Impact of Vegetation on Urban Microclimates

Transcription

1 Impact of Vegetation on Urban Microclimates Paula Shinzato PhD Student at Faculty of Architecture University of São Paulo Brazil com Denise H. S. Duarte Professor at Faculty of Architecture University of São Paulo Brazil Summary The main purpose of this research is the effect of vegetation in urban microclimate conditions, under the premise that the presence of vegetation influences the microclimate and its surroundings. Considering different forms of green space distribution, one could quantify the intensity and the spatial distribution for the microclimate effects by a vegetated area. The modifications for air temperature and the surface temperature at the level of the pedestrians were analyzed based on energy balance and computer models. Parametric simulations on ENVI-met were carried out to investigate the effect of green areas distribution (linear forms, large mass of vegetation and small groups of trees) and were compared with on-site measurement data. The simulation results showed the impact of urban vegetation for climatic aspects. The outputs indicated that the vegetation effect is local and do not have a significant influence beyond the limits of the green area at pedestrian level. Regarding to the intensity of vegetation effect, the average difference between air temperature under the trees and above the streets is 1.5ºC, while shadowing of a dense tree (LAI=10) showed an average difference of 23ºC for superficial temperatures between green spaces and the street. This result pointed out that green area is an important strategy to mitigate the heat island effect in cities, as it can prevent asphalt and concrete from heating and releasing long wave radiation during night time. Keywords:urban vegetation, urban microclimate, ENVI-met. 1. Introduction Urban structure, building envelope and materials have a direct influence in the atmospheric conditions near the ground [1]. These urban transformations result in an increasing air temperature, causing the urban heat island effect in cities which is defined as the measurable temperature difference between adjacent urban and rural regions [2]. Vegetation is an important design element in improving urban microclimate and outdoor thermal comfort in urban spaces in hot climates [3]. Due to urbanization, however, vegetation is scarce in many tropical cities. There has often been a tendency to replace natural vegetation and permeable soils with impervious surfaces such as asphalt and concrete, which leads to more sensible than latent heat flux [4].

2 The reduction of green areas contributes to the intensity of urban heat island effect as it influences the mechanisms for controlling the air temperature in cities: shading, retention of pollution particles, wind direction, evapotranspiration process. Vegetation mitigates the heat island not by cooling the air, but by warming the air less [5]. Green areas reduces air temperature by direct shading of surfaces as well as moderating solar heat gain through evapotranspitration of the plants, which combines the evaporation of cropped soils and transpiration of the leaves, reducing the air temperature and creating the oasis effect [6]. The lack of green areas is caused by the increasing of impervious areas and building constructions and illegal occupations of "favelas"(slums) in public spaces. That fact resulted in considerable changes in the microclimate of metropolitan areas. According to the Environmental Agency Atlas [7], in SÄo Paulo, the reduction of green areas has caused various problems such as flooding, which happen in more than 400 points in the city, thermal discomfort in central areas, air pollution and the heat island effect. In this sense, it is necessary to carry on predictive studies, which could quantify the effect of urban vegetation on these specific local microclimates as a way to support landscape and urban planning decisions. 2. Methodology 2.1 Area of Study SÄo Paulo is a sprawling South American megacity which metropolitan area houses almost 19 million inhabitants, distributed in an area of 8051km 2. Based on reports from the Environmental Agency [7], SÄo Paulo s economy represents 31% of the national Gross Domestic Product (GDP). However, this economic development has led to a significant degradation of the urban environment, a common situation in many large cities in developing countries. Today SÄo Paulo is characterised by a heterogeneous urban structure, which has been caused by the rapid growth of the city during the last century. One of the effects of this growth is the social conflict of high-rise office towers close to poor informal settlements (favelas). According to the architect Oscar Niemeyer, The first lesson SÄo Paulo offers is that no city should grow so arbitrarily and the second lesson of SÄo Paulo is that its people, and the people of cities in poor countries elsewhere, should have the right to a habitat that is more graceful [8]. For decades, downtown has suffered a continuous degradation process and has gradually been abandoned by its residents. The metropolitan area of SÄo Paulo experiences simultaneously massive urban sprawl in its peripheries and population decrease in its central parts (the population of the central borough of RepÑblica diminished by 28% between 1980 and 1999). Now the average density in the city centre is approximately only 64 inhab/ha [7] in spite of a high built density. The Environmental Atlas of SÄo Paulo [7] includes a map showing surface temperatures in the city of SÄo Paulo. The highest temperatures were found in the central area without green areas and the lowest in urban parks. Similar temperature patterns were found by Duarte and Souza [9]. The central area of SÄo Paulo is formed by the Old Town (SÜ), the New Town (RepÑblica) and another 10 boroughs [10]. In the downtown area, there are seven urban parks. This study focuses on the area of Luz, which is situated in the city centre. 2.2 Field Measurement in São Paulo Downtown Climate monitoring was carried out on 19 th of December 2006 in the area of Luz. This day can be considered a typical hot summer day. The main goal of these measurements was to create an initial database for air temperature, relative humidity, solar radiation, surface temperature, wind direc-

