Thermal Performance of Green Roofs:
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
Thermal Performance of Green Roofs: A Case Study in a Tropical Region Caroline Santana de Morais1 and Maurício Roriz2 1
Architect, MSc Student in Civil Construction Department of Civil Engineering, Federal University of Sao Carlos, Brazil Washington Luís, SP-310, Km 235, São Carlos, São Paulo Tel: 55-16-260-8262 E-mail: [email protected] 2
PhD, MSc, Architect, Adjunct Professor, Post-Graduation Program in Civil Construction Department of Civil Engineering, Federal University of Sao Carlos, Brazil Washington Luís, SP-310, Km 235, São Carlos, São Paulo Tel: 55-16-260-8262 E-mail: [email protected]
ABSTRACT: Caretakers and others employees of apartment building usually have to stay during long periods of time in shelters that do not offer minimum conditions of thermal comfort. The terrible urban violence induces many designers to conceive these spaces as kinds of warlike shelters, almost without ventilation and with an excess of glass windows that, without appropriate protection, still heat even more the indoor climate. However, in countless examples, the low thermal performance of roofs originates the main source of the uncomfortable conditions. Some of these roofs are simply constituted of a flagstone, without any sun-protection. There are many serious cases, wherein the flagstone is made waterproof using asphalt, in which absortivity reaches the possible maximum. This article shows a case study carried out in the summer, wherein the performances of the roofs of two shelters were compared, one in dark common flagstone and another protected by vegetation. The advantages of the green roofs confirmed the theories on the subject. Conference Topic: Appropriate technologies, Materials and building techniques, Case studies Keywords: Thermal Comfort, Green Roofs, Energy
1. INTRODUCTION The growing concentration of built area and impermeable paving, associated with the concentration of heat and pollution sources, worsen the already precarious environmental condition of urban centers. The proportions between green areas and built areas are becoming more and more distant from the recommended international standards. Therefore the designers need to find alternatives to improve the urban life quality. Increasing the green areas is one of these alternatives because vegetation has low cost and is an extremely efficient option to improve the urban climate and to reduce the energy waste when cooling or heating the buildings. The roofs are the most exposed part of the building envelopes to the solar radiation. Many roof systems are developed to minimize the energy flows through the roof [1]. The vegetation applied to the roofs can protect them from the direct solar radiation and even reduce their temperatures, through evaporative cooling effect [2]. In hot climates, like it is in a large part of Brazilian territory, it can be taken advantage from this effect by reducing the heat that
penetrates indoor. Some studies about garden roofs have been carried out in others countries that have severe summers as well. In Singapore, Wong et al [3] analysed green roofs effects on the energy consumption of commercial buildings and they showed a reduction by 15% in the annual consumption. In Brazil, Pouey et al [4] compared outdoor surface temperatures between a green roof and a terrace, the results showed 35,9 ºC in the roof with vegetation against 48,9 ºC in the terrace. This paper compares the summer thermal performance of two roof systems used in entrances of buildings, one built with vegetation and another bare roof, without any protection against solar radiation. Both buildings are located in São Carlos city, state of São Paulo, Brazil.
2. THE GREEN ROOF SYSTEMS Green roofs, garden roofs or ecological roofs are all terminologies to roof systems with vegetation. According to theirs building typology, they can be classified in intensive or extensive.
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
The intensive green roof is characterized by: layers of soil greater than 20cm; plants, bushes and trees of medium and large bearing; structure reinforced (due to the shipments between 700 and 1200kg/m²) and rigorous maintenance [5]. The extensive green roof adopts a thinner layer of soil, with thickness between 8 and 12cm and equivalent shipment to 100kg/m², small-scale plants, as the autochthones. In this case, small-scaled plants are used due to their performance, wherein low or even none maintenance is required [5]. In descending order, the main building materials used in the layers of green roofs are: vegetation (depending on the climate, kind of soil and structure bears); soil (bears solid, inert or not, unlike soil or from the natural land); filter membrane (a blanket geotextil of 150g/m²); drainage layer (of main importance for the system, constituted by break, expanded clay or rolled pebble); water retention layer (absorbent panels of synthetic stuff); thermal insulation layer (chosen in function of the thermal transmittance); impermeable membrane (asphalt or synthetic paints) and the structural support.
