1.0    Adapting Urban Areas to Climate Change with Vertical Gardens
2.0    From Heated Rockpiles to Forest Ecosystems: Redesigning Cities in a Restorative Economy
3.0    Notes
4.0    Article 4
5.0    Article 5
6.0    Article 6

An abbreviated version of this paper was submitted to the Alternatives Journal on July 28, 1999.

Abstract

Increasing vegetation in urban areas will reduce the urban heat island, and the impacts of other urban environmental problems, which will be exacerbated under climate change. Reducing the urban heat island will also reduce the energy demand for space conditioning, and hence greenhouse gas emissions. Plants directly reduce the urban heat island through evaporative cooling but further reduce energy consumption through shading. The most common strategy to increase urban vegetation is to plant trees at ground level. However, where space is not available for trees, vegetation can be grown on building roofs, but walls offer far more space, hence vertical gardening is a viable alternative.

Data were collected on the evaporative cooling capacity of different vertical garden designs at three different sites at the University of Toronto. An infrared pyrometer was used to measure the surface temperature of a vertical garden, a light-coloured wall and a black surface. The data were compared using an analysis of variance, and in all cases, the vertical garden was significantly cooler than the other two surfaces. This reduction in surface temperature could reduce the urban heat island, providing a strategy for mitigation and adaptation, and achieve other environmental and social benefits.

Introduction

Climate change is expected to result in warmer summer temperatures and more frequent heat waves.

For Canada, climate change models predict an increase in both winter and summer temperatures which is expected to reduce the demand for space heating while increasing the demand for space cooling.

The effect at night could be exacerbated further under climate change as, since 1950, the observed increases in minimum land surface temperatures has been about double those in the maximum temperatures.

Climate change is also expected to impact the supply of energy. Water supplies that are used for hydroelectricity and cooling of thermal and nuclear plants, and water bodies that are used for shipping fuel, such as the Great Lakes, are sensitive to changes in many climatic variables.

These impacts will be exacerbated in urban areas, due to the lack of vegetation and the large amount of concrete and asphalt. Vegetation is important because it makes moisture available for evapotranspiration, the combination of evaporation of water from plant tissue and the evaporation of water from the soil. Evapotranspiration utilizes a significant amount of incoming solar energy cooling both the leaf surface as well as the air. The energy used for evapotranspiration is embodied in the water vapour, which is removed from the surface.

A lower fraction of vegetative cover in the city reduces the available moisture to direct incoming solar radiation towards evapotranspiration. The non-vegetated surfaces absorb the incoming solar radiation and reradiate it as heat. This heat artificially elevates urban temperatures, a phenomenon known as the urban heat island.

The higher temperatures increase the demand for air conditioning which pumps more waste heat into the environment, increasing the heat island. Air conditioning requires electricity that is often generated by the burning of fossil fuels leading to increased greenhouse gas (GHG) emissions. Without air conditioning, the higher temperatures will increase the incidents of morbidity and mortality due to heat stress.

The urban heat island can also exacerbate other atmospheric problems associated with health. Warmer surfaces heat the overlying air, causing it to rise and carry pollutants higher into the atmosphere. As it rises it cools and disperses over a wider area, dispersing any pollutants that may have been present, over a wider area. With increased use of air conditioning comes the increased risk of releasing additional chlorofluorcarbons into the atmosphere, chemical compounds that are responsible for the reduction of the stratospheric ozone that is necessary to protect us from harmful amounts of ultra-violet radiation. The combustion of fossil fuels to generate electricity contributes other pollutants into the atmosphere, such as sulfur dioxide, nitrous oxides and particulate matter, which have been linked to respiratory health problems.

The replacement of vegetation by hard surfaces in urban areas has other environmental impacts. Rainwater and snowmelt cannot penetrate these surfaces, thus they runoff, eventually taxing the capacity of the sewage system. As water runs off over these surfaces, it also absorbs a range of chemicals that either enter the sewage system or local groundwater supplies. Plants can filter some of the pollutants out of water. In addition, leaves filter some pollutants out of the atmosphere. Thus the urban heat island is a serious environmental issue, and increasing vegetation in urban areas is a desirable strategy for mitigation of GHG emissions and adaptation to a broad range of atmospheric and other environmental issues.

