ELSEVIER

Energy and Buildings 29 (1998) 83-92

Reductions in air conditioning energy caused by a nearby park
Vu Thanh Ca *, Takashi Asaeda, Eusuf Mohamad Abu a
“Saitam

Universi~,

Department

of Civil and Environmental

Engineering,

255 Shima-okubo,

Urawa,

Saitama 338-8570,

Japan

Received 7 November 1996; accepted 27 May 1998

Abstract
Field observations were carried out to determine the influence of a park on the urban summer climate in the nearby areas. The possibilities
of reduction in air conditioning energy were investigated. Air temperature, relative humidity and other meteorological factors were measured
at many locations inside a park and in the surrounding areas in the Tama New Town, a city in the west of the Tokyo Metropolitan Area, Japan.
The observations indicated that vegetation could significantly alter the climate in the town. At noon, the highest temperature of the ground
surface of the grass field in the park was 40.3”C, which was 19°C lower than that of the asphalt surface or 15°C lower than that of the concrete
surface in the parking or commercial areas. At the same time, air temperature measured at 1.2 m above the ground at the grass field inside the
park was more than 2°C lower than that measured at the same height in the surrounding commercial and parking areas. Soon after sunset, the
temperature of the ground surface at the grass field in the park became lower than that of the air, and the park became a cool island whereas
paved asphalt or concrete surfaces in the town remained hotter than the overlying air even late at night. With a size of about 0.6 km2, at noon,
the park can reduce by up to 1.5”C the air temperature in a busy commercial area 1 km downwind. This can lead to a significant decrease of
in air conditioning energy in the commercial area.
0 1998 Elsevier Science S.A. All rights reserved.
Keywords:

Air conditioning

energy;

Urban thermal

environment;

Thermal

comfort;

1. Introduction
It is well-known that the developmentof cities significantly
changesthe climate of the city and surrounding areas.The
increasein the impervious portion of the ground surface in
the urban area causedby pavements, roads and buildings
significantly reducesevaporation from the ground surface,
and consequently increasesthe underground heat storage.
This undergroundheat storagemakesthe ground surfacetemperature higher than that of the previously vegetated surface,
thus increasing the sensible heat exchange between the
ground surface and the atmosphere,and the upward long
wave radiation. On the other hand, tall buildings with shading
effect in the ground/wall surfacesmay causea reduction in
air temperature at noon. However, even with the shading
effect, in many cases,the heat releasedfrom the hot surfaces
together with anthropogenicheat releasedfrom industry and
human activity make the air temperature in the urban area
significantly higher than that in the surrounding rural areas,
causingthe so called ‘urban heat island’ (Oke [ 121, Asaeda
and Vu [ 11, Harazono et al. [ 71) . The intensity of the urban
heat island, defined as the difference in air temperature
* Corresponding
author. Tel.: + 81 48 8583567;
e-mail: vuca@env.civil.saimata-u.ac.jp
0378-7788/98/$
- see front matter 0
Z’ZZSO378-7788(98)00032-2

1998 Elsevier

fax:

+ 81 48 8587374;

Vegetation

between the city and surrounding rural areas,is more significant at night.
With increasing urbanization, the urban heat island will
affect a larger numberof urban residents.Thus in areaswhere
there is thermal stress,it makes senseto develop ecological
approachesto mitigation.
It has been reported that green areas in cities not only
improve the urban landscape,but also can regulate the urban
climate by increasing the moisture content in the air and
reducing the urban air temperature (Harazono et al. [7],
Honjo and Takakura [ 81) . With the presenceof a vegetated
surface, evapotranspiration can transform a large portion of
incoming solar radiation to the surface, which otherwise
would contribute to the underground heat storage,into latent
heat and makesthe ground surface cooler. Thus in making a
development plan of a city, the effect of vegetation on the
urban thermal climate shouldbe studiedcarefully.
Previous studiesregarding this problem mainly focus on
the effects of green areason the building thermal climate
(Givoni [ 61, Parker [ 13-151, Hoyano [9], Harazono et al.
[7] ). Honjo and Takakura [ 81 studied thermal effects of
urban green areas on their surrounding areas based on a
numerical model.
The purposeof this study is to give a detailed investigation
on the effects of vegetated areason the thermal climate of a

Science S.A. All rights reserved.

84

V.T. Ca et al. /Energy

and Buildings

new city in the Tokyo Metropolitan area, the Tama New
Town, through field observations of the ground surface temperature, atmospheric temperature and other atmospheric
conditions during hot summer days.

