1

WIND RESISTANT DESIGN OF TALL BUILDINGS IN JAPAN

YUKIO TAMURA
Wind Engineering Research Center, Tokyo Polytechnic University
1583 Atsugi, Kanagawa, Japan


Yukio Tamura is a professor
of the Tokyo Polytechnic
University, and the Director of
the 21st Century Center of
Excellence (COE) Program
named "Wind Effects of
Buildings and Urban
Environment" authorized by
the Ministry of Education,
Culture, Sports, Science and
Technology of Japan. He
currently holds several
responsible positions in the
wind engineering society, including Chairman of the
Sub-Committee on Wind Loading of the Architectural
Institute of Japan, Chairman of the Steering and
Academic Committee of the Japan Association for
Wind Engineering, and Convener of WG-E (Dynamic
Responses) of the Codification Working Group of the
International Association for Wind Engineering.




Yukio Tamura
Abstract

This paper introduces the current status and special
features of wind resistant design of tall buildings in
Japan. It describes the Building Standard Law of Japan
(BSLJ) and Recommendations for Loads on Buildings
published by the Architectural Institute of Japan (AIJ-
RLB). It also introduces details of the new version of
the AIJ-RLB (to be published in March 2004). One of
the most significant revisions is the introduction of
wind directional factor to realize more rational and
economic design of tall buildings, which was difficult
in tropical-cyclone-prone regions. As a result, the
combined effects of along-wind, crosswind and
torsional forces become more significant for precise
estimation of wind loads. These are also introduced
clearly in the provisions of the AIJ-RLB. Some other
relevant design issues such as wind environmental
problems for pedestrians, habitability under vibrations
for building occupants, design damping values, and
damping devices are also discussed.
2
INTRODUCTION

A special feature of building design is the external forces imposed on structures. In the Kobe Earthquake, which
occurred on January 17, 1995, a maximum acceleration of 818cm/s
2
was recorded, and fatalities numbered 6,432.
Typhoon No.19 in 1991 made a landfall on September 27 and traveled through the Japanese islands. New wind
speed records were registered at various weather stations, e.g. maximum peak gusts of 88.8m/s and 58.1m/s were
recorded at Shimokoshiki Island and Hiroshima City, respectively. It caused severe damage to buildings, power
transmission towers, agricultural and forestry areas, etc., and property insurance of 5 billion USD was paid out. This
was a world record for insurance money paid out for one natural disaster at that time. Tall buildings in Japan have to
resist two different kinds of extreme design forces. More recently, Typhoon No.14 attacked southern part of Japan
on September 11, 2003, and a maximum peak gust of 74.1m/s was officially recorded at the meteorological station
on Miyakojima Island. It is interesting that an anemometer at an experimental site for power transmission cables on
the Miyakojima Island recorded a peak gust of 96m/s, and the calibration of this anemometer is now being checked
to ensure that this reading is correct. The special feature of building design in Japan is that they need to resist
extremely strong winds and earthquakes.

It is well known that light, flexible buildings are favorable for resisting seismic forces, while heavy, stiff buildings
are favorable for resisting wind forces. Thus, tall buildings in Japan have to satisfy these two opposite design criteria,
and this is one of the most difficult design issues for tall buildings in Japan. In addition, there are many important
issues to be solved in the design stage, the construction stage, and even in the maintenance stage after completion of
a tall building. Many wind engineers have devoted themselves to finding solutions to these issues, and to exploring
new methods and technologies.


BUILDING STANDARD LAW OF JAPAN


Design Approval Procedure
The design of buildings in Japan should be based on the Building Standard Law of Japan (BSLJ), which specifies
the minimum building design requirements. The BSLJ was completely revised in 1998 2000, and the basic
structural analysis method was shifted to Performance Based Design (PBD). However, this requires greater accuracy
for structural design. PBD also requires more accurate calculation methods reflecting the real situation in evaluating
wind loads.

The wind load provisions of the current BSLJ are similar to those of the Recommendations for Loads on Buildings
published in 1993 by the Architectural Institute of Japan (AIJ-RLB-1993, hereafter). The BSLJ includes the
following items based on the Gust Loading Factor (Davenport, 1961):
- Basic wind speed depending on geographic location
- Four terrain roughness categories
- Simplified Gust Loading Factor based on building tip displacement
- Different wind load estimation methods for structural frames and cladding
- Different aerodynamic factors for structural frames and cladding
- Design velocity pressure at building roof height

The required performance and design wind load levels specified in BSLJ are listed in Table 1. Design of buildings
higher than 60m shall be approved by the Minister of Land, Infrastructure and Transport (MLIT). This approval
procedure is carried out on behalf of MLIT by a designated organization such as the Building Center of Japan (BCJ).
Each designated organization has formed special committees consisting of experts in various building engineering
fields from universities and the Japan Structural Consultants Association.

Essentially, structural designers are required to comply with the BSLJ. Design of buildings higher than 60m is
required to be based on dynamic response analyses in the time domain, as shown in Table 1. As AIJ-RLB is not a
law, they have only been used or consulted by structural designers requiring more sophisticated building designs or
compensating parts not covered by BSLJ. Thus, structural designers of tall buildings commonly use AIJ-RLB.


3
Table 1 Required performance and design wind load levels (BSLJ)
Wind Load Levels Medium-level Winds Very rare Strongest-level Winds
Recurrence Period 50 years 500 years
Required Performance
- No damage to main frames
- Cladding does not fall down
Buildings never collapse
Design Methods
- Allowable Stress Method
- Limit Strength Method
Limit Strength Method
Load Factor 1 1.6
Buildings lower than
60m (H 60m)
Wind Loads Along-wind shall be checked.
Design Method Dynamic response analyses in time domain
Wind Speed Factor 1 1.25
Buildings higher than
60m (H > 60m)
Wind Loads
Along-wind, crosswind, torsional, and vertical loads shall be
checked.


Municipal Bylaws on Environmental Assessment
Many matters need to be clarified in the design and the construction stages of a tall building, e.g. obstacle to radio
waves, solar shadow, and noise damage during construction. Additionally, in order to prevent harming the wind
environment around the tall building, causing discomfort and risk to pedestrians and other buildings, the Metropolis
of Tokyo has a municipal bylaw on the environmental assessment requiring wind tunnel or CFD predictions and
field measurements before and after construction of a building higher than 100m and with a total floor area larger
than 10
5
m
2
. It is recommended to evaluate the wind environment using either the method proposed by Murakami et
al. (1983) or that proposed by Nakamura et al. (1986). The former evaluates the change to the wind environment at a
particular site near a tall building on the basis of the annual maximum peak gust, while the latter evaluates it mostly
on the basis of the annual mean wind speed.

This bylaw has been encouraging building owners and designers to conduct additional wind tunnel tests for wind
loads on main frames and cladding, and occasionally on wind-induced responses as well as for environmental
assessment. This is one of the reasons why there are currently more than 100 wind tunnels for civil engineering use
in Japan.


AIJ RECOMMENDATIONS FOR LOADS ON BUILDINGS


Revision of AIJ-RLB
Since the 1993 version of AIJ-RLB, many relevant studies have been carried out and revision work has continued to
incorporate these study results. In the revised version to be published in March 2004, the criteria classifying
buildings and assessing necessary procedures for wind load estimation for a building of interest are given as the
same as those of the AIJ-RLB-1993. Simplified Procedure for small rigid buildings, Detailed Procedures for along-
wind load and roof wind load for structural frames for general buildings, crosswind load and torsional load for wind
sensitive buildings, and vortex-resonance and aerodynamic instabilities for particularly wind-sensitive buildings are
recommended on the basis of aspect ratio H/B, natural frequency f
1
, design wind speed U
H
and so on. The revised
wind load provisions are as shown in the following, which briefly explains the main frame of the new version and
important revision points.


Design Wind Speed
The design wind speed U
H
defined at the average roof height (reference height) H is given as:
rW H D H
k E K U U
0
= (1)
where, U
0
: basic wind speed, K
D
: wind directionality factor, E
H
: wind speed profile factor, and k
rW
: return period
conversion factor.

