4.7
DEVELOPMENT OF COOL PAVEMENT WITH DARK COLORED HIGH ALBEDO COATING
1
2
3
4
Tsuyoshi Kinouchi *, Tamotsu Yoshinaka , Noriyuki Fukae and Manabu Kanda
2
Public Works Research Institute, Tsukuba, Japan, Nippo Corporation Co., Ltd, Tokyo, Japan
3
4
Nagashima Special Paint Co., Ltd, Tokyo, Japan, Tokyo Institute of Technology, Tokyo, Japan
1
1. INTRODUCTION
investigated by applying canopy energy balance
model in a real urban canopy setting.
It is considered that the dark surface of buildings
and pavements is one of the major heat source
causing the urban heat islands, as it absorbs more
heat from the sun. In Japan, several kinds of cool
pavement have been developed and their cooling
effects were investigated. However, they have never
been widely used in practice due to a limited effect in
mitigating urban thermal environment.
Some of the heat island mitigation studies deal with
the albedo increase of road and building surfaces
(Sailor, 1995). It is not difficult to raise the albedo of
the building roof as it can be attained by bright colored
painting. For pavements, however, the brighter
surface is not allowable for reasons of driving safety
and visibility of white line, unless the brightness is less
than that of the conventional asphalt and concrete
pavements.
In this study, a new type of pavement is developed
to satisfy both high albedo and low brightness by
introducing the paint coating technology. By applying
newly developed durable paint coating with high
albedo and low brightness to the conventional asphalt
pavement, the effectiveness of reducing pavement
temperature and sensible heat flux is investigated by
field measurements. Then, the impact of introducing
high albedo pavement on the overall canopy albedo,
canopy surface temperature and energy balance was
Apply high albedo dark colored
paint coating on pavement surface
遮熱コート材を舗装表面に塗布
Asphalt
mixture
表層
Asphalt遮熱性舗装
mixture
（低騒音舗装への
基層
Sub-base
2. DEVELOPMENT
2.1 Concept
A new type of pavement is developed to satisfy both
high albedo and low brightness with the innovative
paint coating technology. The function of this
pavement is based on a thin paint coating on the
surface of the conventional dark asphalt pavement,
which gives quite high reflectivity for the near infrared
and low reflectivity for the visible (Figure 1). This
results in the dark colored pavement surface while
achieving much higher albedo. The fine hollow
ceramic particles are included in the paint to expect
additional effects on reducing thermal conduction and
heating of the coat.
As the high brightness of road surfaces deteriorates
the visibility of painted lane markings, the target
brightness as represented by the L* value (a
brightness index) is set to approximately 40 and
under.
2.2 Results of laboratory experiment
In the laboratory experiments, trial paint coatings of
more than a hundred types with the combination of
different pigments and modifiers were tested. By this
procedure, we found some types of pigment and paint
coating structure are effective in achieving higher
reflectivity, lower brightness and suppressing the
surface temperature.
The reflectivity is measured for a number of metal
plates coated with different types of paint. The highest
albedo was more than 50% with the L* value around
40. Figure 2 indicates the reflectivity with wavelength
Solar
radiation
日射のうち，
Lo w reflection
for the visible
High
reflect ion for the near infrared
可視光を吸収・低反射
近赤外線を高反射
Hollow中空セラミック微粒子
ceramic particle
Visible
Near infrared
Asphalt
アスファルト
Aggregate
骨材
熱反射性特殊顔料
Highly
reflective pigment
Fig.1. Schematic view of paint-coated asphalt pavement
*Corresponding author address: Tsuyoshi Kinouchi,
Public Works Research Institute, Tsukuba, 305-8516,
Japan; e-mail: [email protected]
Albedo (%)
80
遮熱コート材の塗膜層
High albedo
*
40，黒色系）
（明度L
paint
coating
アスファルト混合物の
Asphalt pavement
骨材粒子
100
60
No.113
No.105
No.104
Prototype
StAs 60/80
40
20
0
300
600
900
1200
1500
Wavelength (nm)
1800
Fig.2. Reflectivity with wavelength
2100
Target L*<40
70
<Conventional pavement>
<Paint-coated pavement>
5cm
Asphalt mixture
Asphalt mixture
5cm
Gravel (20mm)
Gravel (20mm)
15cm
Macadam (C-40)
Macadam (C-40)
5cm
Filtering sand
Filtering sand
Albedo (%)
No.113, No.120
60
No.105
Baseline
30cm
Prototype N-60
50
40
No.104
200cm
Prototype N-40
30
30
40
50
60
Brightness index (L*)
Dark
Bright
Okinawa, 7/24/2003
Target L*<40
No.113, No.120
18
Baseline
No.105
16
Prototype N-60
No.104
65
60
55
50
45
40
35
30
25
20
Paint-coated pavement
Normal pavement
Air temp.
