

Zone in
Description Handbook Level
IS 875.3 *
1 Tripura, Mizoram, Ladakh IV

Coastal strips of Tamil Nadu (including

2 Chennai), Andhra Pradesh, Orissa, Gujarat, III
West Bengal (including Calcutta), Assam
Northern India including Delhi, central Tamil
3 III
Nadu
Coastal strip on Arabian Sea, including
4 II
Mumbai, inland Madhya Pradesh, Orissa
5 Most of southern India II
6 Inland Karnataka, including Bangalore I
* Refer to map in IS 875: Part 3, for details of zones

COPYRIGHT
 35 HB 212—2002

NOTE: Refer to the map in IS 875: Part 3, for details of zones.

FIGURE 11 INDIA

COPYRIGHT
HB 212—2002 36

APPENDIX A
EXTREME WIND ANALYSIS

A1 Extreme Value Analysis of Historical Data

Generally, meteorological data is obtainable from meteorological offices in
the respective countries. In some cases papers have been written, and are
in the public domain, which contain analyses for data from a particular
country.
Since the 1950s onwards, extreme value analysis has been commonly
used to make long-term extrapolations of extreme wind speeds for
structural design from historical data.
In the 1920s, Fisher and Tippett (1928) identified the mathematical forms
of three limiting extreme value distributions—thereafter known as Type I, II
and III. Jenkinson (1955) showed that they can be conveniently
represented in a single common mathematical form, known as the
Generalized Extreme Value distribution (G.E.V.). The most commonly used
distribution of these three is the Type I, also commonly called the 'Gumbel
Distribution'. It has the advantage of having only two adjustable
parameters and being relatively simple to apply. The Generalized Extreme
Value distribution has three parameters, one of which is the shape factor,
k.
The cumulative probability distribution function F U(U) of the maximum wind
speed in a defined period (e.g., one year) may be given as follows:
1/k
F U(U) = exp{ −[1 − k (U − u)/a] } . . . A1

In Equation A1, U is a location parameter and a is a scale factor. When

k < 0, the G.E.V. is known as the Type II Extreme Value Distribution; when
k > 0, it becomes a Type III Extreme Value Distribution. As k tends to 0,
Equation A1 becomes the Type I Extreme Value Distribution, or Gumbel
Distribution.
When historical wind data is analysed, and the correct procedures are
adopted, it is usually found that small positive values of shape factor, k, in
the range 0.1 to 0.3 are obtained. One of the necessary procedures is to
separate out extreme winds caused by storms of different types, such as
thunderstorms and typhoons. If this is not done, the resulting distribution
may be a combination of two, or more, different independent distributions,
and give misleading, and incorrect results, when extrapolated to design
values.
A positive value of k corresponds to a distribution with a theoretical upper limit
to the wind speed that can be obtained from it. It may be possible to calculate
an upper limit to the gust wind speed that can be generated by a typhoon, or
tropical cyclone, from thermodynamic arguments, although it may be quite
difficult to do so. The highest recorded wind speed in the Asia–Pacific region,
is reported to be a gust wind speed of 110 m/s in a typhoon at Casiguran, east
th
coast of the island Luzon in the Philippines on 10 September, 1970.

COPYRIGHT
 37 HB 212—2002

There is extensive literature on methods of extreme value analysis for wind

speeds. A useful review of much of this has been given by Palutikof et al
(1999). The method of 'peaks over threshold' analysis, which has been used
recently in Australia, is described by Holmes and Moriarty (1999), Holmes
(2001), Holmes et al (2001) and Holmes (2002).

A2 Return Period and Design Wind Speeds

The Return Period, R, is simply the inverse of the complementary
cumulative distribution of the extremes.

i.e., Return period, R = 1/(Probability of exceedance)

= 1/(1 − F U(U)) . . . A2

Thus, if the annual maximum is being considered, then the return period is
measured in years. A 50 year return period wind speed has a probability of
exceedance of 0.02 (1/50) in any one year. It should not be interpreted as
recurring regularly every 50 years.
The most common choice of return period in national Standards is
50 years, although the Australian/New Zealand loading Standard uses a
value of 500 years for normal structures (as the design event). This is the
same design event used by earthquake Standards around the world and is
coupled with a load factor for combinations of 1.0.
There should be no confusion between return period, R, and expected
lifetime of a structure, N. The return period is just an alternative statement
of annual risk of exceedance, e.g., a wind speed with a 50 year return
period is one with an expected risk of exceedance of 0.02 (1/50) in any
one year. The risk, r, of exceedance of a wind speed over the lifetime of
the structure, may be determined by assuming that all years are
statistically independent of each other.
Then,
N
r = 1 − (1 − (1/R)) . . . A3

