Energy and Buildings 55 (2012) 6676

Contents lists available at SciVerse ScienceDirect

Energy and Buildings

journal homepage: www.elsevier.com/locate/enbuild

Cool and green roofs. An energy and comfort comparison between passive
cooling and mitigation urban heat island techniques for residential buildings in
the Mediterranean region
M. Zinzi , S. Agnoli
ENEA Italian National Agency for New Technologies, Energy and Sustainable Economic Development Technical Unit for the Energy Efciency, Via Anguillarese, 301, 00123 S. Maria di
Galeria, Rome, Italy

a r t i c l e

i n f o

Article history:
Received 12 May 2011
Received in revised form 29 July 2011
Accepted 12 September 2011
Keywords:
Cool materials
Green roof
Thermal comfort
Passive cooling
Energy performance

a b s t r a c t
The increase of peak and energy demand during the cooling season is becoming a crucial issue, as well
as the intensication of the urban heat island effect. This trend is observed at several latitudes, including
areas where overheating was unknown at building and urban levels. This phenomenon involves different
issues: reduction of greenhouse gases, quality and comfort in outdoor and indoor environment, security
of energy supply, public health. The building sector is directly involved in this change and adequate
solutions can provide great benet at energy and environmental levels. Roofs in particular are envelope
components for which advanced solutions can provide signicant energy savings in cooled buildings
or improve indoor thermal conditions in not cooled buildings. Cool materials keep the roof cool under
the sun by reecting the incident solar radiation away from the building and radiating the heat away
at night. Roofs covered with vegetation take benets of the additional thermal insulation provided by
the soil and of the evapo-transpiration to keep the roof cool under the sun. These two technologies
are different in: structural requirements, initial and lifetime maintenance costs, impact on the overall
energy performance of buildings. This paper presents a numerical comparative analysis between these
solutions, taking into account the several parameters that affect the nal energy performances. By means
of dynamic simulations, the paper depicts how cool and green roofs can improve the energy performance
of residential buildings in different localities at Mediterranean latitudes.
2011 Elsevier B.V. All rights reserved.

1. Introduction
The effects of global warming and climate changes are of
relevant concern for environment and human activities in the
Mediterranean area. The average air temperature rise of 2 C represents a critical limit beyond which dangerous climate changes
should occur by 2030 [1]. More than 90 million people live in the
twenty most populated Mediterranean metropolitan areas; according to the actual trend other 70 million of people are expected to
move to leave the countryside towards the urban area by 2025
[2]. The global warming and the urban sprawl causes a number
of environmental hazards, the urban heat island (UHI) is one of
these.
This phenomenon is dened as the air temperature rise in
densely built environments respect to the countryside surroundings. The main cause is the modication of the land surface
in the urban area, where the vegetation is replaced by exten-

Corresponding author. Tel.: +39 06 3048 6256/4188; fax: +39 06 3048 3930.
E-mail address: michele.zinzi@enea.it (M. Zinzi).
0378-7788/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.enbuild.2011.09.024

sively built surfaces (typically paved roads and buildings surfaces),

characterised by high solar absorption, high impermeability and
favourable thermal properties for energy storage and heat release,
as well as several anthropogenic. The UHI was rst monitored in
London back to the 19th century [3]; many studies were performed
during the past decades [410], showing the quantitative effects of
the phenomenon and the correlation with the previously enounced
causes. Daily mean UHI typically ranges between 2 and 5 C, while
UHI intensities (dened as maximum difference between urban
and background rural temperatures) up to 12 C were registered
under particular conditions. This UHI impacts important issues
such as: the quality of life; the public health, especially for the most
vulnerable population; the environmental hazards.
Roof surfaces of the building accounts for the 2025% of the total
urban surfaces, hence they can successfully used to reduce the air
and surface temperature of urban area [11]. Cool and green roof,
widely described in the next paragraph, are used to mitigate the UHI
and the impact was proved by several studies [1215]. These techniques can also have signicant benets on the energy performance
of buildings, providing passive cooling to the built environment.
This topic is of special interest because of the rapid increase of the