Vegetation can reduce the urban heat island through evapotranspiration and shading a building surface. With a larger fraction of vegetated surface, more of the incoming solar energy is consumed by evapotranspiration, and less is absorbed and reradiated as heat. Shading reduces both the heat flow as well as the incoming solar energy. Light-coloured surfaces can also reduce the urban heat island by reflecting a high percentage of the incoming solar energy so that it is not absorbed and reradiated as heat. However, removing dark colours from the urban environment will not address the other environmental problems that have arisen due the removal of vegetation.

Planting trees at ground level is the most common strategy for restoring vegetation. In addition to reducing the urban heat island and GHG emissions through reducing energy consumption, shade trees can further offset GHG emissions through the sequestration of carbon in their woody mass. However, higher land use densities or space restrictions in some parts of the city restrict the space required to allow trees to reach their full potential. In these cases, rooftop gardens could provide many of the same benefits, and they would take advantage of the unused roof space that is available in most urban areas. In addition, depending on its colour, the heat from bare rooftops can exacerbate the urban summer temperatures and the associated air quality problems.

Reducing the rooftop temperatures would further reduce GHG emissions in both the summer and the winter. In the summer, a typical insulated, gravel-covered rooftop temperature can vary between 60oC and 80oC.

These temperatures increase the cooling load on a building in two ways. Since the internal temperature underneath the roof is lower, the heat will always flow from the roof into the building, thus the need for air conditioning. In addition, modern high-rise buildings are constantly exchanging the internal and external air. Because of the high roof temperatures, the temperature of this external air that is brought into the building’s ventilation system is very high, requiring additional energy for cooling.

Evapotranspiration from rooftop vegetation could reduce temperature of the roof, maintaining it at 25oC on a hot, sunny day.

This would reduce the amount of heat flow into the building through the roof. The lower rooftop temperature would also reduce the temperature of the external air that is exchanged with the building’s air. The external air temperature could also be reduced if the rooftop garden is designed so as to shade the intake valves.

These lower rooftop temperatures could reduce the KWh used for air conditioning by 10%, and this is a very conservative estimate. The actual value would depend on building size, design, orientation, insulation, window type and other internal factors that contribute to heat gain or dissipation. To maintain the cooler roof temperatures it is important to install enough rooftop gardens so as to minimize the potential horizontal movement of hot air from adjacent rooftops that are still bare. During the winter, the rooftop garden would provide additional insulation, which would reduce the flow of heat through the roof. One typical type of installation in Europe, consisting of 20 cm of growth substrate and 20-40 cm of grass would provide the equivalent of 15 cm of rockwool, or and insulation factor of R2011, which is enough insulation to provide a significant reduction in heating. These reductions in summer cooling and winter heating suggest that rooftop gardens could play a significant role in reducing GHG emissions in addition to their role as an adaptation strategy.

An even greater amount of space is available for vegetation on the exterior walls of the buildings in urban areas, and we propose growing vegetation on walls to create a vertical garden. A vertical garden refers to vines growing adjacent to a wall, window shades that utilize plants, or a vertical hydroponic system. Vertical gardens increase the amount of vegetative surface in urban areas, increasing evapotranspiration, and can be used for direct shading as well. In areas that are suitable for trees, they can also to be used to cover windows that cannot be shaded by trees due to the height or specific design features of a building.

Despite the large amount of available space, there are very few studies on vertical gardening. Rooftop gardens are receiving increasing coverage, particularly as a green roof industry has been developing in Germany, Austria and France, yet no similar industry has emerged to promote vertical gardening. The lack of attention to vertical gardens is also evident in a recent report on status of rooftop and vertical gardening in Canada. Most of the examples dealt with roofs, as the requisite information on vertical examples was not available.

To address this gap, three different sites and different types of vertical gardens were evaluated at the University of Toronto between 1996 and 1998. The vertical gardens were cooler than a black surface, but they were also cooler than the light-coloured walls at each site. Before discussing our results, we will define vertical gardens, provide some simple examples of how they might work, and review previous studies on the cooling effect of vertical gardens.

