Summertime thermal and energy performance of a double-skin green facade: A case study in Shanghai

Summertime thermal and energy performance of a double-skin green facade: A case study in Shanghai

Sustainable Cities and Society 39 (2018) 43–51 Contents lists available at ScienceDirect Sustainable Cities and Society journal homepage: www.elsevi...

3MB Sizes 0 Downloads 4 Views

Sustainable Cities and Society 39 (2018) 43–51

Contents lists available at ScienceDirect

Sustainable Cities and Society journal homepage:

Summertime thermal and energy performance of a double-skin green facade: A case study in Shanghai


Feng Yanga,b, , Feng Yuana,b, Feng Qiana,b, Zhi Zhuanga,b, Jiawei Yaoa,b a b

Department of Architecture, College of Architecture and Urban Planning, Tongji University, Shanghai 200092, China Key Laboratory of Ecology and Energy-Saving Study of Dense Habitat, (Tongji University), Ministry of Education, Shanghai 200092, China



Keywords: Vertical greening system Double skin green facade Thermal comfort Building orientation Shading

Vertical greening system(VGS) for buildings is drawing more and more research, because that it provides passive cooling for both indoors and outdoors without occupying valuable urban land, and thus can be a remedy for the deteriorating thermal environment of today’s high density urban areas. This paper reports an investigation on the summertime cooling performance of double-skin green facade (DSGF) of a recently-renovated 5-storey administrative building in a university campus in Shanghai. The results show that, the VGS created a distinctive microclimate in the cavity: air temperature Ta averagely dropped by about 0.4 °C in a daily circle, and maximally 5.5 °C for the Southern facade; and by 0.2 °C in average and 3.3 °C maximally for the Northern facade. The mean exterior surface temperature Ts reduction of the Southern facade by the VGS is 1.5 °C and maximally ∼9 °C, whereas the mean interior Ts reduction is 1.2 °C and maximally 2 °C. The corresponding figures of the Northern facade is 0.5 °C, 4.2 °C, 0.5 °C and 1.3 °C, respectively. The indoor thermal improvement by the VGS, evaluated by operative temperature (Top), is 1.1 °C averagely and 2.7 °C maximally on the South-facing office, and 0.6 °C averagely and 1.9 °C maximally on the North-facing office. The initial findings suggest clear potential of VGS in thermal comfort improvement and cooling energy saving, further study is needed to evaluate the yearly energy and comfort effects of VGS on radiation, convective and conduction heat transfer and on daylighting through building envelop, for comprehensive energy assessment considering cooling, heating and artificial lighting.

1. Introduction Urban green space plays an important role in maintaining a good urban thermal and ecological environment. In high-density Asian megacities such as Shanghai, it is a challenge to maintain a reasonable development density while enhance the ecological service of urban green space. Integration of greenery with urban buildings can contribute to partially restoration of lost green land (Xue, Gou, & Lau, 2017). Vertical greening system (VGS) emerges as a promising strategy in high-density urban development, because it provides valuable “green space” for dense urbanized area and associated environmental, social and economical benefits without occupying much land area (Dunnett, 2004). There are studies demonstrating the benefits of VGS with respect to urban heat island mitigation (Afshari, 2017), microclimate modification (Ip, Lam, & Miller, 2010) and improving pedestrian thermal comfort (Morakinyo, Lai, Lau, & Ng, 2018). VGS is also considered a potential strategy to improving building thermal and energy performance. Building consumes about one third of the overall energy use by human society (International Energy Agency,

2013), thus building energy efficiency is important in coping with global climate change, regional and urban warming (urban heat island) and deteriorating urban environment (Baker and Steemer, 2000; Lechner, 2015). As urbanization in China is entering a new phase of the so-called “New Normal”, in many cities, new construction is on the decrease whereas more and more existing buildings require renovation to meet up-to-date energy codes and indoor environmental quality standards. In fact, compared with developed countries, the overall environmental and energy performance of building stock in Chinese cities is far from satisfactory (Tu, 2011). Under the hot-humid subtropical climate of Shanghai, VGS is not only an attractive face-lift surgery for aged buildings but could also remediate heat flux through building envelope, improving thermal comfort and decreasing cooling load and energy consumption (Wong et al., 2010). Although there are controversial opinions on various forms of VGS regarding costs, complexity and aesthetic value, deeper understanding on its sustainable application requires more systematic research on the tangible and intangible benefits that VGS brings to the urban society (Riley, 2017), and continuous evaluation of VGS environmental

Corresponding author at: Department of Architecture, College of Architecture and Urban Planning, Tongji University, Shanghai 200092, China. E-mail address: [email protected] (F. Yang). Received 8 November 2017; Received in revised form 22 January 2018; Accepted 28 January 2018 Available online 08 February 2018 2210-6707/ © 2018 Elsevier Ltd. All rights reserved.