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

energy consumption and peak demand for cooling in Europe and

in the Mediterranean basin [16].
This study aims at analysing and comparing how these UHI mitigation techniques can also improve the energy performance of
cooled buildings and the thermal comfort of not-cooled buildings
at Mediterranean latitudes. Special attention is paid to residential buildings, where accurate design concepts and technologies
can strongly reduce the use and the installation of cooling systems without affecting the occupants comfort expectations and
help achieving international environment targets.
2. Building applications of cool and green roofs
Cool roofs are characterised by materials having: high solar
reectance (SR) and high thermal emittance (TE). The former
expresses the ability of the materials of reecting most of the incident solar radiation during daytime, keeping their surfaces cooler
respect to conventional construction materials. The high thermal
emittance allows the materials to radiate away the heat stored in
the structure, mainly during night time. This thermal behaviour
allows the roof to reduce the heat transfer to the built environment.
Roofs characterised by low emittance values tend to not dissipate
the stored heat at night and can be considered cool only if they
have a very high solar reectance. White mortars and plaster were
widely used in ancient massive Mediterranean dwellings, in order
to create a more comfortable built environment during the hot season. The coastal villages of Greece, Italy and Spain still witness
this construction technique, which emerged again as an efcient
solution during the recent years.
Several numerical studies were carried out in the past years to
assess the energy performance of buildings equipped with cool
roofs. The impact of cool roofs on a single oor detached house
placed in different climatic zones of the planet world was calculated for insulated and not insulated dwellings [17]. The cooling
energy consumption reduction was 18% and 93% increasing the roof
solar reectance from 20% to 85%. Three typical building models
were developed respectively for: a residential building, an ofce
and retails store, differentiated by age (before and after 1980). The
impact of cool roof ensured global energy savings from 7% to 25%
according to the different age and building type for several US climates [18]. Other studies were focused to limited geographical
sites, as Jordan or Honk Kong [19,20]. Other studies faced the cool
roof positive impact evaluated as an additional thermal insulation
[21]. The results of the analysis revealed that the integrated daily
roof heat gain was not dependent on its thermal mass. An energy
analysis run proved that the daily heat ow in a roof with SR of 0.65
and a thermal resistance (R-value) of 1.1 m2 K/W was equivalent to
the ow in a roof with SR 0.3 and R-value 2.2 m2 K/W.
Limited data from real building application are available. A eld
campaign was carried out in one house and two school bungalows

67

in Sacramento, California. Cooling energy savings of 2.2 kWh/day

were measured increasing the solar reectance of the roof from 0.18
to 0.73 [22]. The energy savings in the school buildings was about
35%. A study in an experimental building in Rome, Italy, proved that
the air temperature of an attic room decreased by 2 C increasing
the albedo of the roof from 14% to 85% and this room was found
cooler than an identical room at the oor below, which had no roof
at all [23].
Green roofs, also called eco-roofs, use the foliage of plants to
protect the building environment. The thermal loads due to the
solar radiation and the air temperature are limited before entering
the buildings by the vegetation layer. This depends on the absorption of the solar radiation by the plants to support their life-cycle,
including: photosynthesis, evapo-transpiration, respiration. Moreover, the soil layer gives an added insulation to the building roof
and the water content increases the thermal inertia of the structure.
The vegetation characteristics affect, in addiction, the convective
and radiative heat transfer through the roof surface.
Green roofs were once typically used in northern climates to
improve the insulation performance of the building envelope, but
they are also an opportunity in warm climates, because of their
thermal behaviour under the solar radiation. Several studies were
produced during the past years trying to quantify the effect of
green roofs on the energy performance of buildings. The noticeable impact of green roofs during the hot and the cold seasons
was analysed in a nursery school in Athens, founding out that
energy savings up to 49% could be obtained [24]. The energy
and water issues were analysed in two experimental setups in
Italy; particular attention was paid to the impact of the foliage
on the radiation and the air temperature proles insisting on the
building roof, respect to the undisturbed values [25]. Combined
measurements and calculation analyses were performed in order
to assess and predict the 60% reduction of the heat ux through
a green roof respect to a conventional roof in a hospital building
in northern Italy [26]. A case study in Brasil demonstrated that a
green roof in an experimental building reduced the heat ux by
9297% compared to a ceramic and a metallic conventional roof
[27]. Specic studies on the substrate materials, foliage characteristics and vegetal species demonstrated the variability of the green
roof performances as a function of the adopted technical solutions
[2830].
3. Methodology
The scope of this work consists in the assessment of the energy
performances of residential buildings using different roof solutions: standard, cool and green roofs. The study is focused on
the Mediterranean area, a mild climatic zone with differences
in rainfall levels and air temperature proles that can lead to
different choices of building technologies to achieve the opti-

Table 1
Air temperature and solar radiation data of the selected localities.
Month

T ( C)
Barcelona

H (kJ/h/m2 )

RH (%)

T ( C)
Palermo

H (kJ/h/m2 )

RH (%)

T ( C)
Cairo

H (kJ/h/m2 )

RH (%)

Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec

8.2
9.4
11.1
13.1
17.0
20.9
23.5
24.1
21.6
17.3
12.1
9.9

288
409
553
744
879
873
1004
858
603
444
287
250

71
68
73
72
74
74
68
71
74
82
78
65

12.7
11.9
13.8
15.7
19.2
22.8
25.5
27.0
24.1
21.6
17.2
13.9

312
454
625
843
980
1090
1099
993
741
549
319
272

76
71
79
71
77
71
76
73
66
74
69
78

14.0
14.5
16.6
21.8
24.7
28.0
28.2
27.9
26.6
23.8
19.0
15.3

439
597
736
913
1052
1142
1118
1008
898
630
493
430

67
58
59
45
40
45
56
60
56
56
61
64

68

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

Table 3
Thermal properties of dwellings envelope components.