4
Basic Wind Speed
The basic wind speed U
0
is defined as the 100-year-recurrence wind speed under the standard meteorological
conditions, i.e. the 10min mean wind speed 10m above the ground for open flat terrain. It is given in a map for the
geographical location. Strong winds in Japan are mainly caused by typhoons, but the effects of synoptic winds and
others cannot be neglected, especially in the north part of Japan. The basic wind speed was estimated considering
the effects of both typhoon winds and synoptic winds. The typhoon winds were estimated by "Typhoon Simulation"
using the Monte-Carlo technique. The synoptic winds were estimated from meteorological wind data. The relation
between the probability of exceedence and the synoptic wind speed was evaluated by the Modified Jensen-Franck
method (Cook, 1983) and the Gumbel Distribution. The combined probability of both wind climates was derived for
the calculation of basic wind speed.


Wind Directionality Factor
Davenport (1969) first pointed out wind directional effects on wind load estimation, and Holmes (1981, 1990), Cook
(1983), Melbourne (1984) and Simiu and Heckert (1998) discussed some important problems related to wind
directionality. Melbournes results are reflected in AS1170.2(1989) and AS/NZS1170.2(2002), in which different
design wind speeds are explicitly given for different wind directions, although the application scope is limited to
non-tropical cyclone regions.

One of the major revisions is the introduction of wind directionality factor. In AIJ-RLB-1993, a constant wind speed
is given for the design wind speed regardless of wind direction, because it was difficult to reasonably estimate wind
directionality in tropical-cyclone-prone regions such as Japan. The main cause of the extremely high wind speed in
most parts of Japan is typhoons. Meteorological stations in Japan have approximately 75 years of reliable records at
most. However, the annual average number of typhoon landfalls on mainland Japan is only three, so the number of
typhoons included in the records of a particular site is very limited. When the records are divided into 8 or 16 sectors
of azimuth, each sector has very few typhoon data, and the sampling error becomes very large.

Tamura et al. (1995) proposed to estimate virtual typhoon wind data at any meteorological station using the
correlation between observed meteorological wind records and those by the Schlmer equation commonly used in
Typhoon Simulation (Mitsuta and Fujii, 1979, Matsui et al., 1997) based on atmospheric pressure data and typhoon
paths. The proposed method uses the correlations between the meteorological records at a meteorological station and
friction-free winds at the same station calculated by an objective analysis using meteorological data of typhoon
paths and accompanying atmospheric pressure distributions surrounding meteorological stations. The friction-free
wind speed and wind direction evaluated by a Typhoon Simulation are converted to wind speed and wind direction
at the height of the anemometer at the target meteorological station on the basis of conditional probability by the
Monte-Carlo technique. The various effects such as terrain roughness, topography, orography, and characteristics of
typhoon paths around the site are all reflected in the virtual typhoon data. The virtual typhoon data were generated at
153 meteorological stations in Japan for 5,000 years. Then, sufficient virtual typhoon wind speed data were
accumulated for 8 azimuth sectors, and an extreme value analysis was made for each sector (Tamura et al., 2003b).

The equivalent return period was estimated at around 150 - 200 years on average for various conditions (Tamura et
al., 2003b). This is because the stress in the main structural frame is generally dominant in only one or two wind
directions for many buildings. Consequently, the Wind Directionality Factor K
D
was estimated as the ratio of the
average directional wind speed U
D
to the 100-year-recurrence basic wind speed U
0
. The Wind Directionality Factors
are to be tabulated in AIJ-RLB-2004 for major cities in Japan.


Wind Speed Profile Factor
The wind speed profile factor E
H
, which accounts for the change in wind speed with height, surface roughness and
topographical features, is given by the following formula by setting the height Z = H.
g r
E E E = (2)
Here, E
r
is the exposure factor for flat terrains, and is given as:
5

|
|
.
|

\
|
<
|
|
.
|

\
|
=
b
G
b
G b
G
r
Z Z
Z
Z
Z Z Z
Z
Z
E

7 . 1
7 . 1
(3)
where, Z
G
: gradient height, Z
b
: interfacial layer height, and : power-law index. These parameters are listed in Table
2 for five flat terrain subcategories.


Table 2 Parameters for Exposure Factor
Flat Terrain
Subcategories
I II III IV V
b
Z (m)
5 5 10 20 30
G
Z (m)
250 350 450 550 650
0.1 0.15 0.2 0.27 0.35


E
g
in Eq.(2) is the topography factor for mean wind speed, and is given as:
( )
1 2 3 2 3
1 1 exp 1
g
s s
Z Z
E C C C C C
H H
| | | |

= + +
` ` | |
\ . \ . ) )
, ( 1
g
E ) (4)
where, H
S
is the height of an escarpment or a ridge, and C
1
, C
2
, and C
3
are parameters depending upon the angle of
inclination
S
of the upwind slope and the distance X
S
from the upper edge of the escarpment or the ridge. These
parameters are tabulated in AIJ-RLB-2004 for escarpments and ridges. The topographic effects on turbulence
intensity are also specified in the new AIJ-RLB.


Return Period Conversion Factor
The return period conversion factor k
rW
is given as:
9 . 3 9 . 2 ln ) 1 ( 62 . 0 + =
U U rW
r k (5)
0
500
U
U
U
= (6)
Here, U
500
is the 500-year-recurrence wind speed for the meteorological standard conditions, and U
0
is the basic
wind speed (100-year recurrence). The contour map is also given for U
500
in AIJ-RLB-2004.


Along-wind Loads for Ordinary Buildings
The Gust Loading Factor (GLF) used in AIJ-RLB-1993 is based on the tip displacement following the original GLF
proposed by Davenport (1961). Holmes (1994, 1996) proposed a more sophisticated GLF for lattice towers, which
enables estimation of any load effects such as bending moments and shear forces at any elevation. This is also
applicable to general prismatic buildings (Holmes, 2002). Zhou and Kareem (2001) proposed to use the GLF based
on the base bending moment (BBM) to provide more realistic equivalent static wind load (ESWL). Recently, an e-
database of aerodynamic wind loads was introduced by Zhou et al. (2003).

Reviewing the recent development of the GLF method, AIJ-RLB-2004 will adopt the BBM-based GLF rather than
the traditional GLF. However, it is not equal to that proposed by Zhou and Kareem (2001). The ESWL proportional
6
to the mean wind load distribution will still be adopted in AIJ-RLB-2004, for a political reason considering the
timing just after the BSLJ newly adopted the traditional GLF in the law in 2000. The ESWL at height Z is given as:
A G C q W
D D H D
= (7)
where, q
H
= U
H
2
/2: design velocity pressure at the mean roof height H (reference height), C
D
: aerodynamic factor at
height Z, G
D
: BBM-based gust loading factor, and A: projected area at height Z. The BBM-based GLF is given as:
'
2
1 1
g
D D D D
g
C
G g R
C
= + + (8)
where, g
D
: peak factor, C
g
and C
g
: fluctuating and mean coefficients for along-wind overturning moment,
D
:
correction factor depending on the mode shape of along-wind vibration, and R
D
: resonance factor. The formulae for
calculating these parameters are given as functions of dimensions and dynamic characteristics of the building and
various wind parameters. The correction factor
D
depending on the mode shape is given as:
D
D
M
M

=
2
ln 4 . 0 1
(9)
Here, the vibration mode shape is approximated by:

|
.
|

\
|
=
H
Z
. (10)
and M: total building mass excluding underground part, and M
D
: generalized mass of fundamental mode of along-
wind vibration.


