climate

Article
Influence of Urban Green Area on Air Temperature of
Surrounding Built-Up Area
Hideki Takebayashi ID

Department of Architecture, Kobe University, Kobe 657-8501, Japan; thideki@kobe-u.ac.jp; Tel.: +81-78-803-6062

Academic Editors: Michele Zinzi and Matheos Santamouris

Received: 17 July 2017; Accepted: 2 August 2017; Published: 7 August 2017

Abstract: In this investigation, a numerical model expressing advection and diffusion effects is used
to examine air temperature rise in urban areas that are on the leeward side of green areas. The model
results are then verified by comparison with measurement results. When the measurement point is at
a distance of 30 m or more from a green area, the air temperature of the urban area is not affected
by the green area. An isotropic diffusion model and a model incorporating buoyancy were applied
for the vertical diffusion term. Results of air temperature rise with distance from the green area
were compared for both calculated and measured values. The rise in air temperature due to the
development of the urban boundary layer in the area near a green space is expressed using the
sensible heat flux from the ground surface, the distance from the green area and the wind velocity.
We considered an approximation of air temperature rise in order to express the following situation:
when entering the urban area, air temperature rises sharply, and when reaching a certain distance
from a green area, it becomes almost constant.

Keywords: green area; built-up area; air temperature; measurement; calculation

1. Introduction
Urban greenery is one of the main measures for mitigating the thermal environment in urban
spaces. Givoni [1] has organized the functions and impacts of urban planted areas through a review
of research papers and has presented climatic guidelines for hot-dry regions, hot-humid regions
and cold regions. A summary of climatic guidelines for park design is as follows: it is to provide
ample shade and to protect from dust for hot-dry regions; it is to provide shade, to minimize wind
blockage, to improve the ventilation and to minimize floods for hot-humid regions; it is to provide
wind protection without blocking the winter sun for cold regions. He summarized that the influence
of city parks and open spaces on the urban climate is limited to the conditions prevailing within these
areas themselves, and extends only a short distance into the surrounding, densely built, urban area.
On the other hand, Honjo and Takakura [2] explained that the range of the effects of urban green
areas extends to about 100 to 300 m into the surrounding urban area. They also explained that 300 m
along the main wind direction is the ideal length for an urban green area, based on two-dimensional
analysis results.
In recent years, interest in this field of study has increased. How to quantify the range of the air
temperature reduction effect of an urban green area on the surrounding urban area is a question that
has been frequently asked by administrative officials responsible for organizing urban green spaces.
Moriyama et al. [3] have conducted numerical simulations to examine increases and decreases in air
temperature in urban areas adjacent to green areas. They used the following conditions: an inflow
upper wind velocity of 2 to 6 m/s at 50 m above the ground, a ground surface temperature difference
of 1 to 5 ◦ C between green and urban areas, and a roughness parameter of 0.1 to 1.0 m for green areas
and 0.5 to 1.0 m for urban areas. The evaluation height was 3.25 m above the ground. They concluded

Climate 2017, 5, 60; doi:10.3390/cli5030060 www.mdpi.com/journal/climate

Climate 2017, 5, 60 2 of 12
Climate 2017, 5, 60 2 of 12

concluded that the influence of the green space extends to a distance of about 150 m from the
urban-green
that the influenceboundary.of theThe green above-mentioned
space extends toHonjo a distanceet al. of [2]
about have150 carried
m from out numerical
the urban-green
simulations
boundary. under the condition that
The above-mentioned Honjoan inflow
et al. [2]upper wind velocity
have carried is 4 m/ssimulations
out numerical at 200 m above underthe the
ground, a ground surface temperature difference is 4 °C between green
condition that an inflow upper wind velocity is 4 m/s at 200 m above the ground, a ground surface and urban areas, and a
roughness
temperature parameter
difference is is ◦
0.24 mCfor both the
between greengreen
andarea
urban and the urban
areas, area. The evaluation
and a roughness parameter is height
0.2 min for
this
bothcase
thewas 2m
green above
area andthetheground.
urban area.They Theconcluded
evaluation that evenin
height a green
this casearea
was with
2 m100 m size
above the affects
ground.
the areaconcluded
They within a distance
that evenofa about
green area300 m from
with 100the
m urban-green
size affects the boundary.
area within a distance of about 300 m
fromThere are a few studies
the urban-green boundary.focusing on air temperature reduction in urban areas around a green
area [4].
ThereCa etareal.a [5]
fewhave
studiescarried out field
focusing on airmeasurements to determine
temperature reduction the cooling
in urban influence
areas around of a
a green
park
areaon[4].the
Casurrounding
et al. [5] havearea in the
carried outTama
field New Town, a city
measurements in the west
to determine theofcooling
Tokyo.influence
With theof size of
a park
0.6
onkmthe2, surrounding
a park can reduce area in thetheairTama
temperature
New Town, by up to 1.5
a city °C at
in the noon
west time in With
of Tokyo. a leeward
the size of 0.6 km2 ,
commercial
area at distance
a park of 1 km.
can reduce Yu temperature
the air and Hien [6] by have to 1.5 ◦out
upcarried C attemperature
noon time and in a humidity measurements
leeward commercial area
inattwo big city
distance of 1green
km. Yu areas
and(36 Hienha [6]
andhave
12 ha) in Singapore.
carried A three-dimensional
out temperature non-hydrostaticin
and humidity measurements
model
two big (Envi-met)
city greenwas areasapplied
(36 ha andfor the simulation
12 ha) of Surface-Plant-Air
in Singapore. A three-dimensional interactions inside urban
non-hydrostatic model
environments. Horizontal air temperature profiles in both the green area
(Envi-met) was applied for the simulation of Surface-Plant-Air interactions inside urban environments.and surrounding area are
calculated
Horizontal byair
thetemperature
Envi-met model. profiles in both the green area and surrounding area are calculated by the
Yagi and
Envi-met model. Takebayashi [7] have performed measurements at four urban areas in Kobe City. The
spatialYagi
variation of the vertical[7]airhave
and Takebayashi temperature
performed gradient between 4.0
measurements m and
at four 1.5 mareas
urban is large in urban
in Kobe City.
areas, since variation
The spatial air temperature reduction
of the vertical effect in urban
air temperature gradient areas is different
between 4.0 m anddepending
1.5 m ison thein
large
circumstances
urban areas, aroundsince air thetemperature
measurement point. Since
reduction effect seainbreezes
urban dominate in summer
areas is different days in many
depending on the
cities in Japan, air
circumstances temperature
around reduction due
the measurement to Since
point. advection effects dominate
sea breezes is expected inin regionsdays
summer leeward
in manyof
urban green areas. In this study, the characteristics of air temperature
cities in Japan, air temperature reduction due to advection effects is expected in regions leeward in the urban area on the
leeward
of urban side of green
green areas.areas
In this arestudy,
considered using a numerical
the characteristics model incorporating
of air temperature in the urbanadvection
area on andthe
diffusion,
leeward and side verified
of greenbyareas comparison with measurement.
are considered using a numericalThe objective
model of this study isadvection
incorporating to clarify theand
characteristics
diffusion, andofverifiedair temperature
by comparison rise inwith
an urban area on the
measurement. Theleeward
objective sideof of
thisa study
green is area, as a
to clarify
contribution to the practical
the characteristics planning ofrise
of air temperature urban greening.
in an urban area on the leeward side of a green area, as a
contribution to the practical planning of urban greening.
2. Measurements
2. Measurements
2.1. Study Site
2.1. Study Site
Mobile measurements were carried out in Higashi-yuen Park (about 2.7 ha, green coverage rate,
which Mobile measurements
is the ratio of the canopywerearea
carried
to theoutpark
in Higashi-yuen Parkand
area: about 45%) (about 2.7 ha, greenbusiness
a neighboring coveragearea
rate,
at 13:00 and 17:00 on 2 August 2012, in Ishiyagawa Park (about 4 ha, green coverage ratio: aboutat
which is the ratio of the canopy area to the park area: about 45%) and a neighboring business area
13:00and
42%) anda 17:00 on 2 August
residential area at 2012,
13:00 in
andIshiyagawa
17:00 on 4 Park (about
August 2012,4 ha,
andgreen coverage ratio:
in Okurayama about 42%)
Park (about 7.9
and a residential area at 13:00 and 17:00 on 4 August 2012, and in Okurayama
ha, green coverage ratio: about 70%) and a residential area at 13:00 and 17:00 on 8 August 2012. Park (about 7.9 ha,
green coverage ratio: about 70%) and a residential area at 13:00 and 17:00 on 8
These parks are all located in Kobe city, Japan. Mobile measurement points and aerial photographs August 2012. These
parks
are are all
shown inlocated
Figurein1.Kobe
The city,
gridJapan.
lines Mobile measurement
are spaced pointsThe
50 m apart. andgreen
aerial color
photographs aregreen
indicates shown
in Figure
coverage. 1. The grid lines are spaced 50 m apart. The green color indicates green coverage.