Table 2
Rainfall data of the selected localities.
Month

Barcelona rainfall
(mm)

Palermo rainfall
(mm)

Cairo rainfall
(mm)

Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec

39.50
40.60
47.36
47.36
53.27
44.10
30.00
51.54
69.26
93.45
61.89
46.10

67.60
66.30
59.70
43.50
26.00
14.40
7.80
12.96
40.78
94.54
92.00
78.63

6.94
4.00
4.00
2.06
0.00
0.00
0.00
0.00
0.00
1.05
2.97
4.94

mal energy performances. The comparison is carried out by

means of a numerical analysis performed using the Design
Builder interface, which relies on the Energy Plus calculation
engine, able to perform energy balance calculations with hourly
time step. The tool carries out accurate thermal analyses and
allows very detailed inputs, including: climatic data (including air
temperature, solar radiation, relative humidity hourly proles);
construction materials and components in dedicated libraries or
manually edited; energy systems specications; time schedules
(systems management, occupancy, electric lighting, ventilation,
etc.). Energy Plus also features a validated mono-dimensional
green roof model developed taking into account the evapotranspiration of the vegetation layer, the time dependent soil
thermal properties (conduction and inertia), the radiative and
convective heat exchanges [31,32]. The tool also allows a complete description of any construction material, since thermal
conductance, solar reectance and thermal emissivity can be modelled.
In order to depict a wider overview of these techniques impact
on the energy performances a number of variables are taken into
account. A number of building variants are, hence, dened and the
cooling and heating demands are calculated for each of them. The

Envelope component

Not-ins U (W/m2 K)

Ins U (W/m2 K)

Wall
Roof
Ground oor
Window glass
Window frame

1.4
1.4
1.7
2.8
5.9

0.7
0.6
0.8
1.8
4.7

comparison is carried out for the net energy, without considering the energy systems efciencies, since the main objective is the
optimisation of the envelope energy performances. The variants
considered in the analysis are described in the next paragraphs.

3.1. Denition of the reference localities

Three localities were selected, typical of different regional areas:
Barcelona for the north rim, Cairo for the south rim and Palermo for
the centre basin. This criterion also responds to a strict relationship
between buildings and climate. The north rim is heating dominated
for not insulated buildings, while heating and cooling are both relevant in buildings in the centre of the Mediterranean region. The
south rim is cooling dominated for any building thermal characteristics. Investigating these three localities, allowed mapping with a
good accuracy the roong techniques efciency according to the
relevant Mediterranean climatic conditions.
The climatic conditions of these three areas referred to air
temperature and relative humidity, as well as the global solar horizontal radiation. These data are taken by the Meteonorm database,
embedded in Design Builder. Table 1 reports the monthly daily
mean air temperature and the solar radiation, expressed as global
irradiance on the horizontal. The difference within the three localities can be easily inferred in this case, since air temperatures and
solar radiation decreased with latitude. The relative humidity is also
reported in Table 1, these parameters are important because of the
impact on the energy use of the building and the comfort condi-

Fig. 1. Layout of the row house with the two thermal zones: ground level day zone and rst oor night zone.

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

69

Table 5
Characteristics of the vegetation and soil layers of the selected green roofs.

Fig. 2. Layout of the single oor detached house with the two thermal zones: day
(right) and night zone.

tions of occupant, even if it does not show signicant relationship

with the latitude.
Table 2 reports the monthly rainfall data expressed in millimetres as taken by the Meteonorm database. Rainfall is not embedded
into the software; hence the data was inputted through an external schedule. The table depicts the different water availability in
the three cities, critical parameter for green roofs applications.
Cairo is practically dry through all the year. Palermo and Barcelona
have different trends: the former has rainfall practically constant
throughout the year, while Palermo has higher rainfall in winter
but gets almost dry during summer months.
3.2. The reference buildings
The analysis is carried out on two residential building typologies
widely used in the Mediterranean region: row houses and detached
single family houses. The buildings are considered with two different envelope congurations: insulated and not insulated, in order
to have signicant results for old and recent constructed buildings.
The rst is a two oors row house, whose geometry is presented in Fig. 1. The building is divided into two thermal zones:
the day zone at the ground level, where almost all the daytime
activities take place; the night zone at the upper oor, hosting
bedrooms and services. The houses is in contact with the external
environment through the roof, the ground oor and the north and
south facades, while the east and west walls are adiabatic, assuming they are boundary layers between adjacent row houses. The
main thermo-physical data are summarised in Table 3. The not
insulated conguration considers a conventional double glazing
unit, whose g-value is 0.75, this value decreases to 0.65 for the
Table 4
Geometry of the selected buildings.

V volume
A net gross area
S total external surface
Sr roof surface
S/V
Sr /V
Sr /S

Unit

Row house

Detached house

m3
m2
m2
m2
m1
m1

427
116
211
68
0.50
0.16
0.32

369
100
364
112
0.98
0.30
0.30

Parameter

Value

Height of plant
Leaf area index
Leaf solar reectance
Leaf emissivity
Minimum stomatal resistance
Max volumetric moisture content of the soil
Min volumetric moisture content of the soil
Initial volumetric moisture content of the soil
Density of the soil
Specic heat of the soil
Conductivity
Soil layer thickness