Setting both R and N equal to 50 years in Equation A3, a value for r of

0.64 is obtained. There is thus a 64% chance that the 50 year return
period wind speed will be exceeded at least once during a 50 year lifetime.
Alternatively, if we accept a risk of 10% that a particular wind speed will be
exceeded at least once during a 50 year lifetime, then Equation A3 (solved
for R) will give a return period of:
0.02
1/(1 − (0.9) ) = 475 years . . . A4

The risk of exceedance of a particular wind speed for design is also related
to the load factor γW, which is applied to the calculated wind loads when
calculating the structural resistance (using combinations of actions). Thus,
if a wind load factor of 1.5 is applied to a nominal 50 year return period
value of wind speed in the design process, a design wind speed equal to
0.5
(1.5) times the 50 year return period value is, in effect, being used.

A3 Errors In Estimation of Extreme Winds

There are several types of errors that inevitably occur when estimating
extreme wind speeds for design. These can be summarised as follows.

COPYRIGHT
HB 212—2002 38

Observation Errors
Observation errors arise from several different sources. Over a period of
40 or 50 years at a measurement station, the instrument used to measure
wind speeds is unlikely to have remained the same. For example, until
about 1990, most recording stations in Australia were equipped with a Dine
pressure tube type of anemometer. Since then, most of these have been
replaced by Automatic Weather Stations (AWS) with small 3 cup
anemometers. Several different types of the latter have been used,
depending on the successful tenderer for the AWS contract. Studies of the
response of these different instruments indicate that they should both be
capable of recording gust wind speeds with a 2-3 s averaging time at
design wind speeds with adequate accuracy (e.g., Sachs, 1978); however,
on some occasions when both instruments have been operating in close
proximity to each other, significant differences in the maximum gust
readings have been recorded. If regular calibration (both static and
dynamic) of these instruments is not carried out, significant errors can
arise from this source.
Siting of the tower or mast supporting the anemometer is another potential
source of error. For many years, it was quite common to mount
anemometer masts on, or near, buildings of significant size. The
aerodynamic influence of the building may be corrected by the use of wind-
tunnel tests, but such corrections may not be reliable if they are large. It is
not advisable to use wind data obtained from a city centre anemometer,
unless there are none available from a better-sited station such as an
airport.
Urban development over a long period of years may justify correction of
anemometer records for the effect of boundary-layer changes, so that the
corrected values are representative of those obtained over fully developed
open country terrain, at 10 m height. Such corrections are difficult to make
when a wind gust is produced by a severe downdraft from a thunderstorm. For
downdrafts, the effects of segments of varying terrain roughness in the upwind
fetch are expected to be small, and correction may not be justified.

Distribution and Fitting Errors

In Paragraph A1, the various forms of the Generalised Extreme Value
distribution are discussed. There are many choices in the method of fitting
of the G.E.V., to historical wind data. Many of these methods are based on
the ab initio assumption that the shape factor, k, is equal to zero (i.e., a
Type I or Gumbel distribution is assumed). The choice of method of fitting,
including selection of the shape factor, is a subject of continuing debate in
the technical literature.
Comparisons of different approaches applied to the same data indicate
that differences in predictions for return periods (Paragraph A2) of the
same order of magnitude as the length of the data, resulting from different
methods of fitting, are small (e.g., Holmes and Moriarty, 1999). This is not
the case when extrapolations are made to high return periods, say 500 to
1000 years. The latter values of return period are relevant when estimating
wind speeds with small risks of exceedance during the lifetime of a
building, for ultimate strength design (Paragraph A2).

COPYRIGHT
 39 HB 212—2002

As already discussed (Paragraph A1), significant errors will also arise if

extreme winds from different storm types are not separated and analysed
independently.

