Research ArticleArticles

Green Infrastructure with Actual Canopy Parameterization: A Simulation Study for Heat Stress Mitigation in a Hot-Humid Urban Environment

Muhammad Zeeshan, Zaib Ali and Qasim A. Ranjha
Arboriculture & Urban Forestry (AUF) September 2023, 49 (5) 247-267; DOI: https://doi.org/10.48044/jauf.2023.016

Abstract

Background The urban heat island (UHI) phenomenon, resulting from rapid urbanization and aggravated by persistent climate change, is intensifying heat stress and temperature anomalies inside the urban microclimate, requiring the implementation of suitable adaptation measures for sustainable development. The integration of street trees inside the urban landscape is a strategy to alleviate the thermal stress of pedestrians. However, trees have variable potential for the regulation of thermal comfort depending on their different canopy shapes/drag. Therefore, a holistic understanding of tree plantings and species with respect to a particular climate is necessary for urban sustainability.

Methods In this study, computational fluid dynamics (CFD) that employ unsteady Reynolds-averaged Navier-Stokes (URANS) equations were performed using FLUENT solver to analyze the cooling potential of isolated tree species based on 5 morphological characteristics and canopy shapes (i.e., tree height, trunk height, crown width, crown height, and leaf area density) in an urban area.

Results Results revealed a variable temperature regulation (i.e., 0.6 to 1.2 °K) depending on the tree species. Overall, the cooling effect was only observed in the vicinity of the tree canopy. This was due to the availability of shading and increased moisture content provided by the canopy foliage, which blocked shortwave radiation from the sun, as compared to its surroundings.

Conclusions The study findings show that leaf area density is the morphological trait that has the greatest impact on thermal comfort, as it results in low ambient air temperature irrespective of the type of urban density. Additionally, the most effective way to reduce thermal stress is to implement taller trees with uniform foliage density, which will produce a well-ventilated environment.

Keywords

Introduction

During the last century, persistent global warming and increased urbanization have intensified ambient temperatures of urban areas by as much as 4 °K due to the proliferation of impervious surfaces. Extensive growth in urbanization and climate change have severely worsened the thermal environment of urban areas (Lai et al. 2019; Morakinyo et al. 2020). Urban areas are generally degraded by air pollution, surface materials, inadequate green infrastructure, and ventilation (Akbari et al. 2016), all of which trap longwave solar radiation inside urban streets and prevent the dispersion of absorbed energy during the daytime, thus affecting the surface eneigy balance of local urban microclimates. As a result, urban areas experience larger ambient air temperatures than their rural surroundings, giving rise to the urban heat island (UHI) phenomenon (Oke 1982). In turn, a greater UHI effect worsens heat waves and heat stress caused by climate change (Mazdiyasni et al. 2019). Higher urban temperatures directly impact electricity demand as the need for air conditioning grows within urban microclimates, increasing as much as 13.5% (Santamouris et al. 2015). Rising outdoor temperatures and extensive growth in urbanization require suitable adaptation measures for sustainable urban development (Rizwan et al. 2008; Gromke et al. 2015). Adaptation measures include street vegetation (Toparlar et al. 2018; Morakinyo et al. 2020), water bodies (Yang et al. 2020; Zeeshan and Ali 2022a), urban geometry (Jamei et al. 2016), surface materials (Santamouris and Yun 2020; Zeeshan and Ali 2022b), and the reduction of anthropogenic heat sources (Rizwan et al. 2008).

Street vegetation is strongly advocated to address current and future global warming due to its attenuation of solar radiation absorption (Rahman et al. 2020), shade creation (Massetti et al. 2019), and evapotranspiration production (Gromke et al. 2015; Zhang et al. 2019). However, these studies were performed by modeling the tree as a cuboid-shaped porous zone, characterized by the pressure loss and permeability coefficients and aerodynamic turbulence, in computational fluid dynamics (CFD) simulations due to the challenging nature of discretizing the complex tree canopy shapes. This modeling difference led to an underestimation of actual drag offered by tree canopy, resulting in inaccurate modeling of its aerodynamic effects. However, Zeeshan et al. (2022a) studied the effects of actual canopy drag of a specialized tree by modeling the actual tree performance under representative hot-humid climatic conditions and found a significant difference in air temperature, surface temperature, and apparent temperature (0.8 °K, 4 °K, and 0.6 °K, respectively) when compared with simulation results obtained with a tuned value of canopy drag coefficient as employed by previous studies.

In reality, the microclimatic benefits of vegetation, such as reduced heat absorption, shade, and transpiration rate, are strongly affected by tree species and morphological characteristics (i.e., leaf density, tree height, crown width, tree architecture, and proximity to each other) alongside tree size and shape (Berry et al. 2013). This poses a dire need for research and analysis to move away from the generalized tree to site-specific trees for better thermal conditions in the urban environment. Various studies, both numerical and experimental, have been performed in recent years to observe the impact of species characteristics on thermal comfort. Morakinyo et al. (2020) studied the impact of various tree species and their configurational parameters on thermal comfort in simple canyons and found a varying cooling effect ranging from 0.3 to 1.0 °K in the daytime temperatures and 0 to 2.0 °K in the nighttime. This study reported the leaf area density (LAD), tree crown, and trunk height as the most influential parameters. Teshnehdel et al. (2020) studied the effect of different tree species and foliage cover on microclimate and found a significant reduction of 0.29 °K in air temperature (Ta) with an increase in coverage fraction during the summer. The varying cooling potential by species owes its effectiveness to morphological and structural variability, which has not been extensively studied.

