International Journal of the Physical Sciences Vol. 6(1), pp.

65-73, 4 January, 2011

Available online at http://www.academicjournals.org/IJPS
ISSN 1992 - 1950 2011 Academic Journals

Full Length Research Paper

Study on rooftop outdoor thermal environment and slab

insulation performance of grass planted roof
Han-Hsi Liang1 and Kuo-Tsang Huang2*
1

Department of Architecture, National United University, No. 1, Lien-Da, Miaoli 36003, Taiwan.
Department of Landscape Architecture, National Chiayi University, No. 300 University Rd., Chiayi City 60004, Taiwan.

Accepted 31 December, 2010

The benefits of rooftop greening are ascertained in thermal perspective. It effectively reduces the near
surface ambient temperature and the reflected irradiation from the rooftop slab, thus increasing the
thermal comfort environment above the lawn and reduces the heat transferred through the rooftop slab.
The former effect enables the potential usage of the rooftop spaces for human activity and helps to
alleviate the urban heat island effect while the latter is able to reduce the interior space cooling load
especially in subtropical climate. A field experiment was carried out in the subtropical central Taiwan to
quantify these two performances. Indices of Mean Radiant Temperature (MRT), Wet Bulb Globe
Temperature (WBGT), and heat flux rate were used to explain and discuss the performance of thermal
comfort and heat flux of the rooftop lawn. The results confirmed that rooftop lawn contributes benefits
both on its outdoor surrounding environment and the indoor energy beneath.
Key words: Grass planted roof, thermal environment, slab thermal insulation.
INTRODUCTION
Taiwan which central is located right on the Tropic of
Cancer is a fast developed and over populated island.
The climate here is of subtropical climate characteristic.
High concentration of buildings is commonly seen in
many urban districts, which result in various urban
environmental issues, such as lacking enough recreation
areas, low urban greenery cover ratios, and moreover,
the urban Heat Island Effect (UHI). Green roofs (or
planted roofs) as an extension of urban green areas and
recreational open spaces not only provide visual
aesthetic, they also contributes to the thermal benefits in
buildings and their surrounding environments.
There are direct and indirect effects of green roof from
thermal perspective. The direct effects of green roof are
their thermal benefits in reducing surface temperatures of
roofs and heat transfer into the rooms underneath. It will
directly contribute to improving the indoor thermal
environment and thermal performance of buildings.

*Corresponding author. E-mail: [email protected]. Tel:

+886-5-2717636. Fax: +886-5-2717634.

Indirect effects of green roofs refer to its potential thermal

impacts on surrounding environment. It will contribute to
creating better outdoor thermal environment and
mitigating the UHI effect (Wong et al., 2003).
Besides, because most of the rooftop zones are private
estate, low maintenance is usually required to sustain a
rooftop garden. Extensive roof greening style such as
lawn or turf greening is a much favorable and feasible
solution due to its lower maintenance and lower initial
cost benefits. Incentives from the government were
recently also proposed to encourage rooftop greening in
Taiwan. This is why the lawn planted rooftop is therefore
chosen for studying in this research. Although several
studies on green roofs regarding its thermal impacts on
the environment have been carried out worldwide
(Niachou et al., 2001; Theodosiou 2003; Wong et al.,
2003; Kumar and Kaushik, 2005; Lazzarin et al., 2005;
Wong and Li, 2007; Wong et al., 2007; Dvorak and
Volder, 2010; Jim and He, 2010; Teemusk and Mander,
2010), they are limited to certain locations, climate and
planting types. These data are not directly applicable to
the subtropical environment.
The objectives of the research are as follows:

66

Int. J. Phys. Sci.

Figure 1. A five meter square lawn planted area with 13 cm substrates underneath were created at the
centre of a buildings rooftop.