(a) (b)

Figure 1. Cont.
Climate 2017, 5, 60 3 of 12
Climate 2017, 5, 60 3 of 12

N
◆ urban point ◆ green point ◆ fixed point

(c)
Figure 1.
Figure 1. Mobile
Mobilemeasurement
measurementpoints
pointsandandaerial photograph,
aerial photograph,all all
located in Kobe
located city,city,
in Kobe Japan. (a)
Japan.
Higashi-yuen Park and business area; (b) Ishiyagawa Park and residential area; (c) Okurayama
(a) Higashi-yuen Park and business area; (b) Ishiyagawa Park and residential area; (c) Okurayama Park
and residential
Park area.area.
and residential

Higashi-yuen Park and the business area are located in the center of Kobe City. There are
Higashi-yuen Park and the business area are located in the center of Kobe City. There are public
public buildings such as Kobe City Hall and general offices etc. in the business area.
buildings such as Kobe City Hall and general offices etc. in the business area. Middle-high-rise
Middle-high-rise buildings are dominant. Ishiyagawa Park and the neighboring residential area are
buildings are dominant. Ishiyagawa Park and the neighboring residential area are located in an urban
located in an urban area at the southern foot of Rokko Mountain on the east side of Kobe City.
area at the southern foot of Rokko Mountain on the east side of Kobe City. There are mainly detached
There are mainly detached houses and small scale collective houses in the residential area. Low-rise
houses and small scale collective houses in the residential area. Low-rise buildings are dominant.
buildings are dominant. Okurayama Park and its residential area are located in the urban area at
Okurayama Park and its residential area are located in the urban area at the southern foot of Rokko
the southern foot of Rokko Mountain on the west side of Kobe City. There are detached houses,
Mountain on the west side of Kobe City. There are detached houses, hospitals, etc. in the residential
hospitals, etc. in the residential area. Low-rise building and middle-rise building are mixed.
area. Low-rise building and middle-rise building are mixed.
2.2. Outline of Measurements
2.2. Outline of Measurements
The elements measured are air temperature, wind direction, wind velocity at a height of 1.5 m,
The elements measured are air temperature, wind direction, wind velocity at a height of 1.5 m,
and surface temperature. The measuring device and method are shown in Table 1. Wind velocity
and surface temperature. The measuring device and method are shown in Table 1. Wind velocity was
was sampled every second at each mobile measurement point which is indicated as an urban point
sampled every second at each mobile measurement point which is indicated as an urban point and
and green point in Figure 1 and the averaged value for 30 s was recorded. Wind direction was
green point in Figure 1 and the averaged value for 30 s was recorded. Wind direction was recorded
recorded based on the direction with the highest frequency in the 30 s. Measurement results for air
based on the direction with the highest frequency in the 30 s. Measurement results for air temperature
temperature at the fixed measurement points are shown in Figure 2. It was continuously measured
at the fixed measurement points are shown in Figure 2. It was continuously measured only at fixed
only at fixed points. Thermistor sensors were installed in a natural ventilation-type solar radiation
points. Thermistor sensors were installed in a natural ventilation-type solar radiation shielding device
shielding device and were set on the roof of the Kobe City Hall No. 3 building (47 m above the
and were set on the roof of the Kobe City Hall No. 3 building (47 m above the ground, flat concrete roof
ground, flat concrete roof with the usual waterproof sheet finish) and the trunks of trees in the
with the usual waterproof sheet finish) and the trunks of trees in the Higashi-yuen Park, Ishiyagawa
Higashi-yuen Park, Ishiyagawa Park, and Okurayama Park (3 m above the ground). Kobe City Hall
Park, and Okurayama Park (3 m above the ground). Kobe City Hall and Higashi-yuen Park are close
and Higashi-yuen Park are close to each other. The distances from Ishiyagawa Park and
to each other. The distances from Ishiyagawa Park and Okurayama Park to Kobe City Hall are about
Okurayama Park to Kobe City Hall are about 5.7 km and 2.1 km, respectively.
5.7 km and 2.1 km, respectively.
Although it took a maximum of 1.5 h for the mobile measurements at each site to be made, no
Although it took a maximum of 1.5 h for the mobile measurements at each site to be made,
sudden changes in weather were confirmed as compared with the results of the fixed-point
no sudden changes in weather were confirmed as compared with the results of the fixed-point
measurements, so no correction was made to the results of the mobile measurements. For the
measurements, so no correction was made to the results of the mobile measurements. For the analysis
analysis in the next section, I used the difference between air temperature from the mobile
in the next section, I used the difference between air temperature from the mobile measurements in the
measurements in the urban area and the air temperature of the fixed-point measurements in the
urban area and the air temperature of the fixed-point measurements in the park at that time.
park at that time.
Climate 2017, 5,
Climate 2017, 5, 60
60 44 of
of 12
12

Table 1. Measuring device and method.