3 tion and speed for the summer period. Researchers from the Laboratory of Environment and Energy Studies LABAUT, from the Faculty of Architecture and Urban Planning at the University of Sao Paulo have collaborated in the measurements as part of a major ongoing study that considered ventilation, urban noise, thermal comfort and green areas. Three main types of urban spaces were defined for the measurements: a park, an urban canyon and an open air square. These three areas were chosen considering the main characteristics of the site and its surroundings (Fig. 1). The particular interest was to compare the effect of vegetation in the park with the other two situations. The sky view factor of the three sites varies considerably (Fig. 2). Figure 1: Location of the three measurement points in Luz borough. Source: Google Earth. May (1) (2) (3) Figure 2: Hemispheric pictures of the sky view taken with a Nikon 4500 equipped with a 180à fish eye lens for park (1), canyon (2) and open square (3). Source: LABAUT The canyon is located in a dense area with 5-10 storeys high commercial buildings. The street has an intense flux of vehicles and pedestrians attracted by the local commerce of electro-electronics devices. The average block size in Luz is 60m by 100m. The street canyon has an average width of 12m and the building height is on average 26m (corresponding to eight-storey buildings) which gives a height to width ratio (H/W) of 0.5. There are very few trees in the street due to narrow sidewalks and problems with aerial electric cables and telephone lines. The open square, which has white stone paving is more exposed to direct solar radiation than the other sites due to the lack of vegetation and shading.

4 Three meteorological stations (Huger WM 918 and 968, as well as ELE MM900) were used to measure air temperature, humidity and wind speed simultaneously at 1.10m height at the three locations. All measurements were recorded on data loggers every ten minutes between 7:00 and 19:00 local daylight saving time. On the square, additionally global solar radiation was measured with a pyranometer. Surface temperatures of various construction materials and natural surfaces were measured at the square using an infrared thermometer TFA Besides the measurements, interviews with pedestrians regarding their thermal comfort sensation and their opinion on benefits and problems with vegetation in the city were carried out. The results, which are shown at Local Time (in this case equal to daylight saving time), gave important information about the microclimatic differences among the three urban sites. The ambient air temperature profiles show that the cooling effect of the park is on average 2ºC compared to the open square with peaks up to 6ºC. Compared to the canyon, the air temperature of the park is about 2.5ºC lower around noon. The relative humidity was about 10% higher in the park than in the other two sites (Fig. 3). The absolute humidity (g/m³) is also higher in the park. The lower temperature and higher humidity in the park is due to shading and evapotranspiration. Temperature ( C) :00 9:00 11:00 13:00 15:00 17:00 19:00 Local Time Temp Park Temp Canyon Temp Square Gráfico 1: Comparativo das temperaturas do ar obtidas na primeira etapa de medição (em ºC). Figure 3: a) Results of air temperature distribution for the park, the canyon and the open square. b) Results of relative and absolute humidity distribution for the park, the open square and the canyon. 2.3 ENVI-met Parametric Simulations without Vegetation RH Park RH Canyon RH Square Abs. hum. Park 10 Abs. hum. Canyon Abs. hum. Square 0 7:00 9:00 11:00 13:00 15:00 17:00 19:00 Local Time The micro scale model ENVI-met [11] was chosen for this study due to its advanced approach on plant-atmosphere interactions in cities. The numerical model simulates aerodynamics, thermodynamics and the radiation balance in complex urban structures with resolutions between 0.5m and 10.0m according to the position of the sun, urban geometry, vegetation, soil and various construction materials by solving thermodynamic and plant physiological equations. The model calculates the energy balance for long and short waves, considering total radiation, air fluxes, air temperature, humidity, local turbulence, reflection from buildings and vegetation. It also determines the superficial temperature (pavement and building envelope), water exchange, soil temperature and biometeorological parameters such as Effective Temperature and Predicted Mean Vote - PMV [12]. Preliminary studies were done considering field measurements data for the initial simulations in ENVI-met. These parametric studies were important to investigate the model main variables and verify the influence for each parameter in the input data. Before inserting the green areas, it was fundamental to adjust the model values to the same climatic conditions for São Paulo. After the calibration process similar daily temperature and humidity curves were achieved. The comparison between measured and simulated air temperature is shown in Fig. 4. Relative Humidity (%) Absolute Humidity (g/m3)