1. To investigate the environment thermal comfort

enhancement over the rooftop lawn greening.
2. To understand thermal insulation performance of the
rooftop lawn greening which contributes to the reduction
of interior heat gain.
The first subject intend to quantified the reduction of
ambient temperature and the surrounding long-wave
radiation due to the rooftop lawn, which is essentially the
two key factors that affect the peoples willingness of
using rooftop for recreational purpose. The second
subject attempts to quantify the interior heat gain
reduction amount on its insulation perspective.

corridor.
Cynodon dactylon was chosen as the rooftop lawn to plant over
the 10 cm-thick Finland KEKKIL peat moss studied herein,
because of its drought and heat enduring as well as fast growing
features. Furthermore, an additional 3 cm perlite layer for drainage
was placed below the growing medium separated by a thin
nonwoven layer. Therefore a minimum of 13 cm thick lawn grass
layer paved over the original concrete rooftop was constructed.
Manual irrigation was performed consecutively in the past one week
when there was no natural precipitation to wet the grass.
The experiment last for over a year and was carried out from 8th
September, 2008 to 9th October, 2009. The comparative set (that
is, control set) of the experiment without rooftop greening located
about five meters away from the lawn planted field was also carried
out simultaneously (Figure 2).
Deployment of instruments

METHODOLOGY
An experiment field of lawn greening were established on top of a
four-story student center building located in National Chiayi
University at Chiayi City, Central Taiwan, where its latitude is of
23.6N. Five meter square lawn planted area with 10 cm growing
medium was created at the center of the rooftop to avoid any
rooftop object shading overlaid (as shown in Figure 1). The surface
material of the rooftop exposed slab is grey concrete tile which is a
very common material used for flat roofs surface construction in
Taiwan. Material layers that comprise the original rooftop
construction from top to down are 5 cm concrete tile, 2 cm
polystyrene insulation, 1 cm polyurethane membrane, and 18 cm
reinforced concrete. The overall thermal conductivity (U-value) of
the original rooftop slab is estimated 0.99 W/mK with horizontal
outdoor and indoor near-surface air-film resistance included. Space
underneath the rooftop experiment field is a public unconditioned

The experiments in the research comprises two sections, one is for

studying outdoor thermal comfort and another is for understanding
the effect of indoor heat flux reduction.
For experiment of outdoor thermal comfort, two sets of
instruments were placed above the lawn and the exposed slab
respectively. Each set includes sensors for measuring globe
temperature (Tg), dry-bulb temperature (that is, ambient
temperature)(Ta), relative humidity, wind velocity (W V) and naturally
ventilated wet-bulb temperature (Tnwb). All the sensors were
horizontally aligned at one meter height above the ground surface,
as shown in Figure 3. Instruments used herein includes calibrated
T-type thermocouples placed in the white louver shelters for
measuring ambient temperature, black globe thermometers for
measuring Tg, omnidirectional hotwire air speed transmitters for
measuring wind velocity. Naturally ventilated wet bulb temperature
could be obtained via placing cotton-wrapped Pt100 sensor within

Liang and Huang

67

Figure 2. The comparative set of the experiment.

SR
A
B

A
B

S
S
S

S
Q

Figure 3. Sensors used and their deployment in the field. (A: globe temperature, B: ambient temperature and
relative humidity, C: wind velocity, D: naturally ventilated wet bulb temperature, S: surface temperature, Q: heat
flux, SR: solar radiation).

two third water filled vacuum stainless bottles. Moreover,

pyranometer was installed to record the incident total horizontal
solar radiation that reaches to the experiment field. All the
measured values from the above instruments were simultaneously
recorded at an interval of 10 s by a Delta GL-800 data logger.
For experiment of indoor heat flux, another two sets of
instruments were deployed above and under the floor slab both with
and without lawn conditions. Thin heat flux sensors were
seamlessly attached beneath the floor slab with aluminum foil cover

over it to prevent heat intervention from the corridor making sure

the measured values only accounts for heats that transferred
through/from the floor slab. Moreover, to further understand the
fluctuations of the surface temperature of the slab due to lawn
greening, several T-type thermocouple wires were placed above
and beneath the floor to fetch values of their surface temperature,
as shown in Figure 3. The indoor ambient temperatures were also
taken measured by hanging thermocouples five centimeters under
the floor slab ceiling.

Int. J. Phys. Sci.