tile shards of 5cm, filter blanket, vegetable soil and two species of plants, Duranta repens and Asparagus densiflorus. Table 1: Terminology adopted in this study STS: Superficial thermal stability coefficient ATI: Average temperature increment coefficient DHMax: Thermal time-lag of maximum temperatures ∆Sup: Surface temperatures amplitude ∆Tbs: Outdoor air temperatures amplitude FATH: Timetable factor of temperature adjustment FATHS: Timetable factor of surface temperature adjustment FATHT: Timetable factor of air temperature adjustment HsupMax: Time of maximum surface temperature HtbsMax: Time of maximum air temperature RH: Relative Humidity (%) SupHS: Timetable surface temperatures SupMax: Maximum surface temperature SupAvg: Average surface temperature SupMin: Minimum surface temperature TbsH: Timetable value of air temperatures TbsMax: Maximum air temperature TbsAvg: Average of air temperature TbsMin: Minimum air temperature
4. FIELD MEASUREMENTS Figure 1: View from the ordinary roof
Field measurements were recorded by using a HOBO electronic data logger system, manufactured by Onset Computer Corporation. It was registered outdoor air temperature, relative humidity and indoor surface temperature of the slabs data, in open th th surrounding, during 7 days, from February 14 to 20 in 2003. Table 2: Embrapa station climate between the 14 and 20th February 2003.
th
Day TbsAvg(°C) TbsMax(°C) TbsMin(°C) RH
Figure 2: View from the green roof
3. THE ROOFS STUDIED The ordinary roof (Fig. 1), with area of 7.22 m², is part of a residential entrance in a condominium and it is built in solid concrete slab (thickness: 20cm) with impermeable layer of asphalt paint. The green roof (Fig. 2), with area of 4.44 m², integrates the entrance of a security company building and is built in solid concrete slab (thickness: 10cm) impermeable layer, drainage layer of pottery
14 15 16 17 18 19 20
22.5 23.5 24.0 21.6 24.5 23.5 25.1
26.0 28.0 29.0 23.2 29.0 29.4 29.6
19.0 19.0 19.0 20.0 20.0 17.5 20.5
96 99 97 100 94 97 92
Table 2 shows the measured climate in an adjacent climatological station of Embrapa [6], the Agriculture Brazilian Company, during the same days. Table 3 shows a summary of the temperature data of the two roofs, during February 17th to 20th.
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
o Table 3: Summary of the temperature data ( C) in the th th two roofs (February 17 to 20 )
17
18
19
20
22.2 23.7 20.7 3.0 23.1 24.4 21.7 2.7
24.0 27.0 21.0 6.0 24.7 27.8 21.7 6.1
26.2 32.8 19.5 13.3 28.5 37.3 19.8 17.5
27.2 32.3 22.2 10.1 30.0 36.0 24.0 12.0
22.1 23.5 20.7 2.8
23.6 26.7 20.6 6.1
24.6 30.6 18.7 11.9
25.5 29.7 21.3 8.4
SupAvg
22.5
23.3
24.2
25.3
SupMax
23.2
25.1
27.9
27.5
SupMin
21.7
21.6
20.6
23.1
∆Sup = Max-Min
1.5
3.5
7.3
4.4
Ordinary roof TbsMax TbsAvg TbsMin ∆Tbs = Max-Min SupAvg SupMax SupMin ∆Sup = Max-Min Green roof TbsMax TbsAvg TbsMin ∆Tbs = Max-Min
In figures 3 and 4, empty points indicate air temperatures and dark points indicate surface temperatures from the concrete slab indoor face. The values schedules of these temperatures are arithmetic average from that recorded in each 10 minutes. The gray spot indicates the interval between maximum and minimum air temperatures observed at climatological station of Embrapa (Table 2).
5. DATA ANALYSIS AND DISCUSSION The recorded differences between the outdoor air temperatures in both measured places confirm conclusions of other studies about the existence of several microclimates in a same city. These differences also affect the indoor temperatures and complicate the comparison of the thermal effects caused directly by the roofs. For this reason, a method was proposed to foresee which would be the internal surface temperatures, if both buildings were submitted to a same microclima. 5.1 Performance indicators Firstly, it was defined two performance indicators for the roofs, one regarding the thermal stability of the slabs (STS: superficial thermal stability coefficient), and other to the average temperatures values (ATI: average temperature increment coefficient). STS = ∆Sup /∆ Tbs ATI = (SupAvg - TbsAvg) / ∆Tbs
[ Eq. 1 ] [ Eq. 2 ]
Comparing the values of STS and ATI for the two roof systems the following values were achieved: STS ordinary roof = 1.17 ATI ordinary roof = 0.24
Figure 3: Hourly average of the temperatures measured in the ordinary roof.