Sampling Errors
Sampling errors are errors in predicted wind speeds of specified return
periods arising from uncertainties in the parameters of the distribution
used to make the predictions. These errors arise when data from relatively
short data records are used to make predictions to high return period. Note
when assessing sampling errors, it is assumed that the correct probability
distribution has been chosen—the errors arise from the uncertainties in the
parameters of the chosen distribution.
Simiu et al (1978) have studied the sampling errors resulting from fitting
the G.E.V. distribution to extreme wind speeds (also discussed in Simiu
and Scanlan, 1996). Simiu et al found that the sampling error in estimating
a wind speed with a 50 year return period from 25 years of data, with a
68% confidence level, is about 7%. The error in estimating the 1000 year
return period value from 25 years of data is calculated to be 9%. In respect
of the latter value, it should be noted that this estimate is based on an
assumption of a shape factor, k, fixed equal to 0 (Type I extreme value
distribution). Any error in the shape factor estimation, or selection, is not
included in the sampling errors obtained by Simiu et al.

COPYRIGHT
HB 212—2002 40

APPENDIX B
BASIC STRONG WIND TYPES

B1 General
Wind is air movement relative to the earth, driven by several different
forces, especially pressure differences in the atmosphere, which are
themselves produced by differential solar heating of different parts of the
earth's surface, and forces generated by the rotation of the earth. The
differences in solar radiation between the poles and the Equator, produce
temperature and pressure differences. These, together with the effects of
the earth's rotation set up large-scale circulation systems in the
atmosphere, with both horizontal and vertical orientations. The result of
these circulations is that the prevailing wind directions in the tropics, and
near the poles, tend to be easterly. Westerly winds dominate in the
temperate latitudes.
Local severe winds may also originate from local convective effects
(thunderstorms), or from the uplift of air masses produced by mountain
ranges (downslope winds). Severe tropical cyclones, known in some parts
of the world as hurricanes and as typhoons, generate extremely strong
winds over some parts of the tropical oceans and coastal regions, in
latitudes from 10 to about 30°, both north and south of the equator.

B2 Synoptic Winds (gales)

In the mid-latitudes from about 40° to 60°, the strongest winds are gales
generated by large and deep depressions or (extra-tropical) cyclones, of
synoptic scale. They can also be significant contributors to winds in lower
latitudes. Navigators, particularly in sailing ships, are familiar with the
strong westerly winds of the 'roaring forties'.
These systems are usually large in horizontal dimension—they can extend
for more than 1000 km, so can influence large areas of land during their
passage (several countries in the case of Europe). These winds may take
several days to pass, although they may not blow continuously at their
maximum intensity during this period. They tend to be quite turbulent near
the ground, as the flow has adjusted to the frictional effects of the earth’s
surface over hundreds of kilometres. The direction of the winds remains
quite constant over many hours.

B3 Monsoonal Winds
The Asian summer monsoon, which affects most of south and east Asia
arrives from the east, but produces southerly or south-westerly winds at
ground level. These are generally quite light but can sometimes produce
significant monthly maximum wind gusts.

COPYRIGHT
 41 HB 212—2002

B4 Thunderstorm Winds
Thunderstorms, both isolated storms, and those associated with advancing
cold fronts, are small disturbances in horizontal extent, compared with
extra-tropical depressions and tropical cyclones, but they are capable of
generating severe winds, through tornadoes and downbursts. They are the
main source of high winds in the equatorial region (within about 10° of the
Equator), although their strength is not high in these regions.
Thunderstorms derive their energy from heat. Warm moist air is convected
upwards to mix with the drier upper air. With evaporation, rapid cooling
occurs and the air mass loses its buoyancy, and starts to sink.
Condensation then produces heavy rain or hail, which falls dragging cold
air with it. A strong downdraft reaches the ground, and produces a strong
wind for a short period of time—perhaps 5 to 10 min. The strongest winds
produced by this mechanism are known as downbursts, which are further
subdivided into microbursts and macrobursts, depending on their size. The
strongest winds produced by these events have a large component of wind
speed due to the forward motion of the convection cell.

B5 Tornadoes
Tornadoes are vertical funnel-shaped vortices that are often created in the
strongest convection cells called 'supercells'. They are typically small in
diameter, around 100 m, but can travel long distances with destructive
winds.
They occur mainly in large continental plains in countries such as the USA,
Argentina, Russia and South Africa.