0
14
8/21
0
8/22
0
8/23
Time (hour)
0
8/24
0
8/25
0
Fig.6. Surface temperature variation of conventional
and developed asphalt pavement in August, 2003.
12
Prototype N-40
10
20
Dark
30
40
50
Brightness index (L*)
60
25
70
Bright
Fig.4. Relation between L* value and surface
temperature reduction (∆Ts)
for selected trial paint coatings. The reflectivity is low
for the visible and it changes significantly across the
wavelength 750nm, achieving stably high reflectivity
beyond it. The results indicate that the average
reflectivity of No.113, for example, is 86% for the near
infrared (750-2100nm wavelength) and 23% for the
visible (350-750nm). The albedo and the L* value of
those selected coatings range from 44 to 51% and 37
to 41, respectively.
The trial plates underwent the ultra-violet radiation
exposure test, which enables the evaluation of
changes in color, brightness and reflectivity after the
exposure to the ultra-violet radiation for the duration
equivalent to 3,000 hours. It is found that the changes
in albedo and the L* value are very limited.
2.3 Field measurement
The temperatures of asphalt coated by selected
paints were measured in the field yards located in
Tsukuba and Okinawa, the latter has been exposed to
larger solar radiation. In the Okinawa field
measurement, test pieces with dimensions of 30cm
square and 5cm thick were placed on existing asphalt
surfaces. The surface temperature was continuously
measured using thermocouples.
Figure 3 summarizes the relation between the L*
value and corresponding albedo for each coating of
the test piece. Two prototypes are the initially
developed coatings, and other circles represent
Surface temperature (°C)
Surface temperature reduction ∆Ts (°C)
Fig.3. Relation between L* value and albedo
20
Fig.5. Schematic diagram of vertical structure
70
Surface temperature (°C)
20
Styrene foams
Paint-coated pavement
Normal pavement
20
15
Air temp.
10
5
0
-5
-10
0
1/6
0
1/7
0
1/8
Time (hour)
0
1/9
0
1/10
0
Fig.7. Surface temperature variation of conventional
and developed asphalt pavement in January, 2003.
further developed test coatings. Blacked circles are
considered to have high quality in terms of brightness
and albedo. The maximum reduction of surface
temperature for each piece (∆Ts) from the
conventional asphalt pavement on a sunny clear day
(24th of July, 2003) was plotted against the L* value
(Figure 4). It is judged that pieces labeled as No.104,
105, 113 and 120 exhibit relatively high performance
in reducing the surface temperature.
Another field measurement has been conducted in
Tsukuba City located about 50 km north-east of Tokyo.
Larger test pieces with more realistic layered structure
were installed as shown in Figure 5. Meteorological
elements such as solar radiation, atmospheric
radiation, air temperature, relative humidity and wind
velocity at a height of 300 cm above the pavement
were measured along with the substrate temperatures
at several depths in the pavement.
It is found that the paint-coated asphalt pavement
shows about 15°C lower surface temperature than
that of the conventional one at the maximum (Figure
6). It must be noted that even in the nighttime the
surface of the paint-coated asphalt is cooler for more
Heat flux (W/m )
400
200
0
-200
2
Heat flux (W/m )
800
600
400
200
0
-200
Rn
3
H
G
0
2
Table 1. Summary of parameters
(a)
6
12
18
0
6
12
18
0
6
12
18
0
Floor
2.2
0.7
0.98
0.1-0.7
Walls*
2.1
1.6
0.98
0.2
Roof*
2.1
1.6
0.98
0.2
Rn
(b)
H
G
0
Heat capacity(J/cm /K)
Conductivity(W/m/K)
Emissivity
Albedo
*Nunez and Oke (1976)
6
12
18
0
6
12
18
Time (hour)
0
6
12
18
0
Fig.8. Energy balance of (a) conventional pavement
and (b) paint-coated pavement for August, 2002.
than 2°C. In the winter, the surface temperature of the
paint-coated pavement is slightly lower than that of
the conventional one even below the freezing point
(Figure 7), and it may result in the delay of snowmelt.