Defining Vertical Gardens

Vertical gardening is a comprehensive term referring to any manner in which plants can be grown on, up, or against the wall of a building such as a vine, as part of a window shade, as a balcony garden, or in a vertical hydroponic system.

As a window shade, plants can be grown in a planter box installed below a window, or hanging plants can be suspended above a window and used as a part of an awning. Since a building's façade typically encompasses four times the area of the roof, vertical gardening may offer a greater potential impact than rooftop gardening, without the additional concerns about weight. For a high-rise building, the area of the walls is typically twenty times the area of the roof. To allow some natural light into the room, the vertical garden could be installed on a moveable louvre, or installed as part of adjustable awning, so that it could be maneuvered to intercept only direct sunlight. Additional design considerations are also required to cope with high wind speeds for plants and planter boxes located above eight stories.

More specific engineering considerations are beyond the scope of this article.

How Vertical Gardens Reduce Air Conditioning and the Urban Heat Island

There have been very few studies on the potential effect of vertical gardening on reducing either energy consumption or the urban heat island. In the summer, air conditioning is reduced through shading of windows and evapotranspiration. Since many buildings are already insulated, the largest impact of shading is on reducing the solar energy transmitted through the windows as this energy can be absorbed by objects in the room and reradiated as heat.

The effect of shading a window can be estimated by computing the monthly avoidance of cooling load with the following formula:



MCLA = monthly coolant load avoidance
A = area of window (m2)
D = shading coefficient difference (%)
S = normal sunshine hours per month
G = average hourly heat gain for that month (Wm-2)

The value of S should reflect the hours of direct exposure to the sun. For example, in Toronto, there may be 16 hours of sunshine during the summer months, but the western face of a building is exposed to direct sunlight for only six hours.

To calculate the heat gain, G, requires information on the incoming solar energy. A typical amount of radiation in Toronto would be 600 Wm-2. For a wall, this would be reduced, usually by a cosine function of the angle of the sun. An alternative is to reduce the number of hours. Thus 6 hours of sunlight on a vertical western exposure is equivalent to 2.5 hours of 600 Wm-2 on a horizontal surface.

The variable G is the estimated fraction of incoming solar energy on a wall multiplied by the percent transmission of solar energy through the window. The percentage of incoming solar energy that is transferred through the window varies from 76% for a single pane window with no protection to 26% for gold glass heat-reflecting windows. For double paned windows reduce the amount of transmitted energy varies from 64 to 25%.

For plants, D is typically 80%, but it could be as high as 90%.

Depending on the orientation, the size of the widows, the size of the building, and the number of windows, the potential reductions in air conditioning, and thus GHG reductions, from adding a vertical garden, with a shading coefficient of 80%, could be quite significant. This formula does not include the effect of evapotranspiration on heat flow, so the actual reductions would be somewhat larger.

Evapotranspiration is the most direct way in which vegetation reduces both the urban heat island and heat flow into a building. Assessing a reduction in the urban heat island requires sophisticated models of the atmosphere and the land surface. However, the contribution of any surface can be estimated if we know add a few more components of the radiation balance and calculate the total amount of radiation absorbed.

The important components are the incoming solar radiation (Rs) and surface reflectance of Rs, called albedo and often represented by the Greek symbol alpha. What produces heat is the longwave or infrared radiation. Longwave radiation is radiated from the surface by it is also produced in the atmosphere and re-radiated to the surface (L*). The total radiation absorbed (Rabs) by a surface without vegetation or moisture is computed as below

Rabs = (1 - alpha) Rs + L*

Assuming a dark surface with and albedo of 0.3, a typical July Rs of 600 Wm-2, the downward longwave radiation is 300 Wm-2 at 15oC.




The surface temperature, T(oC) can be computed with the formula relating energy to temperature developed by Stefan and Boltzmann.



The value of 5.67 x 10-8 is the Stefan-Boltzmann constant, which relates the radiance of a black body to temperature, and is symbolized by the Greek letter sigma. T + 273 changes degrees Celsius to degrees Kelvin or K. Thus the above formula could be rewritten as



or



and in this case,



This figure may be somewhat high since we have ignored the other smaller components of the radiation budget.