<60 cm
1.2
0.25
0.9
120 m/s
0.32
0.01
0.15
960 kg/m3
1500 J/kgK
0.34 W/mK
0.12 m

insulated conguration, where a low-e double glazing units is considered. The windows are provided of an external shading device
(30% solar transmittance and 10% solar absortpance). Assumptions are also made regarding the occupancy prole, the electric
and the appliances loads. The natural ventilation/inltration was
xed at 0.5 ACH. Since the net energy has to be calculated, the
following hypotheses are assumed: the set-point temperature is
continuously maintained constant and heating and cooling system
generators of unlimited power are implemented to keep this condition. The air exchange rate is increased to 3 volumes per hour for
the free oating analysis, that is carried out to assess the impact
of the roof solutions on the indoor thermal comfort of not cooled
buildings.
The detached single family house consists of a single oor building and it is divided in two thermal zones: day (mainly facing east)
and night (mainly facing west). The same thermo-physical and
operational data of the row house are applied to this building. A
schematic layout is presented in Fig. 2.
These typologies were selected because they also dene different thermal behaviours related to the building geometry, as
inferred from Table 4. Both houses have similar roof surfaces compared to the total surface area, but the roof surface of the row house
is about half of the detached house. These congurations imply a
higher roof solar gain ratio in the detached house. Conversely this
house has a wide open geometry (S/V ratio is 0.95 m1 ) with consequent high thermal losses whenever the indoor air temperature
is higher than outdoor and vice versa. The row house has a more
compact geometry, hence tends to maintain the heat in the inside.
These differences imply different cooling and heating loads proles
during the year.
3.3. The roof technologies
Several studies proved the impact of the thermal insulation and
the thermal mass on the heat balance of the roof and, more in
general, of the building envelope [3337]. Even if these properties affect the energy performance of buildings during the cooling
and heating season, this study is, conversely, focused on assessing
the energy performance of residential buildings when existing at
roofs are equipped with UHI mitigation techniques.
It is worth noting that both green roofs and cool roofs can be
implemented using different solutions, corresponding to different
thermo-physical and biological parameters. In order to consider
a limited number of variables, the selected roofs were dened
according to good quality standards related to green and cool roofs.
The following describes the selected roof solutions:
ST Conventional roof standard product with thermal insulation dened in Table 3; the external layer has solar
reectance 0.25, typical for most construction materials,
and thermal emittance 0.9.

70

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

Fig. 3. Not-insulated row house energy performance for different roof solutions in Palermo.

CR White cool roof same layers of the previous roof nished with an elastomeric white coating having solar
reectance 0.8 and thermal emittance 0.9.
CR low-e Metallic reective coating as above with a metallic layer
with solar reectance 0.65 and thermal emittance 0.4.
GR Green roof, whose main properties are summarised in
Table 5. The structure of the green roof is more complex.
The adopted solution is dened according to typical values of the green roof design in Mediterranean area. It is
important reminding that the green roof model implemented in Energy Plus, is mono-dimensional add based
on several assumptions. To be noted that the roof insulation thickness is lower than conventional roof insulation,
to take into account the insulation effect of the green roof.
It is also noted that, according to typical green roof systems, the green area correspond to 80% of the total roof
surface, the remaining is surface dedicated to footpaths,
here assumed to be made of concrete.

4. Results
Results refer to the set of simulations carried out including the variables above dened. A rst set of calculation was
made to evaluate the performance of the green roof under different moisture conditions. Following the energy simulation results
are presented, considering the green roof performances under
effective rainfall. The results presented the demand for heating, cooling and energy, the latter expressed as simple sum of
the two energy uses. The calculations refer to the net energy
demand, without considering the energy systems efciencies, in
order to focus on the envelope behaviour. The same simulations
are presented in free oating conditions, in order to evaluate
the improvements of comfort conditions using cool and green
roofs.

Fig. 4. Not-insulated row house energy performance for different roof solutions in Barcelona.

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

71

Fig. 5. Not-insulated row house energy performance for different roof solutions in Cairo.

4.1. Impact of the green roof water content on the energy

comparison
Conventional and cool roofs are static elements, whose performances do not change according to the climatic conditions.
Green roofs are different: climatic conditions affect the water
content and, as a consequence, the cooling and heating performances. This rst set of simulations refers to the not insulated

dwellings only. Figs. 35 report the results for the three localities
and include the heating and cooling net demand for six different congurations: ST, CR, CR low-e, GR with actual rainfall, GR
dry, GR wet. The latter condition implies that the roof is continually watered and the soil reached the maximum humidity content.
This condition can be reached only under the hypothesis of a well
designed irrigation system, with additional costs and resources
uses that are not part of this investigation. The bars in the gure

Table 6
Calculation results: heating, cooling and total energy demand, energy savings compared to the standard roof (the rst column includes: locality insulation level roof
technique).
Cooling
(kWh/m2 /y)

Energy
(kWh/m2 /y)

Sav. to ST (%)

Heating
(kWh/m2 /y)
Detached house

Cooling
(kWh/m2 /y)

Energy
(kWh/m2 /y)

Sav. to ST (%)

City and roof

technique

Heating
(kWh/m2 /y)
Row house

Bar-ins-ST
Bar-ins-CR
Bar-ins-CR
low-e
Bar-ins-GR
Bar-not ins-ST
Bar-not ins-CR
Bar-not ins-CR
low-e
Bar-not ins-GR
Pal-ins-ST
Pal-ins-CR
Pal-ins-CR
low-e
Pal-ins-GR
Pal-not ins-ST
Pal-not ins-CR
Pal-not ins-CR
low-e
Pal-not ins-GR
Cai-ins-ST
Cai-ins-CR
Cai-ins-CR
low-e
Cai-ins-GR
Cai-not ins-ST
Cai-not ins-CR
Cai-not ins-CR
low-e
Cai-not ins-GR

