-
COLD AIR SEEPING FROM AN URBAN GREEN SPACE, IMPERIAL PALACE, IN
CENTRAL TOKYO
Ken-ichi Narita*1, Hirofumi Sugawara*2, Hitoshi Yokoyama*3,
Ikusei Misaka*4 and Dai Matsushima*5 *1Nippon Institute Technology,
Saitama, Japan; *2National Defense Academy, Kanagawa, Japan
*3 Tokyo Metropolitan Research Institute for Environmental
Protection, Tokyo, Japan *4 R&D Institute, Takenaka
Corporation, Chiba, Japan;*5 Chiba Institute of Technology, Chiba,
Japan
Abstract In this paper, results are shown from
micro-climatological observations performed in and around a largest
green space in central Tokyo, “Imperial Palace”, during summer. Air
temperature was measured at 1-minute intervals between the boundary
of the green space and the Tokyo Station. In order to catch the
phenomena of cold air seeping-out in calm conditions, several sonic
anemometers were set along the green space boundary. In a clear
calm night, flow-out wind directions from green spaces to
surrounding built-up areas were discerned at all measuring points.
The obvious shift of wind direction appeared at the beginning of
the flow-out with a sharp temperature fall of 1 or 2 degrees. Key
words: radiative cooling, cold air drainage, urban park 1.
INTRODUCTION The mitigation effect of green space in urban area has
been expected to be comparable to that of water surfaces. In Japan,
its cooling effect in hot and humid summer is especially important
and it is considered as an important method available for city
planning. In a clear calm night, ground inversion layer is formed
in a green space by radiative cooling, and some accumulated cold
air near green surface flows out gravitationally into surrounding
city area. Such “cold air seeping” was firstly observed in
“Shinjyuku-Gyoen Park” in Tokyo (Narita et al. 2002, Honjyo et al.
2003 and Narita et al. 2004). Different from the daytime advection
by prevailing wind, the turbulent mixing is very weak in nighttime
seeping phenomena because of the stable atmospheric condition.
Therefore, the cooling effect for ambient air, that is the negative
sensible heat flux, was very small, but the flow-out air was kept
cool as far as the seeping front, 80-90 m from the boundary in that
case. In this paper, results are shown from micro-climatological
observations performed in and around a largest green space in
central Tokyo, “Imperial Palace”, during summer. It was once the
Edo-Castle, then surrounded by a moat. Its area is 230ha, and it
located adjacent to one of the main CBD, “Maru-no-uchi”, which
includes Tokyo Station. The topography around the Imperial Palace
is inclined toward east; the elevation of west side is 25 m higher
than east side. Then, the east side water surface in the moat is
only few meters lower than ground level, but the depth of west side
moat is more than 20 m. The core area of the Imperial Place is
covered by dense forest.
2. OBSERVATION METHODS The observations were performed during
summer in 2007 and 2008. The main purpose in 2008 was to make clear
the effect of topography on the cold air seeping, and that in 2007
was to grasp the seeping front in east side CBD area. Air
temperature was measured at 1-minute intervals at totally 27 points
in and around the Imperial Palace including west side moat slope in
2008 (Fig. 1), and at totally 50 points between the east side
boundary of the green space and the Tokyo Station in 2007. In order
to catch the cold air seeping-out in calm conditions, several sonic
anemometers were set along the green space boundary in 2008, and
also in adjacent built up area in 2007. Measuring height was 2.5m
above the ground and sampling frequency 1Hz. In addition, the
vertical air temperature profile was measured using a lift car in
2007 to reveal the depth of seeping-out cold air.
Fig.1 Observation points around Imperial Palace.
*Corresponding author address; Ken-ichi Narita, Department
ofArchitecture, Nippon Institute of Technology, Gakuen-dai, 4-1,
Miyashiro, Saitama, Japan; e-mail: [email protected]
Imperial Palace
S2S1
S7
S6
S5 S4S3
S8
S9
S10
S11
S12 S13S14
S15
S16
F1
P1～5
C1～5
ThermometerSonic (within palace)Sonic anemometer
Slope Temp.