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

walls had reduced wind speed immediately adjacent to the facade, ranging 0–43% depending on facade orientation and presence of nearby obstructions, and the ivy layers were estimated to averagely reduce 10% in heat flux through opaque walls, and reduce 4–12% in air infiltration rates per unit wall area (Susorova, Azimi, & Stephens, 2014). An experimental study investigated the key traits contributing to cooling effects of vine plant traditional green facades, and identifies leaf percentage coverage and leaf solar transmittance as the two key traits (Koyama, Yoshinaga, Hayashi, Maeda, & Yamauchi, 2013). A study monitored a direct green facade in Nanjing China and found out that the cooling effect is linearly related to percentage of green coverage while plant thickness and volume of green facade were power function distributions (Yin et al., 2017). A two-year experiment with South-facing green facades under the Mediterranean climate indicates a yearround thermal- mediating effect from the greenery, and that the highest cooling occurs with ambient wind speed of 3–4 m/s, relative humidity of 30–60% and solar radiation higher than 800 W/m2 (Vox, Blanco, & Schettini, 2018). A numerical simulation study also supports that leaf density and thickness, indicated by leaf area index and shading coefficient, improve building thermal insulation and reduce air-conditioning electricity consumption (Wong, Tan, Tan, & Wong, 2009). Larsen et al. analyzed the DSGF’s thermal regulating mechanisms involving evapotranspiration process, and proposes a simplified method to simulation DSGF with standard transient simulation software (Flores Larsen et al., 2015). Susorova et al. developed a mathematical model for a climber-covered exterior wall. The model takes into account plant stomata resistance in transpiration process (Susorova, Angulo, Bahrami, & Brent, 2013). A comparative study on energy performances between LWS and DSGF in Spain shows a 58% cooling electricity reduction by LWS and 34% by DSGF (Coma et al., 2017). In China, an experiment examined the impacts of a number of design parameters on a Westfacing living wall system in Wuhan, and tested sensitivities of cavity depth and cavity ventilation rates with the cooling effect during hothumid summer months (Chen, Li, & Liu, 2013). He et al. developed a coupled heat and moisture transfer model of a living wall system and validated it by field experiments in Shanghai (He et al., 2017). The above two experiments employed rooftop test chambers made of light structure (e.g. thin foam sandwich panel in the case of Shanghai), therefore the measured envelope and indoor thermal condition can be affected by the low heat capacity, and differ with that of normal buildings. A theoretical analysis on DSGF on high-rise residential buildings in Hong Kong indicates a total cooling energy savings of up to 76% (Wong and Baldwin, 2016). The simplified calculation aims at understanding the potential energy savings and feasibility of applying DSGF to residential buildings at the urban scale. At the building scale, more detailed examination on transient heat transfer process would be necessary in energy savings analysis. Previous research has indicated that for the cities in the Northern hemisphere, the cooling efficiency of VGS on indoor environment can be profound for West-oriented facades in summer (Kontoleon and Eumorfopoulou, 2010), and North-oriented facades in winter (He et al., 2017). However, the dominant orientation of residential and small (< 20,000 M2 in TFA) non-residential buildings in Southern Chinese cities is South-North. This is due to the wide application of passive design to utilize solar heating and wind-driven indoor ventilation (Yang, Lau, & Qian, 2011). Therefore, when considering the total thermal and energy benefits of building stock in Southern Chinese cities, performance of VGS on South and North facades deserve equal, if not prioritized, research effort. Previous research investigated various types of VGS in cities mostly located in Europe and North America. There are very few studies that reported site-monitored summertime cooling performance of DSGF on real-world buildings in Asian megacities with a hot-humid Sub-tropical climate. The present paper reports an empirical study on a recentlyrenovated office building in a university campus in Shanghai, the largest city in China. During summer months, the thermal and