Wind Loads on Roof Structures
The expression of wind loads on roof structures follows the essential concept of the GLF as shown in Eq.(11),
describing the maximum load effect by multiplying the GLF by the mean wind load effect. However, because there
are cases where the difference between internal and external pressure coefficients becomes zero, the GLF method is
not always valid. In such cases, the wind load calculation takes into account only the fluctuating component of
pressure. There are cases where the downward wind load becomes critical, considering the combination with fixed
load, snow load, and so on, even if its absolute value is small. Then, the ESWL for roof structures of buildings
without any predominant openings is given as:
R R R H R
A G C q W = (11)
pi pe R
C C C = (12)
where q
H
: design velocity pressure, C
pe
: external pressure coefficient, C
pi
: internal pressure coefficient, G
R
: gust
loading factor, and A
R
: subjected area for a roof beam.
The GLF is given three different situations as:
( )
c
c
R
r
r R r
G

+ +
=
1
3 . 0 1 3 . 12
1
2 2
Re Re
, for the case C
pi
= 0.4 and C
R
0 (13a)
( ) 3 . 0 1 3 . 12 25 . 0
2
Re
+ + =
Re
R r G C
R R
, for the case C
pi
= 0.4 and C
R
= 0 (13b)
7
( )
2 2
Re
3 . 0 1 3 . 12 1
c R
r R r G + + =
Re
, for the case C
pi
= 0 (13c)
Here, the three parameters r
c
, r
Re
, and R
Re
in Eqs.(13a), (13b) and (13c) are given as the functions of span/height
ratio, turbulence intensity, external pressure coefficient and the dynamic characteristics of roof beams. They also
vary with the roof beam direction.


Crosswind Loads and Torsional Loads
Crosswind and torsional wind loads should be examined in design of slender, flexible buildings to ensure that they
satisfy:
3
BD
H
(14)
where, H: building height, B and D: building width and depth.

The crosswind load at a local height Z is given by:
L L L L H L
R g
H
Z
A C q W
2
1 3 + = (15)
where C
L
= 0.0082(D/B)
3
0.071(D/B)
2
+0.22(D/B): fluctuating crosswind overturning moment coefficient, A:
projected area of the building, g
L
: peak factor,
L
: correction factor depending on the mode shape of the crosswind
vibration, and R
L
: resonance factor. The formulae for calculating these parameters are given as functions of
dimensions and dynamic characteristics of the building and various wind parameters. The correction factor
L
is
given as:
( )

ln 4 . 0 1
3
1

|
.
|

\
|
=

H
Z
M
M
L
L
(16)
where, M
L
: generalized mass of fundamental mode of crosswind vibration.

The torsional wind load at height Z is given by:
2
1.8 1
T H T T T T
Z
W q C AB g R
H
= + (17)
where, C
T
= {0.0066+0.015(D/B)
2
}
0.78
: fluctuating torsional moment coefficient, A: projected area of the building,
B: breadth of the building, g
T
: peak factor,
T
: correction factor depending on the mode shape of the crosswind
vibration, and R
T
: resonance factor. The formulae for calculating these parameters are given as functions of
dimensions and dynamic characteristics of the building and various wind parameters. The correction factor
T
is
given as:
( )
) ln 4 . 0 1 (
36
1
2 2

|
.
|

\
|
+
=

H
Z
I
D B M
T
T
(18)
where, M: total building mass excluding underground part, and I
T
: generalized inertial moment of the fundamental
mode of torsional vibration.


Along-wind, Crosswind and Torsional Acceleration Responses
The maximum acceleration responses for along-wind and crosswind directions, and the maximum torsional
acceleration angle are given as:
8

D
D g H aD H
D
M
R C BHC g q
a
) ln 4 . 0 1 (
max

= (19a)

L
L L aL H
L
M
R C BH g q
a
) ln 4 . 0 1 (
max

= (19b)

T
T T aT H
T
I
R C H B g q
a
) ln 4 . 0 1 ( 6 . 0
2
max

= (19c)
Here, a
Imax
and g
aI
are the maximum peak acceleration and the peak factor of the I (D, L and T ) component.


Vortex Resonance and Aerodynamic Instabilities
Vortex resonance and aerodynamic instabilities should be examined for particularly wind-sensitive buildings and
structures satisfying the following conditions. The condition for structures with a rectangular plan is given by:
4
BD
H
and
|
|
.
|

\
|

* *
83 . 0 or 83 . 0
Tcr
T
H
Lcr
L
H
U
BD f
U
U
BD f
U
(20a)
That for structures with a circular plan is given by:
7
m
H
D
and 4.2
H
L m
U
f D
(20b)
Here, f
L
and f
T
: fundamental natural frequencies of crosswind and torsional vibrations, respectively; and U
*
Lcr
and
U
*
Tcr
: non-dimensional onset wind speeds for crosswind and torsional instabilities. U
*
Lcr
and U
*
Tcr
are tabulated for
structures with a rectangular plan, and vary with the side ratio D/B, the mass-damping-parameter , and the terrain
categories. These values can be calculated and are to be tabulated in AIJ-RLB-2004. In design of these types of
buildings or structures, some appropriate investigations including wind tunnel tests are recommended. Incidentally,
AIJ-RLB-2004 specifies wind load formulae only for buildings and structures with a circular plan or with cylindrical
members.


Combinations of Wind Load Components
In AIJ-RLB-1993, the wind load combinations are not strongly recommended, because the same design wind speed
estimated from annual maximum wind speeds regardless of wind direction has to be applied for the most severe
wind direction. However, if wind directionality is adopted in AIJ-RLB-2004, it will become important to take the
wind load combinations into account in design.

Wind load combinations for low- medium- and high-rise building models have been investigated comprehensively
(Tamura et al., 2002a, 2002b, Tamura et al., 2003a, Hibi et al., 2003), and the combination method is clearly shown
in the revised version. AIJ-RLB-2004 gives two methods.
The first is applicable even without information on crosswind or torsional responses. For buildings not satisfying the
condition given by Eq.(14), the crosswind force W
LC
given by the following formula should be applied
simultaneously with the along-wind load W
D
given by Eq.(7).
D LC
W
B
D
W 35 . 0 = (21)
This formula is based on studies by Tamura et al. (2002a) and Hibi et al. (2003). Tamura et al. (2002a) reported that
the peak normal stresses in the columns could be almost 30% underestimated on average for both low-rise and
9
middle-rise building models if only the along-wind force component is taken into account in structural design. Hibi
et al. (2003) discusses the wind load combination effects based on comprehensive pressure data for various building
models in terms of the peak normal stress in columns, including dynamic resonant effects. He proposes a wind load
combination factor defined as W
LC
= W
D
, which is the crosswind load applied with the long-wind load.

For buildings satisfying the conditions of Eq.(14), AIJ-RLB-2004 applies a more detailed method, based on
analytical results using the correlation method by Asami (2000, and 2002) for the combination of load effects by
along-wind, crosswind, and torsional actions. Asami (2000) proposed wind load combination methods considering
the correlations of along-wind, crosswind and torsional responses based on the spectral modal technique, where 8
design points are used to approximate the response trajectory. The load combination factor is defined as a function
of the correlation coefficient between the crosswind response and the torsional response (Asami, 2002). As is well
known, Solari and Pagnini (1999) proposed a wind load combination method using 12 design points enveloping the
possible elliptic trajectory of the along-wind and crosswind responses. Table 3 gives the wind load combinations to
be considered.

W
D
, W
L
, and W
T
in Table 3 are given by Eqs.(7), (15), and (17).
LT
in Table 3 is the correlation coefficient between
the crosswind response and the torsional response, and is tabulated as a parameter depending upon the side ratio D/B,
the frequency ratio f

/f
L
, and the reduced frequency f
1
B/U
H
. Here, f

and f
L
are the fundamental natural frequency of
the torsional vibration and the crosswind vibration, respectively. f
1
is the smaller of f

and f
L
.


Table 3 Wind Load Combinations for High-rise Buildings
Combination Along-wind Load Crosswind Load Torsional Load
1
D
W
L
W 4 . 0
T
W 4 . 0
2
|
|
.
|

\
|
+
D
D
G
W
6 . 0
4 . 0
L
W ( )
T LT
W 1 2 2 +
3
|
|
.
|

\
|
+
D
D
G
W
6 . 0
4 . 0 ( )
L LT
W 1 2 2 +
T
W


Wind Loads for Components and Cladding
The wind loads for components and cladding are expressed by:
C C H C
A C q W

= (22)
where,
C
is a peak wind force coefficient given by:
*

pi pe C
C C C = (23)
Here,
pe
: peak external pressure coefficient, and C
*
pi
: coefficient accounting for the effect of the internal pressure
fluctuation.
pe
is given for various building surface locations, and is a function of a tributary area A
C
. For C
*
pi
, 0
and 0.5 should both be considered for buildings without any dominant openings.