Table 1. Measuring device and method.
Device Method Accuracy of Device
Air Thermistor withDevice
solar Method Accuracy of Device
Averaged for 5 min by sampling every 5 s ±0.5 K
temperature
Air temperature radiation shield
Thermistor with solar
Averaged for 5 min by sampling every 5 s ±0.5 K
radiation shield
Wind direction Windsock Highest frequency in 30 s by visual inspection
Wind direction Windsock Highest frequency in 30 s
Hot-wire Averaged for 30 s by sampling everyby visual inspection
±2% of indicated
Wind velocity Averaged for 30 s by sampling every
Wind velocity anemometer
Hot-wire anemometer second value
±2% of indicated value
second
Measured on ground and wall surface, a
Measured on ground and wall surface,
SurfaceSurface representative material surface at each
±1.0 K ±1.0 K
InfraredInfrared
thermometer a representative material surface at each
thermometer
temperature
temperature measurement
measurementpoint wasmeasured
point was measured several
several
timestimes to obtain
to obtain stabledata
stable data

38

36
Air temperature (oC)

34

32

30

28

26
measurement period
24
0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00
City office Aug. 2 Higashi-yuen park Aug. 2
City office Aug. 4 Ishiyagawa park Aug. 4
City office Aug. 8 Okurayama park Aug. 8

Figure 2. Measurement results of air temperature

temperature at
at the
the fixed
fixed measurement
measurement points.
points.

The boundary between the green area and the urban area is set to 0, and the following analysis
The boundary between the green area and the urban area is set to 0, and the following analysis
is carried out, focusing on the relationship between the horizontal distance from the boundary and
is carried out, focusing on the relationship between the horizontal distance from the boundary and
the air temperature in the urban areas. Since the main wind direction was southwest in the case of
the air temperature in the urban areas. Since the main wind direction was southwest in the case of
measurements around Higashi-yuen Park, the measurement results in the northeastern urban area
measurements around Higashi-yuen Park, the measurement results in the northeastern urban area
were used for analysis. The distance to the park was calculated by drawing a straight line in the
were used for analysis. The distance to the park was calculated by drawing a straight line in the
southwest direction from each mobile measurement point. Similarly, since the main wind direction
southwest direction from each mobile measurement point. Similarly, since the main wind direction
was east in the case of measurements around Ishiyagawa Park, the measurement results in the west
was east in the case of measurements around Ishiyagawa Park, the measurement results in the west
urban area were used for analysis. In the case of measurements around Okurayama Park, the main
urban area were used for analysis. In the case of measurements around Okurayama Park, the main
wind direction was south-southeast, so the measurement results in the northern urban area were
wind direction was south-southeast, so the measurement results in the northern urban area were used
used for analysis. Figure 3 shows the distance from the green area to each mobile measurement
for analysis. Figure 3 shows the distance from the green area to each mobile measurement point in
point in the urban area and the air temperature rise. This is the difference to the air temperature
the urban area and the air temperature rise. This is the difference to the air temperature measured in
measured in the windward side green area. Air temperature rise is large in a weak wind case.
the windward side green area. Air temperature rise is large in a weak wind case. Strong wind and
Strong wind and weak wind were classified by the upper wind velocity of 5.5 m/s measured at the
weak wind were classified by the upper wind velocity of 5.5 m/s measured at the Kobe meteorological
Kobe meteorological observatory. A measurement point where the distance from the green area is
observatory. A measurement point where the distance from the green area is about 30 m or more was
about 30 m or more was considered representative of the urban area’s air temperature, without
considered representative of the urban area’s air temperature, without being affected by the green area.
being affected by the green area. It is considered that air temperature in the urban area in a weak
It is considered that air temperature in the urban area in a weak wind case is fluctuating due to the
wind case is fluctuating due to the influence of local ventilation and solar radiation shielding. The
influence of local ventilation and solar radiation shielding. The wind velocity was measured by mobile
wind velocity was measured by mobile measurement at a height of 1.5 m above the ground. In the
measurement at a height of 1.5 m above the ground. In the urban area, this was 1.0 to 1.3 m/s at 13:00
urban area, this was 1.0 to 1.3 m/s at 13:00 and 0.7 to 1.1 m/s at 17:00 in a strong wind case and 0.5 to
and 0.7 to 1.1 m/s at 17:00 in a strong wind case and 0.5 to 1.0 m/s at 13:00 and 0.4 to 1.1 m/s at 17:00
1.0 m/s at 13:00 and 0.4 to 1.1 m/s at 17:00 in a weak wind case. The wind velocity in the urban area
in a weak wind case. The wind velocity in the urban area fluctuated because of the influence of the
fluctuated because of the influence of the surrounding buildings.
surrounding buildings.
Climate 2017, 5, 60 5 of 12
Climate 2017, 5, 60 5 of 12

(a) (b)
Figure
Figure 3.
3. Distance
Distance from
from the
the green
green area
area to
to each
each mobile
mobile measurement
measurement point
point in
in urban
urban area
area and
and air
air
temperature
temperature rise
rise (a)
(a) at
at 13:00;
13:00; (b)
(b) at
at 17:00.
17:00.

3. Results
3. Results

3.1.
3.1. Outline
Outline of
of Calculations
Calculations
Calculations
Calculations were werecarried
carriedout outby by Computational
Computational FluidFluid Dynamics
Dynamics (CFD). (CFD).
For theFor the turbulence
turbulence closure
closure
model, amodel,
standard a standard
k-ε modelk-ε was model
used. wasThisused.
is the This
most is the most
common common
model used inmodel
CFD to used in CFD
simulate meanto
simulate mean flow characteristics for turbulent flow. The outline
flow characteristics for turbulent flow. The outline of the calculation model is shown below. of the calculation model is shown
below.
∂ui
=0 (1)
∂xi 0 (1)
( !)
∂ui ∂ui u j ∂π ∂ ∂ui ∂u j g
+ =− + (νt + ν) + + i (Θ − Θ0 ) − 2 3i,j,k Ω j uk (2)
∂t ∂x j ∂xi ∂x j ∂x j ∂xi ΘΘ 0 Θ 2∋,, Ω (2)
( !) Θ !

∂k ∂kui ∂ νt ∂k ∂ui ∂u j ∂ui νt ∂Θ
+ = +ν + νt + − ε − gβ (3)
∂t ∂x j ∂x j σ1 ∂x j ∂x j ∂xi ∂x j PrtΘ∂xk (3)
(  !) !
∂ε ∂εui ∂ νt ∂ε ε ∂ui ∂u j ∂ui kε
+ = +ν + C1 νt + − C2 (4)
∂t ∂x j ∂x j σ2 ∂x j k ∂x j ∂xi ∂x j νt
(4)
(  !)
∂θ ∂
∂ νt ∂θ Q
+ θui = +ν + (5)
∂t ̅ ∂x j ∂x j Prt ∂x̅j Cp ρ
̅ (5)
For the application of the turbulence model to urban space, Ashie and Ca [8] have proposed a
modelForthat
theexpresses
application theofeddy viscosity coefficient
the turbulence model toνurbant as a function
space, Ashie of theandfluxCa
Richardson number Raf
[8] have proposed
(Equations
model that (8)–(11)).
expressesThey do this
the eddy by aggregating
viscosity coefficientthe νt buoyancy
as a function effect intoflux
of the the Richardson
vertical eddy viscosity
number Rf
model coefficient C and the turbulent Prandtl number P rt . In this
(Equations (8)–(11)). They do this by aggregating the buoyancy effect into the vertical eddy
µ study, we used both Equation (7)
and the conventional
viscosity model coefficient Equation Cµ(6).
andEquations (8)–(11)
the turbulent are used
Prandtl in calculating
number the vertical
Prt. In this study, weeddy viscosity
used both
coefficient(7)νtand
Equation on the
the right side of Equation
conventional Equation(7). (6).Calculation
Equations (8)–(11) conditions and the
are used outline of the
in calculating calculation
vertical
conditions
eddy are shown
viscosity coefficientin Table
νt on2the
andright
Figure 4. of Equation (7). Calculation conditions and the outline
side
(Conventional
of calculation isotropic
conditions diffusion
are shown in model)
Table 2 and Figure 4.
(Conventional isotropic diffusion
k2 model)
vt = CD (horizontal and vertical diffusion) (6)
ε
ν horizontal and vertical diffusion (6)
(Model incorporating buoyancy effect)