B6 Downslope Winds
'Downslope' is a generic name for a variety of strong wind types in
mountain areas that are influenced by the large topography of a region.
Katabatic wind is a downslope wind of cold dense air at high elevations
draining into valleys. Maximum velocity occurs just before sunrise.
The Föhn or Chinook wind is a strong gusty dry, warm wind that develops
on the lee side of a mountain range. They may be produced by lee waves
developed in the stable upper atmosphere behind a long mountain range.
These types of wind occur in New Zealand on the eastern side of the Alps
in the South Island, as well as in the Northern Alps of Europe, and the
Rocky Mountains of North America.

B7 Cyclonic Winds (typhoons)

Tropical cyclones are intense cyclonic storms that occur over the tropical
oceans, mainly in late summer and autumn. They are driven by the latent
heat of the oceans, and require a minimum sea temperature of about 26° C
to sustain them; they rapidly degenerate when they move over land, or into
cooler waters. They will not form within about 5° of the equator, and do not
reach full strength until they reach at least 10° latitude. They are usually at
full strength when they are located between 20 and 30° latitude, but can
travel to higher latitudes if there are warm ocean currents to sustain them.

COPYRIGHT
HB 212—2002 42

The strongest tropical cyclones have occurred in the South China Sea,
where they are called typhoons, and off the north-west coast of Australia.
Areas of medium typhoon and tropical cyclone activity are the South
Pacific, southern Japan, Bay of Bengal and the Coral Sea (off eastern
Australia). Regions of lesser activity or weaker storms are the Arabian
Sea, the Gulf of Thailand, and the north coast of Australia (including the
Gulf of Carpentaria).

COPYRIGHT
 43 HB 212—2002

APPENDIX C
WIND LOADING CODES AND STANDARDS

C1 General
Many countries in the Asia–Pacific region have their own loading
Standards or codes with wind loading components, or stand-alone wind
loading documents. Others are in the process of developing their own
Standards.

C2 ISO 4354
ISO 4354, Wind actions on structures, is the International Standard for
wind actions, published by the International Organization for
Standardization. This international Standard is not a true wind code for use
in design. It provides a framework for national code writers to use in
producing wind Standards. It gives definitions of three factors for use in
calculating wind pressures and gives the equation for their use. It does not
provide wind speeds for individual countries, and does not consider the
special aspects of tropical cyclone and thunderstorm winds.

C3 China
The Load code for the design of building structures (GBJ 9-87), a National
Standard of the Peoples Republic of China, is available as an English
translation. The wind pressure map in this Standard is discussed in
Paragraph 5.10.
The 1994 document is published by New World Press, 24 Baiwanzshuang
Road, Beijing, 100037, China.
The recent revision, GB 50009—2001, has been published by the Building
Industry Press, at the same address.

C4 India
The Indian Code of Practice for design loads (other than earthquake) for
buildings and structures Part 3 Wind Loads IS 875 (Part 3):1987 was
reaffirmed with minor revisions in 1997 with a wind map dated 1993. This
Standard should be relatively easy to use for engineers from Australia and
New Zealand, as it has a similar appearance to former versions of
AS 1170.2 and NZS 4203. It is noted in the foreword that account had
been taken of both AS 1170:2—1983 and NZS 4203:1976 during the
preparation of IS 875 (Part 3):1987, as well as the British and American
Standards. The basic wind speed is a 3 s gust with a 50 year return period.
Terrain categories are defined identically with those in AS/NZS 1170.2.
The gust factor calculations for along-wind dynamic response of tall
structures is the graphically based method used in the pre-1989 versions
of AS 1170.2.
As discussed in Paragraph 5.13, India is divided into six colour-coded
zones with a considerable range of design wind speeds.

COPYRIGHT
HB 212—2002 44

C5 Japan
The Recommendations of the Architectural Institute of Japan were revised
in 1993 (English language edition published in 1996) and are a
comprehensive loading code including the effects of dead, live, snow,
seismic, temperature, earth and hydraulic pressure, as well as wind loads.
Chapter 6 on wind loads comprises 20 pages, with 30 pages of
Commentary. This is a comprehensive and advanced wind loading
document, although the Recommendations have no legally binding
standing in Japan. The Building Law of Japan has a separate set of
simplified wind loading rules.