5 T (àc) Measured Simulated 22 07:00 09:00 11:00 13:00 15:00 17:00 19:00 Figure 3: Comparison between measured and simulated air temperature for the existing street canyon. 2.4 ENVI-met Parametric Simulations with Vegetation A vegetation data basis was created based on previous data survey that pointed out the main tree species used in the city and their canopy characteristics quantified by Leaf Area Density LAD. This parameter is defined as the total one-sided leaf area (mâ) per unit layer volume (mä) in each horizontal layer of the tree crown. In this research, LAD values were calculated based on the Peper and McPherson [13] methodology that uses photographic images to determinate the leaf density. In these studies, a dense tree canopy (T4) was chosen for ENVI-met. The leaf area indices (LAI) for T4 was 10 mâ/mâ and the tree type was 10m high and its leaf density started 3m above the ground to avoid obstruction for wind flux at the pedestrians level. The parametric tree model had ellipsoid leaf area distributions with maximum Leaf Area Densities (LAD) located in the middle of the crowns (Fig. 5). Figure 4: LAD (Leaf Area Density) (mâ/mä) in 10 layers of the tree models T4. The input data for the simulations are shown in Table 1 and it is based on the results obtained in the parametric studies without vegetation Table 1: Input configuration data applied in the ENVI-met simulations. Start Simulation at day Wind Speed in 10m ab. Ground [m/s] 0.8 Wind Direction (0:N/ 90:E/ 180:S/ 270:W) 170 Initial Temperature Atmosphere [K] 297 Specific Humidity in 2500m [g Water/ kg air] 9

6 Relativ Humidity in 2m [%] 70 Initial Temperature Upper Layer (0-20cm) [K] 295 Initial Humidity Upper Layer (0-20cm) [%] 50 Firstly a Base Case model was created without vegetation and the input area domain was formed by 9 blocks with 9.600m 2 each, 0.66 for plot ratio and buidings with 24m high in a perimetral shape. Based on this form, another four different scenarios were defined varying the green area distribution: Scenario 1 with a central park occupying one entire block; Scenario 2 presents a park in a linear form with a stream in the middle, occupying three blocks; Scenario 3 with small pocket parks in every block and Scenario 4 with trees only in the sidewalks. Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4 The air temperature results were simulated for December 19 TH at 2p.m. All results are shown at Local Time, which in our case is equal to daylight saving time (summer time) that is local time +1 h. Base Case Scenario 1 Scenario 2 Legend Scenario 3 Scenario 4

7 Regarding superficial temperatures the following results were obtained: Base Case Scenario 1 Scenario 2 Legend Scenario 3 Scenario 4 3. Analysis of Results Comparing the Base Case to the scenarios with green areas we observe that vegetation reduced up to 1.5ºC the air temperature. Over paved areas the simulations showed an average air temperature of 27.5ºC at 14h and 26.8ºC for concrete. The lower air temperature occurred in the Scenario 2, where the linear form of the park contributed to wind channeling and removed part of the storaged heat. The average air temperature in this case was around 25ºC and absolute humidity was approximately 15g/m 3. The results for superficial temperature indicated that trees have a significant impact by lowering the surface temperature by up to 29ºC. The tree canopy blocks part of the direct solar radiation and avoid the rapidly surface warming up. In the street, the superficial temperature varies from 55ºC to 58ºC, while in the green areas, it goes between 28,5ºC and 31ºC. 4. Final Considerations According to Givoni [14], large urban parks often have an important role in establishing the image of a city and in providing areas for large gatherings and social activities. From the climatic aspect it should be noted that the range of the effect of parks, even very large ones, on the climatic conditions within the surrounding built-up areas is rather limited at pedestrian level. Santamouris [15] measured green areas in the city of Athens, Greece and the results indicated that the variation between the air temperature inside and outside a park was up to 3 ºC. Bruse [16] studied the effect of a green park in Melborne, Australia using ENVI-met model to simulate the air temperature in the green area and demonstrated that the difference was up to 2 ºC. In this research, based on microclimatic measures at pedestrian level and energy balance models, particularly ENVI-met, the impacts of vegetation were simulated, aiming to determine the intensity and the spatial distribution of those impacts in the surroundings over the reduction of air tempera-