(Model incorporating kbuoyancy

2 effect) k2
vt = CD (horizontal diffusion), vt = Cµ (vertical diffusion) (7)
ε ε
ν horizontal diffusion , ν vertical diffusion (7)
Climate 2017, 5, 60 6 of 12

Climate 2017, 5, 60 viscosity

: eddy constant (0.09), : eddy viscosity coefficient 6 of 12

0.8∅ 0.5 ∅ 0.53 0.94

CD : eddy viscosity constant (0.09), Cµ : eddy viscosity coefficient (8)
0.8 ∅ ∅ 1
0.8∅γ − 0.5γ(∅θ − γ) 0.53 − 0.94R f
Cµ = 1.59 1.5∅ 2.82 (8)
0.8γ + ∅(∅θ − γ) 1 − Rf (9)
1.59 3∅ 5.22
1.59 − R f (1.5∅θ + 2.82)
1.59
Prt = Prt0 5.22 (9)
∅ 1.59 + R, ∅f (3∅
0.2,
−∅ 5.22) 1⁄3.2 (10)
1.59 2.82
1.59 − 5.22R f
γ = ∅θ , ∅ = 0.2, ∅θ = 1/3.2 (10)
1.59 − 2.82R f ⁄
0.195 0.6588 h 0.1776 0.3221 0.03156 (11)

1/2 i
βg ∂Θ R g ≤ 0.195 R f = 0.6588 R0.195 R g 2 − 0.3221R g + 0.03156
g + 0.1776 −0.191
Rg = h 2
∂z
2
i (11)
( ∂U
∂z ) + ( ∂z )
∂V
R g ≥ 0.195 R f = 0.191
The composition
The composition of of the
the model
model was was set
set according
according to to Moriyama
Moriyama et et al.
al. [3],
[3], and
and the
the calculation
calculation
conditions were set based on weather conditions at the time of measurement.
conditions were set based on weather conditions at the time of measurement. The mesh size The mesh size in
in the
the
horizontal direction was set to 50 m in correspondence with the selection
horizontal direction was set to 50 m in correspondence with the selection policies of the measurement policies of the
measurement
points. points. The
The calculation calculation
condition as showncondition
in Figureas4 shown
expressesin the
Figure 4 expresses
phenomenon the phenomenon
flowing out from the
flowing out from the green area to the urban area in three dimensions. The vertical
green area to the urban area in three dimensions. The vertical air temperature profile in the air temperature
green area
profile
was in the green
uniformly givenarea
forwas uniformly
the inflow given for
condition. theupper
The inflowwind
condition. The
velocity at upper
50 m highwindwas
velocity at 50
relatively
m high was relatively large, as it was measured under conditions
large, as it was measured under conditions where a sea breeze was dominant. where a sea breeze was dominant.

Table2.
Table 2. Calculation
Calculation conditions.
conditions.

13:00 17:00
13:00
17:00
Inflow air temperature with uniform vertical profile 33 °C 31 °C
Inflow air temperature with uniform vertical profile ◦
33 C5.6 m/s Large:31 ◦
Large: 4.7Cm/s
Inflow wind velocity at 50 m high with logarithmic vertical profile Large: 5.6 m/s Large: 4.7 m/s
Inflow wind velocity at 50 m high with logarithmic vertical profile Small: 4.1 m/s Small: 4.2 m/s
Small: 4.1 m/s2 Small: 4.2 2m/s
Sensible heat from ground surface 314 W/m 196 W/m 2
Sensible heat from ground surface 314 W/m2 196 W/m
Roughness parameter
Roughness parameter 0.5
0.5 m m
Horizontal
Horizontal meshmesh
size size 5050
mm
Vertical
Vertical meshmesh
size size 3m 3m

500m
500m

Green area Urban area 500m Green area Urban area 50m

Calculation area
x: 500m, y: 500m x: 500m, z: 50m
Horizontal: 50m mesh Vertical: 3m mesh

Figure 4. Outline
Figure 4. Outline of
of calculation
calculation conditions.
conditions.

3.2. Results
3.2.
Calculation and
Calculation andmeasurement
measurementresults resultsforfor
airair temperature
temperature riserise
withwith distance
distance fromfrom the green
the green area
area are shown in Figures 5 and 6. The results in both isotropic and non-isotropic
are shown in Figures 5 and 6. The results in both isotropic and non-isotropic diffusion models are diffusion models
are shown.
shown. In the
In the isotropic
isotropic diffusion
diffusion model,
model, thethe horizontal
horizontal andvertical
and verticaleddy
eddyviscosity
viscositycoefficients
coefficientsννtt
are given
are given by by Equation
Equation (6).
(6). In
In the
the non-isotropic
non-isotropic diffusion
diffusion model,
model, thethe eddy
eddy viscosity
viscosity coefficient
coefficient ννtt in
in
the vertical
the vertical direction
direction is given
given byby the
the formula
formula of Equation
Equation (7) when considering the buoyancy effect.
Distance from the green area and the heat flux component of the calculation
Distance calculation result, at 13:00
13:00 in
in the
the
mesh near the ground surface, is shown in Figure 7. In the incorporated buoyancy
mesh near the ground surface, is shown in Figure 7. In the incorporated buoyancy model, the sensible model, the
sensible
heat flux heat flux from
supplied supplied from the
the ground ground
surface surface is transported
is transported in the
in the vertical vertical
direction direction
due due to
to the vertical
the vertical
diffusion diffusion
effect, effect, so air in
so air temperature temperature
the mesh near in the
the mesh
groundnear the ground
surface does not surface does
rise. The notofrise.
part the
 velocity is small, the calculation result in the isotropic diffusion model, in which the diffusion effect
in the vertical
velocity direction
is small, is not prominent,
the calculation result in is
the close to thediffusion
isotropic measurement model, result of thethe
in which air temperature.
in the vertical direction is not prominent, is close to the measurement result of the airdiffusion effect
temperature.
in theTherefore,
vertical the calculation
direction is not result by is
prominent, theclose
previous
to the study, using the
measurement isotropic
result
Therefore, the calculation result by the previous study, using the isotropic diffusion model, of the diffusion
air model,
temperature.
may Therefore,
be matched thewith the measurement
calculation result by the result when study,
previous the inflow
usingwind velocity diffusion
the isotropic is small. This is
may be matched with the measurement result when the inflow wind velocity is small. model,This is
shown
may in the
be inmatchedright-hand
with the panels of Figures
measurement 5 and 6. Even if the distance from the green area isThis
150 m
shown the right-hand panels of Figures result
5 and 6.when
Eventhe inflow
if the wind
distance velocity
from is small.
the green area is 150 is
m
or more,
shown air
2017,in5,the
temperature
right-hand panels rises more and
of Figures its effect
5 and may
6. Even thus extend
if theextend
distanceto over 200 m. Since the inflow
Climate
or more, 60 temperature
air rises more and its effect may thus to from the green
over 200 m. Sinceareathe
is 150 7m
of 12
inflow
wind
or velocity
more, air is small, therises
temperature vertical
more diffusion
and its effect may
effect is also
thus small,
extendandtothe air 200
over temperature
m. Since rise inflow
the in the
wind velocity is small, the vertical diffusion effect is also small, and the air temperature rise in the
urban
wind area is thus larger than whendiffusion
the inflow wind velocity is large. Atairthis time air temperature
urbanvelocity
area is is small,
thus the
larger vertical
than when the inflow effect
windis also small,
velocity isand
large.the temperature
At this rise in the
time air temperature
in
urban the
urban urban
area more area varies
than 50 m considerably
from the because
green area of
is the influence
dominated of
by local
the ventilation,
diffusion solar
effect in radiation
the vertical
in thearea
urban is thus
area larger
varies than when thebecause
considerably inflow windof thevelocity
influence is large.
of local Atventilation,
this time airsolartemperature
radiation
shielding,
direction
in over
the urban etc.
theThis can be
advection seen from
effect. the measurement results in the right-hand panel of Figure 5.
shielding, etc.area
Thisvaries
can beconsiderably
seen from the because of the influence
measurement results in of thelocal ventilation,
right-hand panelsolar radiation
of Figure 5.
shielding, etc. This can be seen from the measurement results in the right-hand panel of Figure 5.