23.2
27.3
22.3

4.2
1.5
3.2

27.4
28.9
25.5

0.0
5.3
6.8

33.2
40.6
33.3

8.4
3.2
6.5

41.6
43.8
39.8

0.0
5.4
4.3

21.3
44.1
54.2
44.3

4.4
6.0
0.9
3.7

25.7
50.1
55.1
48.0

6.2
0.0
10.0
4.2

32.5
71.9
89.6
71.9

5.9
7.9
1.1
4.8

38.3
79.8
90.7
76.7

7.8
0.0
13.7
3.8

38.9
8.2
10.9
8.4

5.4
10.1
4.7
8.2

44.3
18.3
15.6
16.6

11.6
0.0
14.9
9.1

68.5
11.4
16.0
11.7

3.6
19.2
9.5
16.0

72.1
30.6
25.5
27.6

9.6
0.0
16.7
9.7

7.0
18.6
25.5
19.1

10.7
14.4
3.6
10.1

17.7
32.9
29.1
29.2

3.2
0.0
11.7
11.4

11.4
30.4
42.6
31.1

18.0
20.7
5.8
14.8

29.4
51.1
48.4
45.9

3.8
0.0
5.3
10.2

15.6
3.2
4.9
3.3

13.8
19.7
10.7
17.0

29.3
23.0
15.6
20.3

10.8
0.0
32.2
11.5

29.9
4.1
6.9
4.2

16.8
37.0
22.2
32.4

46.6
41.1
29.1
36.6

8.9
0.0
29.0
10.9

2.6
8.4
13.3
8.6

19.9
27.7
7.7
21.6

22.5
36.1
21.0
30.2

2.0
0.0
41.7
16.4

3.2
13.0
21.8
13.1

36.7
44.2
18.2
35.5

39.9
57.2
40.0
48.5

2.8
0.0
30.1
15.1

6.6

24.8

31.4

13.0

10.0

39.2

49.2

13.9

72

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

Fig. 6. Heating, cooling and total energy demand of the row house. The different building congurations, specied on the x-axis as a function of: locality insulation level
roof technique.

represent the cooling, heating and total net energy demand. This
set of calculation is performed in order to show the impact of water
content on the energy performance of buildings and to stress the
importance of irrigation strategies in order to optimise the GR performance respect to other static cool techniques. The results are
expressed as percentage reduction respect to the standard roof
performances.
The results obtained for Palermo, Fig. 3, show that the best
performances are obtained with the green roof always wet, with
heating demand comparable with the conventional roof, but with
the cooling demand reduced by more than the half. The total energy
savings are 24% with this conguration. CR reduces the cooling
demand by 75%, but the increase of the heating demand lowers

the total energy savings close to 12%. The metallic cool roof
produces global energy savings closer to CR, due to heating performances similar to ST and improved cooling performances. The
green roof has similar performances in dry or actual rainfall conditions. Energy savings are around 11% and are obtained in winter,
thanks to the higher insulation level of the roof, but low improvements are reached in the cooling season, because of the limited
advantage of the dry vegetation layer.
Barcelona has a cooler climate and this impact the performances of the different roong systems in a different way respect
to Palermo; see results in Fig. 4. The not insulated envelope induces
a high heating demand. The best result is obtained by the dry GR,
because of the insulation effect produced by the soil layer in winter

Fig. 7. Heating, cooling and total energy demand of the detached house. The different building congurations, specied on the x-axis as a function of: locality insulation
level roof technique.

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

73

Fig. 8. Operative and ambient air temperature proles in the not insulated detached house in July. The operative temperatures are presented for the for different roof
techniques: ST, CR, CR low-e, GR.

and the natural water content in summer. The total energy demand
is reduced by 14%. Slightly worse are the performances of the
green roof with effective rainfall or continuously wet, with energy
savings between 10% and 11.6%. According to the above considerations, CR registers the worst performance, with a 10% total energy
increase. The low emittance cool roof improves the overall energy
performance of the building, because of the limited heating penalties and the 38% of the cooling demand reduction.
The results of the simulation for Cairo, Fig. 5, show the climate
dependence of the south Mediterranean area. Heating demand is
low and the best solutions are cooling efciency driven. CR lead to
40%, while moderated advantages are calculated for the metallic
roof High differences are found for the green roof congurations:
the wet green roof is the best performing solution, thanks to a

cooling demand similar to the cool roof and a heating demand

slightly higher than the conventional roof. This conguration leads
to 45% total energy savings. Actual rainfall and dry GR performances
practically give the same performance with 13% energy savings
respect to ST, with a 10% reduction of the cooling demand.