Almost all previous studies focused on species performance in open environments, parks, and generic symmetric canyons with fixed sky view factor (SVF) rather than real canyons with variable SVF across asymmetrical canyons and miscellaneous street orientations. Such orientation and scenarios strongly affect solar access due to shadowing effects, trapping of longwave radiation, humidity increases, and variable wind-flow patterns. Moreover, the cooling effect of tree vegetation with species-specific canopy drags modeled has not yet been investigated numerically in real urban areas with hot-humid climates. Furthermore, the comparative analysis of results from simulations, based on realistic case studies in an isolated street surrounded by buildings with different height to width ratios, can provide a better understanding of the real contributions and potential limitations of each species to the overall thermal profile of a place (Morakinyo et al. 2017). Such locations are always challenging due to variation over time and space in microclimatic boundary conditions, which affects vegetation effectiveness due to impaired transpiration. Thus, a detailed study regarding the evaluation of different tree species in hot-humid climates is essential.

In this context, the current study aimed to fill this gap by providing detailed insight into the effectiveness of different tree species, with their actual canopy drag modeled, toward heat stress mitigation under representative hot-humid climatic conditions. This was accomplished through undertaking various CFD simulations employing the finite volume method with unsteady Reynolds-averaged Navier-Stokes (URANS) equations using a well-validated CFD tool (i.e., ANSYS FLUENT). The study results helped in making appropriate selections of the best tree species under representative climatic conditions in a real urban environment (I. I. Chundrigar Road, Karachi, Sindh, Pakistan). In addition, this research suggested more suitable and adaptable tree species for urban settings.

Description of Study Area

The heat-mitigation potential of street vegetation was evaluated in an urban area of Karachi, Pakistan, the 12th most densely populated urban area in the world (Sajjad et al. 2009; Qureshi 2010; Sajjad et al. 2015). One reason for studying this city was its vulnerability to severe heat indexes and frequent heatwaves resulting from the rapid increase in urbanization and climate change (Kovats and Akhtar 2008). The subarea of Karachi, I. I. Chundrigar Road as portrayed in Figure 1, was selected for the current study because it is among the hottest regions in Karachi during the summer, with high humidity levels resulting from seawater evaporation. The selected area is approximately 1.2 km2 as a circular subdomain, and the highest building stands at 120 m. This location (67 °N, 24.8 °E) has a time zone of+5 Greenwich mean time. The climatic parameters, used in current simulations, were measured on an hourly basis for the heatwave period of 2015 by the Pakistan Meteorology Department (PMD) from the Kiamari observatory and are tabulated in the Appendix.

Figure 1.
Figure 1.

(A) Location of urban microclimate understudy in Karachi, Pakistan; (B) Aerial view of the area with surroundings. Red circled zone modeled in CFD simulation (Zeeshan and Ali 2022b; Zeeshan et al. 2022b).

Methods

Numerical Setting (CFD)

The CFD simulations were performed using a finite volume method (FVM) along with a realizable kinetic energy-dissipation rate (k-ε) turbulence model (Shih et al. 1995) in the commercially available tool ANSYS FLUENT 16.2 (ANSYS 2016). The governing flow equations for velocity and turbulence were solved through the URANS equations model. Moreover, the urban flow best practice guidelines (Tominaga et al. 2008; Franke et al. 2010) were used for performing the CFD simulations. A computer-aided design (CAD) tool was adopted to model the buildings with their configurations for the studied urban environment. The standard wall function, as proposed by Launder and Spalding (1974) with modified sand grain-based roughness, was imposed on wall type boundaries to resolve the turbulence at the wall. An aerodynamic roughness length (Zo) of 0.03 was set for all buildings, including ground inside the inner circular subdomain, and ranged from 0.03 to 1 for ground surface outside of the circular subdomain. Moreover, the equations of turbulence, energy, and mean flow were resolved with the use of a second-order discretization scheme to avoid numerical diffusion (divergence) in solution variables. The Pressure-Implicit with Splitting of Operators (PISO) algorithm was used for its stability and application for transient flow, and the discrete ordinates (DO) radiation model was used (Gromke et al. 2015; Toparlar et al. 2018) for its ability to model transmissivity, or transparency, of windows and mirrors. The solar irradiation and radiative transfer were handled with a solar-ray tracing model and converges were set for study results at 10−4 for all flow variables for each time step.