STS green roof = 0.55 ATI green roof = 0.01
According to these values, the green roof presented better thermal performance. 5.2 Occurrence Time of Maximum Temperatures Table 4: Time-lags in maximum roofs temperatures Ordinary roof HSupMax HTbsMax ∆HMax Green roof HSupMax HTbsMax ∆HMax
17
18
19
20
Average
20 16 4
18 17 1
18 17 1
19 17 2
18.6 16.3 2.3
19 17 2
18 17 1
18 16 2
20 15 5
18.4 16.7 1.7
Figure 4: Hourly average temperatures measured in the green roof.
Obs: Being considered only entire hours, the timelags (∆HMax) are rounded in 2 hours.
The figures 3 and 4 show the evolution of the outdoor air temperatures and internal surface temperatures of the slabs, during four days of th th measurement (February 17 to 20 ).
The characteristic thermal time-lag of each roof system can be taken regarding the occurrences of the maximums outdoor air temperatures and the internal faces of these roofs. Being HSupMax to hour of occurrence of the maxims surface temperatures and
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
HTbsMax to occurrence of the maximums air temperatures, this time-lag (∆HMax) is done by equation 3. Table 4 illustrates the thermal time-lag in maximum roofs temperatures. ∆HMax = HSupMax-HTbsMax
[ Eq. 3 ]
5.3 Typical temperatures in a summer day (February) in São Carlos Table 5: Typical February temperatures in São Carlos. Average values between 1961 and 1990. TbsMax TbsMin TbsAvg = (TbsMax+TbsMin) / 2 ∆Tbs = TbsMax-TbsMin
(°C) 27.2 17.7 22.5 9.5
The typical curve of the air temperature variation (TBS), for the same month, was identified through the analysis of the scheduled data of the climate from the city. This curve was defined through the timetable factor of temperature adjustment (FAT), calculated for each hour (H) of the day: FATH = (TBSH-TbsMin) / ∆Tbs
[ Eq. 4 ]
In cases when only the maximum and minimum temperatures are known, through the equation 5 it is possible to calculate the TBS hourly values. The temperatures presented in table 6 are resultants from this equation and from the typical climatic month data. TBSH = TbsMin + FATH (∆Tbs)
[ Eq. 05 ]
5.4 Estimates of the roofs performance in a typical summer day (February) The surface temperatures in the lower faces slabs can be estimated by the following procedures: A) The hourly oscillation curves were obtained by the application of the timetable factor of temperature adjustment (FAT), considering the time-lags observed in the superficial maximums (∆HMax) (Equation 6). FATHS = FATHT
[ Eq. 6 ]
Wherein HS is the surface temperature occurrence hour (varying between 0 and 24) and HT the air temperature occurrence hour corresponding to time-lag from the maximum surface temperature. (HT = HS - ∆HMax). Negative values are corrected adding them to 24. B) The amplitudes of the surface temperatures variation (∆Sup) were calculated from the superficial thermal stability coefficient (STS) and from the air temperatures amplitudes (∆Tbs) (Equation 7). ∆Sup = STS . ∆Tbs
[ Eq. 7 ]
According to February climatological typical data from the city, the average air temperature amplitude is 9.5°C. Applying the equation 7 to the values of STS and ∆Tbs are found the results showed in the table 6. Table 6: Amplitudes of the surface temperatures oscillation Ordinary roof 1.174 9.5°C 11.15°C
STS ∆Tbs ∆Sup
Green roof 0.553 9.5°C 5.25°C
C) The average surface temperatures were estimated through the application of the respective average temperature increment coefficient (ATI) and this also allowed to calculate the minimums surface temperatures (Equations 8 and 9). Table 7 shows the average and the minimum surface temperatures. SupAvg = TbsAvg + (ATI . DTbs)
[ Eq. 8 ]
SupMin = SupAvg – (DSup/2)
[ Eq. 9 ]
Table 7: Average and minimum surface temperatures TbsAvg ATI ∆Tbs SupAvg ∆sup/2 SupMin
Ordinary roof 22.5°C 0.242 9.5°C 24.8°C 5.58 19.2°C
Green roof 22.5°C 0.017 9.5°C 22.7°C 2.63 20.0°C
D) Finally, it was obtained the values of the surface temperatures, according to the Equation 10: SupHS = SupMin + (FATHS . ∆Sup)
[ Eq. 10 ]
Tables 8 and 9 illustrate the surface temperatures on the internal faces of slab on the ordinary roof and in on the green roof, respectively. Table 8: Surface temperatures on the internal face of slab in ordinary roof Hour 0 1 2 3 4 5 6 7 8 9 10 11
(SupH) (°C) 24.4 23.2 22.4 21.3 20.4 19.8 19.3 19.2 19.4 20.1 21.5 23.3
Hour 12 13 14 15 16 17 18 19 20 21 22 23
(SupH) (°C) 25.6 27.8 29.3 30.1 30.4 30.2 29.8 28.9 28.3 27.3 26.5 25.3
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
Table 9: Surface temperatures on the internal face of slab in the green roof Hour 0 1 2 3 4 5 6 7 8 9 10 11
(SupH) (°C) 22.5 21.9 21.5 21.0 20.6 20.3 20.1 20.0 20.1 20.4 21.1 22.0
Hour 12 13 14 15 16 17 18 19 20 21 22 23
(SupH) (°C) 23.0 24.1 24.8 25.1 25.3 25.2 25.0 24.6 24.3 23.8 23.4 22.9
are slightly smaller than that one of the green cover. However, at 4 pm the internal surface of the ordinary roof reaches 3.7°C over the TBS and this difference continues increasing practically until the end of the day, reaching at 9 o'clock the maximum value of 4.4°C.