performance in comparison to other construction solutions could lead to increased application of VGS and reduced costs (Manso and CastroGomes, 2015). Based on growing substrate and support system, VGS can be classified into green facades and living walls (Pérez, Coma, Martorell, & Cabeza, 2014). Living walls are vegetation grown out of usually-modularized panels attached to building facades either directly or by supportive structures. Green facades can be further classified into traditional green facade, in which climber plants are attached to existing building walls, and double-skin green facade (DSGF), in which climber plants are supported by an added frame to the original building wall and act as a new skin of the old building. From the perspective of building physics, VGS moderates the indoor thermal environment by manipulating the radiative, convective and conductive heat transfer processes (Eumorfopoulou and Kontoleon, 2009). Kerez et al. summarize the four fundamental mechanisms of VGS as a passive building design system: shading, insulation, evapotranspiration and variation of wind effect (Pérez, Rincón, Vila, González, & Cabeza, 2011). Shading is the most significant approach to reduce indoor heat gain by intercepting direct and diffuse (which can be significant in humid climates) solar radiation. Factors influencing the shading effect of VGS include leaf density (e.g., leaf area index), leaf solar transmittance, leaf area coverage etc. (Raji, Tenpierik, & van den Dobbelsteen, 2015). But VGS differs with other man-made shading materials, in that vegetation is a living component, and its evapotranspiration diverts a portion of sensible heat into latent heat flux, therefore leaf surface temperature is reduced compared to metal shading devices, and so is the radiative heat transfer between leaves and walls and leaves and interior (through windows and openings) (Hoelscher, Nehls, Jänicke, & Wessolek, 2016). Factors influencing evapotranspiration performance of VGS include climatic condition, type and exposure of plant, substrate moisture content, leaf stomata characteristics, etc. (Pérez et al., 2011; Davis and Hirmer, 2015). To model the cooling mechanism of evapotranspiration by vegetation is not straightforward (Flores Larsen, Filippín, & Lesino, 2015), A study in Berlin indicates that the cooling effect of direct green facades (climbers on brick wall) in summer was contributed mainly by shading in clear days but transpiration became more prominent in cloudy days (Hoelscher et al., 2016). A energy balance analysis on a living wall system under Shanghai climate indicates that evapotranspiration can account for more than 50% of heat release in summer (He, Yu, Ozaki, Dong, & Zheng, 2017). Insulation effect of green facade is associated with foliage density and air change rate of the buffer space in between wall and greenery (Pérez et al., 2011). An experimental study in the UK using brick cuboids array suggests 21–37% heating energy savings achievable by applying direct green facade (Hedera helix) (Cameron, Taylor, & Emmett, 2015). For living walls, heat capacity and thermal resistance of the substrate should also be taken into account, and dependent on configuration, location and season, the thermal insulation is variable (Tudiwer and Korjenic, 2017; He et al., 2017). When designing a VGS, modification on the nearby air movement also needs careful consideration to optimize the energy implication. Air speed within the cavity between wall and plant has strong influences on the convective heat transfer coefficient of the wall (Ottelé, Perini, Fraaij, Haas, & Raiteri, 2011). In general, plant species (deciduous or evergreen) with regard to facade orientation and spacing between foliage and wall should be taken into account, to encourage summertime air circulation while reduce wintertime wind speed. A comparative study indicates that, in cold climates, direct green facade and living wall system are more effective wind barriers compared to indirect (double skin) green facade system. And for the DSGF to effectively stagnant near-wall air movement, the cavity depth should be minimized, e.g., 40–60 mm (Perini, Ottelé, Fraaij, Haas, & Raiteri, 2011). A series of summertime field measurements on four pairs of walls of different plant coverage and orientations has found that, the ivy-covered 44

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

Fig. 1. Site plan and perspective of the building under investigation.

used in this reference station. The North pair was measured from August 04 to August 18, and the south pair was measured from August 18 to September 1, 2015. During the measurement, the windows and doors were kept closed and all electrical devices shut down. As it was summer vacation, the rooms were not occupied during the measurement, and so they were in free running mode, i.e. no air-conditioning or internal heat gain.

biometeorological parameters of DSGF-covered rooms were measured against reference rooms, and the thermal comfort and energy saving potentials of DSGF were estimated. The impact of weather and design parameters of the DSGF on its thermal performance are analyzed in order to inform better climate-responsive building design and renovation. 2. Methodology

3. Analysis and results The field-investigation was carried out on the summertime cooling performance of DSGF on a recently-renovated 5-storey administrative building in a university campus in Shanghai (Figs. 1 and 2). Two pairs of office rooms on the 4th floor (∼4.2 m × 3.6 m × 3.6 m in L-W-H) were monitored. Each pair features one DSGF and one normal masonry facade (reference) with the same facade orientation. The building was constructed in the 1980s. The wall is 300 mm in width, constructed of brick and mortar, with 20 mm cement plaster, 240 mm brick, and 40 mm granitic plaster, from interior to exterior. The DSGF is planted with an evergreen plant called Mucuna sempervirens ‘Hemsl’. One pair faces Southwest South and other faces Northeast North. The leaf area index is 3.0-3.5 for both the South and North green facade, calculated using the sampling method suggested by Liu and Chen (2004). Each pair was continuously monitored for around fifteen days. Measured thermal parameters include surface temperatures (Ts) of outer vegetated skin of the DSGF and masonry facade, cavity air temperature (Ta) and relative humidity(RH) in between VGS and masonry facade, and indoor thermal comfort conditions, including Ta, RH, wind velocity (WV) and globe temperature (Tg). Table 1 lists the specification of the equipment. Fig. 3 shows the structure of the DSGF and placement of sensors. The reference weather station is mounted on the rooftop of a nearby mid-rise building in the same campus. Hobo weather station is

3.1. Time series of DSGF-induced bioclimatic development For the South pair, data collected from August 26 to September 1 is used for analysis, excluding influences from adverse climate conditions such as rains (when the thermal effect of the VGS was largely diminished). During the period of South pair measurement, the weather condition was mainly sunny and partially cloudy (Fig. 4a). The mean Ta is 27.2 °C, and the highest recorded Ta is 33.3 °C. The highest recorded global solar radiation (GSR) is 873 W/m2. The Mean WV is 0.4 m/s, and highest recorded WV is 3.6 m/s. The ambient air temperature (Ta) of the cavity between the VGS and Southern building facade (with VGS) is lower than that near the bare Southern facade (w/o VGS), by averagely 0.4 °C and maximally 5.5 °C during the measurement period (Fig. 4b). The relative humidity (RH) with VGS is higher than that w/o VGS, by averagely 3.6% and maximally 22.3% during the measurement period. The surface temperature (Ts) of the Southern facade with VGS is shown in Fig. 4c. In average, Ts of exterior VGS is 0.4 °C higher than that of the masonry wall behind. The exterior surface of VGS is warmer than exterior surface of the masonry wall during daytime hours, by maximally 9.2 °C, but cooler in the evening and at night, by maximally

Fig. 2. The facade with DSGF and the original bare facade (the South pair).


Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

Table 1 Instrument specification Model



Operating Range

Sensor locations

Temperature/RH Smart Sensor:

Ta, RH

± 0.2 °C (0–50 °C);

−40 °C to 75 °C; RH ≤ 95%

DSGF cavity, indoor


± 2.5% RH (10–90%) ± 5% of reading or ± 0.05 m/s (15–35 °C)

0–10 m/s



± 0.2 °C (0–50 °C)

−40 °C to 100 °C



± 0.7 °C

−260 °C to 1370 °C

Wind velocity sensor (with Temp & RH sensors): Testo 0635 1535

Ta, RH, WV

± 0.3 °C; ± 2%RH(+2 ∼ 98%RH); ± 0.03 m/s

Globe temperature sensor: Testo 0602 0743


−20 ∼ +70 °C 0 ∼ +100%RH 0 ∼ +20m/s 0 ∼ +120 °C

Surfaces of DSGF and walls Indoor

Hobo S-THB-M002 Wind velocity sensor: Cambridge Accusense sensor T-DCIF900-S-P Temperature smart sensor: Hobo S-TMB- M002 (in a 40 mm matt-grey vinyl ball) Hobo UX120-014M data logger with K-type thermocouple

−2.1 °C. Southern facade wall Ts with and without VGS is compared in Fig. 4d. The mean exterior surface temperature reduction by VGS during the whole measurement period is 1.6 °C, while the maximum cooling effect reached 8.7 °C during the day, but the effect can be reversed at night under certain weather conditions (cloudy, humid or rainy), when the VGS-covered facade can be maximally 0.9 °C warmer than that without VGS. For the interior wall surface, the facade covered with VGS is consistently cooler than that without VGS, by averagely 1.2 °C during the measurement period, and maximally 2 °C, occurred in late afternoon hours. Indoor thermal parameters including Ta, Tg and RH is plotted for comparison (Fig. 4e), while WV is not plotted because the air was generally stable, with measured WV consistently under 0.1 m/s in both rooms. In both rooms, Tg is close to Ta during most of the time, although Tg can be a little higher than Ta during the day while slightly lower at night. Ta is constantly lower in the VGS-covered room, by averagely 1.2 °C and maximally 2.5 °C; Tg largely follows the Ta pattern, with VGS-covered room averagely lowered by 1.1 °C and maximally 2.7 °C, occurred in afternoon hours. RH is consistently higher in the VGS-covered room, by 5% on average and 9% maximum. Indoor thermal comfort is evaluated using operative temperature (Top). The VGS-covered room is consistently cooler than the room without VGS. The mean Top reduction is 1.1 °C and the maximum is 2.7 °C. The North pair analysis adopts data measured from August 13 to August 17, 2015. During the period, the weather condition was mainly partially cloudy and it was one shower of rain during August 16 (Fig. 4f). The mean Ta is 28.3 °C, and the highest recorded Ta is 33.5 °C. The highest recorded GSR is 896 W/m2. The Mean WV is 0.4 m/s, and highest recorded WV is 3.3 m/s. The Ta between the VGS and northern building facade is lower than that near the bare Southern facade during the day but higher at night (Fig. 4g). Overall, VGS made the cavity Ta averagely 0.2 °C lower than bare masonry facade, maximally 3.3 °C occurred in early afternoon. The relative humidity (RH) is on the opposite: RH with VGS is higher during the day but lower at night. Averagely VGS resulted in a little lower RH than that w/o VGS, by 0.6%, and the maximum difference is 9.2%. The Ts pattern of Northern facade with VGS is different to that of the southern facade (Fig. 4h). Ts of exterior VGS almost consistently remained lower than that of the exterior wall surface. During early daytime, the difference of Ts between VGS and masonry wall was very small, with the wall slightly warmer than the VGS. After 3 p.m. till next sunrise, the Ts of leaf surface dropped clearly lower than the VGS interior frame surface and the exterior wall. Ts difference between VGS and exterior wall surface is in average 0.5 °C during the measurement period, with maximum 1.6 °C at night. The Ts comparison between Northern-facing VGS and bare masonry facade (Fig. 4i) reveals that, for exterior wall surface, the VGS-covered facade were clearly cooler than that of bare facade during daytime, the