Computational Domain and Grid Discretization

The computational domain along with its grid discretization were compiled using the best practice guidelines of Blocken (2015). With the inner circular domain of 500 m and an average building height of 60 m as referenced in Figure 2A, the domain computed had a length, width, and height of 1700 m, 1100 m, and 360 m, respectively. The nonphysical boundaries of the domain were kept at least 5 times the height of the highest building (Hmax) from the circular subdomain, except for the outflow boundary of computational domain which was 15 times Hmax (Tominaga et al. 2008). The domain constructed with buildings for CFD simulations had a frontal area of less than 3% in the computational domain in its flow direction (Franke et al. 2010; Blocken 2015). Pointwise V18.1 (Pointwise 2017), a commercially available tool, was employed to mesh the domain using best practice guidelines of Blocken (2015). The grid was first created on surfaces of building envelopes, roofs, ground, and sides of the computational domain with at least 10 cells on the smallest side/surface. The volume grid/mesh was then composed using a grid extrusion technique with a maximum stretching ratio of 1:2. High resolution of the mesh (i.e., up to 1-m edge length) was imposed on the inner circular domain, resulting in 21,600 triangular elements and 732,500 quadrilateral elements (Figure 2B). This resolution was intentionally decreased on the nonphysical boundaries to reduce the computational cost without compromising the discretization quality. Volume mesh on the whole computational domain was then composed using surface grid extrusion techniques, resulting in 23 million cells. Grid independency results are referenced in the Appendix.

Figure 2.
Figure 2.

(A) Computational domain; (B) Mesh for the circular subdomain, buildings, grounds, and vegetation zones.

Boundary Conditions

The simulation was performed with the following inlet velocity and turbulence profiles, proposed by Richards et al. (1993), which were selected based on their application with various turbulence models, and is mathematically described as: Embedded Image 1 Embedded Image 2 Embedded Image 3 where Embedded Image and where Cμ, U*, and Uref referred to the von Kármán constant with a value of 0.42, the friction velocity, and velocity of wind at the reference height (height of PMD Observatories).

The symmetry boundary conditions were applied at the lateral and top sides of the computational domain and had zero gradients of all parameters. The pressure outlet condition was imposed at the outflow. The buildings inside the circular sub-domain and ground of the computational domain corresponded to the wall. The building was modeled implicitly with an equivalent thickness of 0.35 m, assigned brick material, enabled conduction equations, and treated as air-conditioned at a temperature of 296.15 °K. The thermal and radiation conditions imposed on walls had a heat transfer coefficient of 0.5 W/m2/K, a free stream temperature, and a radiation temperature of 296.15 °K. The free stream temperature and radiation temperature governed the building’s internal conditions and ground conditions at a depth of 10 m. The detailed material properties for all surface elements in terms of thermal conductivity, specific heat, density, emissivity, and absorptivity are shown in Table 1. Absorptivity refers to albedo values of particular materials. Regarding the selected material for the surface elements, earth with concrete material was imposed on the ground beneath the building while shaded ground refers to zone directly under the canopy zone.

View this table:
Table 1.

Material specifications and components used in this study (Lin et al. 2019).

Modeling the Effects of Vegetation

The vegetation effect on airflow in CFD simulations was modeled by employing the momentum (SUi), turbulent kinetic energy (Sk), and dissipation rate’s sink/source terms (Sε) into transport equations through user-defined functions. These terms were computed by Green (1992), Liu et al. (1996), and Sanz (2003) and are given below. Embedded Image 4 Embedded Image 5 Embedded Image 6

In these equations, tree characterizing parameters are the drag coefficient (cd,cfd and leaf area density (LAD). The tuned value of this coefficient (cd,cfd is 0.15 (Gromke et al. 2015; Toparlar et al. 2018). However, the actual form drag coefficient is used based on the canopy shape of tree species. The used value of model coefficients βd is 5.0 and βp ranges from 0 to 1, and the empirical coefficient Cε4 and Cε5 is 0.9. LAD’s value is 3.0 m2/m3 (National Parks Board 2017) for common tree species (Guaiacum offinale, Azadirachta indica, and Peltophorum pterocarpum) present in the studied urban environment. Another species (Bauhinia × blakeana) with a LAD of 4.41 m2/m3 was also studied to simulate the effect of dense foliage density. The density of air was ρ. Ui and U was the velocity of fluid in “i” direction and flow velocity, respectively. K was the turbulent kinetic energy.

The vegetation effect on temperature was also modeled in this study by using a time-dependent energy source term (i.e., volumetric cooling power [Pc][W/m3]) to consider its unsteady behavior (Toparlar et al. 2018). This was incorporated into the energy equation through a user-defined function (UDF), given as follows: Embedded Image 7 where ETP = (0.0252Ti – 0.078)ETeq,h and ETeq, h = Rh/λρ denoted the potential evapotranspiration and equivalent hourly evapotranspiration in m/h. Rh (MJ/m2/h) was the incoming hourly solar radiation per unit area. The latent heat of vaporization was λ and λwh in Wh/kg and Watt-hours terms. The density of water was ρ. The shading of the tree canopy was modeled with reduced absorptivity of the ground (Equation 8). Embedded Image 8 where αshaded and αopen referred to the ground surface solar absorptivity with and without trees below its canopy. Shading factor (SF), responsible for the regulation of solar radiation absorbed and reflected from the tree canopy, had average value of 0.88 (Nowak 1996).