C6 Vietnam
The loading code of Vietnam is TCVN 2737—1995. This is now based on a
3 s gust wind speed with a 20 year return period (previous versions were
based on a 2 min averaging time). It has several advanced features,
including methods for accounting for the dynamic effects of wind.

C7 Malaysia
Malaysia has a Code of Practice on Wind Loading for Building Structures,
CIS:5. This document has drawn on the American Standard (ASCE-7), and
International Standard ISO 4354, as well as AS/NZS 1170.2:2002, the joint
Australia/New Zealand Standard. In fact large parts of CIS:5 are derived
from AS/NZS 1170.2, and this Code can be easily utilised in conjunction
with this Handbook.
CIS:5 is published by Construction Industry Development Board, Level 8,
Grand Seasons Avenue, No. 72, Jalan Pahang, 53000 Kuala Lumpur,
Malaysia .

COPYRIGHT
 45 HB 212—2002

APPENDIX D
ANALYSIS OF WIND SPEEDS FOR LEVEL I AND HONG KONG

D1 Equatorial Level
Historical wind gust data from six stations in Malaysia, two in Indonesia,
two stations in Singapore and Port Moresby (Papua New Guinea) have
been composited, fitted to an extreme value distribution, and plotted
together on an extreme value plot. This gave a total of 352 station-years.
The resulting plot is given in Figure D1.
A single fit has been made to the data in Figure D1; however, closer
inspection of the data indicates that there may be two different
populations—with the lower speeds caused by monsoons, and the higher
values produced by thunderstorms.

FIGURE D1 EQUATORIAL WINDS

The predictions obtained from all 352 values, and those from the top 12
gusts are tabulated in Table D1. It would be advisable to use the more
conservative values derived from the top 12 gusts.
The following equation, for basic wind speed as a function of return period
(3 s gust speed at 10 m height in open country terrain), fits the data from
the equatorial area:
− 0.1
V R = 70 − 56R . . . D1

COPYRIGHT
HB 212—2002 46

TABLE D1
PREDICTED GUST WIND SPEEDS (m/s) FOR EQUATORIAL REGIONS
(3 s gust, 10 m height, open country terrain*)

Return Period From all 352 values From top 12 values

10 25.6 25.1
20 27.7 27.7
50 30.4 31.1
100 32.4 33.6
200 34.5 36.1
500 37.2 39.4
1000 39.2 41.9
* Open country terrain is terrain with a roughness length of 0.02.

D2 Hong Kong
Hong Kong is fortunate in having an excellent long-term record of extreme
wind speeds commencing in the nineteenth century from the Royal
Observatory located in Kowloon.
Maximum winds were recorded at the station from 83 typhoons between
1884 and 1957. The maximum recorded gust was 67 m/s in 1937. From
1884 to 1910 a cup anemometer was used and only hourly mean values
were obtained. These were corrected to equivalent peak gust speeds, by
multiplying by a gust factor of 1.8, an average value obtained from later
years.
The data was corrected to the equivalent values at a height of 50 m over
flat ground in urban terrain (Melbourne 1984), and then analysed by the
'peaks over threshold' approach (Holmes and Moriarty, 1999). This
analysis gave a shape factor, k, of 0.255, and predictions of wind speed as
a function of return period listed in Table D2.

TABLE D2
PREDICTED GUST WIND SPEEDS AT HONG KONG
(3 s gust, 50 m height, urban terrain)

Predicted gust speed

Return Period
(m/s)
10 50.3
20 54.5
50 59.0
100 61.7
200 64.0
500 66.5
1000 68.1

A similar analysis of typhoon gust data from the station at Waglan Island
(1953–1999) gives values of 50 year return period gusts about 3 m/s higher
than those from the Observatory, reflecting the more exposed location of
Waglan (Holmes et al, 2001).
The following equation, for basic wind speed as a function of return period
(3 s gust speed at 50 m height in urban terrain), fits this data:
− 0.1
V R = 95.5 − 54R . . . D1