2. Observational

procedure

and equipment

The study area is located within the Tama New Town, a
new, well-planned city in the West of the Tokyo Metropolitan
area (Fig. 1). The construction of the town began in 1963.
At present, the town has an area of 30.2 km2 with a population
of about 410 000. The study area, which is near the Tama
Central Station, has a size of about 1.2 km X 1.2 km (Fig. 1) .
A large portion of the ground surface in the study region is
covered by commercial and office buildings together with
asphalt paved parking lots. There are several green areas in
the region with the largest one at the Tama Central Park as
indicated in Fig. 1.
Observations were carried out during the summer season
from August to September 1994. During the observations,
hourly values of air temperature, relative humidity, wind
velocity, foliage temperature, and temperatures of the grass
surface, wall and pavement surfaces were measured at many
locations in the area by six groups of students riding on
bicycles and cars. Solar radiation, downward longwave radiation, net radiation, albedo, together with soil temperature at
the ground surface, 5 and 10 cm depths were measured at the
center of a wide grass field in the center of the park. Air
temperature, relative humidity and wind velocity were also
measured at a height of 113.6 m above the ground surface on
top of the Fukutake Publishing House building, or 2 m above
the roof. The values measured above this building can be
considered representative for the study region.
Although observations were carried out several times during the summer of 1994, for the purpose of this study, all
observational periods except August 26-27 were rejected
because of variable weather. During some observational periods, the weather was fine during the morning, and then
became cloudy or rainy in the afternoon. Thus, the data collected were not representative of hot summer days in this area
of Japan. As the purpose of this study is to investigate the
effect of vegetation on the thermal environment of an urban
area, only results obtained during the fine days of August 2627 when the effects were notable are reported here.
Fig. 1 also depicts the observational points in the study
area during August 26-27. Details of the observational locations, items and equipment are shown in Table 1. In this table,
‘measured band’ in the last column means the sensitive band
for wave length.
The sampling interval is 5 min for soil temperature and
radiation and 30 min during the day and 1 h at night for the
other measurements. It was not possible to have simultaneous
measurements at all observational points, and values for a
physical quantity at all points at a given time were obtained
by interpolation.

29 (1998)

83-92

For the measurements by students riding on cars, to avoid
any possible contamination by engine heat from the car,
measurements were taken when the car was stopped and at a
distance of more than 5 m in front of the car.
Since various equipment were used for the measurement
of air temperature, relative humidity, ground surface temperature and wind velocity at different locations, all equipment
were calibrated before and after the observation in the laboratory and the field data were adjusted according to the calibration results.
The data of air temperature and wind velocity on top of the
Fukutake publishing house were verified by comparing with
data at nearby meteorological observational stations. A general agreement confirms that the obtained data can be considered representative for the observational area.
3. Effects of grass and wooded areas on the heating and
cooling
Atmospheric conditions such as solar radiation, atmospheric downward longwave radiation, net downward radiation recorded at the grass field inside the park are depicted in
Fig. 2. Fig. 3 depicts relative humidity, air temperature and
wind velocity, measured simultaneously at the same place. It
can be seen that during the day, the weather was fine with a
downward maximum solar radiation flux higher than 780 W/
m2 at 12 a.m. The highest air temperature at the grass field
(32.4”C) was recorded at 2 p.m. while the lowest air temperature (23.5”C) was recorded at 6 a.m. Relative humidity
reached 100% at night and about 60% at noon. It was relatively calm at night, but the wind became strong from 9 a.m.
until 7 p.m. with maximum wind velocity of 7.3 m/s,
recorded at 3 p.m., and predominantly from the south.
Fig. 4 depicts the time variation of the temperature at
ground surface in the grass field, on a nearby asphalt road
and concrete pavement, the temperature under the woods, the
foliage temperature above the woods and the temperature of
the water in the pond for 24 h, from 0O:OOAugust 26 to 0O:OO
August 27. The surface temperature of the asphalt road and
concrete pavement at noon rose up to 55°C at noon, almost
23°C higher than the air temperature measured at 1.2 m above
the surface. At the same time, the grass surface temperature
reached a maximum of 40.3”C, only about 8°C higher than
the air temperature. The foliage temperature above the woods
reached 37.8”C or 5.4”C higher than that of the air above the
grass surface, while the air temperature in the woods reached
31.o”c.
Fig. 5 depicts the time variation of temperature at the surface, 5 and 10 cm below the surface of the grass field. During
the observational period, there was no water supply to the
grass field, which was usually well irrigated. It is clear that
even when the surface temperature at the grass surface rose
to 40.3”C, the temperature at a depth of 5 cm reached a
maximum of only 30.2”C at 3 p.m., and the temperature at
10 cm depth reached a maximum of 28.5”C at 5 p.m. The
steep variation of soil temperature with depth near the grass

V.T. Ca et al. /Energy

and Buildings

29 (1998)

83-92

85

tnce
House

Fig. 1. Map of the study area and observational

Table 1
Observational

locations,

Location
Partenol

items and equipment
Observational

street

Air temperature,
velocity

Items
relative

Observational
humidity

wind direction,

wind

Streets in front of the
wk
Inside city

Air temperature, relative humidity, temperatures
of the sunlit
and shaded ground surface, tree foliage, grass surface
Air temperature,
relative humidity

2 m above the roof of
111.6 m above ground
surface
Road in the Park

Air temperature,
relative
velocity and direction

Grass field in the park

In the wood

points.

humidity

atmospheric

pressure,

wind

Air temperature,
relative humidity, temperatures
of the sunlit
and shaded ground surface, tree foliage, grass surface
Air temperature,
relative humidity, wind velocity and direction,
downward
longwave radiation, Albedo, Net radiation, Solar
radiation, Ground temperature
at the surface and 5 cm, 10 cm
depth, water temperature
in a nearby pond
Ground

surface temperature,

air temperature

Equipment

(accuracy

or measured

band)