Vegetation was modeled as a cuboid zone in CFD simulation due to complex discretization involved with tree canopy. This simplified domain results in the inflated volume of green vegetation and an underestimation of the drag coefficient. Thus, a drag correction factor for different canopy-shaped trees was incorporated to model the realistic canopy shape (Zeng et al. 2020) and is described as: Embedded Image 9 where F was the corrective factor for each canopy shape and was taken from the study of Zeng et al. (2020). The value of drag corrective factors for representative paraboloid-shaped, cone-shaped, and spherical-shaped canopy trees is compiled in Table 2.

View this table:
Table 2.

Canopy characteristics.

Tree Classification Characterization Scheme

This study applied a generalized classification scheme based on 5 important structural parameters, assessed mainly by the landscape tree databank (National Parks Board 2017). Based on this scheme, 5 cases were compiled to model trees in ANSYS FLUENT. Each of the morphological parameters is divided into 2 classes: foliage density (sparse ≤ 3 and dense), crown height (short vs. tall), crown width (narrow and wide ≤ 9), and trunk height (high and low ≤ 2). For the selected tree species (i.e., Guaiacum offinale, Azadirachta indica, Peltophorum pterocarpum, and Bauhinia × blakeana), the average values of their morphological characteristics (i.e., tree height [HT], trunk height [TH – height of tree trunk from ground base to the bottom of live crown], crown height [CH], crown width [CW], and leaf area density [LAD]) are tabulated in Table 3. All selected tree species present inside the studied urban area were evergreen and represented a specific canopy shape structure.

View this table:
Table 3.

Physical configuration of the studied tree species. HT = tree height; TH = trunk height (from ground base to bottom of the crown); CW = crown width; CH = crown height; LAD = lead area density (National Parks Board 2017).

In addition to common species, the effect of the Bauhinia × blakeana tree was also simulated with its greater LAD and was proposed for planting in Karachi for its capability of reducing temperature. The LAD distribution for Bauhinia × blakeana was 0 m2/m3, 0.17 m2/m3, 0.30 m2/m3, 0.52 m2/m3, 0.87 m2/m3, 1.27 m2/m3, 1.28 m2/m3, 0.03 m2/m3, and 0 m2/m3 at crown height of 2 m, 3 m, 4 m, 5 m, 6 m, 7 m, 8 m, and 9 m, respectively. In ANSYS, vegetation was modeled as a porous zone incorporated with canopy shape correction (i.e., form drag coefficient and heat transfer occur through its interaction with the surroundings through energy absorption and evapotranspiration).

It is pertinent to mention that Guaiacum offinale is recognized with sparse foliage density, low trunk height, short crown height, and narrow crown width (Tree Case [TC]-1), while Bauhinia × blakeana is characterized by dense foliage density, low trunk height, short crown height, and narrow crown width (TC-2)(Table 4). By contrast, Azadirachta indica has sparse foliage density, high trunk height, tall crown height, and narrow crown width (TC-5), while Peltophorum pterocarpum has sparse foliage density, low trunk height, tall crown height, and wide crown width (TC-3 and TC-4)(Table 4).

View this table:
Table 4.

Description of tree species study cases. HT = tree height; TH = trunk height (from ground base to bottom of the crown); CW = crown width; CH = crown height; LAD = lead area density; CD = coefficient of drag.

Heat Stress (Thermal Comfort)

A simple bioclimatic index (apparent temperature) was adopted to represent heat stress or thermal comfort to study an adjustment of temperature at different humidity levels. Mathematically, it was a function of vapor pressure, velocity, relative humidity, and air temperature (Steadman 1984) and is given as: Embedded Image 10 where Embedded Image represented vapor pressure and RH represented the relative humidity. Q was the net radiation absorbed per unit area of the body surface in W/m2. U, Ta, and TA represented flow velocity, air temperature, and apparent temperature, respectively.

Results

The evaluation zones and points selected for species analysis were based upon surface temperature distribution of the reference case which was simulated for a 5-day heatwave period from 2015 June 18–22. First, the base case for vegetation (TC-1)(Table 4) was modeled for 2015 June 19 by employing the tuned value of drag coefficient and energy source terms for Equations 7 and 8 in the flow governing equations through user-defined functions. The same vegetation model was first validated using the sub-configuration method, for which the experimental study of Shashua-Bar et al. (2011) was referenced, to simulate its transpiration cooling power (Appendix). After the base case, separate simulations, as articulated in Table 4, were performed with different morphological parameters. However, the results were compared as averaged data of the evaluation parameters, due to the localized vegetation effect, and as box plots based on discrete data taken at some discrete locations.