(a) (b)
(a) (b)
Figure 5. Calculation(a) (b)at 13:00 according to
results and measurement results of air temperature rise
Figure 5. Calculation results and measurement results of air temperature rise at 13:00 according to the
Figure 5. Calculation results and measurement results of air temperature rise at 13:00 according to
the distance
Figure from the green
5. Calculation area.
results and(a) strong windresults
measurement case; (b)
ofweak wind case. rise at 13:00 according to
airwind
temperature
distance from
the distancethe green
from area. (a)
the green strong
area. windwind
(a) strong case;case;
(b) weak
(b) weak case.case.
wind
the distance from the green area. (a) strong wind case; (b) weak wind case.

(a) (b)
(a) (b)
Figure 6. Calculation(a) (b)at 17:00 according to
results and measurement results of air temperature rise
Figure 6. Calculation results and measurement results of air temperature rise at 17:00 according to
Figure
the Calculation
6. 6.
distance
Figure from theresults
Calculation resultsand
green andmeasurement
area. (a) results
strong wind
measurement ofofair
case; (b)
results airtemperature
weak wind case.rise
temperature riseatat17:00
17:00according
accordingtotothe
the distance from the green area. (a) strong wind case; (b) weak wind case.
distance from the
the distance fromgreen area. area.
the green (a) strong windwind
(a) strong case;case;
(b) weak
(b) weakwind case.
wind case.

(a) (b)
(a) (b)
(b)
(a) the green area and heat flux component of the calculation
Figure 7. Distance from result at 13:00. (a)
Figure 7. Distance from the green area and heat flux component of the calculation result at 13:00. (a)
isotropic
Figure diffusion model; (b) incorporating buoyancy model.
Figure 7. 7.Distance
isotropic
Distance from
from the green
the(b)
diffusion model; green area
areaand heat
and
incorporating
flux
heat component
flux
buoyancy
of the
component
model. of calculation resultresult
the calculation at 13:00. (a)
at 13:00.
isotropic diffusion model; (b) incorporating buoyancy model.
(a) isotropic diffusion model; (b) incorporating buoyancy model.

When the inflow wind velocity is large, the calculation result in the incorporated buoyancy model
tends to coincide with the measurement result of the air temperature. When the inflow wind velocity
is small, the calculation result in the isotropic diffusion model, in which the diffusion effect in the
vertical direction is not prominent, is close to the measurement result of the air temperature.
Therefore, the calculation result by the previous study, using the isotropic diffusion model, may be
matched with the measurement result when the inflow wind velocity is small. This is shown in the
right-hand panels of Figures 5 and 6. Even if the distance from the green area is 150 m or more,
air temperature rises more and its effect may thus extend to over 200 m. Since the inflow wind velocity
is small, the vertical diffusion effect is also small, and the air temperature rise in the urban area is thus
larger than when the inflow wind velocity is large. At this time air temperature in the urban area
Climate 2017, 5, 60 8 of 12

Climate 2017, 5, 60 8 of 12
Climate 2017, 5, 60 8 of 12
varies considerably because of the influence of local ventilation, solar radiation shielding, etc. This can
4. Discussion
be seen from the measurement results in the right-hand panel of Figure 5.
4. Discussion
In addition to the measurement results in Kobe City, the calculation results in Figures 8 and 9
In addition to the measurement results in Kobe City, the calculation results in Figures 8 and 9
4. Discussion
were also compared to the measurement results in the urban area around Koishikawa park in
were also compared to the measurement results in the urban area around Koishikawa park in
TokyoIn by
addition
Kato etto al.
the[9],
measurement
and severalresults
parks ininKobe
Osaka City,
citythe
bycalculation
Moriyama results in Figures
et al. [10]. In the 8results
and 9
Tokyo by Kato et al. [9], and several parks in Osaka city by Moriyama et al. [10]. In the results
were also compared to the measurement results in the urban area around Koishikawa
measured in Tokyo and Osaka, air temperature does not rise as it enters the part of the urban area park in Tokyo
measured in Tokyo and Osaka, air temperature does not rise as it enters the part of the urban area
by Kato
more et al.
than 50 [9], and several
m from the greenparks
area.inOn
Osaka city by
the other Moriyama
hand, Honjo et al. Takakura
and [10]. In the[2]results measured
explained in
that the
more than 50 m from the green area. On the other hand, Honjo and Takakura [2] explained that the
Tokyo
range ofand Osaka,
the effectsairoftemperature
urban greendoes
areasnot rise as to
extends it enters
about the100 part of the
to 300 urban
m into thearea more thanurban
surrounding 50 m
range of the effects of urban green areas extends to about 100 to 300 m into the surrounding urban
from
area. the green
Since theyarea.
usedOn thethe other hand,
isotropic Honjomodel,
diffusion and Takakura [2] explained
it is recognized that itthat
wasthe range ofonly
a finding the effects
in the
area. Since they used the isotropic diffusion model, it is recognized that it was a finding only in the
of urban
case green
of weak areas extends to about 100 to 300 m into the surrounding urban area. Since they used
wind.
case of weak wind.
the isotropic diffusion model, it is recognized that it was a finding only in the case of weak wind.

Figure 8. Distance from the green area and the air temperature rise in several urban areas in the daytime.
Figure 8.
Figure Distancefrom
8. Distance fromthe
thegreen
greenarea
areaand
andthe
theair
airtemperature
temperaturerise
risein
inseveral
severalurban
urbanareas
areasininthe
thedaytime.
daytime.