Digital thermohygrometer
(DTH-3, Tokyo Keisoku)
( + 0.2”C
and + 0.5%); Microanemometer
(KC-101 Makino Oyo
Sokuki Kenkyujo)
Thermohygrometer
(TRH-CZ
Jinei) ( + 0. 1°C and + 0.5%) ;
Pyranometer
(IT2-60, Keyence)
(8-16 pm)
Digital thermohygrometer
(Climomaster
Model 65 1,
Kanomax)
( + O.l”C and + 0.5%)
Meteograph
No. 8-P (Ota Keiki Seisakusho)
( + 1°C and
+ 5%); Anemometer
(KC-101,
Makino Oyo Sokuki
Kenkyujo)
Thermohygrometer
(TRH-CZ,
Jinei) ( +O.l”C
and +0.5%);
Pyranometer
(Thermoflow
EM-101, Eko) (3-30 pm)
Thermohygrometer
(TRH-CZ,
Jinei) ( f O.l”C and +0.5%);
Anemometer
(KC-101, Makino Oyo Sokuki Kenkyujo)
;
Pyrgeometer
(MS-200,
Eko) ( 5-50 pm) ; Albedometer
(MR22, Eko) (0.3-3 pm) ; Net-radiometer
(CN-11, Eko) (0.3-50
pm) Pyranometer
(Ota Keiki) (0.36-2 pm) ; Thermocouple
Thermography
(TVS2000LW)
(Nippon Abionix)
(8-12
pm) ; Thermohygrometer
(TRH-CZ,
Jinei) ( + O.l”C and
&0.5%)

86

V.T. Ca et al. /Energy

-200

0

and Buildings

29 (1998)

83-92

the north of the park, together with the surface temperature
of a nearby grass area, a shaded area, and the foliage temperature. Fig. 7 depicts air temperature and relative humidity
measured at the same place at 1.2 m above (the surface). As
shown in Fig. 6, the temperature of the asphalt surfacereached
a maximum of 59°C at about 1 p.m. At the same time, the
temperature of the nearby grass surface reached a maximum
of only 39.7’T, almost 20°C lower than that of the asphalt
surface. The temperature of the leaves reached a maximum
value of 39.3”C and fell below that of the air from 5 p.m.
onwards. Comparing the air temperature and relative humid-

6

12
18
24
Local Time (HE.)
Fig. 2. Radiation conditions at the grass field, 0 incoming solar radiation,
0 downward infrared longwave radiation, 0 downward
net total radiation.

8.0

Local

Local Time (hrs)
Fig. 3. Meteorological
conditions at the grass field, 0 air temperature,
relative humidity, 0 wind velocity.

Local

Time

Time

(hrs)

Fig. 5. Time variation of soil temperature
at the grass field inside the park,
- 0 - at the soil surface, - 0 - 5 cm below surface, - 0 - 10 cm below
surface.
0

(hrs)

Fig. 4. Time variation of temperature
at the - 0 - asphalt surface, - 0 concrete surface, - X -grass surface, - A -under the wood, - . - foliage,
- - 0 - - water surface inside the park.

surface is due to the small thermal conductivity of the soil.
This means that for this surface, the heat conducted to deep
soil layers, or the below surface heat storage is small. Even
before sunset, at 5 p.m., the temperature of the grass surface
fell below that of the air and the grass surface began to cool
the atmosphere.
Other studies ( Asaeda et al. [ 21, Doll et al. [ 51) revealed
that in contrast to the grass surface, paved surfaces store a
large amount of heat during the day and continue to heat the
atmosphere throughout the day and night.
Fig. 6 depicts the time variation of surface temperature of
an asphalt paved parking lot inside the commercial area in

Fig. 6. Time variation of temperature
of the - 0 - asphalt surface, - 0 grass surface, - - - foliage, - X - shaded surface at the parking lot in the
commercial area.

Local Time (hrs)
Fig. 7. Air temperature
and relative humidity at 1.2 m above surface at the
parking lot in the commercial
area, - 0 - air temperature,
- 0 - relative
humidity.