4.2. Building energy comparison for the different roong

techniques
This section presents the results of the different roong solutions: ST, CR, CR low-e and GR, with the water content determined
by the rainfall only. The complete set of results is presented in
Table 6, where heating, cooling and total net energy demand are

74

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

Table 7
Cumulative distribution of the number of hours exceeding three reference operative temperatures.
Roof system

hours > 26 C []
Insulated

ST
CR
CR-low-e
GR

Barcelona
794
366
659
428

ST
CR
CR-low-e
GR

Palermo
1992
1333
1790
1571

ST
CR
CR-low-e
GR

Cairo
3635
3133
3520
3289

hours > 28 C []

hours > 30 C []

hours >26 []
Not insulated

hours >28 C []

hours >30 C []

68
0
26
5

0
0
0
0

931
246
727
495

227
0
60
9

0
0
0
0

759
281
599
428

24
0
0
0

2165
1101
1865
1487

997
177
677
345

215
0
47
0

2393
1530
2156
1713

875
299
654
420

3658
2806
3494
3403

2545
1217
2213
1858

1222
214
806
525

summarised. The relative energy savings compared to the standard

roof are also presented for each building conguration.
Fig. 6 presents the results of the row house for Barcelona,
Palermo and Cairo. Barcelona is the cooler among the three selected
cities and this affects the nal results. The most efcient solution for
the insulated conguration is the metallic cool roof a total energy
saving close to 7%; slightly lower savings (6.2%) are calculated for
GR. Green roof is the most effective technique for the not insulated
conguration, with an energy demand reduction of close to 12%;
due to the extra insulation of the soil and vegetation layer. Cool
roof are by far the best cooling efcient technique for both congurations, with almost no need of cooling systems for the two
congurations. The impact on the energy demand is negative for
the two congurations, because of the increase of heating demand
due to the reduced solar gains through the roof. CR is the most
performing solution for the insulated row house in Palermo, with
global savings close to 15%. The CR low-e application gives a 10%
energy savings, while limited advantages are obtained with GR.
Very close results, around 11% of energy savings, are obtained for
the three roof solutions for the not insulated conguration. The
energy reduction is all in cooling mode for CR, while metallic cool
and green roof reduce both, the heating and the cooling demand.
Cairo is cooling dominated and CR gives the best results in summer
and in the whole year, with a reduction of total energy demand
between 30% and 40% for the two building congurations. CR lowe and GR ensure energy savings between 11% and 18%, except for
GR for the insulated conguration, since the extra insulation does
not lead to cooling reduction in hot climates.
The detached house presents slightly different results compared
to the row house and they are shown in Fig. 7. Because of the higher
surface exposed to the outdoor environment, this house typology is
characterised by higher heating and cooling energy demands. The
heating demand is even more predominant in Barcelona. Green roof
reduce both, the heating and cooling demand, with annual savings
of about 810% respect to ST, for the insulated and not insulated
conguration. The CR low-e ensures small energy savings on annual
basis, around 4% for both congurations. The impact of CR is negative on yearly basis but cooling savings between 60% and 85% are
calculated. Different results are obtained for the Palermo and Cairo
calculations; the high reecting roof techniques ensure in all the
situations a total net energy saving of 17% and 10% respectively
for the insulated and not insulated congurations. GR reduces the
heating and the cooling demand with a 10% energy savings for the
not insulated conguration, while similar results are obtained by
the CR low-e for the insulated conguration. Cairo is characterised
by a cooling dominated climate and CR has the best performances

with a 30% reduction of the annual net energy demand. Signicant

savings are obtained also by CR low-e and GR, the latter only for the
not insulated conguration, with annual energy savings between
10% and 15%.
It is worth noting that some cross checks demonstrated that
the specic results, intended as energy demand normalised to the
building square meters, can be applied also to the last oor of
multi-storey dwellings with good accuracy. This aspect is of relevance when assessing the energy saving potentials of this building
typology, often used in the densely built urban area.

4.3. Thermal comfort in not cooled buildings

This section analyse the evolution of thermal comfort conditions in not cooled buildings as a function of the adopted roof
solution. The operative temperature is selected as relevant indicator, the most signicant to express the indoor thermal comfort.
The qualitative impact of the different roofs can be inferred from
Fig. 8, reporting the operative temperature proles of 4 days in July
for the not insulated detached house in the 3 selected localities.
The most effective solutions is given by CR, while CR low-e and GR
improve the indoor conditions with similar thermal proles. It is
worth repeating that with actual rainfall, GR will be dry for a significant part of the cooling period. Table 7 summarises the cumulative
distribution of the hours with operative temperatures above 26, 28
and 30 C. The results are presented as an average of the row and
detached houses, in order to present data in a more compact way.
The section presents the results of the different roong solutions:
ST, CR, CR low-e and GR, with the water content determined by the
rainfall only.
The impact of CR is by far the most effective for the improvement
of summer thermal comfort. The hours with operative temperature
higher than 26 C in Barcelona are reduced to 26% and 46% of the
ST hours, for the insulated and not insulated congurations. The
metallic cool roof reduce the number of hours of about 20% for both
congurations; while the hours above 26 C are halved respect to ST
when GR is used in both congurations. The number of hours above
28 C is negligible for CR and GR, while it is strongly reduced with
the low emittance cool roof. No hours with operative temperature
above 30 C are calculated.
The application of the three advanced roof techniques is signicant in Palermo and Cairo, even if these climatic conditions strongly
increase the operative temperature levels and the number of discomfort hours. Cool roofs have a noticeable impact in reducing the
number of hours with operative temperatures above 26 C, while