Total horizontal solar radiation (W/m2)

68

Figure 4. Surface temperature variation above slab in different seasons.

Indices for interpreting thermal comfort

There were two indices used to describe thermal environment. The
mean radiant temperature (MRT) is a key variable in thermal
calculations for the human body. It is the uniform temperature of an
imaginary enclosure in which radiant heat transfer from the human
body equals the radiant heat transfer in the actual non-uniform
enclosure (ASHRAE, 2005). The simplified calculation method of
MRT can be easily obtainable using several physical measured
parameters as described in Equation1. The MRT index is closely
related to parameters including Tg, Ta and W v. It can efficiently
responses the long-wave radiation from surrounding environment
enclosure, which is considered the most influential factor to human
comfort sensation. Furthermore, the energy balance of one's body
is highly responsive to changes in MRT and is more suitable and
direct for describing human's thermal sensation.
MRT=Tg+0.237(W v)0.5(Tg-Ta)

(1)

The wet bulb globe temperature (WBGT) index is widely used for
estimating the heat stress potential of industrial environments
concerning activity status. It is a composite temperature used to
estimate the overall effect of temperature, humidity, and solar
radiation on humans. Parameters of Ta, Tnwb, and Tg are each
multiplied by a corresponding weighted values to calculate WBGT
index as in Equation 2 (Dukes-Dobos and Henschel, 1973; ISO
7243, 1989).
WBGT=0.7Tnwb+0.2Tg+0.1Ta

(2)

RESULTS AND DISCUSSION

Although the experiment lasts for one year, it is rather

difficult to process such a huge amount of data; therefore,

typical day was selected for analysis. The day in which
ten days before it was without precipitation or cold draft or
other exceptional weather conditions is considered a
typical day to rule out unusual weather influences.
Moreover, the daily averaged differences of dry-bulb
temperature, wind velocity and total horizontal solar
radiation of the typical day should all fall between 1%
margin compare to long-term recorded seasonal data.
According to the above criteria, 10th December, 2008 and
10th June, 2009 were selected as typical days for further
analysis each representing winter and summer
respectively.
Surface temperature variation
The slab surface diurnal temperature variation of the
typical days selected for winter and summer were plotted
as shown in Figure 4. The daily averaged total horizontal
solar radiation intensity were 176.1 (W/m) and 288.4
(W/m) for winter and summer cases. It showed that the
variation patterns from the two seasons, either with or
without lawn, were of similar shape. The major difference
was its absolute values. By comparing the surface
temperature drop in winter, temperature decreased more
on the exposed slab surface than on the lawn soil
surface. With the similar variation pattern observed from
Figure 4, therefore, for the discussion, only the selected
typical day of summer was discussed afterward.

Liang and Huang

69

Between lawn and slab

Surface temperature (C)

Total horizontal radiation (W/m2)

Total horizontal solar radiation

Figure 5. Diurnal surface temperature variation in summer.

Figure 5 revealed that while without lawn greening, the

maximum temperature of the exposed slab surface could
reach 61.3C which occurred at seven minute later when
solar radiation was at its daily high of 1015 W/m at
12:52. The maximum daily variation of surface
temperature was 35.4C. For lawn planted area, the
surface temperature measured during daytime was not
as high as that of the exposed slab surface. The
maximum surface temperature of lawn field was around
32.4C and the maximum daily variation of surface
temperature was only 3.5C which was much lower than
those measured on the exposed slab surface. The reason
could be due to the combination effect from the grass leaf
shading and the evaporation of moisture in the soil.
As for the surface temperature beneath the exposed
floor slab, maximum surface temperature of 41.1C occur
at 17:06, at which time it was delayed for about four
hours when its top surface temperature reaches its
maximum value. For lawn planted area, maximum
surface temperature of 32.3C occurs at 23:04, at which
time it was delayed for around 10 h in comparison to that
of the exposed slab. It indicates that with a 13 cm
substrate layered lawn greening on top of the roof could
extend the time lag of sub-surface temperature to around
six hours. With this effect it could also alleviate the
daytime indoor heat gain by delaying heat transfer into
space.