V.T. Ca et al. /Energy

and Buildings

ity at the parking lot in Fig. 7 with that at the grass field in
the park in Fig. 3, it is evident that air temperature at the
parking lot reached the highest value of 34S”C at 2 p.m.,
about 2°C higher than that above the grass surface inside the
park at the same time. The relative humidity here reached a
maximum of 90% at 5 a.m. and a minimum of 48% at noon,
compared with that of 100% and 61%, respectively at the
grass field in the park.
In the afternoon, the temperature of the grass surface
decreased very quickly and became the same as the air temperature at about 6 p.m.; from 7 p.m. onwards, it was lower
than that of the air. Thus at night, the grass surface cooled
the atmosphere. On the other hand, throughout the day, the
temperature of the asphalt surface was always higher than
that of the air, even at night and in the early morning and this
surface always heated the atmosphere.
There are various reasons for the higher air temperature in
the parking lot compared with that in the grass field inside
the park. One of the reasons could be the fact that the place
is located inside the commercial area and large values of
anthropogenic heat, released due to air conditioning, traffic
and other human activities, which might heat the air. However, evaluated values of released anthropogenic heat in the
area based on land use and traffic data obtained from the
Tama City Government Office indicate that the daily average
of anthropogenic heat for this site did not exceed 60 W/m2.
On the other hand, the sensible heat released from the asphalt
paved surface (later in Fig. 8 (a) ) was evaluated at more than
360 W/m2 at noon, which was about 200 W/m2 higher than
that released from the bare soil surface (Fig. 8(b) ) . Comparing this value to the 60 W/m2 of anthropogenic heat (or
the values of 245-265 W/m2 in Section 4.2), it is reasonable
to remark that the contribution of the heat released from the
pavement surface is larger than that of the anthropogenic heat.
The lower relative humidity in this location is due to the
high air temperature and decreased evaporation at the ground
surface because of the pavement.
The heated ground surface heats the atmosphere not only
by sensible heat exchange, but also by the absorption of
longwave radiation emitted from the surface by the atmosphere. The upward longwave radiation emitted by the asphalt
pavement surface is 130 W/m2, higher than that by the bare
soil surface as depicted later on in Fig. 8 (a,b) , and is absorbed
within 100 m of the lower atmosphere ( Asaeda et al. [ 21) ,
increasing air temperature near the ground.
The high air temperature over the asphalt surface in the
parking lot, compared with that over the grass surface inside
the park, is due to sensible heat exchange with the paved
surface, the absorption of longwave radiation emitted from
the surface, and anthropogenic heat.
To make a quantitative evaluation of the contribution of
the pavement in the commercial and parking area on the
heating of the near surface air, and to study in detail the heat
balance at each surface, a numerical model developed by
Asaeda and Vu [l] has been employed. The heat balance
equation at the surface is

29 (1998)

83-92

87

o-

0

Time

6

(Hours)

12
Time

18

24

(Hours)

Fig. 8. Components
of heat fluxes at the surface of (a) asphalt pavement;
(b) the grass field in the park, total downward
solar radiation, - - net upward longwave radiation, - - - sensible heat flux, - - - latent heat
flux, - - downward
ground heat flux.

-KZ

=S(l-a)+R,“+H-L,

(1)

where K is the thermal conductivity of the ground surface; T
is the temperature of the ground surface; S and RLn are the
solar and net longwave radiation to the surface, respectively;
His the sensible heat flux; Le is the latent heat flux; and z is
the vertical coordinate. The left hand side of Eq. ( 1) represents the heat flux to the ground at the ground surface.
The solar radiation in Eq. ( 1) is evaluated based on the
measured solar radiation at the grass field and the partitioning
of this radiation into direct and diffuse components is estimated following Davies and Hay [ 41. The direct solar radiation to the surface under consideration is then estimated
based on the solar zenith angle and the height of the obstructing building. The incoming diffuse solar radiation from the
sky to the surface is evaluated as the product of total diffuse
solar radiation and the sky view factor from the point. Similarly, the incoming longwave radiation to the ground surface
is evaluated as the sum of total downward longwave radiation
from the sky, multiplied by the sky view factor, and the

88

V.T. Ca et al. /Energy

and Buildings

longwave radiation emitted by the surrounding walls. The
contribution of longwave radiation emitted by the surrounding walls to the net longwave radiation is estimated based on
the measured wall surface temperature and the view factor
from the surface for the wall. The view factors from the
horizontal surface for the sky and walls are determined following Johnson and Watson [ lo] _
It is extremely difficult to determine the sensible and latent
heat exchange between the ground surface and the air because
of the complexity of the surface conditions at the observational site. However, as the wind velocity, air temperature
and relative humidity were measured at many locations, it
was attempted to use the values measured at each place to
evaluate the heat and moisture exchange between the ground
surface and the air at the place. This can minimize possible
errors due to the following assumption of local similarity law
for the vertical distribution of potential temperature and wind
velocity near the ground surface. With this assumption, the
sensible heat exchange between the surface and the atmosphere can be evaluated using the following formula (Louis

[ill).
H=u*T,“=

83-92

bRiB
l+clRi,l”2

l-

1
(l+b’RiB)’

F,=C,*F,,

where AT, = TP- TPs.The drag coefficient a* is defined as

a2=k2

for stable case

where the coefficient C1* is assumedconstant and can be
determinedfrom the calibration of the model by comparing
the computed and measuredtemperaturesof the soil surface
until a best fit between them. C,* wasdeterminedby Asaeda
and Vu [ 11 as0.68.
In Eq. (6), b = 2b’ = 9.4, and c is defined as
l/2

c=C*a2

4
“0ZCl

(8)

where C* is a constant equal to 5.3 for both vapor and heat
fluxes (Louis, 1979). Once F, is determined, the turbulent
sensibleheat flux can be determinedas
(9)

(3)

T,U;

(6)

(7)

n

Ri =gzdTp

for unstable case

The functions F,, and F, are assumedto have the same
form. Thus,

zUa(TP-T,,)F,

where U*and TP*are the scalingvelocity andvirtual potential
temperature,respectively; a2 is the drag coefficient in neutral
conditions; n is a constant equal to 0.74; U, is the wind
velocity at height Z; T,, is the virtual potential temperatureat
height z; TPqis the virtual potential temperatureat the surface
roughnessheight z,; Fh is a function, accounting for atmospheric stability conditions near the ground surface; and RiB
is the bulk Richardsonnumber of the layer defined as