Water Temp.
Sakashita Gate
Imperial Palace
S2S1
S7
S6
S5 S4S3
S8
S9
S10
S11
S12 S13S14
S15
S16
F1
P1～5
C1～5
ThermometerSonic (within palace)Sonic anemometer
Slope Temp.
Water Temp.
Sakashita Gate
The seventh International Conference on Urban Climate, 29 June -
3 July 2009, Yokohama, Japan
-
3. RESULTS Observation period in 2008 was totally 65 days from
July 27 to September 29. The weather condition of this period was
rather anomalous, the number of no rainfall day amounts only 19.
Figure 2 shows the example of seeping-out phenomena of cold air in
calm clear night around Imperial Palace. At east-side point (S16),
wind direction is almost fixed in north-west after 22:00 on Aug.
14. At the west-side inner point (F1), obvious shift of wind
direction appeared at 1:00 on Aug. 15 with a sharp temperature fall
of 2 degrees. Before this event, time variation of temperature and
wind direction is similar to that of the east-side. These imply the
westerly wind after 22:00 was synoptic scale phenomena and it
resulted in rather rapid temperature drop at all measuring points.
Then, three hours later (at 1:00), this westerly wind became weak
and cold air drainage from the Imperial Palace was started even in
the west side. The wind directions at 1:30 at each sonic point show
the cold air flows out to the surrounding city area for all
direction (illustrated by arrows in Figure 2). While there was no
traffic near the points (F1) and (S16), the other four sonic
anemometers were clung to the roadside light poles and probably
affected by traffic disturbance. That is the reason why the wind
direction was fluctuated and the sharp temperature drop was not so
clear at these points. In order to examine the topographic effect
of the deep moat, we focused the difference between inner side
point (F1) and outer side point (S7) of the west side moat. In the
outer side, the magnitude of sharp temperature drop became smaller
than that of the inner side, but there can be seen no significant
time lag in both side temperature time-variations. This suggested
that the cold air from the Imperial Place flows across the deep
moat at once; it dose not constitute the barriers against the cold
air seeping.
Outer Slope
25
26
27
28
29
30
31
32
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p. �
��
Water Temp.
A�e�A�S9
C5C3
C1 1.5m7.5m
7.5m
7.5m
7.5 m
S9
C5C4
C3
C2
C11.5m7.5m
7.5m
7.5m
7.5 m
S9
C5C4
C3
C2
C1
Fig.3 Time variation of air temperature along the moat
slope.
25
26
27
28
29
30
31
32
33
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p���
�
0
1
2
3
4
5
6
7
8
Win
d S
peed
(m/s
)
N
W
S
E
N
Air Temp.
Wind Speed
Wind Direc.
Water Temp.
25
26
27
28
29
30
31
32
33
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p���
�
0
1
2
3
4
5
6
7
8
Win
d S
peed
(m/s
)
N
W
S
E
N
25
26
27
28
29
30
31
32
33
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p���
�
0
1
2
3
4
5
6
7
8
Win
d S
peed
(m/s
)
N
W
S
E
N
25
26
27
28
29
30
31
32
33
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p���
�
0
1
2
3
4
5
6
7
8
Win
d S
peed
(m/s
)
N
W
S
E
N
25
26
27
28
29
30
31
32
33
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p���
�
0
1
2
3
4
5
6
7
8
Win
d S
peed
(m/s
)
N
W
S
E
N
Air Temp.
Wind Speed
Wind Direc.