maximum Ts difference is 4.2 °C. The difference became smaller at night, and the VGS-covered facade became warmer than bare facade, the maximum difference can reach 1.2 °C at night. During the whole measurement period, VGS caused a wall Ts reduction of 0.5 °C. For interior wall surface, the VGS-covered facade was consistently cooler than the bare one, averagely by 0.5 °C and maximally by 1.3 °C. In both rooms, Tg was close to Ta most of the time (Fig. 4j). Tg was a little higher during the day and slightly lower at night. Ta was constantly lower in the VGS-covered room, by averagely 0.5 °C and maximally 1.5 °C; Tg largely follows the Ta pattern, with VGS-covered room averagely lowered by 0.6 °C and maximally 1.9 °C; RH was consistently higher in the VGS-covered room, by 1.5% on average and 9% maximum. Regarding indoor thermal comfort, the VGS-covered room is consistently cooler than the room without VGS, the mean Top reduction is 0.6 °C and the maximum is 1.9 °C. 3.2. Statistical comparison of VGS cooling effects Southern wall surface temperature reductions by VGS is plotted in Fig. 5b. On a daily basis, VGS generally reduced the exterior Ts of the masonry wall more than it does on the interior Ts of the wall, when there was sufficiently high solar radiation and air temperature. Specifically in this case, exterior Ts reductions were only smaller than interior Ts reductions on day 4 and day 6, when GSR dropped below 300 W/m2 and mean Ta below 27 °C (Fig. 5a). If consider only the working hours (i.e., 8 a.m.–6 p.m.) for this office building, then the reductions in exterior Ts were significantly larger than the daily values, and they were always larger than those in interior Ts of the masonry wall, indicating the influence of solar radiation on surface temperature elevation and the shading effect of VGS. Whereas the reductions in the wall interior Ts was kept at around 1.2 °C. Averagely, during the working hours, exterior Ts was reduced by about 3.4 °C and interior Ts by about 1.2 °C. While during the whole period, exterior Ts reduced by around 1.6 °C whereas interior Ts by around 1.2 °C. As can be seen from Fig. 4d, the magnitude of change in the interior Ts was much lower than exterior Ts. This helps maintain a stable radiant thermal indoor environment. The indoor Ta&Top reductions by VGS is plotted in Fig. 5c. The daily mean Ta reductions ranges 0.8-1.5 °C during the measurement period, and averagely ∼1.2 °C. The Ta cooling effect is a little better if consider only the working hours, nearly 1.5 °C. It is shown from Fig. 5d that, on a daily basis, for the VGS-covered wall, the exterior surface was cooler than the interior surface (the values being negative), the mean △Ts being around −0.4 °C; while the exterior surface was warmer than the interior surface for the bare wall, the mean △Ts being ∼0.7 °C. If considering only working hours, the temperature difference was significantly higher than that of daily values for bare wall room, with the mean △Ts being about 3 °C, whereas for the VGS-covered room the △Ts is still below zero. 46

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

Fig. 3. Exploded axonometric showing the structure of DSGF and locations of sensors.

wall lost heat to the outside through conductive heat transfer, the mean △Ts being ∼ −1 °C. Considering the working hours, heat flux was reversed, indicating conductive heat gain, with the mean △Ts being ∼0.1 °C for VGS-covered wall and 0.8 °C for bare wall.

The North-facing wall Ts reductions by VGS is plotted in Fig. 5f. Like the South-facing scenario, surface cooling effect of VGS seems positively related with solar radiation and ambient air temperature. But unlike the South-facing scenario, the cooling effects do not seem significant difference between the daily circle period and working hours only. This may be due to the very limited amount of diffuse radiation received by the Northern facade. On average, Ts were reduced by ∼0.5 °C for exterior and interior wall surface, for working hours and for daily circle. The indoor Ta&Top reductions (Fig. 5g) were also positively related to solar radiation and ambient temperature, and the cooling range was much narrower than south-facing rooms, with about 0.5-0.7 °C for both working hours and for daily circle. Regarding the exterior-interior △Ts of the North-facing wall (Fig. 5h), it is shown that, on a daily basis, both VGS-covered and bare

4. Discussion The VGS on the South-oriented facade created a distinctive microclimate in the cavity between the VGS and the masonry wall (Fig. 5b). In comparison, the North-oriented VGS in present study lowered the cavity Ta to a lesser extent (Fig. 5g). This relatively cooler air skin indicates a theoretical energy saving benefit and a more accurate boundary as well, if the cavity Ta is used to replace long-term Ta monitored at some suburb observatory, which is widely used for building energy simulation. When individual AC unit is used and the 47

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

Fig. 4. Measurement plots on the South and North DSGF and reference station.

of VGS can provide possible energy saving potential due to lowered Ta. A study in Mediterranean climate suggests that the cavity air of a living wall system could be cooled by 5–6 °C, and the makeup air could lead to an cooling energy saving by 26% (Perini, Bazzocchi, Croci, Magliocco,

outdoor AHU unit is installed within this cavity, as applied in the building under investigation, the cooler air help increase the cooling efficiency and reduce electricity consumption. If individual fresh air supplier is used to extract immediate outdoor makeup air, the presence 48

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

Fig. 5. Daily mean thermal performance of the South and North DSGF.