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

the metallic CR and the green roof cause a reduction lower than
10% respect to ST.
The number hours with operative temperature higher than 28 C
are reduced of 73% and 82% respect to ST for CR applications in
Palermo; this happens for the insulated and not insulated congurations. The low emittance cool roof reduces of about the 20%
the hours above 28 C, while the reduction increases to about 28%
for the green roof. The three roong systems reduce to negligible
numbers the hours above 30 C.
The hours with an operative temperature higher than 28 C in
Cairo are reduced to 48% and 64% respect to ST when applying
CR, for the insulated and not insulated congurations. Reductions
around 10% and 28% are calculated for respectively CR low-e and
GR. The number of hours with operative temperatures higher than
30 C is signicant in Cairo and the passive techniques can improve
the indoor comfort conditions. The number of hours is strongly
reduced using cool roofs (between 18% and 34% respect to ST),
while GR reduce to the half that numbers. Moderate advantage
is achieved with the metallic cool roof, with a number of hours
reducing between 66% and 75%.
To be noted that being the green roof almost dry in Palermo and
Cairo, most of the peculiarities of the green roof are not working.
The benets of the evapo-transpiration of the vegetation layer are
party lost because of the extra insulation soil layer, which tends
keeping the heat stored inside the building.

5. Conclusions
This study presented a comparison among different roong
techniques able to reduce the cooling demand of residential buildings while mitigating the urban heat island. The analysis is carried
out using a validated tool; hence the results acceptance goes along
with the model accuracy. Even if energy optimisation strategies
of the roof cannot prevent from taking into account the thermal
insulation and thermal mass, the results show that the mitigation strategies of the urban heat island, currently planned by the
metropolitan area authorities, con positively impact the energy
performance of dwellings on annual basis. The upgrade of conventional hot roong systems has net energy advantages, especially
considering the new insulation standards adopted throughout the
European Mediterranean countries.
Cool roof are very effective for the cooling and (excluding the
northern area of the basin) energy savings. Cool roofs are the most
effective solutions for the centre and southern areas of the Mediterranean basin. Not insulated house might have excessive increase
in heating demand but, on the other side, cool roofs practically
may avoid the installation of the cooling systems, because of the
very low cooling energy demand. Low emittance cool roofs perform worse than cool roofs, because of the reduced radiative losses
at night time, but improve the performance of conventional roofs.
For the same reason, metallic cool roofs have also limited heating
penalties respect to conventional cool roofs. They might represent
an acceptable compromise in the coolest Mediterranean area.
Green roofs are very difcult to be modelled and correctly
inputted in calculation tools, because of the high number of variables which enter into the heat transfer mechanisms and because
of a general lack of information related to the input data required
by the adopted model. The study highlighted a rst very important issue: green roofs performances strongly depend on the water
content of the systems with the adopted model. A well wet green
roof has good cooling performance, but relaying on the rainfall does
not ensure effective energy performances during the dry Mediterranean hot season, especially in the centre and the south east of the
basin. Green roofs improve the heating performances as well, when
compared with the conventional roofs. The limited water content