6. CONCLUSIONS Figure 5 indicates some differences observed between the outdoor temperatures of the two buildings. These differences confirm conclusions of other studies about the existence of several microclimates in the same city and how they also affect the indoor temperatures, hindering ( not allowing) to compare the thermal effects caused directly by the roofs. In order to solve this problem, a method was proposed that allows to foresee what would be the indoor temperatures when both constructions were submitted to the same microclimate.
Figure 5: Differences between outdoor air temperatures in the two roofs The figures 6 and 7 show that during the day the ordinary roof surface temperatures accumulate (s?)65.2 degrees-hour of warmth over the outdoor air temperatures, while the green roof accumulates less than 40% of this value. Furthermore, in hotter schedules of the day, the green roof contributed to cool down the environment, because its surface temperatures kept lower temperatures compared to the outdoor air temperatures since the 9 to 6 pm. The bigger difference, 3.3°C, happened at noon, when the surface temperature was 23.0°C and TBS reaches to 26.3°C. Due to the higher amplitude caused by the ordinary roof, in the beginning of the morning (between 5 and 8 o'clock), its surface temperatures
Figure 6: Differences between surface and outdoor air temperatures in ordinary roof
Figure 7: Differences between surface and outdoor air temperatures in a green roof Hence it is possible to conclude that the garden roofs are economically and technically viable alternatives for the local climate, not only the building type studied, but also for wider possibilities. Apart from these aspects, the environmental benefits from green roofs systems overstep the building itself and cover the whole area around the same. It improves the air quality and allows the integration and harmony between vegetation and the built areas. Surprisingly, the green roof focused on this study is the only one in São Carlos while the bare roofs, without any thermal protection, are widely used.
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
7. REFERENCES [1]
[2]
[3]
[4]
[5]
[6]
MACHADO, M.V.; BRITTO, C.; NEILA, F.J. (1999) Thermal behaviour simulation of models with ecological roofs. . In: XVIII PLEA – International Conference on Passive and Low Energy Architecture. Brisbane, University of Queensland, Austrália, September 22-24. DEL BARRIO, Elena Palomo.(1998) Analysis of the green roofs cooling potential in buildings. Energy in Buildings, v.27, nº.2, p. 179-193, 1998.] WONG, N. H. et al.(2003). The effects of rooftop garden on energy consumption of a commercial building in Singapore. Energy and Buildings, v.35, nº. 4, p. 353-364, 2003. POUEY, M. T.F.et.al.(1998). Coberturas verdes: analise de desempenho térmico. In: VII Encontro Nacional de Tecnologia do Ambiente Construído. Florianópolis. Anais. NPC/UFSC. p. 473-481. CORREA, Celina B.; GONZÁLEZ, Neila F. Javier.(2002). O uso de coberturas ecológicas na restauração de coberturas planas. In: Núcleo de Pesquisa em Tecnologia da Arquitetura e Urbanismo. São Paulo. Anais: FUPAM/FAUUSP. p.686-696. EMBRAPA (2003) “Dados Climáticos Diários da Estação Agrometeorológica do Centro de Pesquisa de Pecuária do Sudeste” Embrapa São Carlos SP. [www.cppse.embrapa.br/].
ACKNOWLEDGEMENT The authors thank FAPESP, "Fundação de Amparo à Pesquisa do Estado de São Paulo", for the support granted to the development of this research.