releases towards cooler surfaces of the VGS due to its irradiation cooling to ambient environment and sky. The Ts profile of the Northoriented DSGF showed a less clear pattern during the day, but similar with the South counterpart at night. Due to solar geometry, diffuse and reflected radiation warmed the North DSGF to a less extent compared to the Southern side, whereas the nocturnal cooling process is not disturbed. Although all orientations other than North receive higher radiation and thus should be prioritized when considering VGS (Pérez, Coma, Sol, & Cabeza, 2017), the North-oriented VGS showed tangible

& Cattaneo, 2017). No such study is found on a DSGF, and this can be included in the authors’ future work. Examining the surface temperatures of the DSGF system, a positive Ts gradient from exterior to interior is revealed at the South-oriented one during the day, while this gradient was reversed at night, with absence of solar radiation. Daytime Ts pattern is largely because of the conduction and long-wave radiation heat transfer from exterior VGS to interior surface of the wall, due to increased leaf surface temperature by intercepting solar radiation; and at night, stored heat in the wall 49

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

taken into account, as energy for artificial lighting is also a significant part of office building energy consumption, and the VGS can lower daylight illuminance level and thus increase energy used for artificial lighting. The overall energy performance should be assessed taken the above-mentioned factors into account. Therefore, based on understanding on the preliminary findings as well as limitations of present study, the next stage should carry out a more comprehensive long-term experiment covering a more complete set of environmental variables that impact building energy use and indoor thermal comfort. For instance, air speed in the DSGF cavity should be monitored to understand the cavity depth and air change rate of the cavity (Hunter et al., 2014). And energy consumptions of the AC devices of test cells should be monitored as well. Parametric simulations using validated software are to be followed, to investigate approaches to design optimization of VGS.

summer cooling in this case. In addition, its potential to provide wind buffer and insulation in winter is not studied at this stage, which may further justify its applicability on the Northern facade. Ts comparison between the VGS-covered and bare facade in the South orientation indicates a mean reduction of exterior Ts by 1.5 °C and the peak value of around 9 °C. At night the VGS-covered wall could turn slightly warmer than the bare one, dependent on weather conditions. For the interior Ts, a mean reduction of 1.2 °C and maximally 2 °C. The North pair followed the pattern with that of South, but the magnitudes of reduction were shrunk to about one half of that of the South pair. Daily variation of the cooling effect largely followed the variation in solar radiation intensity and ambient air temperature. Note that the cooling performance of VGS on wall surfaces depends on a number of factors including vegetation characteristics (e.g., species, coverage, LAI), wall material properties (e.g., emissivity, colour, heat capacity, etc.), spacing between vegetation layer and wall, in addition to orientation and weather conditions. Further studied is needed to explore the effect of design variables, preferably by parametric study using validated numerical tools. Weather condition has clear impact on the DSGF cooling effect. Higher ambient air temperature and radiation seems to have led to enhanced wall Ts and indoor Ta and Top reductions by vertical greenery (Fig. 5), except that for the South pair measured in this case study, the cooling effect on the indoor environment seems negatively related to the solar radiation and ambient air temperature. Indoor warming occurred on sunny days was found in a green facade study in Hong Kong, under a subtropical climate, and the authors attributed this to the result of solar penetration through windows (Lee and Jim, 2017). Indoor Ta reduction of averagely 1.2 °C were observed for the Southoriented VGS covered room. Considering working hours only obtains a slightly improved cooling of ∼1.5 °C. while the VGS cooling effect was reduced to 0.5-0.7 °C for the North pair. But, regarding the surface temperature differentials of facade walls, for the South pair, the exterior-interior differential during working hours was overall positive for the bare wall (∼3 °C) while it was around zero for the VGS-covered wall. Similar pattern was observed for the North pair, albeit to a lesser extent. Heat flux through opaque walls is positively related to the surface temperature difference, given the same thermal resistance of the walls. Therefore, other energy transfers being equal (i.e. radiation through window and convection through fenestrations, etc.), and assuming no change in stored heat of the masonry wall, the exterior-interior Ts difference can give a proximate estimate of cooling energy load. So theoretically indoor Ta reduction and Ts differentials both indicate potential cooling energy savings (Eumorfopoulou and Kontoleon, 2009; Perini et al., 2017). It is shown from Fig. 5d that, on a daily basis, theoretically, heat was fluxed from inside to outside for the VGS-covered room, indicating negative cooling load due to conductive heat loss; while heat was fluxed from outside to inside room for the bare wall room, indicating a positive cooling load due to conductive heat gain. If considering only working hours, the temperature difference was significantly higher than that of daily values for bare wall room, with the mean △Ts being about 3 °C, whereas for the VGS-covered room this is still negative, indicating no stationary cooling load during working hours. Therefore, it can be seen that VGS in this case study has clear energy-saving potential in summer days for the office building schedule. For the North facade, VGS showed no clear cooling effect on a daily basis, although it did show certain surface cooling effect during the daytime. The measured data suggests that the VGS on the North facade may not be an efficient surface cooling strategy for housing, but may play a better role in reducing solar heat gain (on the diffuse component) for other building types, for instance, office buildings. It seems not feasible to apply a simplified method of cooling energy calculation, without considering radiation and convective heat transfer through fenestrations (i.e. windows). In addition, the shading effect of VGS on indoor thermal and illumination conditions should also be