Outdoor thermal comfort performance

On the discussion of outdoor thermal comfort of spaces
between above lawn and above exposed concrete slab,
Tg, MRT index, WBGT index, ambient temperature and
relative humidity measured at one meter height above
both conditions were compared. The daily averaged
ambient temperature above the lawn was 0.24C lower
than the exposed slab condition. After sunset there was
significant reduction of ambient temperature above the
lawn planted area and lasted till the next day sunrise. It
indicated that the lawn constantly lowered the ambient
temperature during non-solar-radiation period. Maximum
difference of 2C was observed at around 7:00.
The comparison of relative humidity measured at one
meter height above exposed slab surface and lawn
planted area is shown in Figure 6. When at night time,
lower relative humidity was observed above the lawn
planted area with a maximum difference value of 10.48%
in comparison to that of the exposed slab. But when in
daytime it was a bit higher by around 5%, it could be due
to the grass vigorous transpiration in daytime that slightly
moisturize the surrounding air.
Based on the Ta, Tg, and W V measured at one meter
height, the MRT above the exposed slab surface and
lawn area were calculated. The Tg represents the
integrated effects of radiation and wind. The measured Tg

Int. J. Phys. Sci.

Global temperature (C)

70

Figure 6. Diurnal relative humidity and globe temperature variation in summer.

and calculated MRT were plotted in Figures 6 and 7

respectively. There were obvious differences in Tg and
MRT between both conditions. Maximum differences of
Tg and MRT, 2.6C and 3.19C, occurred at 13:18 when
the solar radiation was the strongest. It was due to the
fact that surface temperature on the exposed slab was
relatively higher than that of the lawn planted surface,
thus it caused long-wave radiation to be emitted a lot
more from the exposed slab surface resulting in higher Tg
and MRT. This indicated that lawn planted field was able
to provide more comfortable environment for outdoor
activities than with exposed slab conditions during
daytime.
On the other side, during night time, heat dissipation
was more quickly on the exposed slab surface than on
the lawn planted area, which led to drastic surface
temperature drop on the exposed slab surface. Without
solar radiation, the radiation part of Tg and MRT mainly
depended on the amount of long-wave radiation emitted
from surrounding surfaces. When in early morning,
around 6:00, surface temperature on the exposed slab
surface reached to a maximum difference of 3C lower
than the lawn planted area, and subsequently resulted in
0.7C of Tg and 0.87C of MRT lower than that of planted
area.
WBGT was also studied in this research, providing
more overall view on the heat stress issue that relates to
thermal comfort. The calculated WBGT was drawn in

Figure 7 together with MRT overlaid. According to

regulations from Taiwan Council of Labour Affairs,
maximum WBGT of 30.6C for continuously light work
load activity is recommended. With the above criteria,
from sieving the calculated WBGT during working hours
(that is, from 8:00 to 18:00), there was 52.3% for planted
condition and 63.5% for exposed slab conditions that
would fell out of the criteria. That is, there was 11.2%
higher probability unable to meet the criteria on above
exposed slab surface than on above lawn planted area.
Furthermore, according to Hwang and Lin (2007), a
research of defining semi-outdoor thermal environment
comfort zone for Taiwan, the result reveals that while
MRT is between 21 to 47C is considered comfortable.
From this point of view, there was a probability that of
24.79% the MRT within a day above the lawn would fall
out the comfort zone. When compared to the MRT above
the exposed slab, there was 31.74% probable of falling
out the comfort zone, which is 6.95% higher. All the fellout period was between 8:00 to 16:00.
Indoor heat flux fluctuation
On indoor heat flux, a considerable rate of heat flux was
observed under the floor slab with exposed slab on top.
The maximum heat flux, 48.5 W/m, was found at 17:07,
which was five hours later when daily maximum total

71

Temperature (C)

Liang and Huang

Solarradiation
radiation (W/m
(Wm 2))
Total horizontalSolar
solar
radiation

flux (Wm
(W/m2)2)
Heat flux

Figure 7. Calculated MRT and WBGT variation on a summer day at the height of 1 m.