One-Year-Recurrence Wind Speed
In order to assess the habitability of buildings subject to wind-induced vibrations, the Guidelines for the Evaluation
of Habitability to Building Vibration (AIJ-Guidelines, 1991) was published by AIJ, in which the one-year-
recurrence peak acceleration was applied for the evaluation. The evaluation method for wind-induced acceleration
and the one-year-recurrence directional wind speeds in the major cities of Japan are to be given in the Appendices.
Incidentally, the AIJ-Guidelines is going to be revised at almost the same time as the AIJ-RLB-2004 in the spring of
2004.

10
GUIDELINES FOR EVALUATION OF HABITABILITY TO BUILDING VIBRATION


Habitability to Building Vibrations
Habitability to wind-induced vibration of buildings is an essential criterion in design of tall buildings. The main
reason for this is the longer duration of winds and frequent storm events compared to earthquakes. As mentioned
before, the seismic force is predominant for almost all Japanese tall buildings, and they tend to be designed as light-
weight and flexible, thus being vulnerable to wind-induced vibrations. Thus, building habitability to wind-induced
vibrations is an important design issue.

There have been many studies on human comfort or habitability under building vibrations. Reed (1971), Jeary et al.
(1988), Fujimoto and Ohkuma (1988), Tamura et al. (1988), Nakata et al. (1993), and Denoon (2000) studied human
responses under full-scale conditions. Chen and Robertson (1973), Kanda et al. (1988, 1990), Shioya et al. (1992),
and Shioya and Kanda (1993) conducted experimental studies under laboratory conditions using an artificial room
mounted on an exciting facility.

In order to define the most appropriate parameter for habitability to randomly exciting wind-induced events, the
following problems should be solved: effects of visual cues; effects of acoustic cues; effects of postures; effects of
mental and physical conditions; wide dispersions in individual differences in human comfort; effects of social and
economic situations; almost no field data on the compliance rate; and so on. The problem is still far from being
solved, although there are several guidelines relevant to the evaluation of habitability under horizontal vibrations of
tall buildings: ISO6897 (1984), BS6841 (1987), ECCS (1987), AS1170.2-1989 Commentary (1990), AIJ-Guidelines
(1991), NBCC (1990), and ISO2631-1 (1997).

In Japan and some Asian countries, the AIJ-Guidelines (1991) have been widely used for design. It adopts peak
acceleration as an index, although ISO6897 refers to the root mean square (rms) value for evaluation of habitability
under vibration. The main reason for this is that the peak or maximum value is more easily understood by designers,
owners and occupants than the rms value. As both values can be converted by using a peak factor, there is no
substantial difference between them. However, ISO6897 uses 5-year-recurrence rms acceleration for the evaluation,
the AIJ-Guidelines (1991) uses 1-year-recurrence peak acceleration, valuing daily comfort rather than some special
event.


Probabilistic Perception Threshold and Design Criteria
Tamura (1998b) summarized the probabilistic human perception thresholds by connecting the results for the low-
frequency region from 0.125Hz to 0.33Hz by Shioya et al. (1992) and the high frequency region from 1.0Hz to
6.0Hz by Nakata et al. (1993). For the low frequency region tests, a test room with a 4.0m3.1m plan and a 2.6m
height was supported by laminated rubber bearings to cut off higher frequency motions, and acoustic insulation was
installed in the walls and ceiling of the test room. The test room was mounted on an electro-hydraulic servo-
controlled 6DOF shaking table. For the high frequency region tests, a test room with a 4.2m2.45m plan and a 2.4m
height was mounted on the 5th floor of an actual 7-story building. The building was excited by two exciters installed
on the 6th floor. The isolation from acoustic and high frequency disturbances was almost exactly the same as for the
low frequency region tests, and the conditions were most realistic because the room was settled in the actual
building. These results are consistently connected between 0.33Hz and 1.0Hz, as shown in Fig.1. This fact can be
evidence of the high quality of the data, because the two sets of results were obtained under different conditions and
for different subjects. In these experiments, noisy high-frequency vibrations, acoustic cues and visual cues were cut,
and the perception thresholds to horizontal vibrations were obtained at various frequencies for up to 122 subjects.
These experiments were made basically for uni-axial sinusoidal motions, but tests were also conducted for random
motions and for bi-lateral elliptic motions simulating wind-induced building motions (Tamura, 2003). The
randomness did not seem to affect the perception threshold, and the perception threshold for the random motions
was almost the same as that of the sinusoidal motions as shown in Kanda et al. (1990). Thus, the motion perception
in the wind-induced vibration of a building might be simply based on the acceleration amplitude and its predominant
natural frequency.

Jeary (1988) reported field test results on probabilistic perception thresholds using an actual 10-story building.
These results are also shown in Fig.1. Although Jearys data were obtained in the field in an actual building and a
11
slightly wider dispersion is recognized, they almost coincide with the laboratory data. Denoon (2000) also conducted
field measurements on the perception threshold at two air-traffic-control towers and one port-operation tower, and
the average perception thresholds are also shown in Fig.1.




















Figure 1 Probabilistic human perception thresholds for horizontal building vibrations (Tamura, 2003)





















Figure 2 Guidelines and criterion for habitability to horizontal building vibrations (Tamura, 2003)


Perception Thresholds and AIJ-Guidelines (1991)
AIJ-Guidelines (1991) give four lines from H-1 to H-4 as shown in Fig.2, and the standard for office buildings (H-3)
is set at almost 50% perception threshold level, and that for residential buildings (H-2) at 10% - 20% perception
threshold level, by applying the annual peak acceleration and the first mode natural frequency. Peak acceleration is
only used for ease of understanding by designers and building owners as mentioned before, and the guidelines are
adjusted and specified considering the duration effects.

0.1 0.2 0.5 1 2 5 10
Frequency (Hz)
50
10
5
2
1
20
0.5
0.2
0.1
90%
99%
50%
10%
2%
1%
A
c
c
e
l
e
r
a
t
i
o
n

A
m
p
l
i
t
u
d
e

(
c
m
/
s
2
)

2%
10%
90%
(Jeary et al.)
(Denoon: Ave.)
H-1
NBCC (10-year-rec.)
ISO6897 General Buildings
(5-year- rec., r.m.s.3.5)
ISO6897 Min. (r.m.s.2)
0.1 0.2 0.5 1 2 5 10
Frequency (Hz)
50
10
5
2
1
20
0.5
0.2
0.1
P
e
a
k

A
c
c
e
l
e
r
a
t
i
o
n

(
c
m
/
s
2
)

AS1170-2, Commentary
(10-year-rec.)
H-4
ISO6897 Ave. (r.m.s.2)
AIJ Guidelines (1-year-rec.)
(5-year-rec.)
(1-year-rec.)
H-3:Standard for Office Buildings
H-2:Standard for Residential Buildings
12
Figure 2 compares major codes or standards. Incidentally, the ISO6897 minimum and average perception thresholds
in Fig.2 coincide with the probabilistic perception thresholds shown in Fig.1. As ISO6897 recommends 5-year-
recurrence rms accelerations for the acceptable limit, the ISO6897 acceptable limit for general buildings is
multiplied by 3.5 for comparison with the guidelines based on the peak acceleration. NBCC specifies a range from
10cm/s
2
to 30cm/s
2
without frequency dependency for rather low probability events, i.e. 10-year recurrence.
Melbourne and Cheung (1988) and Melbourne and Palmer (1992) proposed a maximum peak acceleration criteria
for any return period, and this was reflected in the AS1170-2 Commentary as shown in Fig.2. The AIJ-Guidelines
show more severe criteria, and the change in slope at 0.2 Hz does not seem to have supporting data.


Revised Guidelines (AIJ-Guidelines-2004)
After publication of the AIJ-Guidelines (1991), a series of experiments on the human perception threshold and
researches on habitability to building vibrations were conducted, and many results were published, e.g. Shioya et al.
(1992), Nakata et al. (1993), Denoon (2000) and Inoue et al. (2003).