Figure 9.
Figure Distance from
9. Distance from the
the green
green area
area and
and air
air temperature
temperature rise
rise in
in several
several urban
urban areas
areas in
in the
the evening.
evening.
Figure 9. Distance from the green area and air temperature rise in several urban areas in the evening.

detail, a recalculation
In order to discuss this in more detail, recalculation was carried out, improving the spatial
In order to discuss this in more detail, a recalculation was carried out, improving the spatial
resolution in the urban area near the green area. An outline of the modified calculation conditions is
resolution in the urban area near the green area. An outline of the modified calculation conditions is
shown ininFigure
Figure10.10.
TheThe horizontal
horizontal mesh
mesh size changed
size was was changed
to 5 m to 5 m
from 50 from 50the
m as in mabove
as in calculation.
the above
shown in Figure 10. The horizontal mesh size was changed to 5 m from 50 m as in the above
calculation.
The The otherconditions
other calculation calculationwere
conditions were not changed.
not changed.
calculation. The other calculation conditions were not changed.
Calculation results of air temperature rise according to the distance from the green area are
Calculation results of air temperature rise according to the distance from the green area are
shown in Figure 11. Sensible heat flux from the ground surface in the urban area was assumed to be
shown in Figure 11. Sensible heat flux from the ground surface in the urban area was assumed to be
236.7 W/m22 for daytime and 28.3 W/m22 for evening. A value of 132.5 W/m22 was also assumed for
236.7 W/m for daytime and 28.3 W/m for evening. A value of 132.5 W/m was also assumed for
their intermediate value. When entering the urban area air temperature rises sharply. The smaller
their intermediate value. When entering the urban area air temperature rises sharply. The smaller
the wind velocity, the larger the distance influenced by the green area, and the larger the air
the wind velocity, the larger the distance influenced by the green area, and the larger the air
Climate 2017, 5, 60 9 of 12

Climate 2017, 5, 60 9 of 12
236.7 W/m2 for daytime and 28.3 W/m2 for evening. A value of 132.5 W/m2was also assumed for
temperature
their rise. As
intermediate the When
value. distance from the
entering the green
urban area increases,
area air air temperature
temperature becomes
rises sharply. constant.
The smaller the
When entering the part of the urban area more than 50 m from the green area, the
wind velocity, the larger the distance influenced by the green area, and the larger the air temperatureair temperature
nearAs
rise. thethe
ground surface
distance fromisthedominated
green areabyincreases,
the diffusion effect in thebecomes
air temperature vertical constant.
direction When
rather entering
than the
advection effect from the green area.
the part of the urban area more than 50 m from the green area, the air temperature near the ground
In is
surface general, air temperature
dominated rise ∆T
by the diffusion (K) in
effect duethetovertical
the development of thethan
direction rather urbantheboundary
advectionlayer is
effect
expressed by Equation
from the green area. (12).
In general, air temperature rise ∆T (K) due to the development of the urban boundary layer is
expressed by Equation (12). ΔT = s
(
2 1 + k HLα )
(12)
C p ρ U
2(1 + k) HLα
∆T = (12)
where k is the ratio of entrainment (0 to 1), H is Cthe p ρUsensible heat flux from the ground surface
(W/m2k),isLthe
where is the
ratiodistance from the
of entrainment boundary
(0 to 1), H is the(m), α is air
sensible heattemperature
flux from thegradient
ground (K/m),
surface C p is the
(W/m 2 ),
specific heat of air (=1000 J/(kgK)), ρ is air density (=1.2 kg/m 3), and U is wind velocity (m/s).
L is the distance from the boundary (m), α is air temperature gradient (K/m), Cp is the specific heat of
Assuming
air α = 0.006 ρ(K/m),
(=1000 J/(kgK)), it becomes
is air density Equation
(=1.2 kg/m3 ),(13).
and U is wind velocity (m/s). Assuming α = 0.006
(K/m), it becomes Equation (13).
(
ΔT = 0.0032 ~ 0.0045 H U L ) (13)
√ √
Air temperature rise ∆T, by∆T Equation (12),∼when
= ( 0.0032 0.0045k)= 0H/U
is shownL in Figure 12 together with(13) the
calculation results. Equation (12) is calculated using the boundary layer thickness
Air temperature rise ∆T, by Equation (12), when k = 0 is shown in Figure 12 together with the h = ∆T/α. Actually,
when the development
calculation of the
results. Equation (12)boundary layerusing
is calculated is not sufficient
the boundary and h isthickness
layer small, α hshould
= ∆T/α.be Actually,
set to be
large. Then, air temperature rise ∆T approximated by Equation (14) is shown
when the development of the boundary layer is not sufficient and h is small, α should be set to be in Figure 13.large.
The
coefficient
Then, a at this time
air temperature is ∆T
rise shown in Table 3. by
approximated It isEquation
larger than
(14)the 0.0032 in
is shown used in Equation
Figure (12).
13. The coefficient a
at this time is shown in Table 3. It is larger than the 0.0032 used in Equation (12).
ΔT = a H U L (14)
√ √
∆T temperature
As described above, the calculated air = a H/U nearL (14)it
the ground surface rises sharply as
enters the urban area. This is because of the sensible heat flux from the ground surface, and when
As described above, the calculated air temperature near the ground surface rises sharply as it
entering the area beyond about 50 m, it becomes almost constant. On the other hand, the
enters the urban area. This is because of the sensible heat flux from the ground surface, and when
approximate value of the air temperature due to the development of the boundary layer
entering the area beyond about 50 m, it becomes almost constant. On the other hand, the approximate
monotonically rises with the distance from the green area. Therefore, we considered an
value of the air temperature due to the development of the boundary layer monotonically rises with
approximation based on the following equation where air temperature rise becomes constant as the
the distance from the green area. Therefore, we considered an approximation based on the following
distance goes above a certain value. Air temperature rise ∆T by Equation (15) is shown in Figure 14
equation where air temperature rise becomes constant as the distance goes above a certain value.
together with the calculation results. When entering the urban area, air temperature rises sharply,
Air temperature rise ∆T by Equation (15) is shown in Figure 14 together with the calculation results.
and when entering the area beyond a certain distance it becomes almost constant.
When entering the urban area, air temperature rises sharply, and when entering the area beyond a
a H U L L < b
certain distance it becomes almost constant.
ΔT = (  √ √
a a HH/U
U L LL<>bb (15)
∆T = √
≈ a ′√ H Up L (L + b )
a H/U L>b (15)
≈ a0 H/U L/( L + b)

100m
100m

Green area Urban area 100m Green area Urban area 50m

Calculation area
x: 100m, y: 100m x: 100m, z: 50m
Horizontal: 5m mesh Vertical: 3m mesh

Figure 10. Outline

Figure 10. Outline of
of modified
modified calculation
calculation conditions.
conditions.
Climate 2017, 5, 60 10 of 12
Climate 2017, 5, 60 10 of 12
Climate 2017, 5, 60 10 of 12
1.4 1.4
1.4 1.4
1.2 1.2
(K) (K) 1.2 1.2