If vegetation was present, evapotranspiration could reduce the absorbed energy by as much as one-half. Using a more conservative estimate for Rabs of 420 Wm-2 results in a temperature closer to 20oC. These temperatures are not so absurdly high and low as might be expected. Temperature measurements of 60oC to 80oC have been recorded for bare rooftops, and temperatures of 25oC have been observed on rooftops with vegetation, under similar summer conditions. 19 Taking 2.5 hours of 600 Wm-2, provides an estimate of Rs on vertical wall of 250 Wm-2, which produces a surface temperature of 29oC and a corresponding decrease with a vertical garden.

Previous observations indicate that vertical gardens do reduce the heat flow into the building, and their surface temperature is lower than a bare wall, which is necessary to reduce the urban heat island. A series of experiments in Japan suggested that vines could reduce the temperature of a veranda with a southwestern exposure.

Vines were effective at reducing the surface temperature of a wall, but the vine surface temperature was slightly higher than the outdoor temperature except during the hottest time of the day.

In Germany, the vertical garden surface temperature was 10oC cooler than a bare wall when observed at 1:30 PM in September, yet still greater than the air temperature.

A series of observations were collected in South Africa on English ivy, Boston ivy, Virginia creeper and grape vines.

All were grown at a cover depth of 200 mm to emulate the thermal improvement to a typical South African house. The plants were installed over steel sheets that were compared to black and white panels. Temperatures collected behind all panels were less than the outdoor temperature, but the largest reduction of 2.6oC was behind the vegetated panel. For a building consisting of two 10mm fiber-cement sheets with 38mm of fiberglass insulation, a computer simulation estimated that a vertical garden reduced summer daytime temperatures on the surface by 5 oC. These results are not as dramatic as the cooling effect on a horizontal surface, such as a roof, but since the ratio of walls to roofs in urban areas is at least 4:1 the potential impact of vertical gardening is expected to be quite dramatic.

Analysis of Vertical Gardens at the University of Toronto

The evaluation of vertical gardens at the University of Toronto focused on the evaporative cooling capacity of vegetation in comparison to light-coloured wall and, at two of the sites, a black surface. The observations were collected at three different sites, at three different times on different plants in window shade configurations and on a vine covered wall. The data were collected with a computerized infrared pyrometer, which converts longwave radiation, i.e. heat, into surface temperatures.

The first data were collected in situ, in August 1996. Three different plants * morning glories, scarlet runner beans and sunflowers * were installed as window shades on the western exposure of a university residence, consisting of beige-coloured brick. The data were collected over two days at 1200, 1330 and 1430 hours for each garden, the adjoining brick wall and a black surface in the same location as the garden. The set-up allowed for the plant and brick temperatures to be taken side-by-side while the black surface was below the plants.

The average temperatures of the vertical gardens were lower than both the black surface and the brick wall (Figure 1). The average wall temperature of 28.3oC is in accord with the estimate of 29 oC from radiation balance calculation. Two statistical procedures, the binomial test and the analysis of variance (ANOVA), were used to evaluate significance of the temperature differences between the plants and the non-vegetated surfaces. The binomial test examines the probability of getting a particular combination of outcomes (for instance heads or tails, cooler or warmer) due to chance alone. The binomial test indicated that the probability of these results being due to chance alone was 2 times in 1000.

The ANOVA test compares the variability within a sample to the variability between samples. If the within sample variability is smaller, the differences are significant, and the ANOVA test evaluates the probability of this occurring by chance. An ANOVA was conducted to evaluate the whether the temperatures of the vertical gardens and the black surface were significantly different, since there were not enough observations for a separate binomial test. The differences were significant except for beans at 1:00 PM. No further tests were required to compare the plant and the wall temperatures, since their collection allowed a side-by-side visual comparison. The plant temperatures were always cooler except for one case where they were equal.