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
Thermal Performance of Green Roofs: A Case Study in a Tropical Region Caroline Santana de Morais1 and Maurício Roriz2 1
Architect, MSc Student in Civil Construction Department of Civil Engineering, Federal University of Sao Carlos, Brazil Washington Luís, SP-310, Km 235, São Carlos, São Paulo Tel: 55-16-260-8262 E-mail: [email protected] 2
PhD, MSc, Architect, Adjunct Professor, Post-Graduation Program in Civil Construction Department of Civil Engineering, Federal University of Sao Carlos, Brazil Washington Luís, SP-310, Km 235, São Carlos, São Paulo Tel: 55-16-260-8262 E-mail: [email protected]
ABSTRACT: Caretakers and others employees of apartment building usually have to stay during long periods of time in shelters that do not offer minimum conditions of thermal comfort. The terrible urban violence induces many designers to conceive these spaces as kinds of warlike shelters, almost without ventilation and with an excess of glass windows that, without appropriate protection, still heat even more the indoor climate. However, in countless examples, the low thermal performance of roofs originates the main source of the uncomfortable conditions. Some of these roofs are simply constituted of a flagstone, without any sun-protection. There are many serious cases, wherein the flagstone is made waterproof using asphalt, in which absortivity reaches the possible maximum. This article shows a case study carried out in the summer, wherein the performances of the roofs of two shelters were compared, one in dark common flagstone and another protected by vegetation. The advantages of the green roofs confirmed the theories on the subject. Conference Topic: Appropriate technologies, Materials and building techniques, Case studies Keywords: Thermal Comfort, Green Roofs, Energy
1. INTRODUCTION The growing concentration of built area and impermeable paving, associated with the concentration of heat and pollution sources, worsen the already precarious environmental condition of urban centers. The proportions between green areas and built areas are becoming more and more distant from the recommended international standards. Therefore the designers need to find alternatives to improve the urban life quality. Increasing the green areas is one of these alternatives because vegetation has low cost and is an extremely efficient option to improve the urban climate and to reduce the energy waste when cooling or heating the buildings. The roofs are the most exposed part of the building envelopes to the solar radiation. Many roof systems are developed to minimize the energy flows through the roof [1]. The vegetation applied to the roofs can protect them from the direct solar radiation and even reduce their temperatures, through evaporative cooling effect [2]. In hot climates, like it is in a large part of Brazilian territory, it can be taken advantage from this effect by reducing the heat that
penetrates indoor. Some studies about garden roofs have been carried out in others countries that have severe summers as well. In Singapore, Wong et al [3] analysed green roofs effects on the energy consumption of commercial buildings and they showed a reduction by 15% in the annual consumption. In Brazil, Pouey et al [4] compared outdoor surface temperatures between a green roof and a terrace, the results showed 35,9 ºC in the roof with vegetation against 48,9 ºC in the terrace. This paper compares the summer thermal performance of two roof systems used in entrances of buildings, one built with vegetation and another bare roof, without any protection against solar radiation. Both buildings are located in São Carlos city, state of São Paulo, Brazil.
2. THE GREEN ROOF SYSTEMS Green roofs, garden roofs or ecological roofs are all terminologies to roof systems with vegetation. According to theirs building typology, they can be classified in intensive or extensive.
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
The intensive green roof is characterized by: layers of soil greater than 20cm; plants, bushes and trees of medium and large bearing; structure reinforced (due to the shipments between 700 and 1200kg/m²) and rigorous maintenance [5]. The extensive green roof adopts a thinner layer of soil, with thickness between 8 and 12cm and equivalent shipment to 100kg/m², small-scale plants, as the autochthones. In this case, small-scaled plants are used due to their performance, wherein low or even none maintenance is required [5]. In descending order, the main building materials used in the layers of green roofs are: vegetation (depending on the climate, kind of soil and structure bears); soil (bears solid, inert or not, unlike soil or from the natural land); filter membrane (a blanket geotextil of 150g/m²); drainage layer (of main importance for the system, constituted by break, expanded clay or rolled pebble); water retention layer (absorbent panels of synthetic stuff); thermal insulation layer (chosen in function of the thermal transmittance); impermeable membrane (asphalt or synthetic paints) and the structural support.