Figure 8 shows the energy balance of the
conventional and paint-coated asphalt pavements.
The sensible heat flux (H) is estimated from Louis’s
scheme (Louis, 1979) with the roughness length
-4
z0=3x10 m, which was verified to give a satisfactory
estimate. The heat flux into the ground (G) is derived
by subtracting H from the measured net radiation (Rn).
The surface albedo was referred to the result of
laboratory experiments. It is found that a significant
reduction can be found by applying high albedo
coatings for the sensible heat flux and the heat flux
into the ground.
3. IMACT ON URBAN ATMOSPHERE
3.1 Outline
The high albedo pavement reflects more solar
radiation back to the sky. If it is widely used for the
urban canopy floor, surrounding buildings absorb part
of the reflected solar radiation, which in turn could
increase the wall temperature and sensible heat
fluxes from the wall. Thus, the impact of introducing
high albedo pavement on the overall canopy albedo
and energy balance was investigated by applying a
canopy energy balance model developed by Kanda et
al. (2004a) to a real urban canopy setting. The
dependency of the overall canopy albedo, sensible
heat flux and canopy surface temperature on the
canopy configuration and floor albedo was
investigated, and the efficiency of introducing the high
albedo pavement is discussed.
3.2 Model explanation
The energy balance model used in this study is the
simple urban energy balance model for meso-cale
simulations (SUMM), which consists of a 3-D
theoretical radiation scheme (Kanda et al., 2004b) and
the conventional heat transfer expression that uses a
network of resistances. SUMM allows one to readily
calculate the energy balance and surface temperature
at each face of the urban canopy (i.e., roof, floor, and
four vertical walls) without time-consuming iterations.
This model was validated by the surface temperature
of canopy facets and heat fluxes measured in a
reduced-scale hardware canopy model experiment. In
addition, the parameters related to the thermal
properties for the asphalt pavement of the canopy
floor were determined by comparing simulation results
of asphalt surface temperature in a horizontally open
situation with the data of field measurement in
Tsukuba (Table 1). Other thermal properties of walls
and roof used in simulations are also listed in Table 1.
The outer meteorological boundary conditions
including the downward shortwave and longwave
radiation were given from Moriwaki and Kanda (2003)
as an hourly average for sunny days in July 2002
measured at the height of 29m over a densely built-up
residential area in Tokyo. The inner temperature of
buildings and the bottom temperature of the canopy
floor are assumed to be a constant.
The urban canopy geometry in the simulation
assumes an infinitely extended regular array of
buildings with square horizontal cross-section and
uniform surface properties, and streets with asphalt
pavement. The canopy geometry is characterized by
the horizontal dimension of buildings (W), the height
of buildings (H) and the width of streets (L). W is set
to 10m in all simulations, and the plane area index
2
2
and
frontal
area
index
(λp=W /(W+L) )
2
(λf=WH/(W+L) ) are used to define canopy geometry.
3.3 Results
Figures 9 and 10 show the total sensible heat flux
(TSH) at the reference height (29m) and the
area-weighted temperature of four wall facets and
floor surface (AT) for four cases of floor albedo (α=0.1,
0.3, 0.5 and 0.7) with the canopy configuration
H=10.5m and W=10m. It is found that both TSH and
AT are significantly reduced with increased albedo for
the smaller plane area index (L=20m, λp=0.1111).
However, the smaller λp case gives larger sensible
heat flux for the floor albedo below around 0.3.