Evaluation Zones Identification and Validation of Surface Temperatures

To make a comparative assessment of the cooling effect of street trees inside a hot-humid urban area, the reference case was first modeled without imposing the representative vegetation zone. The purpose of this modeling scenario was threefold: first, the identification of hotspots for vegetation intervention; second, simulating the in-situ conditions of the heatwave period; and third, validating the surface temperature results obtained from satellite data with the current study. To model this case, the inlet conditions for flow velocity, temperature, and humidity were taken from the meteorological department. Based on the magnitude of these variables, 15 transient scenarios were simulated (Appendix) through a finite volume method using URANS with numerical settings for a complete 5-day heatwave period (2015 June 18–22). The results were reported in terms of surface temperature contours for 2 different times (11 AM and 3 PM) of 2015 June 19 (Figure 3). These time conditions were selected because the flow conditions were relatively stable at these times of day. Each color in Figure 3 represents a particular temperature. Higher surface temperatures were present in isolated spaces (i.e., streets due to proliferation of impervious surfaces) while lower temperatures were observed at building facades and vertical envelopes due to shorter solar access. There was considerable cooling in open spaces due to lower absorptivity and higher wind velocity. Based on the temperature distribution, the isolated streets were chosen for incorporation of tree vegetation for appraising their mitigation effectiveness. The zones evaluated in open spaces are depicted in Figure 4A. Three areas (first, second, and third zones) were selected to analyze the effectiveness of street trees toward modulating the studied area’s variable building configurations and morphology. Two layers of tree vegetation were modeled with center free space in the first zone to predict their effect on pedestrian thermal comfort in narrower and wide-open spaces. In this regard, 17 points were selected (Figure 4B) to measure the effect of tree species and their morphological characteristics on thermal comfort and for making quantitative comparisons of cooling effectiveness. Points 1 to 3 were located on the windward side of vegetation zones while Points 4 to 7 were located on the downward side of the canopy zones. Points 8 to 15 were located on the downwind side of the vegetation zone but were also surrounded by buildings to cater to the effect of the sky view factor, while points 16 and 17 were located between the 2 vegetation zones.

Figure 3.
Figure 3.

Contours of surface temperature: (A) 11 AM (Greenwich Mean Time [GMT+5]); (B) 3 PM (GMT+5)(Zeeshan and Ali 2022a; Zeeshan and Ali 2022b; Zeeshan et al. 2022a).

Figure 4.
Figure 4.

(A) Representation of vegetation zones highlighted in red; (B) The locations of discrete monitoring points; (C) Comparison of surface temperatures between CFD and satellite measurements for the heatwave period (2015 June 18–22); (D) Comparison of air temperature between CFD and Shashua-Bar study (Shashua-Bar et al. 2011).

The validation of surface temperature obtained from current study results was made with Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data (Wan 2015) and was described in detail by Zeeshan et al. (2022). MODIS, with the bandwidth of 3.66 nm to 3.84 nm and pixel size of 1,000 m, validated CFD results in urban microclimates (Zeeshan et al. 2022b) which were then validated with satellite imagery data extracted using the MODIS database. MODIS had a spatial resolution of 1.1 km for monitoring the surface temperature in the form of a Hierarchical Data Format (HDF) file for any specific coordinate. The HDF file was then fed into QGIS 2.18 (Belgacem et al. 2019) to extract the single average value of surface temperatures for the studied location. After obtaining the surface temperature results, the 2 were compared for the 5-day heatwave period (Figure 4C). CFD results were quite consistent with satellite imagery data, as the resulting value of correlation coefficient (R2) from regression analysis was 0.965. However, noticeable deviations (i.e., of 3.4 °K and 2.0 °K at noon on 2015 June 18–20) were observed, which could be attributed to reporting the averaged value of results for different intensities of temperature and velocity, not considering the thermal stratification of the atmospheric boundary layer (ABL) in the CFD simulation, and taking the average value of results at different areas in the numerical and experimental approach.

Validation of the Vegetation Cooling Model

The effect of transpiration on air temperature is usually modeled through a cooling power-based source or sink term in the canopy area. Shashua-Bar et al. (2011) measurement results were used for comparison since they had been used by Gromke et al. (2015) and Toparlar et al. (2018) to validate transpiration cooling power (Appendix). Another reason for using this study was the ease of replicating the boundary/modeling conditions. Shashua-Bar et al. (2011) examined the vegetation effects on the urban environment for constructed, semi-enclosed courtyards with various vegetation types and arrangements. For this validation, steady-state simulations were performed and compared with the measurement data. The comparisons were made for 2007 July 7 in terms of averaged data. The courtyard contained a structured mesh of 0.15-m edge comparable to the one used in Shashua-Bar et al. (2011). On the outer walls of the buildings, an unstructured grid was applied with a gradual increase in cell sizes toward the boundaries of the domain, resulting in a total of 1.9 million cells.

A realizable k-ε turbulence model was used to perform steady-state Reynolds-averaged Navier-Stokes (RANS) simulations. Solar irradiation and radiative transfer were modeled using solar-ray tracing and the P1 radiation model. The simulation used input parameters such as velocity and temperature. The zero fluid velocity at the wall (no-slip condition) was imposed on the ground and building wall and assigned an aerodynamic roughness length Zo = 0.3 m. The aerodynamic effects of trees inside the courtyard were incorporated by employing the transport equation of momentum, turbulent kinetic energy, and turbulence dissipation rate (Equations 4 through 6) via UDF. Moreover, the volumetric cooling power of vegetation was provided as a source term to the porous zone cells, which also varied with leaf area density. At the inflow boundary, a vertical profile for mean velocity, K, and ε were employed as given in Equations 1 through 3. Outlet pressure was prescribed on the outflow side and symmetry boundary conditions on top and lateral sides.