tile shards of 5cm, filter blanket, vegetable soil and two species of plants, Duranta repens and Asparagus densiflorus. Table 1: Terminology adopted in this study STS: Superficial thermal stability coefficient ATI: Average temperature increment coefficient DHMax: Thermal time-lag of maximum temperatures ∆Sup: Surface temperatures amplitude ∆Tbs: Outdoor air temperatures amplitude FATH: Timetable factor of temperature adjustment FATHS: Timetable factor of surface temperature adjustment FATHT: Timetable factor of air temperature adjustment HsupMax: Time of maximum surface temperature HtbsMax: Time of maximum air temperature RH: Relative Humidity (%) SupHS: Timetable surface temperatures SupMax: Maximum surface temperature SupAvg: Average surface temperature SupMin: Minimum surface temperature TbsH: Timetable value of air temperatures TbsMax: Maximum air temperature TbsAvg: Average of air temperature TbsMin: Minimum air temperature
4. FIELD MEASUREMENTS Figure 1: View from the ordinary roof
Field measurements were recorded by using a HOBO electronic data logger system, manufactured by Onset Computer Corporation. It was registered outdoor air temperature, relative humidity and indoor surface temperature of the slabs data, in open th th surrounding, during 7 days, from February 14 to 20 in 2003. Table 2: Embrapa station climate between the 14 and 20th February 2003.
th
Day TbsAvg(°C) TbsMax(°C) TbsMin(°C) RH
Figure 2: View from the green roof
3. THE ROOFS STUDIED The ordinary roof (Fig. 1), with area of 7.22 m², is part of a residential entrance in a condominium and it is built in solid concrete slab (thickness: 20cm) with impermeable layer of asphalt paint. The green roof (Fig. 2), with area of 4.44 m², integrates the entrance of a security company building and is built in solid concrete slab (thickness: 10cm) impermeable layer, drainage layer of pottery
14 15 16 17 18 19 20
22.5 23.5 24.0 21.6 24.5 23.5 25.1
26.0 28.0 29.0 23.2 29.0 29.4 29.6
19.0 19.0 19.0 20.0 20.0 17.5 20.5
96 99 97 100 94 97 92
Table 2 shows the measured climate in an adjacent climatological station of Embrapa [6], the Agriculture Brazilian Company, during the same days. Table 3 shows a summary of the temperature data of the two roofs, during February 17th to 20th.
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
o Table 3: Summary of the temperature data ( C) in the th th two roofs (February 17 to 20 )
17
18
19
20
22.2 23.7 20.7 3.0 23.1 24.4 21.7 2.7
24.0 27.0 21.0 6.0 24.7 27.8 21.7 6.1
26.2 32.8 19.5 13.3 28.5 37.3 19.8 17.5
27.2 32.3 22.2 10.1 30.0 36.0 24.0 12.0
22.1 23.5 20.7 2.8
23.6 26.7 20.6 6.1
24.6 30.6 18.7 11.9
25.5 29.7 21.3 8.4
SupAvg
22.5
23.3
24.2
25.3
SupMax
23.2
25.1
27.9
27.5
SupMin
21.7
21.6
20.6
23.1
∆Sup = Max-Min
1.5
3.5
7.3
4.4
Ordinary roof TbsMax TbsAvg TbsMin ∆Tbs = Max-Min SupAvg SupMax SupMin ∆Sup = Max-Min Green roof TbsMax TbsAvg TbsMin ∆Tbs = Max-Min
In figures 3 and 4, empty points indicate air temperatures and dark points indicate surface temperatures from the concrete slab indoor face. The values schedules of these temperatures are arithmetic average from that recorded in each 10 minutes. The gray spot indicates the interval between maximum and minimum air temperatures observed at climatological station of Embrapa (Table 2).
5. DATA ANALYSIS AND DISCUSSION The recorded differences between the outdoor air temperatures in both measured places confirm conclusions of other studies about the existence of several microclimates in a same city. These differences also affect the indoor temperatures and complicate the comparison of the thermal effects caused directly by the roofs. For this reason, a method was proposed to foresee which would be the internal surface temperatures, if both buildings were submitted to a same microclima. 5.1 Performance indicators Firstly, it was defined two performance indicators for the roofs, one regarding the thermal stability of the slabs (STS: superficial thermal stability coefficient), and other to the average temperatures values (ATI: average temperature increment coefficient). STS = ∆Sup /∆ Tbs ATI = (SupAvg - TbsAvg) / ∆Tbs
[ Eq. 1 ] [ Eq. 2 ]
Comparing the values of STS and ATI for the two roof systems the following values were achieved: STS ordinary roof = 1.17 ATI ordinary roof = 0.24
Figure 3: Hourly average of the temperatures measured in the ordinary roof.