COPYRIGHT
 47 HB 212—2002

APPENDIX E
REFERENCES AND BIBLIOGRAPHY
The following documents are referred to in this Handbook or provide useful
background material:
• CHOI, E.C.C., TANURDJAJA, A., 'Extreme wind studies in Singapore—an
area with mixed weather systems', 5th. Asia–Pacific Conference on Wind
Engineering, Kyoto, Japan, October 21-24, 2001.
• DAVENPORT, A.G., LUKKUNAPRASIT, P., HO, T.C.E., MIKITIUK, M., and
SURRY, D, 'The design of transmission line towers in Thailand',
Proceedings, Ninth International Conference on Wind Engineering, New
Delhi, 9-13 January, 1995, 57-68.
• FISHER, R.A., and TIPPETT, L.H.C., 'Limiting forms of the frequency
distribution of the largest or smallest member of a sample', Proceedings,
Cambridge Philosophical Society Part 2, 24, 1928, 180-190.
• HOLMES, J.D., Wind loading of structures, Spon Press, London, 2001.
• HOLMES, J.D. and MORIARTY, W.W., 'Application of the generalized
Pareto distribution to extreme value analysis in wind engineering', Journal
of Wind Engineering and Industrial Aerodynamics, 83, 1999, 1-10.
• HOLMES, J.D., HITCHCOCK, P., KWOK, K.C.S. and CHIM, J.K.S., 'Re-
analysis of typhoon wind speeds in Hong Kong', 5th. Asia–Pacific
Conference on Wind Engineering, Kyoto, Japan, October, 2001, pp. 21-24.
• HOLMES, J.D., A re-analysis of recorded extreme wind speeds in Region
A, Australian Journal of Structural Engineering, Vol 4, 2002, pp 29-40.
• JENKINSON, A.F., 'The frequency distribution of the annual maximum (or
minimum) values of meteorological elements', Quarterly Journal of the
Royal Meteorological Society, 81, 1995, 158-171.
• LUKKUNAPRASIT P., 'Seismic and wind loading codes in Thailand',
International Workshop on Harmonization in Performance-based building
structural design, Tsukuba, Japan, December 1997, 1-3.
• MELBOURNE, W.H., 'Design wind data for Hong Kong and surrounding
coastline', Third International Conference on Tall Buildings, Hong
Kong/Guangzhou, 10-15 December, 1984.
• PALUTIKOF, J.P., BRABSON, B.B., LISTER, D.H. and ADCOCK, S.T., 'A
review of methods to calculate extreme wind speeds', Meteorological
Applications, 6, 1999, 119-132.
• SACHS, P., Wind forces in engineering, 2nd edition, Pergamon Press,
Oxford, U.K., 1978.
• SIMIU, E. and SCANLAN, R.H., Wind effects on structures—fundamentals
and applications to design, 3rd. edition, John Wiley, New York, 1996.
• SIMIU, E., BIETRY, J. and FILLIBEN, J.J., 'Sampling errors in estimation
of extreme winds', Journal of the Structural Division, ASCE, 104, 1978,
491-501.
• WHITTINGHAM H.E., 'Extreme wind gusts in Australia', Bureau of
Meteorology, Bulletin No. 46, 1964.

COPYRIGHT
HB 212—2002 48

• FABER, S.E., BELL, G.J., 'Typhoons in Hong Kong and building design',
Hong Kong Royal Observatory, Reprint No. 37, 1967.
• CHEN, T.Y., 'Comparison of surface winds in Hong Kong', Hong Kong
Royal Observatory, Technical Note 41, 1975.
• DAVENPORT, A.G., GEORGIOU, P.N., MIKITIUK, M., SURRY, D., and
LYTHE, G., 'The wind climate of Hong Kong', Proceedings, 3rd
International Conference on Tall Buildings, Hong Kong and Guangzhou,
1984.
• MELBOURNE, W.H., 'Design wind data for Hong Kong and surrounding
coastline', Proceedings, 3rd International Conference on Tall Buildings,
Hong Kong and Guangzhou, 1984.
• SHARMA, V.R., SEETARAMULU, K. and CHAUDRY, K.K., 'Studies on
extreme winds in the Indian sub-continent', Preprints, Ninth International
Conference on Wind Engineering, New Delhi, 9-13 January, 1995.

COPYRIGHT
This page is intentionally left blank.
 GPO Box 5420 Sydney NSW 2001
Administration Phone (02) 8206 6000 Fax (02) 8206 6001 Email mail@standards.com.au
Customer Service Phone 1300 65 46 46 Fax 1300 65 49 49 Email sales@standards.com.au
Internet www.standards.com.au

ISBN 0 7337 4817 1 Printed in Australia

This page has been left intentionally blank.