The draft AIJ-Guidelines-2004 are based on those data. They consist of three parts: human- and machine-induced
vertical vibrations; traffic-induced vertical/horizontal vibrations; and wind-induced horizontal vibrations. The
following are in the revised version of the Evaluation of Habitability to Wind-induced Horizontal Vibrations. Major
revisions have been made in the expressions of the guidelines and the expansion of the frequency range up to 5Hz.
The latter were made because of problems occurring in urban areas in Japan in wind-induced vibrations of relatively
low-rise wooden or steel residential buildings.

Instead of giving some recommended line, the AIJ-Guidelines-2004 only gives five curves: H-10, H-30, H-50, H-70
and H-90 shown in Fig.3. The number of each curve indicates the perception probability as a percentage, i.e. 10% of
people can perceive the vibration specified by the H-10 curve.





















Figure 3 Curves given in the draft AIJ-Guidelines-2004


The essential concept of the draft AIJ-Guidelines-2004 is that the criteria for building habitability to vibration
should be decided by a building owner. Although AIJ-Guidelines (1991) specified the H-2 curve and H-3 curve as
the standard level for residential buildings and office buildings, respectively, no specified criteria is recommended in
the draft AIJ-Guidelines-2004. However, it may be difficult to judge the most appropriate vibration level and to
select one of the curves as the design target for a given building. For designers' reference, calculated along-wind and
crosswind responses are compared with the curves in the commentary of the draft AIJ-Guidelines-2004. Wind-
induced responses of 286 buildings were calculated from formulae for the maximum along-wind and crosswind
H-10
H-30
H-50
H-70
H-90
0.1 0.2 0.5 1 2 5
Frequency (Hz)
20
5
2
1
0.5
10
A
n
n
u
a
l

M
a
x
i
m
u
m

A
c
c
e
l
e
r
a
t
i
o
n

(
c
m
/
s
2
)

13
acceleration responses shown in Eqs.(19a) and (19b). In the response calculations, the measured full-scale values of
the natural frequency and the damping ratio in the database collected by the AIJ Sub-Committee on Damping in
Buildings (Tamura et al., 2000) were used. Several figures are plotted to show different building usages, e.g. offices,
hotels and residences, and different building types, e.g. steel buildings, steel encased reinforced concrete buildings,
and reinforced concrete buildings. Owners and structural designers can select the design target or criteria that the
building should satisfy.


DAMPING IN BUILDINGS


Importance of Design Value of Damping
Structural damping is the most important, but most uncertain, parameter affecting dynamic responses of buildings.
This uncertainty significantly reduces the reliability of structural design for dynamic effects. Accurate determination
of structural damping is very important, not only for evaluating structural responses, but also for designing active
and passive auxiliary damping devices to be installed in buildings and structures. However, there is no theoretical
method for estimating damping in buildings. Thus, it has been estimated on the basis of actual measurements of
widely dispersed damping ratios.

High-quality full-scale damping data for buildings and structures are required to establish an accurate damping
predictor for assessing damping ratios for design. Recently, vibration-measuring techniques as well as data-
processing techniques have been significantly improved with the development of microprocessor-based devices and
software. Damping evaluation techniques have also improved (Jeary, 1986, 1992, Tamura and Suganuma, 1996,
Tamura et al., 2002c). Furthermore, a lot of Japanese buildings are equipped with vibration monitoring systems for
active or passive damping or for verifying design assumptions, and their dynamic characteristics during various
types of excitations have become our common property.

As Tamura et al. (2000) reported, data of dynamic properties of buildings, including damping ratios, were collected
by the AIJ Sub-Committee on Damping Database (Tamura et al., 1994, Tamura, 1997, 1998a). The committee
members, representing research institutes, structural design offices, and construction companies, provided their own
recent reliable data that satisfy required conditions in a specified format. Reliable high-quality full-scale damping
data on buildings for various conditions were accumulated from about 40 organizations. Additional damping data
have been collected from many journals and proceedings published mainly after 1970. Then, the Committee
conducted questionnaire studies by mail to literature authors, designers and building owners to confirm the
measured values, and to obtain additional important information, especially on measurement and evaluation
techniques and vibration amplitudes as well as detailed building information. Data were excluded if their amplitudes
were not clear or could not be deduced from the measuring methods. Consequently, many original non-published
data were collected and additional information was added to the published data. Furthermore, the database contains
only data approved by the owners for worldwide distribution. The Japanese Damping Database (JDD) includes
building information and dynamic properties for 137 steel buildings (S-buildings), 43 steel encased reinforced
concrete buildings (SRC-buildings), 25 reinforced concrete buildings (RC-buildings), and 80 tower-like non-
building structures, making a total of 284 buildings and structures.

Tamura et al. (2000) proposed design damping predictors for S-buildings and RC-buildings, and recommended
design values were given for habitability and safety levels.


Amplitude Dependency of Dynamic Characteristics
Figure 4 shows evidence of the amplitude dependency of dynamic characteristics of buildings, and of the
contribution of secondary members. It clearly shows that the damping ratio increases with amplitude, while the
natural frequency decreases (Tamura and Suganuma, 1996). These results are attributed to the same physical reason,
i.e. the contribution of secondary members and their stick-slip phenomena. In the small amplitude regime,
connections, joints or contact surfaces of various building components are stuck to each other. However, if the
amplitude increases, they begin to slip, thus providing friction damping but losing their stiffness. This is the main
reason for the increasing damping and decreasing natural frequency with the increase in vibration amplitude shown
14
in Fig.4. However, if the amplitude reaches a certain level at which all connections do slip, the number of slipping
connections does not increase any more with the increase in amplitude, and the damping ratio shows its peak.














Figure 4 Variation of dynamic characteristics of a 99m-high steel-framed building (Tamura and Suganuma, 1996)


Natural Periods (Natural Frequencies)
Figures 5(a) and 5(b) show the relations between the fundamental natural periods T
1
and building heights H for
SRC/RC-buildings and S-buildings, respectively, in the low-amplitude regime based on JDD (Sasaki et al., 1997).
These results show that the natural period increases with building height, and the regression line approximates the
natural period T
1
(s) as a simple linear function of building height H(m) as follows:
H T 015 . 0
1
= or
H H
f
67
015 . 0
1
1
= = : RC and SRC-buildings (Low-amplitude) (24a)
H T 020 . 0
1
= or
H H
f
50
020 . 0
1
1
= = : S-buildings (Low-amplitude) (24b)














(a) RC and SRC-buildings (b) S-buildings
Figure 5 Fundamental natural periods T
1
(s) and building heights H

(m) in Japanese damping database
(Sasaki et al., 1997)


The measured natural period is almost 20% lower than their design values for the low-amplitude regime. This
difference is also attributed to the contributions of secondary members such as furring strips, studs, partitions,
claddings and finishing, whose contributions to stiffness are difficult to quantify in the design stage (Tamura et al.,
2000). As these contributions become minimized in the high amplitude regime, the natural frequency is likely to
0
1
2
3
0 50 100 150 200
RC
SRC
Building Height (m) H
N
a
t
u
r
a
l

P
e
r
i
o
d







(
s
)

T
1
H, r 0.015 0.94
T
1

Building Height (m)

0
1
2
3
4
5
6
7
0 50 100 150 200 250 300
H
N
a
t
u
r
a
l

P
e
r
i
o
d







(
s
)

T
1
H, r 0.020 0.94
T
1

0 1 2 3 4 5 6 7
Acceleration Amplitude (cm/s
2
)
D
a
m
p
i
n
g

R
a
t
i
o

1

0.03
0.02
0.01




N
a
t
u
r
a
l

F
r
e
q
u
e
n
c
y

f
1

(
H
z
)

0.66
0.65
0.64
15
become closer to the design value. Based on the above consideration, Tamura et al. (2000) recommended the
following formulae for the fundamental natural frequency in the high-amplitude regime.
H H
f
56
018 . 0
1
1
= = : RC and SRC-buildings (High-amplitude) (25a)
H H
f
42
024 . 0
1
1
= = : S-buildings (High-amplitude) (25b)
Ellis (1980) proposed a simple approximation for the fundamental natural frequency f
1
= 46/H for all structural types
of buildings, and this is widely used in some codes or standards. Japanese results for both RC/SRC and S buildings
give a higher natural frequency than Ellis' proposal in the low amplitude regime, as shown in Eqs.(24a) and (24b),
and those for RC/SRC-buildings also give higher values even in the high amplitude regime, as shown in Eq.(25a).
Incidentally, higher mode natural periods are proportional to the fundamental natural period (Sasaki et al., 1997).