B

F,=

29 (1998)

2

(4)

where k is the Von Karman constant.
The atmospheric transport mechanismof water vapor is
quite similar to that of the sensibleheat (Louis [ 111, Brutsaert [ 31) . Thus, it is assumedthat the evaporation rate can
be describedby the sameform asEqs. (2) and (3)
a2
e=Pa ~II,(P,,-P,,Y,
where pais the air density; pvsandpvaare the specifichumidity
of air at the surface roughnessheight z, and at the screen
height, respectively; and F, is also a function, accounting for
atmosphericstability conditions near the ground surface.
The function Fhis supposedto have the form (Louis [ 111)

and the latent heat as

(10)

Fig. 8( a-b) depict componentsof heat fluxes at the asphalt
pavement surfaceand the surface of the grassfield insidethe
park, respectively. As seenin Fig. 8 (a), for the asphaltpavement, the ground heat flux peaksat 9:00 a.m. with a value of
about 320 W/m2 and remains positive until 5:OOp.m. This
large ground heat flux is releasedin the early evening to the
atmospherein the form of sensibleheat andupwardlongwave
radiation. The upward transfer of sensible heat and net
upward longwave radiation peak at 1:00 p.m. with values of
370 W/m2 and 230 W/m’, respectively. The sensibleheat
directly heatsthe air while the upward longwave radiation is
absorbed within a thin air layer near the ground surface
(Asaeda et al. [ 21)) affecting the near surface air temperature. Since the asphaltsurfaceis always hotter than the overlying air throughout the day and night, the surface always
heatsthe air even late a night, contributing to the formation
of the so called ‘nocturnal urban heat island’.
Fig. 8 (b) showsthat for the grasssurface,the largestportion of net radiation to the surfaceis converted to latent heat.
The latent heat increaseswith time from the morning, and at
12:00 a.m. its value is greater than 360 W/m2. At this time,
with the net solarradiation of 610 W /m2 reachingthe surface,
only 150 W/m’ is converted to sensibleheat and 75 W/m2
is converted to net upward longwave radiation. The ground
heat flux at this time is only 25 W/m2. At 1:00 p.m., the latent

V.‘.T. Ca et al. /Energy

and Buildings

heat is 360 W/m2 while the sensibleheat and net upward
longwave radiation are 170 W/m2 and 100 W/m’, respectively. Thus the cooling of the grasssurfaceis mainly due to
latent heat and the direct heating of the air by sensibleheat,
while the net upward longwave radiation is rather small.The
latent heat peaksat 2%) p.m. at a value of 420 W/m’. From
5:OO p.m. onwards, the temperature of the grass surface
becomeslower than that of the air, and the surfacebeginsto
cool the air due to downward transfer of sensibleheat. The
ground heat flux peaksat 1l:OO a.m. with a value of 215 W/
m2, becomingnegative at 1%) p.m. when cooling of the soil
surface begins.
Fig. 9(a) depictscontour lines of the air temperatureconstructed from measureddata at 1.2 m above surface for all
observational points at 0O:OOa.m. The contour lines are
drawn by interpolating air temperature recorded at each
observational point. From Fig. 9(a), even at this time, the air
temperaturedifference betweenthe coolestandhottest places
amountedto 15°C. The highest air temperatureat the Tama
Central Station and a road in a housingareain the South-East
of the park was 25.7%. The reasonfor high air temperatures
here is the heat releasedfrom paved surfacesand anthropogenieheatreleasedmainly dueto air conditioning. The lowest
air temperatureinsidethe park was24.2”C. At this time, there
is no predominant wind and the cooling effects by the park
extends almostequally in all directions. From the figure, one
can seethat the influence areaextendsonly about 100m from
the park.
Fig. 9(b) depicts contour lines of air temperature at 9:00
a.m. As in the figure, the air temperatureat the hottest place
over a parking lot in the commercial areato the North of the
park had reached30.5”C, 2.5”C higher than that at the coolest
place in the park. Since there was no obstruction to the solar
radiation receipt of the parking lot, the surface temperature
of the asphaltpavement hadreachedmore than 40°C; andthe
most likely candidatewhich contributes to the high air temperature in this place is the sensibleheat releasedfrom the
pavement surface. However, sincethis place is located in the
commercial area, anthropogenic heat releasedby air conditioning, traffic and other commercial activities cancontribute
significantly to the heating of air here. The air temperature
measuredin Partenolstreet,which is alsolocatedin the center
of the commercial area is not much higher even with the
expected high anthropogenicheat flux. One reasonfor this is
in the morning, the street is in the shadedue to high rise
buildings located on both sidesof this wide street. This phenomenonhasbeeninvestigated by Vu et al. [ IS], Swaid and
Hoffman [ 171, Sharlin and Hoffman [ 161.The other reason
for the cool air in the Patenol streetis becauseof the influence
of the park. At this time, there was a moderate South-East
wind and as in the Fig. 9(b) , and a cool area in the NorthWest of the park was due to this wind. The cooling effect of
the park extended about 700 m beyond the park boundary.
At 12 a.m., the pattern of temperaturedistribution in the
city is rather complicated, asshownin Fig. 9(c) . The hottest
place is now at the northern entrance of the park with air