STS green roof = 0.55 ATI green roof = 0.01
According to these values, the green roof presented better thermal performance. 5.2 Occurrence Time of Maximum Temperatures Table 4: Time-lags in maximum roofs temperatures Ordinary roof HSupMax HTbsMax ∆HMax Green roof HSupMax HTbsMax ∆HMax
17
18
19
20
Average
20 16 4
18 17 1
18 17 1
19 17 2
18.6 16.3 2.3
19 17 2
18 17 1
18 16 2
20 15 5
18.4 16.7 1.7
Figure 4: Hourly average temperatures measured in the green roof.
Obs: Being considered only entire hours, the timelags (∆HMax) are rounded in 2 hours.
The figures 3 and 4 show the evolution of the outdoor air temperatures and internal surface temperatures of the slabs, during four days of th th measurement (February 17 to 20 ).
The characteristic thermal time-lag of each roof system can be taken regarding the occurrences of the maximums outdoor air temperatures and the internal faces of these roofs. Being HSupMax to hour of occurrence of the maxims surface temperatures and
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
HTbsMax to occurrence of the maximums air temperatures, this time-lag (∆HMax) is done by equation 3. Table 4 illustrates the thermal time-lag in maximum roofs temperatures. ∆HMax = HSupMax-HTbsMax
[ Eq. 3 ]
5.3 Typical temperatures in a summer day (February) in São Carlos Table 5: Typical February temperatures in São Carlos. Average values between 1961 and 1990. TbsMax TbsMin TbsAvg = (TbsMax+TbsMin) / 2 ∆Tbs = TbsMax-TbsMin
(°C) 27.2 17.7 22.5 9.5
The typical curve of the air temperature variation (TBS), for the same month, was identified through the analysis of the scheduled data of the climate from the city. This curve was defined through the timetable factor of temperature adjustment (FAT), calculated for each hour (H) of the day: FATH = (TBSH-TbsMin) / ∆Tbs
[ Eq. 4 ]
In cases when only the maximum and minimum temperatures are known, through the equation 5 it is possible to calculate the TBS hourly values. The temperatures presented in table 6 are resultants from this equation and from the typical climatic month data. TBSH = TbsMin + FATH (∆Tbs)
[ Eq. 05 ]
5.4 Estimates of the roofs performance in a typical summer day (February) The surface temperatures in the lower faces slabs can be estimated by the following procedures: A) The hourly oscillation curves were obtained by the application of the timetable factor of temperature adjustment (FAT), considering the time-lags observed in the superficial maximums (∆HMax) (Equation 6). FATHS = FATHT
[ Eq. 6 ]
Wherein HS is the surface temperature occurrence hour (varying between 0 and 24) and HT the air temperature occurrence hour corresponding to time-lag from the maximum surface temperature. (HT = HS - ∆HMax). Negative values are corrected adding them to 24. B) The amplitudes of the surface temperatures variation (∆Sup) were calculated from the superficial thermal stability coefficient (STS) and from the air temperatures amplitudes (∆Tbs) (Equation 7). ∆Sup = STS . ∆Tbs
[ Eq. 7 ]
According to February climatological typical data from the city, the average air temperature amplitude is 9.5°C. Applying the equation 7 to the values of STS and ∆Tbs are found the results showed in the table 6. Table 6: Amplitudes of the surface temperatures oscillation Ordinary roof 1.174 9.5°C 11.15°C
STS ∆Tbs ∆Sup
Green roof 0.553 9.5°C 5.25°C
C) The average surface temperatures were estimated through the application of the respective average temperature increment coefficient (ATI) and this also allowed to calculate the minimums surface temperatures (Equations 8 and 9). Table 7 shows the average and the minimum surface temperatures. SupAvg = TbsAvg + (ATI . DTbs)
[ Eq. 8 ]
SupMin = SupAvg – (DSup/2)
[ Eq. 9 ]
Table 7: Average and minimum surface temperatures TbsAvg ATI ∆Tbs SupAvg ∆sup/2 SupMin
Ordinary roof 22.5°C 0.242 9.5°C 24.8°C 5.58 19.2°C
Green roof 22.5°C 0.017 9.5°C 22.7°C 2.63 20.0°C
D) Finally, it was obtained the values of the surface temperatures, according to the Equation 10: SupHS = SupMin + (FATHS . ∆Sup)
[ Eq. 10 ]
Tables 8 and 9 illustrate the surface temperatures on the internal faces of slab on the ordinary roof and in on the green roof, respectively. Table 8: Surface temperatures on the internal face of slab in ordinary roof Hour 0 1 2 3 4 5 6 7 8 9 10 11
(SupH) (°C) 24.4 23.2 22.4 21.3 20.4 19.8 19.3 19.2 19.4 20.1 21.5 23.3
Hour 12 13 14 15 16 17 18 19 20 21 22 23
(SupH) (°C) 25.6 27.8 29.3 30.1 30.4 30.2 29.8 28.9 28.3 27.3 26.5 25.3
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
Table 9: Surface temperatures on the internal face of slab in the green roof Hour 0 1 2 3 4 5 6 7 8 9 10 11
(SupH) (°C) 22.5 21.9 21.5 21.0 20.6 20.3 20.1 20.0 20.1 20.4 21.1 22.0
Hour 12 13 14 15 16 17 18 19 20 21 22 23
(SupH) (°C) 23.0 24.1 24.8 25.1 25.3 25.2 25.0 24.6 24.3 23.8 23.4 22.9
are slightly smaller than that one of the green cover. However, at 4 pm the internal surface of the ordinary roof reaches 3.7°C over the TBS and this difference continues increasing practically until the end of the day, reaching at 9 o'clock the maximum value of 4.4°C.