Damping Predictor Based on JDD
Based on JDD and considering the amplitude dependency shown in Fig.4, formulae for predicting the fundamental
damping ratios for RC-buildings and S-buildings were given as:
0018 . 0 470
93 . 0
0018 . 0 470 014 . 0
1 1
+ = + =
H
x
H H
x
f
H H
(RC Buildings) (26a)
0029 . 0 400
65 . 0
0029 . 0 400 013 . 0
1 1
+ + = + + =
H
x
H H
x
f
H H
(Steel Buildings) (26b)
These formulae consist of a frequency dependent term, an amplitude dependent term, and a constant.

It is important that the frequency dependent terms, 0.014

f
1
and 0.013

f
1
in Eqs.(26a) and (26b) cannot be attributed
to the pure frequency effect. It is likely to be due to the soil-structure-interaction (SSI). As shown in Eqs.(24a) and
(24b) , for example, the natural frequency of buildings is expressed with satisfactory accuracy by a simple function
of the height H. A building with a higher natural frequency is basically a lower-height building, and the SSI effects
become more significant, resulting in a higher damping ratio. This is reflected in the frequency dependent term, and
it may be more reasonably expressed by a function of the building height H rather than the natural frequency f
1,
as
shown in the above formulae.

According to the comparison of the full-scale damping ratios in JDD and their predicted values only for data whose
amplitude was exactly known, Equation (26a) coincides closely with all the known amplitude full-scale data of RC-
buildings in the database, and the correlation coefficient was estimated at 88% (Tamura et ail., 2000). However, it
should be noted that Eqs.(26a) and (26b) are empirical results. Therefore, the formulae are essentially limited in
their application. The non-dimensional amplitudes of the original data are within the range x
H
/H 210
5
.


Recommended Design Damping Ratio
Based on the above results, recommendations for design damping ratios were made for "Habitability" level and
"Safety" level (Tamura et al., 2000). The vibration amplitude of the "Habitability" level was set at the H-3 curve in
Fig. 2, and that of the "Safety" level was set at the upper limit x
H
/H

= 210
5
of the scope of application of the
formulae. The natural frequency was estimated for a given building height by Eqs.(24a), (24b), (25a) and (25b). The
damping values calculated by the proposed formulae were made to round numbers and tabulated in Tables 4 and 5
as the "Standard" values. However, considering the dispersion of the full-scale data around the regression line, it is
recommended to use 80% the Standard values for structural design, which corresponds closely to (mean ).
"Recommended" values for "Habitability" are also shown in Tables 4 and 5. These results were adopted in AIJ-
Damping (2000) published by the AIJ for practitioners.


16
Table 4 Design damping ratios for RC buildings (Tamura et al., 2000)
Habitability Safety (Elastic Range)
Damping Ratio
1
Damping Ratio
1

Building
Height
H (m)
Natural
Frequency
f
1
= 1 / 0.015H
Recommended Standard
Natural
Frequency
f
1
= 1 / 0.018H
Recommended Standard
30 2.2 Hz 2.5 3 1.9 Hz 3 3.5
40 1.7 1.5 2 1.4 2 2.5
50 1.3 1.2 1.5 1.1 2 2.5
60 1.1 1.2 1.5 0.93 1.5 2
70 0.95 0.8 1 0.79 1.5 2
80 0.83 0.8 1 0.69 1.2 1.5
100 0.67 0.8 1 0.56 1.2 1.5
a
Less than the "Recommended" values should be used for buildings that have very few interior walls.
b
The "Standard" values can be used for buildings such as hotels and apartment houses, which have many
interior walls.
c
The value for H = 30m should be used for buildings lower than 30m.
d
The tabulated values represent the damping ratios including the soil-structure interaction effects. Thus,
they should not be used for structural damping of the upper structure only.


Table 5 Design damping ratios for Steel buildings (Tamura et al., 2000)
Habitability Safety (Elastic Range)
Damping Ratio
1
Damping Ratio
1

Building
Height
H (m)
Natural
Frequency
f
1
= 1 / 0.02H
Recommended. Standard
Natural
Frequency
f
1
= 1 / 0.024H
Recommended Standard
30 1.7 Hz 1.8 2.5 1.4 Hz 2 3
40 1.3 1.5 2 1.0 1.8 2.5
50 1.0 1 1.5 0.83 1.5 2
60 0.83 1 1.5 0.69 1.5 2
70 0.71 0.7 1 0.60 1.5 2
80 0.63 0.7 1 0.52 1 1.5
100 0.50 0.7 1 0.42 1 1.5
200 0.25 0.7 1 0.21 1 1.5
a
The same notes as Table 4


DAMPING DEVICES


Necessity for Damping Devices to Suppress Wind-induced Vibrations of Tall Buildings in Japan
Vibration control becomes important for slender, flexible tall buildings. Especially in Japan, installation of damping
devices is now very common in tall buildings (Tamura, 2003). As discussed, Japanese tall buildings have to resist
extremely strong earthquakes and extremely strong winds. However, seismic force is generally predominant in the
design of tall buildings in Japan if the building is lower than around 200m. Therefore, almost all tall buildings in
Japan are designed to be light and flexible, making them vulnerable to wind. Thus, habitability under wind-induced
vibrations for more frequent daily events is an important issue. Since the Kobe Earthquake in 1995, there has been
increasing demand by building owners for safer buildings under extremely strong earthquakes. This is why so many
tall buildings have been equipped with damping devices and base-isolation systems in Japan.

A slight change in sectional shapes can cause a significant change in wind-induced responses of tall building models
(Shiraishi et al., 1988, Kwok, 1988, Shimada et al., 1988, Isyumov et al., 1989, Shimada and Hibi, 1995, Tamura,
1998b). Corner chamfering, corner cutting, openings, complex changes in sectional shape along the vertical axis,
and so on can provide significant reduction in wind forces, especially in crosswind forces. It is essential to conduct
studies on the aerodynamic shape of the target building using wind tunnel tests or CFD at an early stage of design.

17
Incidentally, the traditional way of increasing stiffness can reduce displacements, but it is not appropriate for
improving habitability under vibrations, since this accompanies an increase in natural frequency, and does not
provide a sufficient reduction in accelerations (Tamura, 1998b).


Recent Trend
Table 6 compares recently used damping devices and their performance in tall buildings in Japan (Shibuya, 2001).
Almost all are efficient under wind-induced motions, but their efficiency under seismic actions is limited, especially
for extremely strong earthquakes. As a result, the urgency to install AMDs or HMDs in the early 90s seems to have
waned. Considering their cost and maintenance as well as their efficiency, SDs may currently be the most commonly
used devices in Japan due to their promising performance under strong earthquakes. VFDs and VEDs are also
widely used for both earthquakes and winds (Tamura, 2003).


Table 6 Performance of damping devices (Shibuya, 2001)
Damping efficiency
Dampers
Extremely
Strong
Earthquake
Medium Class
Earthquake
Wind
Cost Maintenance
Steel Damper (SD) VG G VG VG
Passive (OD) G G VG G
Oil Damper Semi-
active
(AOD) G VG VG
Viscous Fluid Damper (VFD) G VG VG G G
Visco-elastic Damper (VED) G VG VG G G
Passive (TMD) G G
Mass Damper
Active (AMD) VG
VG: Very good, G: Good


As introduced in Tamura (2003), another new trend in Japan is the use of more than two different types of damping
devices for different design targets. Four different types of damping devices were installed in the Nippon TV Office
(193m, completed in 2003): two TMDs for habitability under wind-induced vibrations; 32 ODs for wind-induced
vibrations and weak/medium-class earthquakes; 64 unbonded bracing dampers (SDs) and 312 link beam dampers
(SDs) for extremely strong earthquakes. AMDs for habitability to wind-induced vibrations and AODs for wind-
induced vibrations and extremely strong earthquakes were installed in the Kajima Tower (172m, completed in 2003).
AMDs and honeycomb dampers (SDs) were installed in the Kyodo News Service Building (172m, completed in
2003) for habitability under wind-induced vibrations and extremely strong earthquakes, respectively. In Roppongi
Hills (238m, completed in 2003), 192 AODs were installed for winds and extremely strong earthquakes and 356
unbonded bracing dampers (SDs) for extremely strong earthquakes. Combinations of different types of devices can
achieve more efficient and reliable damper design by specializing the target performance of each damping device,
thus increasing the reliability of structural design and robustness of auxiliary damping devices.