29 (1998) 83-92

89

temperaturereaching 33.7”C. This openspaceislocated overhead a heavy traffic street and is surroundedby offices and
commercial buildings. The continuous heating of the pavement surfaceby solar radiation and the releaseof anthropogenie heat contributes to the high air temperaturehere.
A cool spot is observed at the Tama Central Park where
air temperature reached 3 1.5”C. The park was exposed to
solarradiation throughout the morning, but evaporation from
the grasssurface and trees and the shadingeffects of trees
caused low ground surface temperature, and consequently
low air temperature. The air temperaturedifference between
the park and the hottest areain the town is 2.2”C. Like that
in Fig. 9(b), there was a cool areajust in the North-West of
the hot areain the northern entranceof the park. At this time,
the South-East wind became very strong, and it can be
observedthat due to this wind, the influence areaof the park
can extend to a distance of about 1 km in the North-West
direction, from the park until the Tama Central Station.
Other hot areasat this time are located in the North-East
and the South of the park where the air temperaturereached
more than 33.5”C. A steephorizontal air temperaturegradient
is observedin the South-East, East and North of the park.
The cooling effect of the park is clearly observed in the
early afternoon as shown in Fig. 9(d), which depicts the
contour lines of air temperature at 2:00 p.m. The park now
becomesa cool islandsurroundedby hot areas.The difference
in air temperature between the park and the hottest place at
the Tama Central Station is more than 2°C. As the wind
direction haschangedto that from the South, the influence of
the park in the North-West direction was diminished. The
cool areaat this placenow becamea hot area.This is because
the heating from the surface of the parking lot and artificial
heat releaseddue to traffic and extensive air conditioning at
this time prevailed over the cooling effect. However, the
influence area of the park now shifted to the Partenol street.
As statedbefore, this is a busy commercialstreet with a wide
paved road, which is fully exposed to solar radiation from
12:00 a.m. At this time, there is also an extensive heating due
to air conditioning andthe hot ground surface.However, even
with this heating, air temperature in the street reached only
32°C well below that in the surroundingareas.The hot area
at the northern entranceof the park is alsomuch weakerthan
it was at 12:00 a.m.
4. Effect of the park on thermal environment and
energy use
4.1. Effect of the park on the thermal comfort
The effects of vegetation on the thermal environment in
the urban area can be evaluated basedon thermal comfort
analysis.In this study, this is doneby employing a two-nodes
model for the humanphysiological regulatory response,proposed by Gagge et al. [ 25,261. The model will be briefly
describedhere.

90

V.T. Ca et al. /Energy

Fig. 9. Contour

lines of air temperature

and Buildings

29 (1998)

83-92

at 1.2 m above surfaces at (a) 0O:OO a.m., (b) 9:00 a.m., (c) 12:00 a.m., (d) 2:00 pm.

V.T. Ca et al. /Energy

and Buildings

The heat exchange between the skin and the environment
at the skin surface Hsk is described by the heat balance equation as
H,k=h’(T,k-T,>+wh,‘(p,,,-p,,,)

(11)

where h’ is the combined heat transfer coefficient over the
temperature gradient; rY:ckis the skin temperature; T,, is the
operative temperature, defined as the average of ambient air
temperature and the mean radiant temperature (MRT),
weighted by their respective linear heat transfer coefficient
h, and h,; w is the skin wetness; h,’ is the latent heat transfer
coefficient; Pssk is the saturated vapor pressure at skin; and
Psdp is the ambient vapor pressure.
The heat flow to the skin surface from the body core is
expressed as
H,,=M-W-(E,,,SC,,,)-0

(12)

where M is the metabolic heat; W is the accomplished work;
E,, and C,, are latent and sensible heat exchange by respiration, respectively; and S is the body heat storage.
The thermal comfort is evaluated by the Predicted Mean
Vote (PMV) , proposed by Fanger [ 271 as
PMV=a[H,,-h’(T,,-T,)-E,iTf-Ecomfl

(13)

where Ediff is the evaporative heat loss caused by diffusion of
moisture through the skin and approximately 6% of the maximum possible evaporative cooling E,,, from the skin surface
when the skin wetness w = 1; and Ecomfis the evaporative loss
by sweating that occurs during the state of comfort and is
evaluated by Fanger [ 271 as:
E .,,,=0.42(M-W-58.2)

(14)

The maximum evaporative heat loss E,,, from the body
surface can be evaluated as (Gagge et al. [ 251)