Wind direction at 1:30, Aug. 15
S5
S7
S11
S1
F1
25
26
27
28
29
30
31
32
33
8/1418
8/1419
8/1420
8/1421
8/1422
8/1423
8/150
8/151
8/152
8/153
8/154
8/155
8/156
Air
Tem
p���
�0
1
2
3
4
5
6
7
8
Win
d S
peed
(m/s
)
N
W
S
E
N
S16
Fig.2 Seeping-out phenomena of cold air in calm night－1-minute
average time variations of air temperature, wind speed and
direction at boundary points.
The seventh International Conference on Urban Climate, 29 June -
3 July 2009, Yokohama, Japan
-
Figure 3 shows the comparison of air temperature variation at
the points along the outer slope. At road side point (S9), the
sharp temperature drop appeared clearly just after 1:00. On the
slope (C1～C5), the temperature decreased gradually from the time
about one hour earlier than road side steep change. Notice the
water temperature in Figure 3. This moat has no sufficient inflow,
so natural rain is the only water supply. Consequently, water
temperature was higher than air temperature during the night;
exceeded 30 degrees Centigrade even early in the morning.
Nevertheless, the air temperature was lowest in the bottom of
slope, that is, the ground inversion layer was formed in the moat
valley. The occurrence of cold air seeping phenomena is limited in
calm night, so the turbulent heat exchange at the water surface is
considered very weak. Figure 4 is the result in 2007, focused on
the seeping front in “Maru-no-uchi” CBD area. Three figures show
the 1-minute average time variations of air temperature, wind speed
and direction, as same manner as Figure 2. The upper graph is the
result of point (S16) that is common to the observation in 2008. In
this night, cold air seeping was started about at 0:50. At the
point beyond the moat (middle graph), and at the edge of CBD beyond
the Route 1 (lower graph), the start of temperature drop appeared
later than point (S16). Figure 5 shows the temperature field during
3:00～4:00, when cold air expanded most broadly. The arrows in the
figure express the wind directions at each sonic point. To
investigate the effect of the moat and wide-street on the seeping
of cold air in detail, the isopleths between the Imperial Palace
(Sakashita Gate) and the Tokyo Station were drawn in Figure 6.
Here, the color means a temperature difference from the
area-averaged temperature around Tokyo Station. Within the “Kokyo
Gaien National Garden”, cold air spread out rapidly. However, the
beginning of invasion of seeping front into CBD area was about two
hours later. After that, the cold air front penetrated the CBD area
and reached near the Tokyo Station. According to the sonic data of
wind direction change, there was not so large time-lag until the
front passed across the moat. These results suggested that the wide
street have a relatively large effect on cold air seeping than the
moat even though water temperature was higher than ambient air
temperature. Figure 7 is the time-height isopleths drawn using
vertical temperature profile collected by a lift car. Here,
temperature was expressed as a difference from the upper layer air
temperature at the height of 110 meters. This special observation
was carried out for three days, and cold air seeping occurred only
one night during these period. At the time of most strongly cooled
period, air temperature was lower than reference upper layer up to
the height of 14 m. Even the top height (25 m above the ground),
the effect of cold air seeping was discernible clearly. 4.
CONCLUSIONS In a clear calm night, the cold air flows out from the
Imperial Palace to the surrounding city area gravitationally. In
the west side, it flows over the valley of moat (20m depth) and
spread out into the built-up area. In the east side, cold air front
sometimes penetrate the adjacent CBD area and reached near the
Tokyo Station.
Fig.5 Temperature field and wind direction when seeping-out
phenomena appeared.
Fig.4 Seeping-out phenomena of cold air in the east side of
Imperial Palace. （same manner as Figure 2）
252627282930313233
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
d Sp
eed(
m/s
)
W
N
S
E
N
252627282930313233
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
d Sp
eed(
m/s
)
W
N
S
E
N
252627282930313233
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
d S
peed
(m/s
)
W
N
S
E
N
Aug. 9～102007
0 300m
S16
Wind Speed
Air Temp. Wind Dire.
Water Temp.