Figure 8. Diurnal heat flux variation comparison.

horizontal solar radiation occurred. Heat gain was

observed nearly all day long, except there was around
five minute periods at 8:50, in which heat loss was
encountered. Diurnal heat flux variation range could
reach up to 50.7 W/m, as indicated in Figure 8.

For the area with lawn planted condition, most heat flux
observed underneath the floor slab during daytime was of
negative values, that is, the space was encountering heat
loss. The heat loss began at approximate 10:00 and
continued till around 21:00 at night, which had positive

72

Int. J. Phys. Sci.

Figure 9. Diurnal heat flux and indoor temperature difference.

influence on reducing indoor cooling loads during work

hours in daytime. Consecutive indoor heat gain occurred
during the time other than heat loss period, which was
mainly at night.
There was apparent time lags regarding heat gains
under lawn planted area, as indicated in Figure 8. Solar
heat gain received on top of the lawn surface during
daytime was lagged for nearly 14 h when it began to
transfer into the interior space. In comparison with the
exposed slab condition, there was only approximate four
hours lag. It indicated that with lawn planted above the
rooftop, the interior space underneath would not instantly
respond to the solar radiation heat gain from the top of
roof during daytime. Thus, the spaces that resulted
underneath the lawn planted area were less responsive
to outdoor climate, and were beneficial to the interior
cooling energy reduction.
The diurnal variation range of heat flux under lawn
condition was 11.4 W/m, which was only around one fifth
amount comparing to the exposed slab condition, and
moreover, the peak heat flux was not apparent. The total
cumulative heat gain of a day was -0.25 W/m with lawn
planted condition and was 19.21 W/m without lawn
planted. The lawn had a significant influence on the
amount of heat flux transferred. This was possibly due to
grass leafs sun blocking effect and the heat dissipation in
substrate layer by both heat absorption and evaporation
effect.
Figure 9 shows the differences of transferred heat flux
and indoor room temperature between with and without
lawn conditions. Maximum indoor temperature difference,
4.2C, was detected at around 16:00 which corresponded
to the maximum heat flux difference, which is around 50

W/m, occurring time. Comparison between Figures 8

and 9, shows that the lawn planted layer provides a good
heat storage as well as heat dissipating means, reducing
the fluctuation range of indoor temperature and
transferred heat flux. The interior space underneath the
lawn planted roof would have more stable indoor climate
which significantly contributed to improve indoor thermal
comfort and also reduced air-conditioning energy
consumption.
Conclusions
In this research, comparisons of outdoor thermal comfort
performance and indoor heat flux between lawn planted
area and exposed slab area were made through
investigating onsite field experiments in subtropical
central Taiwan. Both with and without lawn planted
experiment field on top of roof were established and were
simultaneously measured for a yearlong for studying
various parameter reduction effect. Various parameters
including physical measurements and calculated indices
studied in the research are tabulated in the amount of
reduction term as indicated in Table 1 for better
understanding. Conclusions drawn from this research are
as follows:
1. The planted lawn provides excellent solar radiation
block to the rooftop slab, which greatly reduces transient
heat flux through the rooftop slab and leads to lower
surface temperature under the lawn panted slab.
2. Cooling effect from the lawn was confirmed by
comparing the Ta drop. Lower ambient temperature of

Liang and Huang

73

Table 1. Reduction summary of varies parameters.

Parameters

Unit

Surface temperature above slab

C
Surface temperature beneath slab
C
Outdoor ambient temperature
C
Indoor ambient temperature
C
2
Heat flux
W/m
MRT
C
WBGT
C
Relative humidity
%
Globe temperature (Tg)
C

Whole day
Reduction range
Avg.
-3.6 to 30.1
9.12
-0.3 to 9.5
4.24
-1.09 to 2.00
0.24
-0.90 to 4.2
1.22
-3.9 to 52.3
19.49
-0.87 to 3.19
1.16
-0.16 to 1.78
0.32
-6.26 to 10.48
1.48
-0.7 to 2.6
0.99