When vegetation absorbs the incident solar radiation and releases the latent heat, the surface temperature of the leaves and air temperature amongst the vegetation decreases. The amount of incident solar radiation on the ground under the vegetation also decreases due to the absorption of solar radiation by the vegetation. The ground surface temperature decreases, while the convective heat transfer from the ground also decreases. Finally, the air temperature between the vegetation and the ground surfaces decreases.

Figure 4D shows the comparison of simulation results and measurement data for the bare courtyard for cooling power (PC = 750 W/m3), which indicates that air temperature exhibited a satisfactory relationship between the measurement and CFD results; the resulting value of correlation coefficient (R2) from regression analysis was 0.9966. It is evident from the graphs that CFD predicted values are slightly greater than the measurement data. The maximum difference of 1.5 °K exists between the 2 results for 750 W/m3 cooling power at 3 AM which is 0.5 °K at 3 PM during sunshine hours.

As both the validation studies had a correlation coefficient (R2) in range of 0.965 to 0.9966, where a reasonable agreement was obtained, it is thus obvious that CFD can predict the urban microclimate accurately. The results of CFD can therefore be exploited to identify the problem areas and appraise the impact of mitigation measures.

Base Case for Vegetation (TC-1)

The basic vegetation case was first modeled with transpiration-rated cooling power, according to Equations 7 through 9, with a tuned value of canopy drag coefficient for 2015 June 19. This case served as a basic case for carrying out a comparative analysis based on the cooling effect of different morphological characteristics of tree species. Air temperature contours at pedestrian height were recorded for 2 different periods of the day (i.e., 11 AM and 3 PM)(Figures 5A and 5B). In addition, contours of air temperature difference between the base case of vegetation and reference case, simulated without incorporating the vegetation, have also been reported to ascertain the intensity of the cooling effect of vegetation (Figures 5C and 5D).

Figure 5.
Figure 5.

Contours of air temperature on 2015 June 19 at z = 1.5 m: (A) for 11 AM (GMT+5); (B) for 3 PM (GMT+5). Contours of air temperature difference for vegetation base case minus reference case: (C) for 11 AM (GMT+5); (D) for 3 PM (GMT+5).

From the contours of air temperature difference, the cooling effect was only noticeable near the vegetation zone. This was due to the availability of additional shading and transpired vapor provided by canopy foliage at this location as compared to its surroundings. The intensity of the cooling effect with TC-1 at 3 PM was about 1.2 °K (Figure 5D). This reduction in air temperature was due to the blockage of shortwave radiation by the tree foliage and replacement of the surface with the shaded region below the tree canopy. Previous studies similarly reported a reduction of up to 1 °K (Rahman et al. 2020; Yan et al. 2020). The cooling was also observed in open spaces away from vegetation zones (Figures 5C and 5D). This was owed to the presence of prominent wind velocities at these locations, causing a rapid decrease in temperature on the urban surfaces from sharp spatial heat exchange. This level of cooling in open spaces was also attributed to the lower conduction of heat, resulting in less absorption of solar radiation on surfaces of building elements.

Impact of Morphological Characteristics of Different Tree Species

The potential of the tree species, characterized by their morphological characteristics, toward improving the comfort conditions of the urban area was investigated. Five scenarios were modeled to simulate the impact of various tree configuration parameters (trunk height, crown width, crown height, and leaf area density)(Table 4), as these parameters are highly correlated with the cooling capacity of trees (Morakinyo et al. 2018). The enlisted values of drag coefficient, adopted for these simulation cases, were calculated from the proposed corrective factor of tree species with respect to their relevant canopy shapes.

24-Hour Temperature Distribution

The effectiveness of tree vegetation toward regulating thermal environment was presented as 24-hour data (averaged across the circular domain) and discrete data (located on or near vegetation zones in the urban streets) for the studied tree morphological scenarios.

Figure 6A establishes that TC-2 (highest LAD) provided the largest reduction in air temperature (1.2 °K) followed by TC-3 (0.8 °K) when compared with TC-1 (smallest LAD), while TC-4 and TC-5 provided the smallest reduction in temperature (0.7 °K). For TC-1, temperatures started rising with the amount of solar irradiance and peaked at time of high solar irradiance before decreasing. Throughout the day, the tree’s effectiveness also varied similarly with reduced intensity, owing to continuous changes in shading and evapotranspiration rates that varied with solar access.

Figure 6.
Figure 6.

Results for species morphological characteristics over a 24-hour distribution: (A) air temperature; (B) surface temperature; (C) velocity; (D) apparent temperature.