CONCLUDING REMARKS

This paper has presented some topics relevant to wind resistant design of tall buildings in Japan: special features of
design of the tall building in Japan; Building Standard Law of Japan; AIJ Recommendations for Loads on Buildings;
AIJ Guidelines for Evaluation of Habitability to Building Vibrations; damping in buildings; and damping devices. It
has also implicitly demonstrated that there are many unsolved problems that have a pressing need for solution.
Recent trends of various issues in wind resistant design of tall buildings suggest the future status of tall building
design. Health monitoring systems will be widely used, and an electronic aerodynamic database will be used for
practical design. By combining these databases with the structural design database composed of full FEM models,
the picture of structural design systems and disaster prevention systems will be changed.


18
ACKNOWLEDGEMENTS

The author gratefully acknowledges the members of the AIJ Sub-Committee on Wind Loads (chaired by Y. Tamura),
and those of four Working Groups for their voluntary base efforts on the revision work of wind load provisions of
AIJ-RLB. He also acknowledges the members of the AIJ Working Group on Evaluation of Habitability to Building
Vibration (chaired by T. Ishikawa) and Sub-WG on Wind-induced Vibration (chaired by O. Nakamura) for their
devoted effort in making the draft AIJ-Guidelines-2004.


REFERENCES
AIJ-Damping (2000), Damping in Buildings, Architectural Institute of Japan, Maruzen, pp.287
AIJ-Guidelines (1991), Guidelines for the evaluation of habitability to building vibration, Architectural Institute of
Japan, Maruzen, pp.82.
AIJ-RLB (1993), Recommendations on Loads for Buildings, Architectural Institute of Japan (English version: 1996).
AS 1170.2 (1989), Australian Standard, SAA Loading Code, Part 2: Wind load.
AS1170.2-1989 Commentary (1989), A Commentary on the Australian Standard for Wind Load (by Holmes, J.D.,
Melbourne, W.H. and Walker, G.R.), pp.72.
AS/NZS 1170.2 (2002), Australian / New Zealand Standard, Structural design actions, Part 2: Wind actions.
Asami, Y. (2000), "Combination method for wind loads on high-rise buildings", Proceedings of the 16th National
Symposium on Wind Engineering, Tokyo, Japan, pp.531-534 (in Japanese).
Asami, Y. (2002), "Correlation method for wind load combinations", Document presented at Subcommittee on Wind
Loading, Architectural Institute of Japan (in Japanese).
BS6841 (1987), Measurement and evaluation of human exposure to whole-body mechanical vibration and repeated
shock, British Standards Institute.
Chen, P.W. and Robertson, K.E. (1973), "Human perception thresholds to horizontal motion", Journal of Structural
Division, ASCE, Vol.98, No.ST8, pp.1681-1695.
Cook, N.J. (1983), "Note on directional and seasonal assessment of extreme winds for design", Journal of Wind
Engineering and Industrial Aerodynamics, 12, pp.365-372.
Davenport, A. G. (1961), "The application of statistical concepts to the wind loading of structures", Proceedings of
Institution of Civil Engineers, London, 19, pp.449-472.
Davenport, A.G., (1969), "Structural safety and reliability under wind action", Proceedings of International
Conference on Structural Safety and Reliability, pp.131-145.
Denoon, R.O. (2000), "Designing for serviceability accelerations in buildings", PhD Thesis, The University of
Queensland, Australia.
ECCS (1987), Recommendations for the calculation of wind effects on constructions, TC12, No.52, European
Convention for Constructional Steel Work.
Ellis, B.R. (1980), "An assessment of the accuracy of predicting the fundamental natural frequencies of buildings
and implications concerning the dynamic analysis of structures", Proceedings of the Institution of Civil Engineers,
Part 2, Vol.69, pp.763-776.
Fujimoto, M., Ohkuma, T. and Tamura, T. (1988), "Human response to wind-induced vibration of tall buildings",
Proceedings Vol.I, Symposium/Workshop on Serviceability of Buildings (Movements, Deformations, Vibrations),
Ottawa, pp.270-290.
Holmes J.D. (1981), "Reduction factors for wind direction for use in codes and standards", Colloque, Construire
Avec Le Vent, Nantes, France, pp.VI.2.1-VI.2.15.
Holmes, J.D. (1990), "Directional effects on extreme wind loads", Civil Engineering Transactions, Institution of
Engineers, Australia, CE32.1, pp.45-50
Holmes, J.D. (1994), "Along-wind response of lattice towers: part I - derivation of expressions for gust response
factors". Engineering Structures, 16, pp.287-292
Holmes, J.D. (1996), "Along-wind response of lattice towers: part II - aerodynamic damping and deflections".
Engineering Structures, 18, pp.483-488
Holmes, J. D. (2002), "Gust loading factor to dynamic response factor (1967- 2002)", Symposium Preprints,
Engineering Symposium to Honour Alan G. Davenport for his 40 Years of Contributions, The University of Western
Ontario, (AGD2002), pp. A1-1-A1-8.
Hibi, K. Tamura, Y. and Kikuchi, H. (2003), "Peak normal stresses and wind load combinations of middle-rise
buildings", Summaries of Technical Papers of Annual Meeting, Architectural Institute of Japan, Structures I, pp.113-
114 (in Japanese).
19
Inoue, K., Ishikawa, T. and Noda, C. (2003), "Evaluation curve and material to explain habitability of horizontal
vibration -- Part 1 Evaluation curve based on perception by physical and visual feeling --", Summaries of Technical
Papers of Annual Meeting, Architectural Institute of Japan, pp. 299-300 (in Japanese).
ISO2631-1 (1997), Mechanical vibration and shock - Evaluation of human exposure to whole-body vibration- Part
1: General requirement, International Organization for Standardization.
ISO6897 (1984), Guidelines for the evaluation of the response of occupants of fixed structures, especially buildings
and off-shore structures, to low-frequency horizontal motion (0.063 to 1Hz), International Organization for
Standardization.
Isyumov, N., Dutton, R. and Davenport, A.G. (1989), "Aerodynamic methods for mitigating wind-induced building
motions", Structural Design, Analysis and Testing, Proceedings of Structures Congress 89, ASCE, pp.462-470.
Jeary, A.P. (1986), "Damping in tall buildings. A mechanism and a predictor", Journal of Earthquake Engineering
and Structural Dynamics, Vol.14, pp.733-750.
Jeary, A.P. (1992), "Establishing non-linear damping characteristics of structures from non-stationary response time-
histories", The Structural Engineer, Vol.70, No.4, pp.61-66.
Jeary, A.P., Morris, R.G. and Tomlinson, R.W. (1988), "Perception of vibration - Tests in a tall building", Journal of
Wind Engineering and Industrial Aerodynamics, Vol.28, pp.361-370.
Kanda, J., Tamura, Y. and Fujii, K. (1988), "Probabilistic criteria for human perception of low-frequency horizontal
motions", Proceedings, Symposium/Workshop on Serviceability of Buildings (Movements, Deformations,
Vibrations), Vol.I , Ottawa, pp.260-269.
Kanda, J., Tamura, Y. and Fujii, K. (1990), "Probabilistic perception limits of low-frequency horizontal motions",
Proceedings of the Conference with International Participation, Serviceability of Steel and Composite Structures,