8 ture and surface temperature and over the increase of air humidity, testing different arrangements (linear park, pocket park, etc) in a case study in SÄo Paulo downtown, Brazil. Concerning the intensity of the effect, the average difference among air temperatures inside the green areas and the surrounding streets is about 1.5ÖC. For surface temperatures, the dense tree shading (LAI=10) has shown average differences of 23ÖC under the canopy. In that sense, green areas can be an important strategy to mitigate the urban heat island in cities due to the shading and evapotranspiration process. Regarding the spacial distribution of vegetation, results show that the effect is very local, and the influence of green areas doesn t go very far from the green borders at pedestrian level. Measurements in Luz area showed that the effect for air temperature can be felt under the canopy and approximately 2m away from the canopy. This result indicates that green areas should be better distributed in small groups of trees than a large park. The effect of tree canopy shadow has also a great potential to ameliorate the microclimate and mitigate heat stress in a hot humid climate. On the other hand the shade efficiency of trees will depend on the type of plants and the amount of leaf density. Because of this parameters such as Leaf Area Index and Leaf Area Density are fundamental to improve the positive impact of vegetation, considering the plant local influence. 5. Acknowledgements This research was supported by FundaãÄo de Amparo å Pesquisa do Estado de SÄo Paulo FA- PESP and Conselho Nacional de Desenvolvimento Cientçfico e Tecnolégico - CNPq. The authors are grateful to Labaut staff for their assistance in the field measurements and to IAG Institute of Astronomy, Geophysics and Atmospheric Sciences - USP for assistance with the simulations. 6. References [1] WILMERS F. Effect of Vegetation on Urban Climates and Buildings. Energy and Buildings, No.15, [2] SAILOR D. J. Simulated urban climate response to modifications in surface albedo and vegetative cover. Journal of Applied Meteorology, Vol.34, [3] SPANGENBERG J. Improvement of Urban Climate in Tropical Metropolis A case study in MaracanÄ/ Rio de Janeiro. M. Eng. Thesis, University of Applied Sciences, Cologne, [4] EMMANUEL M.R. An Urban Approach to Climate-Sensitive Design Strategies for the Tropics. Spon Press, [5] YU C., HIEN W.N. Thermal benefits of city parks. Energy and Buildings, Vol. 38, 2006, pp [6] DIMOUNDI A., NIKOLOPOULOU M. Vegetation in the Urban Environment: Microclimatic Analysis and Benefits. Energy and Buildings, Vol. 35, No.1, [7] SVMA. Atlas ambiental de SÄo Paulo [8] ROMERO S. Destination: SÄo Paulo. Metropolis Magazine, October

9 [9] DUARTE D., SOUZA T. Urban Occupation Patterns and Microclimates in SÄo Paulo Brazil. In: SB Sustainable Buildings, 2005, Tokyo, Japan. [10] MEYER R.M.P., GROSTEIN M.D., BIDERMAN C. SÄo Paulo MetrÅpole. Sao Paulo. Editor University of Sao Paulo, [11] BRUSE M. Simulating the Effects of Urban Environmental on Microclimate with a Three- Dimensional Numerical Model. In: Climate and Environmental Change, Conference Meeting of the Commission on climatology, Evora, [12] FANGER P.O. Thermal comfort, analysis and application in environmental engineering. New York: McGraw Hill, [13] PEPER P.J., MCPHERSON E.G. Comparison of five methods for estimating leaf area index of open-grown deciduous trees. Journal Arboriculture, No.24, 1998, pp [14] GIVONI B. Impact of Planted Areas on Urban Environmental Quality- A review. Atmospheric Environment. Vol 25, No. 3, 1991, pp [15] SANTAMOURIS M. Energy and Climate in the Built Environment. James and James, London, [16] BRUSE M., SKINNER C. J. Rooftop Greening and Local Climate: A Case Study in Melborne. In: Biometeorology and Urban Climatology at the Turn of the Millennium, WMO, pp