E,,,=~~,[Ps~--P~ewl

(15)

where K is the Lewis relation and h, is the convective heat
transfer.
Values of the coefficients in the model are selected according to Kanda and Tsuchiya [ 191, [ 201, Gagge et al. [ 25,261
and Fanger [ 271.
The model has been employed for the analysis of thermal
comfort in the area. It found that at 12 a.m., inside the park,
the PMV for a walking person is 2.6. The PMV for the same
person in the upwind direction of the park is 3.2 and the
corresponding value of PMV in the downwind direction of
the park is 2.8.
Thus, with the presence of the park, a significant decrease
in the predicted mean vote is achieved, and the thermal environment in the area is improved.
4.2. Impacts of the park on energy use
Konopacki et al. [ 221, Meier and Taha [ 211 showed that
the for cities in the United States, a l-2°C reduction in spaceaveraged air temperature (at around 2 p.m. local time) leads
to a decrease of up to 10% in peak energy demand. In terms

29 (1998)

83-92

91

of annual costs of energy use, results of computations by the
authors suggested net savings in the order of US$lO-35 per
100 m2 of roof area depending on building type and region.
In this study, a model proposed by Vu et al. [24] was
employed for the evaluation of the reduction in air conditioning energy caused by the park. The model is a coupling of a
meso-scale model for the urban climate which allows a computation of the heat exchange between buildings and the air
in the urban canopy layer and a model for the evaluation of
building air conditioning energy. In the model, the conduction
heat flux to rooms of a building through walls and roof is
evaluated by solving the heat conduction equation for the
walls, roof with an assumed constant room air temperature;
and the air conditioning energy is evaluated based on the heat
exchange between walls, roof with room air, other heat loads
in the room due to machines, lights and human body, and the
cooler’s coefficient of performance (denoted as COP, a ratio
of pumped heat to utilized electricity).
Vu et al. [ 241 and Tanaka et al. [ 231 found that for aircooling type heat pumps or building multi-purpose type heat
pumps, COP decreases with increasing ambient air temperature, and is a function of partial load of the heat pump
(defined as the ratio of needed electricity and the capacity of
the heat pump). Multi-purpose type heat pumps are commonly used for room cooling in commercial and office areas.
Since the area under the influence of the park in this study is
a commercial and office area, this type of heat pump with the
capacity of 13 HP is used for the analysis.
Buildings used for the analysis are assumed as modem
four-storey ones with a fixed room temperature of 26°C. This
room temperature is typical for commercial and office rooms
in Japan during summer (Vu et al. [ 241) . The building ratio
(ratio of the ground surface covered by buildings to the total
land surface) is assumed as 0.5 and the average width of each
building in a mesh is 20 m. Results of computations shows
that at noon with an outside air temperature of 33.5”C and
roof level wind velocity of 3 m/s, the total conduction heat
to the room through walls and roof is 52 W/(m* ground
surface). This is about a half of other heat loads in the room
due to machines, lights and human body for this type of
building, which is approximately equal to 125 W / ( m2 ground
surface). Thus, the total heat to be pumped out of the building
amounted to 177 W / ( m* ground surface).
The COP, computed from outside air temperature and the
heat pump partial load for this case is 3.3. Thus, the electricity
needed for the heat pump is 54 W/ (m’ ground surface) and
the total anthropogenic heat released to the air is 224 W/ (m*
ground surface).
With the outside air temperature of 3 1.5”C, the computed
conduction heat flux to the room is 42 W/ (m* ground surface). With a COP of 3.6, the evaluated electricity needed
for the heat pump becomes 46 W/ (m* ground surface) and
the total anthropogenic heat released to the air becomes 223
W/(m* ground surface). Therefore, 8 W/(m* ground surface), or 16 W/ ( m2 build up area) of electricity for air cooling can be saved, and the reduction of anthropogenic heat

92

V.T. Ca et al. /Energy

and Buildings

release can improve air quality in the area. Thus, it can be
found that a saving of almost 15% in air cooling electricity
at noon is saved. This is a little large than that evaluated by
Konopacki et al. [ 223, and Meier and Taha [ 211. It must be
emphasized that the buildings used for this study are official
buildings, and the conduction heat to the room is larger for
residential buildings.
From observational results, it is found that the area of
influence of the park is almost 0.5 km’. Thus, within 1 h,
from 1 to 2 p.m., with the presence of the park, 4000 kW h
of electricity can be saved, and air quality in the area can be
improved.
The price for electricity in Japan is approximately Y22 per
kilowatt hour for summer. With 4000 kW h of saved electricity, 938 000, or roughly, US$650 can be saved within 1
h.
5. Conclusion
From observational results, it is found that even though
small, the Tama Central Park is significantly cooler than the
surrounding area during the day and at night. With strong
local southern winds during the day, it is expected that the
cool air from the park, being brought to the town, can significantly contribute to the reduction of the heat intensity in the
town.
The presence of the park can significantly improve the
thermal environment in the town. It is estimated that 4000
kW h of electricity for cooling, or US$650 can be saved
within 1 h from 1 to 2 p.m. of a hot summer day. Also, the
anthropogenic heat released to the air can be reduced.
Acknowledgements
The authors are grateful to Professor Milo Hoffman of
National Building Research Institute, Technion Institute of
Technology, Israel, and two referees for their careful reading
of the first draft of this paper and provided many useful
suggestions, even English correction. Many thanks are given
to Dr. Tuan Pham of the University of New South Wales,
Australia for English checking of the draft and many useful
suggestions. The contribution of Dr. Murakami of Nihon
Koei, Dr. Fujino of Saitama University, many people from
Nihon Koei and graduate and undergraduate students of Saitama University in the observation and data processing are
greatly appreciated.
References
[ l]

T. Asaeda, T.C. Vu, The subsurface transport of heat and moisture
and its effects on the environment:
a numerical model, BoundaryLayer Meteorol. 65 (1993) 159-179.
[ 21 T. Asaeda, T.C. Vu, A. Wake, Heat storage of pavement and the effects
on the near surface atmosphere, Atmospheric
Environment,
Part B,
Urban Environment
30 (1996) 413427.
[ 31 W. Brutsaert, Evaporation
into the Atmosphere,
Theory, History and
Applications,
D. Reidel Publ., 1984, 299 pp.