5. Conclusion The paper presents a preliminary study on the thermal and energy performance of designing vertical greening system (VGS) in building renovation in Shanghai, the largest city in China under a warm and humid sub-tropical climate. Thermal environmental parameters and biometeorological indices were measured and the impacts on cooling energy use and indoor thermal comfort analyzed for a double-skin green facade (DSGF) system against the original exposed masonry facades. The findings can be summarized as follows:

• The VGS on the South-oriented facade created a distinctive micro-

climate in the cavity, indicating a theoretical energy saving benefit and a more accurate boundary condition for energy simulation of buildings with VGS. The air temperature averagely dropped by about 0.4 °C in a daily circle, and reached peak value of 5.5 °C. The North-oriented VGS lowered the cavity Ta to a lesser extent, by 0.2 °C in average and 3.3 °C maximally, indicating a marginal cooling during the whole day but still substantial when considering the working hours only. Comparing surface temperatures of VGS and the wall behind it revealed a positive gradient from exterior to interior during the day, while this gradient was largely reversed at night with the absence of solar radiation. The South VGS reduced exterior wall Ts by 1.5 °C in average and ∼9 °C maximally. At night the VGS-covered wall could turn slightly warmer than the bare one, dependent on weather conditions. For the interior Ts, a mean reduction of 1.2 °C and maximally 2 °C. The magnitudes of reduction by North VGS were shrunk to about one half of South one. Indoor air temperature reduction of averagely 1.2 °C were observed for the South-oriented VGS-covered room. Considering working hours only results in a slightly improved cooling of ∼1.5 °C; while the indoor Ta cooling effect was reduced to 0.5-0.7 °C for the North pair. The surface temperature differentials of facade walls during working hours indicate reduced cooling load for VGS-covered rooms. Tentative results would favor VGS, in this case as a DSGF, as one of useful strategies for energy-efficient building renovation. But further study is needed considering the year-round energy and comfort effects of VGS on radiation, convective and conduction heat transfer and on daylighting through building envelop, in terms of cooling, heating and artificial lighting.

Acknowledgments The authors are grateful to Mrs. LI Xia, Mrs. XIAO Hui, and another two anonymous colleagues of Tongji University who kindly allowed instrumentation and measurement of their offices during the summer vacation. M.Arch student Ms. WANG Rui helped make the drawings of Fig. 3. The research is supported by the National Natural Science Foundation of China (NSFC) Project (No.: 51678413) and by the 50

Sustainable Cities and Society 39 (2018) 43–51

F. Yang et al.

National Key R&D Program of China (No.: 2016YFC0700200).

greening in a high-density city: Case study of Hong Kong. Urban Forestry & Urban Greening. In Press, Corrected Proof. Ottelé, M., Perini, K., Fraaij, A. L. A., Haas, E. M., & Raiteri, R. (2011). Comparative life cycle analysis for green façades and living wall systems. Energy and Buildings, 43, 3419–3429. Pérez, G., Rincón, L., Vila, A., González, J. M., & Cabeza, L. F. (2011). Green vertical systems for buildings as passive systems for energy savings. Applied Energy, 88, 4854–4859. Pérez, G., Coma, J., Martorell, I., & Cabeza, L. F. (2014). Vertical Greenery Systems (VGS) for energy saving in buildings: A review. Renewable and Sustainable Energy Reviews, 39, 139–165. Pérez, G., Coma, J., Sol, S., & Cabeza, L. F. (2017). Green facade for energy savings in buildings: The influence of leaf area index and facade orientation on the shadow effect. Applied Energy, 187, 424–437. Perini, K., Ottelé, M., Fraaij, A. L. A., Haas, E. M., & Raiteri, R. (2011). Vertical greening systems and the effect on air flow and temperature on the building envelope. Building and Environment, 46, 2287–2294. Perini, K., Bazzocchi, F., Croci, L., Magliocco, A., & Cattaneo, E. (2017). The use of vertical greening systems to reduce the energy demand for air conditioning. Field monitoring in Mediterranean climate. Energy and Buildings, 143, 35–42. Raji, B., Tenpierik, M. J., & van den Dobbelsteen, A. (2015). The impact of greening systems on building energy performance: A literature review. Renewable and Sustainable Energy Reviews, 45, 610–623. Riley, B. (2017). The state of the art of living walls: Lessons learned. Building and Environment, 114, 219–232. Susorova, I., Angulo, M., Bahrami, P., & Brent, S. (2013). A model of vegetated exterior facades for evaluation of wall thermal performance. Building and Environment, 67, 1–13. Susorova, I., Azimi, P., & Stephens, B. (2014). The effects of climbing vegetation on the local microclimate, thermal performance, and air infiltration of four building facade orientations. Building and Environment, 76, 113–124. Tu, F.-X. (2011). On building energy conservation of China. Beijing, China: Architecture and Building Press. Tudiwer, D., & Korjenic, A. (2017). The effect of living wall systems on the thermal resistance of the façade. Energy and Buildings, 135, 10–19. Vox, G., Blanco, I., & Schettini, E. (2018). Green façades to control wall surface temperature in buildings. Building and Environment, 129, 154–166. Wong, I., & Baldwin, A. N. (2016). Investigating the potential of applying vertical green walls to high-rise residential buildings for energy-saving in sub-tropical region. Building and Environment, 97, 34–39. Wong, N. H., Tan, A. Y. K., Tan, P. Y., & Wong, N. C. (2009). Energy simulation of vertical greenery systems. Energy and Buildings, 41, 1401–1408. Wong, N. H., Tan, A. Y. K., Chen, Y., Sekar, K., Tan, P. Y., Chan, D., et al. (2010). Thermal evaluation of vertical greenery systems for building walls. Building and Environment, 45, 663–672. Xue, F., Gou, Z., & Lau, S. S. Y. (2017). Green open space in high-dense Asian cities: Site configurations, microclimates and users’ perceptions. Sustainable Cities and Society, 34, 114–125. Yang, F., Lau, S. S. Y., & Qian, F. (2011). Urban design to lower summertime outdoor temperatures: An empirical study on high-rise housing in Shanghai. Building and Environment, 46, 769–785. Yin, H., Kong, F., Middel, A., Dronova, I., Xu, H., & James, P. (2017). Cooling effect of direct green façades during hot summer days: An observational study in Nanjing, China using TIR and 3DPC data. Building and Environment, 116, 195–206.