6. CONCLUSIONS Figure 5 indicates some differences observed between the outdoor temperatures of the two buildings. These differences confirm conclusions of other studies about the existence of several microclimates in the same city and how they also affect the indoor temperatures, hindering ( not allowing) to compare the thermal effects caused directly by the roofs. In order to solve this problem, a method was proposed that allows to foresee what would be the indoor temperatures when both constructions were submitted to the same microclimate.
Figure 5: Differences between outdoor air temperatures in the two roofs The figures 6 and 7 show that during the day the ordinary roof surface temperatures accumulate (s?)65.2 degrees-hour of warmth over the outdoor air temperatures, while the green roof accumulates less than 40% of this value. Furthermore, in hotter schedules of the day, the green roof contributed to cool down the environment, because its surface temperatures kept lower temperatures compared to the outdoor air temperatures since the 9 to 6 pm. The bigger difference, 3.3°C, happened at noon, when the surface temperature was 23.0°C and TBS reaches to 26.3°C. Due to the higher amplitude caused by the ordinary roof, in the beginning of the morning (between 5 and 8 o'clock), its surface temperatures
Figure 6: Differences between surface and outdoor air temperatures in ordinary roof
Figure 7: Differences between surface and outdoor air temperatures in a green roof Hence it is possible to conclude that the garden roofs are economically and technically viable alternatives for the local climate, not only the building type studied, but also for wider possibilities. Apart from these aspects, the environmental benefits from green roofs systems overstep the building itself and cover the whole area around the same. It improves the air quality and allows the integration and harmony between vegetation and the built areas. Surprisingly, the green roof focused on this study is the only one in São Carlos while the bare roofs, without any thermal protection, are widely used.
th
PLEA 2003 - The 20 Conference on Passive and Low Energy Architecture, Santiago – CHILE, 9 - 12 November 2003
7. REFERENCES [1]
[2]
[3]
[4]
[5]
[6]
MACHADO, M.V.; BRITTO, C.; NEILA, F.J. (1999) Thermal behaviour simulation of models with ecological roofs. . In: XVIII PLEA – International Conference on Passive and Low Energy Architecture. Brisbane, University of Queensland, Austrália, September 22-24. DEL BARRIO, Elena Palomo.(1998) Analysis of the green roofs cooling potential in buildings. Energy in Buildings, v.27, nº.2, p. 179-193, 1998.] WONG, N. H. et al.(2003). The effects of rooftop garden on energy consumption of a commercial building in Singapore. Energy and Buildings, v.35, nº. 4, p. 353-364, 2003. POUEY, M. T.F.et.al.(1998). Coberturas verdes: analise de desempenho térmico. In: VII Encontro Nacional de Tecnologia do Ambiente Construído. Florianópolis. Anais. NPC/UFSC. p. 473-481. CORREA, Celina B.; GONZÁLEZ, Neila F. Javier.(2002). O uso de coberturas ecológicas na restauração de coberturas planas. In: Núcleo de Pesquisa em Tecnologia da Arquitetura e Urbanismo. São Paulo. Anais: FUPAM/FAUUSP. p.686-696. EMBRAPA (2003) “Dados Climáticos Diários da Estação Agrometeorológica do Centro de Pesquisa de Pecuária do Sudeste” Embrapa São Carlos SP. [www.cppse.embrapa.br/].
ACKNOWLEDGEMENT The authors thank FAPESP, "Fundação de Amparo à Pesquisa do Estado de São Paulo", for the support granted to the development of this research.