Further analysis of the data clearly indicated that these results were not independent of the site of the windows where they were installed. The sunflowers were under considerable stress, perhaps due to the confinement of the roots in a window box and the high temperature of this box, which may explain their high surface temperatures. On the other hand, the morning glories’ were partially shaded which may have accounted for their lower surface temperatures. Nevertheless, a general pattern is evident throughout the data. The vertical gardens were cooler than the neighbouring light-coloured brick wall and markedly cooler than the black surface.

A second set of observations was collected indoors in February, 1997, comparing three vertical gardens * ivy, passionflower and elephant ear * to a white wall. Sunflowers were excluded due to the difficulty in growing them in a window box, the results from the beans did not warrant further testing, and the morning glories were destroyed by spider mites. Ivy was chosen because it grows well in many temperate climates, passionflower is a rapidly growing vine and elephant ear does well in high humidity which is characteristic of many summer urban environments, and specifically Toronto. Data were collected between 2:30 and 3:00 in the afternoon, once a week for five weeks. The average temperatures of the vertical gardens were lower than the white wall (Figure 2), and an ANOVA procedure indicated that these differences were significant. In addition, the analysis revealed that the temperature differences between ivy and elephant ear and between passionflower and elephant ear were also significant. Although there were no visible signs of stress, the humidity levels and temperatures may have been to low for elephant ear.

The final set of observations compared a ficus vine to a light-coloured grey wall and a black surface, and was collected in a greenhouse (Figure 3). Data were collected between 2:30 and 3:00, once a week for four weeks during the fall of 1998. Given the large amount of data, it was possible to compare each weekly average temperature as well as pooled value. An ANOVA test on these data which indicated significant differences between all of the surfaces. Week one was the sunniest of the four weeks, which probably explains the larger temperature differences between the different surfaces.

Discussion

The observations of surface temperature confirmed previous research results that indicated that vertical gardens were cooler than a bare wall, often by 5 or 10 oC. The data analysis confirmed that the probability of these results occurring by chance is extremely remote. Although we did not assess the impact on the urban heat island directly, this research supports previous claims that vegetation can lower peak temperatures in urban areas.

The maximum observed temperatures at the first site were 40oC for the wall and 43oC while the surface temperature of the adjacent vegetation was 27oC. These results, in combination with the significant differences between the plant and light-coloured surfaces at the other sites, suggest that vegetation and thus evaporative cooling is a much more effective strategy to reduce the urban heat island.

Conclusions

Using vertical gardens to increase the vegetation in urban areas is an effective strategy for reducing the urban heat island. This would not only be an effective adaptation strategy for urban areas facing warmer summers and more frequent heatwaves, but it would also reduce GHG emissions. Reducing the urban heat island would also improve air quality through reducing the dispersal of pollutants due to rising bubbles of warm air and through the capacity of leaves to filter out some pollutants from the atmosphere. Vegetation can also reduce stormwater runoff and improving water quality. This is important as several climate change scenarios indicate suggest the possibility of more extreme rainfall events, which would be exacerbated in urban areas due to the low ration of vegetation to hard surfaces.
Vertical gardens are an important part of any urban vegetation strategy in that they reduce the heat gain due to windows. However, they also impart other benefits to the city.

For example, they would protect building facades from persistent climatic threats such as freeze-thaw events, acid rain and ultra-violet radiation. Installing vertical gardens also opens up the possibility of transforming our urban canyons into farms, thus reducing the amount of energy required for growing and transporting food from outside of the city. Despite the successes that we have seen with urban forestry and rooftop gardens, the low roof-to-wall area ratio suggests that it is in the vertical dimension in which vegetation may have the largest impact in adapting urban areas to climate change.

Acknowledgements

The authors wish to express their gratitude to several people who assisted with this research. Professor Kim Pressnail, in the Department of Civil Engineering at the University of Toronto provided the infrared pyrometer. Attila Keszei, the University of Toronto Energy Management Engineer, provided the formula for computing the monthly cooling avoidance load. Tom Doyle, of IRC, Inc. provided advice on the benefits of evapotranspiration from a rooftop garden, and Greg Allen of Allen and Khani Associates alerted us to the importance of the air exchange between a building and the external environment.