(K) (K)
1 1
rise rise

rise rise
1 1
0.8 2m/s 0.8 2m/s
Air temperature

Air temperature
0.8 3m/s
2m/s 0.8 3m/s
2m/s
Air temperature

Air temperature
0.6 4m/s
3m/s 0.6 4m/s
3m/s
0.6 5m/s
4m/s 0.6 5m/s
4m/s
0.4 6m/s
5m/s 0.4 6m/s
5m/s
0.4 6m/s 0.4 6m/s
0.2 0.2
0.2 0.2
0 0
0 10 20 30 40 50 0 10 20 30 40 50
0 0
0 10 Distance from
20 the green30 park (m) 40 50 0 10 Distance from
20 the green30 park (m) 40 50
Distance from the green park (m) Distance from the green park (m)
(a) (b)
(a) 1.4 (b)
1.4
1.2
1.2
(K) (K)

1
rise rise

1
0.8 2m/s
Air temperature

0.8 3m/s
2m/s
Air temperature

0.6 4m/s
3m/s
0.6 5m/s
4m/s
0.4 6m/s
5m/s
0.4 6m/s
0.2
0.2
0
0 10 20 30 40 50
0
0 10 Distance from
20 the green30 park (m) 40 50
Distance from the green park (m)
(c)
(c)
Figure
Figure 11.
11. Calculation
Calculationresults
resultsof
ofair
air temperature
temperaturerise
riseaccording
accordingto tothe
thedistance
distancefrom fromthe
thegreen
green area.
area.
Figure
In the 11. Calculation
cases where the results of
sensible airflux
heat temperature
is (a) rise
236.7 W/maccording to the
2; (b) 132.5 W/m distance
2; (c) 28.3from the
W/m 2. green area.
2
In the cases where the sensible heat flux is (a) 236.7 W/m2 ; (b) 132.5 W/m 2 ; (c) 28.3 W/m 2 .
In the cases where the sensible heat flux is (a) 236.7 W/m ; (b) 132.5 W/m2; (c) 28.3 W/m2.
1.4 1.4
1.4 1.4
1.2 1.2
1.2 2m/s 1.2 2m/s
(K) (K)

(K) (K)

1 1
3m/s
2m/s 3m/s
2m/s
rise rise

rise rise

1 4m/s 1 4m/s
0.8 3m/s 0.8 3m/s
Air temperature

Air temperature

5m/s
4m/s 5m/s
4m/s
0.8 0.8
Air temperature

Air temperature

6m/s
5m/s 6m/s
5m/s
0.6 0.6
2m/s
6m/s in Eq. (12) 2m/s
6m/s in Eq. (12)
0.6 3m/s 0.6
0.4 2m/s in
in Eq.
Eq. (12)
(12) 0.4
3m/s
2m/s in
in Eq.
Eq. (12)
(12)
4m/s
3m/s in
in Eq.
Eq. (12)
(12) 4m/s
3m/s in
in Eq.
Eq. (12)
(12)
0.4 5m/s 0.4
0.2 4m/s in
in Eq.
Eq. (12)
(12) 0.2
5m/s
4m/s in
in Eq.
Eq. (12)
(12)
6m/s in Eq. (12)
5m/s in Eq. (12) 6m/s in Eq. (12)
5m/s in Eq. (12)
0.2 0.2
0 6m/s in Eq. (12) 0 6m/s in Eq. (12)
0 10 20 30 40 50 0 10 20 30 40 50
0 0
0 Distance
10 from
20 the green
30 park (m)
40 50 0 Distance
10 from
20 the green
30 park (m)
40 50
Distance from the green park (m) Distance from the green park (m)
(a) (b)
(a) 1.4 (b)
1.4
1.2
1.2 2m/s
(K) (K)

1
3m/s
2m/s
rise rise

1 4m/s
0.8 3m/s
Air temperature

5m/s
4m/s
0.8
Air temperature

6m/s
5m/s
0.6
2m/s
6m/s in Eq. (12)
0.6 3m/s
0.4 2m/s in
in Eq.
Eq. (12)
(12)
4m/s
3m/s in
in Eq.
Eq. (12)
(12)
0.4 5m/s in Eq. (12)
0.2 4m/s in Eq. (12)
6m/s
5m/s in
in Eq.
Eq. (12)
(12)
0.2
0 6m/s in Eq. (12)
0 10 20 30 40 50
0
0 Distance
10 from
20 the green
30 park (m)
40 50
Distance from the green park (m)
(c)
(c)
Figure 12. Air temperature rise by Equation (12). In the cases where the sensible heat flux is (a) 236.7
Figure
W/m 2; (b)
Figure 12. Air
Airtemperature
12.132.5 temperature
W/m riseW/m
2; (c) 28.3 by by
rise Equation (12).(12).
2. Equation In theIncases
the where the sensible
cases where heat fluxheat
the sensible is (a)flux
236.7
is
W/m 2 ; (b) 132.5
2 W/m2 ; (c) 28.3 W/m
2 2
(a) 236.7 W/m ; (b) 132.5 W/m ; (c) 28.3 W/m .. 2
Climate 2017, 5, 60 11 of 12
Climate 2017, 5, 60 11 of 12
Climate 2017, 5, 60 11 of 12
1.4 1.4

1.4
1.2 1.4
1.2

2m/s 2m/s

rise (K)

rise (K)
1.2
1 1.2
1
3m/s 3m/s
2m/s
4m/s 2m/s
4m/s
rise (K)

rise (K)
1
0.8 1
0.8
Air temperature

Air temperature
3m/s
5m/s 3m/s
5m/s
4m/s
6m/s 4m/s
6m/s
0.8
0.6 0.8
0.6
Air temperature

Air temperature
5m/s
2m/s in Eq. (14) 5m/s
2m/s in Eq. (14)
6m/s
3m/s in Eq. (14) 6m/s
3m/s in Eq. (14)
0.6
0.4 0.6
0.4
2m/s
4m/s in
in Eq.
Eq. (14)
(14) 2m/s
4m/s in
in Eq.
Eq. (14)
(14)
3m/s
5m/s in
in Eq.
Eq. (14)
(14) 3m/s
5m/s in
in Eq.
Eq. (14)
(14)
0.4
0.2 0.4
0.2
4m/s
6m/s in
in Eq.
Eq. (14)
(14) 4m/s
6m/s in
in Eq.
Eq. (14)
(14)
5m/s in Eq. (14) 5m/s in Eq. (14)
0.2
0 0.2
0
6m/s in Eq. (14) 6m/s in Eq. (14)
0 10 20 30 40 50 0 10 20 30 40 50
0 Distance from the green park (m) 0 Distance from the green park (m)
0 10 20 30 40 50 0 10 20 30 40 50
(a)
Distance from the green park (m) (b)
Distance from the green park (m)

1.4
(a) (b)
1.4
1.2

2m/s
rise (K)

1.2
1
3m/s
2m/s
4m/s
rise (K)

1
0.8
Air temperature

3m/s
5m/s
4m/s
6m/s
0.8
0.6
Air temperature

5m/s
2m/s in Eq. (14)
6m/s
3m/s in Eq. (14)
0.6
0.4 2m/s
4m/s in
in Eq.
Eq. (14)
(14)
3m/s
5m/s in
in Eq.
Eq. (14)
(14)
0.4
0.2
4m/s
6m/s in
in Eq.
Eq. (14)
(14)
5m/s in Eq. (14)
0.2
0
6m/s in Eq. (14)
0 10 20 30 40 50
0 Distance from the green park (m)
0 10 20 30 40 50
(c) Distance from the green park (m)