252627282930313233
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
d Sp
eed(
m/s
)
W
N
S
E
N
252627282930313233
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
d Sp
eed(
m/s
)
W
N
S
E
N
252627282930313233
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
d S
peed
(m/s
)
W
N
S
E
N
Aug. 9～102007
0 300m0 300m
S16
Wind SpeedWind Speed
Air Temp.Air Temp. Wind Dire.Wind Dire.
Water Temp.
The seventh International Conference on Urban Climate, 29 June -
3 July 2009, Yokohama, Japan
-
References Narita, K., Mikami, T., Sugawara, H., Honjo, T.,
Kimura, K. and Kuwata, N., 2002.Observation about cool-island
phenomena in urban park, AMS 4th Symposium on the Urban Environment
(Norfolk, VN, USA), 86-87. Honjo, T., Narita, K., Mikami, T.,
Sugawara, H., Kimura, K. and Kuwata, N., 2003. Observation of Cool
Island Effect in Urban Park (Shinjuku Gyoen), Proceedings of ICUC-5
(Lodz, Poland), Vol.2, 261-264 Narita, K., Mikami, T., Sugawara,
H., Honjo, T., Kimura, K. and Kuwata, N., 2004. Cool-island and
cold air-seeping phenomena in an urban park, Shinjyuku Gyoen,
Tokyo, Geographical Review of Japan, 77, 403-420. (in Japanese with
a English summary) Sugawara, H.,Tanaka, T., Narita, K., Nakano, T.
And Mikami, T., 2008. How much cool air dose an urban green park
produce? Geographical reports of Tokyo Metropolitan University, 43,
83-89. Observational study of the heat island - Cooling effect of
urban park, 2008. Newsletter of urban heat island countermeasure,
AIJ, Vol.3.
(http://news-sv.aij.or.jp/tkankyo/s3/Newsletter_vol3_081118.pdf)
Profile
Sonic
S16
0 5 10 15 20 25 30 35 40 45 50 55 6002468
1012141618202224
heig
ht (m
)
024681012141618202224
18 19 20 21 22 23 0 1 2 3 4 5 6Time (JST) [h]
S16
222324252627282930
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
012345678
Win
hd S
peed
(m/s
) W
N
S
E
NAir Temp.
Wind Speed
Wind Direc.
Aug. 6～7 2007
Fig.7 Depth of cold air seeping from Imperial Palace.
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
60000
200
400
600
Dis
tanc
e(m
)
N
25
27
29
31
33
18 19 20 21 22 23 0 1 2 3 4 5 6
Air
Tem
p.(℃
)
0
1
2
3
4
Win
d Sp
eed(
m/s
)Wind Direc.
Wind Speed
Air Temp.W
S
N
E
N
25
27
29
31
33
18 19 20 21 22 23 0 1 2 3 4 5 6A
ir Te
mp.
(℃)
0
1
2
3
4
Win
d S
peed
(m/s
)
Air Temp.
Wind Direc.
Wind Speed
W
S
N
E
N
1� 1� 20 21 22 23 0 1 2 3 4 5 6
-3.8
-3.4
-3.0
-2.6
-2.2
-1.8
-1.4
-1.0
-0.6
-0.2
���������
�������������������������
������������ �������������
Water Temp.
�������������
���������������
������������
������1
��������������
�����������N��������������
�������������������������℃�
��������������
�����
�������
Fig.6 Air temperature difference isopleths between Imperial
Palace and Tokyo Station.
Aug. 9～10 2007
The seventh International Conference on Urban Climate, 29 June -
3 July 2009, Yokohama, Japan
![BOURBON/RYE $6 [24,600] THE LOUNGE MENU · East Imperial Soda Water / East Imperial Burma Tonic Water / East Imperial Grapefruit Tonic / East Imperial Yuzu Tonic / East Imperial Thai](https://img.pdfslide.us/doc/110x75/5e6ccd6cbdaeea1277122949/bourbonrye-6-24600-the-lounge-menu-east-imperial-soda-water-east-imperial.jpg)