75

avoids permanent wet conditions of the soil layer, which actively

increase the thermal resistance of the structure. The dryer the roof,
the lower is the heating demand. Water management need to be
calibrated according to the climate conditions and the main energy
use.
The variability of green roofs as a function of many variable
makes it clear that a denitive comparison among the selected
techniques will require in-depth analyses taking into account,
besides the energy issue, other important aspects: water management and demand, life cycle analysis and costs, environment
impacts on urban comfort and on the urban heat island mitigation.
Acknowledgments
This work was carried out in the framework of the project Cool
Roofs, contract number EIE/07/475/SI2.499428, supported by the
Intelligent Energy Europe (IEE) program SAVE 2007.
References
[1] IEA, World Energy Outlook 20082009, International Energy Agency, Geneva,
2009.
[2] L. Dav, C. Giampaglia, First medlink report: a crossed look on reports and international statistics about development, gender, freedom, conicts and mobility
in Mediterraneum, www.medlinknet.org, 2007.
[3] L. Howard, The climate of London, vol. IIII, Harvey and Dorton, London, 1883.
[4] H.E. Landsberg, The Urban Climate, in: International Geographic Series, vol. 28,
Academic Press, New York, 1981.
[5] H. Takebayashi, M. Moriyama, Surface heat budget on green roof and high
reection roof for mitigation of urban heat island, Building and Environment
42 (8) (2007) 29712979.
[6] M. Kolokotroni, I. Giannitsaris, R. Watkins, The effect of the London urban heat
island on building summer cooling demand and night ventilation strategies,
Solar Energy 80 (4) (2006) 383392.
[7] M. Santamouris, Heat island research in Europe the state of the art, Advances
Building Energy Research 1 (2007) 23150.
[8] H. Taha, S.C. Chang, H. Akbari, Meteorological and air quality impacts of heat
island mitigation measures in three U.S. Cities, Lawrence Berkeley National
Laboratory Report LBNL- 44222, Berkeley, CA, 2000.
[9] M.L. Imhoff, P. Zhang, R.E. Wolfe, L. Bounoua, Remote sensing of the urban
heat island effect across biomes in the continental USA, Remote Sensing of
Environment 114 (2010) 504513.
[10] J.P. Montavez, A. Rodriguez, J.I. Jimenez, A study of the urban heat island of
Granada, International Journal of Climatology 20 (2000) 899911.
[11] H. Akbari, S.L. Rose, H. Taha, Analyzing the land cover of an urban environment
using high-resolution orthophotos, Landscape and Urban Planning 63 (2003)
114.
[12] A. Synnefa, A. Dandou, M. Santamouris, M. Tombrou, N. Soulakellis, On the
use of cool materials as a heat island mitigation strategy, Journal of Applied
Meteorology and Climatology 47 (2008) 28462856.
[13] H. Akbari, S. Konopacki, Calculating energy-saving potentials of heat-island
reduction strategies, Energy Policy 33 (2005) 721756.
[14] H. Akbari, S. Menon, A. Rosenfeld, Global cooling: increasing world-wide urban
albedos to offset CO2 , Climatic Change 94 (2009) 275286.
[15] J. Yang, Q. Yu, P. Gong, Quantifying air pollution removal by green roofs in
Chicago, Atmospheric Environment 42 (31) (2008) 72667273.
[16] W. Ishida, Overview of the world air-conditioning market, Appliances
http://www.appliancemagazine.com/editorial.php?article=
Magazine.com,
1812&zone=207&rst=1, 2007.
[17] A. Synnefa, M. Santamouris, H. Akbari, Estimating the effect of using cool coatings on energy loads and thermal comfort in residential buildings in various
climatic conditions, Energy and Buildings 39 (11) (2007) 11671174.
[18] H. Akbari, S. Bretz, D. Kurn, H. Hartford, Peak power and cooling energy savings
of high albedo roofs, Energy and Buildings 25 (1997) 117126, H.
[19] A. Shariah, B. Shalabi, A. Rousan, B. Tashtoush, Effects of absorptance of external
surfaces on heating and cooling loads of residential buildings in Jordan, Energy
Conversion and Management 39 (1998) 273284.
[20] C.K. Cheung, R.J. Fuller, M.B. Luther, Energy efcient envelope design for high
rise apartments, Energy and Buildings 37 (1) (2005) 3748.
[21] H. Suehrcke, E.L. Peterson, N. Selby, Effect of roof solar reectance on the building heat gain in a hot climate, Energy and Buildings 40 (2008) 22242235.
[22] H. Akbari, R. Levinson, L. Rainer, Monitoring the energy-use effects of cool
roofs on California commercial buildings, Energy and Buildings 37 (2005)
10071016.
[23] M. Zinzi, G. Fasano, Properties and performance of advanced reective paints
to reduce the cooling loads in buildings and mitigate the heat island effect in
urban areas, International Journal of Sustainable Energy 28 (1) (2009) 123139.
[24] M. Santamouris, C. Pavlou, P. Doukas, G. Mihalakakou, A. Synnefa, A. Hatzibiros,
Investigating and analyzing the energy and environmental performance of an

76

[25]

[26]
[27]

[28]

[29]
[30]

M. Zinzi, S. Agnoli / Energy and Buildings 55 (2012) 6676

experimental green roof system installed in a nursery school building in Athens,
Greece, Energy 32 (9) (2007) 17811788.
R. Fioretti, A. Palla, L.G. Lanza, P. Principi, Green roof energy and water related
performance in the Mediterranean climate, Building and Environment 45 (8)
(2010) 18901904.
R.M. Lazzarin, F. Castellotti, F. Busato, Experimental measurements and numerical modelling of a green roof, Energy and Building 37 (12) (2005) 12601267.
S. Parizotto, R. Lamberts, Investigation of green roof thermal performance in
temperate climate: A case study of an experimental building in Florianpolis
city, Southern Brazil, Energy and Buildings 43 (7) (2011) 17121722.
R. Kumar, S.C. Kaushik, Performance evaluation of green roof and shading
for thermal protection of buildings, Building and Environment 40 (11) (2005)
15051511.
C.Y. Jim, S.W. Tsang, Modeling the heat diffusion process in the abiotic layers
of green roofs, Energy and Buildings 43 (6) (2011) 13411350.
T. Emilsson, Vegetation development on extensive vegetated green roofs:
inuence of substrate composition, establishment method and species mix,
Ecological Engineering 33 (34) (2008) 265277.

[31] D.J Sailor, A green roof model for building energy simulation programs, Energy
and Buildings 40 (2008) 14661478.
[32] DOE, Getting Started with EnergyPlusEssential Information You Need about
Running EnergyPlus, U.S. Department of Energy, 2007.
[33] R. Lollini, B. Barozzi, G. Fasano, I. Meroni, M. Zinzi, Optimisation of opaque
components of the building envelope. Energy, economic and environmental
issues, Building and Environment 40 (2006) 10011013.
[34] R.U. Halwatura, M.T.R. Jayasinghe, Inuence of insulated roof slabs on air conditioned spaces in tropical climatic conditionsa life cycle cost approach, Energy
and Buildings 41 (2009) 678686.
[35] A. Hasan, Optimizing insulation thickness for buildings using life cycle cost,
Applied Energy 63 (1999) 115124.
[36] N. Sisman, E. Kahya, N. Aras, H. Aras, Determination of optimum insulation
thickness of the external walls and roof (ceiling) for Turkeys different degree
day regions, Energy Policy 35 (2007) 51515155.
[37] M. DOrazio, C. Di Perna, E. Di Giuseppe, The effects of roof covering on the thermal performance of highly insulated roofs in Mediterranean climates, Energy
and Buildings 42 (10) (2010) 16191627.