(c)
Figure 13. Air temperature rise by Equation (14). In the cases where the sensible heat flux is (a) 236.7
Figure 13.2 Air temperature rise by2 Equation (14). In the cases where the sensible heat flux is
Figure 13. 2132.5
W/m ; (b) Air temperature rise
W/m2; (c) 28.3 by Equation
2 ;W/m . (14). In the cases where the sensible heat flux is (a) 236.7
(a) 236.7 W/m ; (b) 132.5 W/m (c) 28.3 W/m2 .
W/m ; (b) 132.5 W/m ; (c) 28.3 W/m .
2 2 2

1.4 1.4

1.4
1.2 1.4
1.2
rise (K)

rise (K)

1.2
1 1.2
1
rise (K)
Air temperature

rise (K)

1 2m/s 1 2m/s
Air temperature

0.8 0.8
Air temperature

3m/s 3m/s
2m/s 2m/s
Air temperature

0.8
0.6 4m/s 0.8
0.6 4m/s
3m/s
5m/s 3m/s
5m/s
0.6
0.4 4m/s
6m/s 0.6
0.4 4m/s
6m/s
5m/s 5m/s
0.4
0.2 6m/s 0.4
0.2 6m/s

0.2
0 0.2
0
0 10 20 30 40 50 0 10 20 30 40 50
0 Distance from the green park (m) 0 Distance from the green park (m)
0 10 20 30 40 50 0 10 20 30 40 50
(a)
Distance from the green park (m) (b)
Distance from the green park (m)

1.4
(a) (b)
1.4
1.2
rise (K)

1.2
1
rise (K)

1
0.8 2m/s
Air temperature

3m/s
0.8 2m/s
Air temperature

0.6 4m/s
3m/s
5m/s
0.6
0.4 4m/s
6m/s
5m/s
0.4
0.2 6m/s

0.2
0
0 10 20 30 40 50
0 Distance from the green park (m)
0 10 20 30 40 50
(c)
Distance from the green park (m)

(c)
Figure 14. Air temperature rise by Equation (15). In the cases where the sensible heat flux is (a) 236.7
Figure 14. 132.5
W/m2; (b) Air temperature
W/m2; (c) 28.3riseW/m
by Equation
2. (15). In the cases where the sensible heat flux is (a) 236.7
Figure 14.2 Air temperature rise by2 Equation (15). In the cases where the sensible heat flux is
W/m ; (b) 132.5 W/m 2; (c) 28.3 W/m .
(a) 236.7 W/m2 ; (b) 132.5 W/m2 ; (c) 28.3 W/m2 .
Table 3. Coefficient a when it is approximated by Equation (14).
Table 3. Coefficient a when it is approximated by Equation (14).
Table 3. Coefficient a 2whenm/s 3 m/s 4 m/s 5 m/s 6 m/s
it is approximated by Equation (14).
236.7 W/m2 20.021m/s 30.020
m/s 40.016
m/s 50.013
m/s 60.011
m/s
132.5 W/m2 0.021
236.7 0.019 0.020
0.016 0.016
0.012 0.013
0.010 0.011
0.008
2 m/s 22 0.010
28.3 W/m
3 m/s0.007 0.0054 m/s 0.003 0.002
5 m/s 6 m/s
132.5 W/m 0.019 0.016 0.012 0.010 0.008
236.7 W/m 2 0.021
28.3 W/m2 0.0200.007 0.005
0.010 0.016 0.003 0.002
0.013 0.011
132.5 W/m2 0.019 0.016 0.012 0.010 0.008
28.3 W/m2 0.010 0.007 0.005 0.003 0.002
Climate 2017, 5, 60 12 of 12

5. Conclusions
In order to clarify the characteristics of air temperature rise in an urban area on the leeward
side of a green area, mobile measurements and calculations expressing advection and diffusion
effects are made. These calculations were then verified by comparison with the measurement results.
The relationship between the distance from the green area to each mobile measurement point in the
urban area and the air temperature rise is analyzed using the measurement results in Kobe city. At a
measurement point where the distance from the green area is 30 m or more, the air temperature of the
urban area becomes unaffected by the green area.
Calculation results and measurement results for air temperature rise with distance from the green
area are compared when an isotropic diffusion model and an incorporated buoyancy model are applied
for the vertical diffusion term. From the comparison with the measurement results in Kobe City, as well
as in Tokyo and Osaka, it is considered that air temperature does not rise as it enters the part of the
urban area beyond more than 50 m from the edge of the green area. The air temperature rise in the
urban area near the green area, due to the development of the urban boundary layer, is expressed
using the sensible heat flux from the ground surface, the distance from the green area and the wind
velocity. We considered an approximation of air temperature rise in order to express the following
situation: when entering the urban area, air temperature rises sharply, and when passing beyond a
certain distance, it becomes almost constant.

Acknowledgments: I thank the urban planning bureau of Kobe city office for their cooperation with respect to
our measurements. I used the modified source code by Moriyama et al. for the calculations.
Conflicts of Interest: The author declares no conflict of interest.

References
1. Givoni, B. Impact of planted areas on urban environmental quality: A review. Atmos. Environ. 1991, 25B,
289–299. [CrossRef]
2. Honjo, T.; Takakura, T. Simulation of thermal effects of urban green areas on their surrounding areas.
Energy Build. 1990–1991, 15, 443–446. [CrossRef]
3. Moriyama, M.; Takebayashi, H.; Fukumoto, K. Effects of Green Areas on Urban Air Temperature by
Numerical Solution. Mem. Grad. Sch. Sci. Technol. Kobe Univ. 1997, 15, 101–115.
4. Bowler, D.E.; Buyung-Ali, L.; Knight, T.M.; Pullin, A.S. Urban greening to cool towns and cities: A systematic
review of the empirical evidence. Landsc. Urban Plan. 2010, 97, 147–155. [CrossRef]
5. Ca, V.T.; Asaeda, T.; Abu, E.M. Reductions in air conditioning energy caused by a nearby park. Energy Build.
1998, 29, 83–92. [CrossRef]
6. Yu, C.; Hien, W.N. Thermal benefits of city parks. Energy Build. 2006, 38, 105–120. [CrossRef]
7. Yagi, R.; Takebayashi, H. Study on the near ground surface temperature influenced by surface temperature
and wind velocity. Eighth Natl. Conf. Heat Isl. Inst. Int. 2013, 1, 48–49.
8. Ashie, Y.; Ca, V.T. Developing a three-dimensional urban canopy model by space-averaging method:
Development of the urban climate simulation system for urban and architectural planning Part 2. J. Environ.
Eng. (Trans. AIJ) 2004, 69, 45–51. [CrossRef]
9. Kato, T.; Yamada, T.; Hino, M. Spatial structure of air temperature and humidityin urban park forest and its
surrounding. J. Inst. Sci. Eng. Chuo Univ. 2016, 12, 63–71.
10. Moriyama, M.; Kono, H.; Yoshida, A.; Miyazaki, H.; Takebayashi, H. Data analysis on “cool spot” effect of
green canopy in urban areas. J. Archit. Plan. (Trans. AIJ) 2001, 66, 49–56. [CrossRef]

© 2017 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).