Figures 11~13 show the dependency of TSH, AT
and the overall canopy albedo (CA) on the floor
albedo and canopy configuration. TSH, AT and CA is
more sensitive to the floor albedo for the smaller
frontal area index. TSH for the higher floor albedo
(α=0.5) is less sensitive to the plane area index than
the lower albedo case (α=0.1). AT is mostly influenced
by the frontal area index and floor albedo. AT is
reduced with increased albedo even though the wall
surface temperature is slightly increased due to the
600
400
200
100
0
-100
α=0.3
400
α=0.5
300
α=0.7
2
2
300
600
α=0.1
500
200
100
0
-100
0
3
6
9 12 15
Time (hour)
18
21
24
0
3
6
9 12 15
Time (hour)
18
Sensible heat flux (W/m )
α=0.1
α=0.3
α=0.5
α=0.7
500
Sensible heat flux (W/m )
2
Sensible heat flux (W/m )
600
21
24
α=0.1
α=0.3
α=0.5
α=0.7
0
3
6
9 12 15
Time (hour)
18
21
24
40
38
36
34
32
30
28
26
24
22
20
400
300
α=0.5
200
100
Plane area index
0.25
L=10m
0.4444
0.5625
0.1111
0.25
L=20m
0.36
0
0
0.1
α=0.1
0.2
0.3
Frontal area index
0.4
0.5
Fig.11. Total sensible heat flux (TSH) from whole
canopy layer (W=10m, L and H are variable).
α=0.3
α=0.5
α=0.7
0
3
6
9 12 15
Time (hour)
18
44
21
24
Fig.10. Averaged surface temperature of wall and floor
with H=10.5m and W=10m. (Left: L=10m, right: L=20m)
higher incident solar radiation reflected from the
canopy floor. CA for α=0.5 is much dependent on the
plane area index than that for α=0.1. If it is assumed
that the same upper boundary condition holds for any
canopy configurations, it can be said that smaller
plane area index and frontal area index become
effective in increasing overall canopy albedo when the
canopy floor albedo is higher.
Surface temperature (°C)
40
38
36
34
32
30
28
26
24
22
20
Surface temperature (°C)
Surface temperature (°C)
Fig.9. Total sensible heat flux from whole canopy layer
with H=10.5m and W=10m. (Left: L=10m, right: L=20m)
α=0.1
500
38
L=20m
36
0.25
0.4444
0.5625
0.1111
0.25
0.36
34
α=0.5
32
0
0.1
0.2
0.3
Frontal area index
0.4
0.5
Fig.12. Area-averaged surface temperature (AT) of
walls and floor (W=10m, L and H are variable).
The high albedo pavement is considered to reduce
the air temperature near the ground and the longwave
radiation emitted from the pavement surface. On the
other hand, it must be considered that the pavement
surface reflects more solar radiation, and it may
increase the thermal stress on the human body
walking or standing on it. Thus, a preliminary test was
carried out to reveal the impact on the thermal
sensation by letting 6 volunteers stand on the
paint-coated pavement and conventional pavement
under the summer outdoor environment.
Thermal sensation, comfort sensation and
sensation regarding the thermal impact on the feet
were declared. WBGT was also monitored during the
test. It is found that the high albedo pavement gives
cooler sensation than the conventional one, which
may be resulted from the mitigated heat conduction
through the feet and the upward longwave radiation.
0.4
Canopy albedo
0.5
Kanda, M. et al., 2004a: Boundary-Layer Meteorol.
(submitted)
Kanda, M., T. Kawai and K. Nakagawa, 2004b:
Boundary-Layer Meteorol. (in press)
L=10m
40
4. IMACT ON HUMAN THERMAL SENSATION
5. REFERENCES
Plane area index
α=0.1
42
Plane area index
0.25
0.4444
L=10m
0.5625
0.1111
L=20m
0.25
0.36
0.3
α=0.5
0.2
0.1
α=0.1
0
0
0.1
0.2
0.3
Frontal area index
0.4
0.5
Fig.13. Canopy albedo (CA) at noon for whole
canopy layer (W=10m, L and H are variable).
Louis, J-F., 1979: Boundary-Layer Meteorol., 17, 182-202.
Moriwaki, R. and M. Kanda, 2003: J. Japan Soc. Hydrol.
and Water Resour. 16, 477-490 (in Japanese).
Nunez, M. and T. R. Oke, 1976: Boundary-Layer Meteorol.,
10, 121-135.
Sailor, D.J., 1995: Journal of Applied Meteorology, Vol.34,
1694-1704.