Std.D.
11.28
3.11
0.42
0.90
15.73
0.73
0.30
3.04
0.61

10:00-22:00 h
Avg.
Std.D.
16.44
10.24
6.64
2.27
0.25
0.28
1.69
0.79
31.85
12.06
1.44
0.69
0.43
0.31
0.47
2.47
1.21
0.57

22:00-10:00 h
Avg.
Std.D.
1.81
6.53
1.83
1.60
0.23
0.52
0.76
0.74
7.13
6.63
0.88
0.67
0.21
0.25
2.50
3.22
0.78
0.58

th

The calculated values in the above table are from onsite measured data recorded on 10 June 2009. The term of reduction is the values
measured from the exposed slab condition deducted by those measured from the lawn planted area. Avg. denotes to average value; Std.D.
denotes to standard deviation.

interior space under the lawn planted area was observed,

which is capable of providing a more pleasant indoor
thermal environment.
3. The time lag effect of lawn planted area was confirmed
by investigating the diurnal heat flux variations. There
were extra 10 h time lags with lawn planted above the
rooftop, which implies that the lawn planted layer can be
used as an efficient thermal barrier in passive building
design.
4. Less long-wave radiation emitted from the lawn
planted roof was confirmed through comparisons of Tg
and MRTs measured/calculated on site. It indicated that
thermal comfort is improved and is more feasible for
rooftop outdoor activities above lawn planted area.
5. Heat flux transfer was greatly reduced under lawn
planted area. This was also confirmed by comparing heat
flux rate between both conditions. It reveals that great
deal amount of summer cooling loads will be reduce
under planted roof circumstance, and as a result, energy
saving on air-conditioning systems will be achieved.
ACKNOWLEDGEMENT
The authors would like to express their greatest
appreciation to Taiwan National Science Council for
financial support to the research.
REFERENCES
ASHRAE (2005). ASHRAE Handbook - Fundamentals. Atlanta, USA,
American Soceity of Heating, Refrigerating and Air-Conditioning
Engineers, Inc.
Dukes-Dobos F, Henschel A (1973). Development of permissible heat
exposure limits for occupational work. ASHRAE J., 9: 57.
Dvorak B, Volder A (2010). Green roof vegetation for North American
ecoregions: A literature review. Landsc. Urban Plan., 96(4): 197-213.
Hwang RL, Lin TP (2007). Thermal comfort requirements for
occupants of semi-outdoor and outdoor environments in hot-humid

regions. Archit. Sci. Rev., 50(4): 60-67.

ISO 7243 (1989). Hot environments -- Estimation of the heat stress on
working man, based on the WBGT-index (wet bulb globe
temperature), International Organization for Standardization.
Jim CY, He H (2010). Coupling heat flux dynamics with meteorological
conditions in the green roof ecosystem. Ecol. Eng., 36(8): 1052-1063.
Kumar R, Kaushik SC (2005). Performance evaluation of green roof and
shading for thermal protection of buildings. Build. Environ., 40(11):
1505-1511.
Lazzarin RM, Castellotti F, Busato F (2005). Experimental
measurements and numerical modelling of a green roof. Energy
Build., 37(12): 1260-1267.
Niachou A, Papakonstantinou K, Santamouris M, Tsangrassoulis A,
Mihalakakou G (2001). Analysis of the green roof thermal properties
and investigation of its energy performance. Energy Build., 33(7):
719-729.
Teemusk A, Mander U (2010). Temperature regime of planted roofs
compared with conventional roofing systems. Ecol. Eng., 36(1): 9195.
Theodosiou TG (2003). Summer period analysis of the performance of
a planted roof as a passive cooling technique. Energy Build., 35(9):
909-917.
Wong NH, Chen Yu, Ong CL, Sia A (2003). Investigation of thermal
benefits of rooftop garden in the tropical environment. Build. Environ.,
38(2): 261-270.
Wong NH, Li S (2007). A study of the effectiveness of passive climate
control in naturally ventilated residential buildings in Singapore. Build.
Environ., 42(3): 1395-1405.
Wong NH, Tan PY, Chen Y (2007). Study of thermal performance of
extensive rooftop greenery systems in the tropical climate. Build.
Environ., 42(1): 25-54.