References Afshari, A. (2017). A new model of urban cooling demand and heat island—Application to vertical greenery systems (VGS). Energy and Buildings, 157, 204–217. Baker, N., & Steemer, K. (2000). Energy and environment in architecture: A technical design guide. E&FN Spon. Cameron, R. W. F., Taylor, J., & Emmett, M. (2015). A Hedera green façade – energy performance and saving under different maritime-temperate, winter weather conditions. Building and Environment, 92, 111–121. Chen, Q., Li, B., & Liu, X. (2013). An experimental evaluation of the living wall system in hot and humid climate. Energy and Buildings, 61, 298–307. Coma, J., Pérez, G., de Gracia, A., Burés, S., Urrestarazu, M., & Cabeza, L. F. (2017). Vertical greenery systems for energy savings in buildings: A comparative study between green walls and green facades. Building and Environment, 111, 228–237. Davis, M. M., & Hirmer, S. (2015). The potential for vertical gardens as evaporative coolers: An adaptation of the ‘Penman Monteith Equation’. Building and Environment, 92, 135–141. Dunnett, N. (2004). Planting green roofs and living walls. Portland USA: Timber Press. Eumorfopoulou, E. A., & Kontoleon, K. J. (2009). Experimental approach to the contribution of plant-covered walls to the thermal behaviour of building envelopes. Building and Environment, 44, 1024–1038. Flores Larsen, S., Filippín, C., & Lesino, G. (2015). Modeling double skin green façades with traditional thermal simulation software. Solar Energy, 121, 56–67. He, Y., Yu, H., Ozaki, A., Dong, N., & Zheng, S. (2017). An investigation on the thermal and energy performance of living wall system in Shanghai area. Energy and Buildings, 140, 324–335. Hoelscher, M.-T., Nehls, T., Jänicke, B., & Wessolek, G. (2016). Quantifying cooling effects of facade greening: Shading, transpiration and insulation. Energy and Buildings, 114, 283–290. Hunter, A. M., Williams, N. S. G., Rayner, J. P., Aye, L., Hes, D., & Livesley, S. J. (2014). Quantifying the thermal performance of green façades: A critical review. Ecological Engineering, 63, 102–113. International Energy Agency (2013). Technology roadmap. Energy efficient building envelopes. Ip, K., Lam, M., & Miller, A. (2010). Shading performance of a vertical deciduous climbing plant canopy. Building and Environment, 45, 81–88. Kontoleon, K. J., & Eumorfopoulou, E. A. (2010). The effect of the orientation and proportion of a plant-covered wall layer on the thermal performance of a building zone. Building and Environment, 45, 1287–1303. Koyama, T., Yoshinaga, M., Hayashi, H., Maeda, K. I., & Yamauchi, A. (2013). Identification of key plant traits contributing to the cooling effects of green façades using freestanding walls. Building and Environment, 66, 96–103. Lechner, N. (2015). Heating, cooling, lighting: Sustainable design methods for architects (4th ed.). Hoboken, N.J: John Wiley & Sons. Lee, L. S. H., & Jim, C. Y. (2017). Subtropical summer thermal effects of wirerope climber green walls with different air-gap depths. Building and Environment, 126, 1–12. Liu, G.-L., & Chen, Q.-B. (2004). Study on the ecological benefits of four kinds of vertical plant in Chengdu. Journal of China West Normal University (Natural Sciences), 25, 259–262. Manso, M., & Castro-Gomes, J. (2015). Green wall systems: A review of their characteristics. Renewable and Sustainable Energy Reviews, 41, 863–871. Morakinyo, T. E., Lai, A., Lau, K. K.-L., & Ng, E. (2018). Thermal benefits of vertical