Likewise, Figure 6B displays the surface temperature distribution throughout the 24-hour study. A reduction of up to 4.0 °K in surface temperature occurred with street trees for scenario of highest LAD (TC -2) when compared with the base case scenario (TC-1). This reduction resulted in low heat access and caused lesser accumulation of solar energy inside the urban area. LAD was the most influential in reducing the actual surface temperature due to the interception of large solar radiation rays by tree foliage, as evident from diurnal variation in Figure 6B. This was then followed by crown height. There was a temperature difference of 3 °K between the trees having lower LAD and crown height (TC-1) and trees with greater height (TC-3) at 3 PM on 2015 June 19. In reducing the surface temperature, Guaiacum offinale appeared to be the least effective. The tree species having high trunk height (TC-5) and crown width (TC-4) were least effective when compared with Bauhinia × blakeana (TC-2).

The distribution of velocity is portrayed in Figure 6C. It was clearly observed that tree vegetation inhibited wind flow inside the urban areas, resulting from aerodynamic performance of the trees. This lower velocity led to reduced ventilation and amplified heat accumulation phenomenon below the canopy zone and windward/downward sides of vegetation and buildings. This decrease was due to the increased roughness of urban surface with the addition of trees and its drag on airflow, which caused a smooth change in wind flow due to the pressure difference created as a result of porous trees (Oke et al. 2017). Tree species with large LAD (TC-2) offered more resistance to the flow, while a tree with a tall trunk (TC-5) offered the least (Figure 6C). This provided better ventilation at all the monitoring locations due to low obstruction to wind flow at the pedestrian height, thus causing a significant improvement in thermal comfort. Tree species with large crown height (TC-3) and crown width (TC-4) provided intermediate ventilation.

Vegetation can be favorable for the urban microclimate since it tends to reduce thermal stress (Yan et al. 2020) due to its ability to reduce radiation and promote air cooling despite decreasing the wind speed. Figure 6D shows the heat stress reduction potential of different tree species in terms of apparent temperature variation. Bauhinia × blakeana (TC-2) was most influential in reducing the actual apparent temperature at the pedestrian height when compared with Guaiacum offinale (TC-1), due to its greater LAD and low height. This was then followed by Azadirachta indica (TC-5) and Peltophorum pterocarpum (TC-3 and TC-4). Thus, irrespective of sky view factor, crown height and width were the least effective for the reduction of heat stress.

Figure 7 presents the boxplots for each scenario summarizing the maximum, minimum, median, average, and interquartile ranges of air temperature, flow velocity, and apparent temperature values simulated at 3 PM. This box plot was based on the data of discrete points, located only near or on the vegetation zones (Figure 8). The impact or cooling potential of each mitigation scenario on in-canyon apparent temperature reductions was expressed as the difference between the reference case and each mitigation scenario in the same spot. Figure 7 summarizes the maximum and average cooling potential for each mitigation scenario with respect to the existing conditions. When comparing ΔTMAX, ΔTAVG, and rank orders (based on TAVG) from 1 (coolest) to 5 (warmest), large LAD and large crown width showed the greatest heat stress reduction.

Figure 7.
Figure 7.

Boxplots for each scenario summarizes the maximum, minimum, median, average, and interquartile ranges of air temperature, flow velocity, and apparent temperature values simulated at 3 PM (Figure 8). Rank orders represent the cooling potential of scenarios based on TAVG values. The green portion represents the range from first quartile to median, while the red portion represents the range from median to third quartile. The circle represents the mean value.

Figure 8.
Figure 8.

Results at discrete monitoring locations for species morphological characteristics: (A) air temperature; (B) velocity; (C) apparent temperature.

Conclusion

The study examined the thermal comfort improvement potential of vegetation based on the intensity of the cooling effect. Comprehensive CFD analysis using finite volume method with different morphological parameters was performed by solving 3D URANS equations for the heatwave period between 2015 June 18 to 22. The vegetation model was evaluated by replicating a previous experimental study through a numerical approach for accessing the cooling effectiveness of street trees. In this work, CFD simulations were performed to analyze the effect of 5 morphological parameters/characteristics with realistic canopies (tree height, trunk height, crown width, crown height, and leaf area density), characterizing different tree species on their overall cooling potential in a real urban area. The following main conclusions were drawn:

  1. Vegetation can be effectively applied to reduce heat stress experienced by pedestrians and promote thermal comfort conditions. For the considered climate, Bauhinia × blakeana is most influential in reducing the apparent temperature at the pedestrian height due to its great leaf area index and small height, while Guaiacum offinale proves to be the least effective. Bauhinia × blakeana resulted in a 1.2 °K decrease in the average temperature and a 1.0 °K decrease in the apparent temperature. For a case with higher LAD and a small tree height, the mitigation potential is higher even for a greater height canopy.

  2. Large trunk heightened trees with well-ventilated foliage density are the most effective at reducing thermal stress because these produce a well-ventilated environment.

The trees considered in this study are native to Pakistan or Asia with one very common across the globe. This suggests that even though the case study was local, the findings and recommendations can be applied to other parts of the world with such dominant trees of similar physical configuration in urban climates. The magnitude of thermal impact, however, may vary due to differences in prevailing weather conditions and urban morphology. Moreover, the study is expected to inform architects regarding retrofitting mitigation of UHI with tree vegetation.