Pardubice, Czechoslovakia, pp.61-72.
Kwok, K.C.S. (1988), "Effect of building shape on wind-induced response of tall buildings", Journal of Wind
Engineering and Industrial Aerodynamics, Vol.28, pp.381-390.
Matsui M., Meng, Y. and Hibi, K. (1997), "Extreme typhoon wind speeds considering the random variation in a full-
scale observation", Proceedings of ICOSSAR'97, pp.1343-1349.
Melbourne, W.H. (1984), "Designing for directionality", 1st Workshop on Wind Engineering and Industrial
Aerodynamics, Highett, Victoria, Australia, pp.11.
Melbourne, W.H. and Cheung, J.C.K. (1988), "Designing for serviceable accelerations in tall buildings",
Proceedings of the Fourth International Conference, Vol.1, Hong Kong & Shanghai, pp.148-155.
Melbourne, W.H. and Palmer, T.R. (1992), "Accelerations and comfort criteria for buildings undergoing complex
motions", Journal of Wind Engineering and Industrial Aerodynamics, Nos.41-44, pp.105-116.
Mitsuta, Y., Fujii T. and Kawahira K. (1979), "Analysis of typhoon pressure patterns over Japanese Islands",
National Disaster Science, Vol.1, N0.1, pp.3-19.
Murakami, S., Iwasa, Y. and Morikawa, Y. (1983), "Investigation of statistical characteristics of wind at ground
level and criteria for assessing wind-induced discomfort, Part III Criteria for assessing wind-induced discomfort",
Transactions of the Architectural Institute of Japan, No.325, pp.74-84.
Nakamura, O., Yoshida, M., Yokotani, K. and Katagiri, J. (1986), "Characteristics of wind at ground level in
Tokyo", Proceedings of the 9th national Symposium on Wind Engineering, Tokyo, pp.73-78.
Nakata, S., Tamura, Y. and Otsuki, T. (1993), "Study on habitability to horizontal vibration of low rise buildings",
International Colloquium on Structural Serviceability of Buildings, Gteborg, Sweden, IABSE Reports, Vol.69,
pp.39-44.
NBCC (1990), National Building Code of Canada, Part 4 Structural Design, Chapter 4 Commentaries on Part 4 of
the National Building Code.
Reed, J.W. (1971), "Wind induced motion and human comfort", Research Report 71-42, Massachusetts Institute of
Technology.
Sasaki, A., Satake, N., Arakawa, T., Suda, K., Ono, J. and Morita, K. (1997), "Damping properties of buildings
using full-scale data", Proceedings of the Symposium on Damping Technology of the 21st Century, Dynamics and
Design Conference 97, The Japan Society of Mechanical Engineers Centennial Grand Congress, pp.25-28 (in
Japanese).
Shiraishi, N., Matsumoto, M., Shirato, H. and Ishizaki, H. (1988), "On aerodynamic stability effects for bluff
rectangular cylinders by their corner-cut", Journal of Wind Engineering and Industrial Aerodynamics, Vol.28,
pp.371-380.
Simiu, E. and Heckert, N.A. (1998), "Wind direction and hurricane-induced ultimate wind loads", Journal of Wind
Engineering and Industrial Aerodynamics, 74-76, pp.1037-1046.
Shibuya, K. (2001), "Evolving dampers", Nikkei Architecture, No.708, pp.68-75 (in Japanese)
20
Shimada, K. and Hibi, K. (1995), "Estimation of wind loads for a super-tall building (SSH)", The Structural Design
of tall Buildings, Wiley, New York, 4, pp.47-60.
Shimada, K., Tamura, Y. and Fujii, K. (1988), "Effects of geometrical shape on response of tall building", Journal
of Wind Engineering, JAWE, No.41, pp.77-78 (in Japanese).
Shioya, K. and Kanda, J. (1993), "Human perception thresholds of horizontal motion", International Colloquium on
Structural Serviceability of Buildings, Gteborg, Sweden, IABSE Reports, Vol.65, pp.45-52.
Shioya, K., Kanda, J., Tamura, Y. and Fujii, K. (1992), "Human perception thresholds of two dimensional horizontal
motion", Structures Congress 92, Compact Papers, ASCE, San Antonio, USA, pp.480-483.
Solari, G. and Pagnini, L.C. (1999), "Gust buffeting and aeroelastic behavior of poles and monotubular towers".
Journal of Fluid and Structures, Vol.13, pp.877-905.
Tamura, Y. (1997), "Japanese damping database", Proceedings of Structures Congress XV, ASCE, 2, pp.1064-1067.
Tamura, Y. (1998a), "Damping in buildings in Japan", Proceedings on CD-Rom, Structural Engineers World
Congress, San Francisco, T193-2, 8.
Tamura, Y. (1998b), "Application of damping devices to suppress wind-induced responses of buildings", Journal of
Wind Engineering and Industrial Aerodynamics, Vols.74-76, pp.49-72.
Tamura, Y. (2003), "Design issues for tall buildings from accelerations to damping -- Tribute to Hatsuo Ishizaki and
Vinod Modi --", Conference Preprints, The 11th International Conference on Wind Engineering, Vol.1, Lubbock,
Texas, USA, pp.81-114.
Tamura, Y. and Suganuma, S. (1996), "Evaluation of amplitude-dependent damping and natural frequency of
buildings during strong winds", Journal of Wind Engineering and Industrial Aerodynamics, Vol.59, pp115-130.
Tamura, Y., Fujii, K. Sato, T., Wakahara, T. and Kosugi, M. (1988), "Wind-induced vibration of tall towers and
practical applications of tuned sloshing damper", Proceedings Vol.I, Symposium/Workshop on Serviceability of
Buildings (Movements, Deformations, Vibrations), Ottawa, pp.228-241.
Tamura, Y., Yamada, M., Yokota, H. (1994), "Estimation of structural damping of buildings", Proceedings of
Structures Congress XII, ASCE, 2, pp.1012-1017.
Tamura, Y., Goto S. and Watanabe, Y. (1995), "Design wind speed estimation by simulation of typhoon winds -
Evaluation of wind direction factor based on wind-direction correlation method -", Proceedings of the 3rd Japan
Conference on Structural Safety and Reliability, pp.483-486.
Tamura, Y., Suda, K. and Sasaki, A. (2000), "Damping in buildings for wind resistant design", International
Symposium on Wind and Structures for the 21st Century, Cheju, Korea, pp.115-130.
Tamura, Y., Kikuchi, H. and Hibi, K. (2002a), "Quasi-static wind load combinations for low- and middle-rise
buildings", AGD2002 Symposium Preprints, Engineering Symposium to Honor Alan G. Davenport for His 40 Years
of Contributions, The University of Western Ontario, Canada, pp.C3-1 - C3-12.
Tamura, Y., Kikuchi, H. and Hibi, K. (2002b), "Quasi-static wind load combinations for buildings", The Second
International Symposium on Wind and Structures, Busan, Korea, pp.61-68.
Tamura, Y., Zhang, L., Yoshida, A., Nakata, S. and Itoh, T. (2002c), "Ambient vibration tests and modal
identification of structures by FDD and 2DOF-RD technique", Proceedings of the Structural Engineers World
Congress, Yokohama, Japan, T1-1-a-1, pp.8.
Tamura, Y., Kikuchi, H. and Hibi, K. (2003a), "Wind load combinations for middle-rise buildings (Part1 Wind force
combinations)", Summaries of Technical Papers of Annual Meeting, Architectural Institute of Japan, Structures I,
pp.111-112.
Tamura, Y., Kawai, H., Uematsu, Y., Kondo, K. and Ohkuma, T. (2003b), "Revision of AIJ Recommendations for
Wind Loads on Buildings", Proceedings of The International Wind Engineering Symposium (IWES2003), Taipei,
Taiwan, pp.81-96.
Zhou, Y. and Kareem, A. (2001), "Gust loading factor: new model", Journal of Structural Engineering, ASCE,
127(2), pp.168-175.
Zhou, Y. Kijewski, T. and Kareem, A. (2003), "Aerodynamic loads on tall buildings: an interactive database",
Journal of Structural Engineering, ASCE, 129 (3), pp.394-404.