29 (1998)

83-92

[41 J.A. Davies, J.E. Hay, Calculation of the solar radiation incident on a
horizontal surface, in: J.E. Hay, T.K. Won (Eds.), Proceedings of the
First Canadian Solar Radiation Data Workshop,
1978, 32-58.
of subsurface
[51 D. Doll, J.K.S. Ching, J. Kaneshiro, Parameterization
heating for soil and concrete using net radiation data, Boundary-Layer
Meteorol.
32 (1985) 351-372.
I61 B. Givoni, Impact of planted areas on urban environmental quality: a
review, Atmospheric
Environment
25B (1991) 289-299.
S. Teraoka, I. Nakase, H. Ikeda, Effects of rooftop
[71 Y. Harazono,
vegetation
using artificial substrates on the urban climate and the
thermal load of buildings, Energy and Buildings 15-16 ( 1991) 435442.
Is1 T. Honjo, T. Takakura, Simulation of thermal effects of urban green
area on their surrounding
areas, Energy and Buildings 15-16 (1991)
443-446.
uses of plants for solar control and the
I91 A. Hoyano, Climatological
effects on the thermal environment
of a building, Energy and Building
ll(1988)
181-199.
IlO1 G.T. Johnson, I.D. Watson, The determination of view-factors in urban
canyons, J. Clim. Appl. Meteorol.
23 (1984) 329-335.
IllI J.-F. Louis, A parametric model of vertical eddy fluxes in the atmosphere, Boundary-Layer
Meteorol.
17 (1979) 182-202.
I121 T.R. Oke, The energetic basis of urban heat island, Quart. J. R. Met.
Sot. 108 (1982) l-24.
of vegetation on residential cooling,
[I31 J.H. Parker, The effectiveness
Passive Solar J. 2 (1983) 123-132.
1141 J.H. Parker, The use of shrubs in energy conservation planting, Landscape J. 6 (1987) 132-139.
consumpCl51 J.H. Parker, The impact of vegetation on air conditioning
tion, Proc. Conf. on Controlling
the Summer Heat Island, LBL-27872
(1989) 4652.
I161 N. Sharlin, M.E. Hoffman, The urban complex as a factor in the air
temperature
pattern in a mediterranean
coastal region, Energy and
Buildings 7 (1984) 149-158.
varia[I71 H. Swaid, M.E. Hoffman, Prediction of urban air temperature
tions using the analytical C’lTC model, Energy and Buildings
14
( 1990) 3 13-324.
ItsI T.C. Vu, T. Asaeda, M. Ito, S. Armfield, Characteristics of wind field
in a street canyon, J. Wind Eng. Aero-dynamics
57 ( 1995) 63-80.
I191 Chronological Table of Science (in Japanese), p. 266 (in Japanese).
r201 M. Kanda, N. Tsuchiya, Field observation and analysis for estimating
thermal load on a human body, Annual J. of Hydraulic Engr.. JSCE
38 (1994) 4191124.
[211 A. Meier, H. Taha, Mitigation of urban heat islands: meteorology,
energy, and air quality impact, Proc., Int. Symp. on Momitoring
and
Management
of Urban Heat Island, Japan, 1997, 124-163.
M.
I221 S. Konopacki, H. Akbari, S. Gabersek, L. Gartland, M. Moe&
Pomerantz, Energy and cost be&its
from light-colored
roofs in 11
U.S. cities, Lawrence Berkeley
National Laboratory
Report LBL39433, Berkeley, CA, 1996.
I231 M. Tanaka, M. Ozawa, Y. Ashie, A practical use of vaporization heat
of building materials for the improvement
of urban climate, Part 4:
Thermal environmental
load around buildings from air conditioners,
Japanese Conference on Remote Sensing, in Japanese, 1997, 12491252.
~241 T.C. Vu, Y. Ashie, T. Asaeda, Analysis of building air conditioning
energy during summer season, in preparation for submission to Energy
and Building.
~251 A.P. Gagge, J.A.J. Stolwijk, Y. Nishi, An effective temperature scale
based on a simple model of human physiological
regulatory response,
ASHRAE Semiannual Meeting in Philadelphia,
PA, 1970,247-262.
I261 A.P. Gagge, A.P. Fobelets, L.G. Berglund, A standardpredictiveindex
of human response to the thermal environment,
ASHRAE Transactions, Part 2B 90 ( 1986) 70973 1.
in Environv71 P.O. Fanger, Thermal Comfort Analysis and Applications
ment Engineering,
Danish Technical Press, 1970.