Based upon the findings of this study, we propose that tree planting should be subjected to a performance-based approach in different SVF locations for shifting from generalized to site-specific tree species under any climatic conditions. This will afford optimized heat mitigation inside the city’s landscape at a city/building scale.

Limitations and Future Work

The study performed has certain limitations in terms of imposed boundary condition: (1) incorporating the same material for each urban surface (i.e., concrete for building, earth for grounds, and asphalt for pavement) rather than exact used materials for each surface; (2) analyzing the thermal conditions with one wind direction (i.e., the dominant one) for entire heatwave period; (3) the application of neutrally stratified boundary layers inlet profiles for velocity and turbulence rather than for different atmospheric conditions; (4) missing the effect of anthropogenic heat sources. Thus, these missing aspects are proposed to be studied for future research work, since their application would ensure and improve the result integrity.

In addition, a simplified approach was used for modeling radiative shading and its temporal variability. The shading effect of vegetation is modeled on the ground by modeling the fixed area below the tree canopy, which has a lowered absorptivity value, while the shading direction continuously varies with the solar direction throughout the simulation period, requiring the shaded area to move accordingly. Thus, the complicated temporal variability of tree shade should be modeled for more precise results.

Moreover, the study includes 4 important morphological characteristics (CH, TH, CW, HT) along with LAD, but leaf color/shape and texture also contribute to comfort improvement and should be simulated for realistic modeling of the effectiveness of vegetation species.

Conflicts of Interest

The authors reported no conflicts of interest.

Acknowledgements

The authors are greatly gratified to PMD regarding the provision of meteorological data for the studied heatwave period of the studied microclimate.

Appendix.

View this table:
Table S1.

Meteorological data. DBT = dry-bulb temperature.

Grid Sensitivity Analysis

Grid sensitivity analysis was carried out with 3 successive refined grids using computational settings as described in the Results to verify the integrity of discretization for all CFD simulations. In this regard, a 50% increase or decrease of cells in each direction was imposed on the inner circular domain and at the bottom of the rectangular domain while deploying surface discretization. The 3 types of meshes were referred to as coarse, basic, and fine mesh. This refinement factor of 1.5 on surface grid discretization resulted in the overall 19 million and 27 million volume cells with coarse and fine mesh. The results of the grid sensitivity analysis were reported as the difference of temperature and velocity variable between the base case minus coarse/fine mesh and are given in Figures S1C and S1D. As portrayed from mesh results, no significant improvement was noticed after the basic mesh, declaring basic mesh as a suitable mesh distribution for performing further CFD simulations.

Temperature at Discrete Points

Figure S2 shows the distribution of air temperature, velocity, and apparent temperature for the aforementioned 5 cases at various evaluation points. At the downwind located points (P4 to P7), the smallest temperature was observed with TC-2, followed by TC-3 and TC-1, while TC-4 and TC-5 provided the highest air temperature. At locations present on the downward side of the vegetation zones and surrounded by the nearby buildings (P8 to P15), TC-2 provided the largest decrease in air temperature, followed by TC-4 and TC-3. TC-5 gave the largest increase in air temperature after TC-1, which shows the influence of the H/W ratio (SVF) besides the morphological characteristics of the tree. Overall, LAD was most influential in reducing the actual air temperature at the pedestrian height due to its solar offsetting capacity, as the ambient air temperature significantly decreased with the increase in LAD (Figure S2A). However, this may not be an optimal solution, as this also resulted in increased obstruction to the outgoing longwave radiation and the wind flow, which can cause an increased loss of longwave radiation and a decrease in the heat removal through convection, causing a rise in ambient air temperature (Tsoka 2017; Yuan et al. 2017).

Figure S1.
Figure S1.

Mesh independence study results in (A) air temperature, (B) air temperature difference, (C) surface temperature difference, and (D) velocity difference.

Bauhinia × blakeana (TC-2), characterized by high LAD, largely affected the flow velocity in downward and windward directions of the vegetation zone that was surrounded by low- and high-rise buildings (Figure 7B). Due to its obstructive nature, this resulted in reduced ventilation across the area, which caused heat accumulation (Figure S2C). On the contrary, tree species with large crown width (TC-4) and large trunk height (TC-5) provided better ventilation at all the monitoring locations due to low obstruction of wind flow at the pedestrian height, thus causing a significant improvement in thermal comfort. Tree species with large crown height (TC-3) and low LAD (TC-1) provided intermediate ventilation.

The effectiveness of various tree species was different across the different densities of the studied microclimate. The trend of temperature variation is depicted in Figure S2C. Bauhinia × blakeana (TC-2) and Peltophorum pterocarpum (TC-3) provided large comfort improvement at the pedestrian height in windward and downward directions, which were surrounded by high- and low-rise buildings when compared to other tree species (Figure S2C). At such locations, building shade outweighed the tree shading effect as witnessed in other previous studies (Morakinyo et al. 2017; Morakinyo et al. 2020). In comparison, Guaiacum offinale (TC-1) provides the least reduction in heat stress caused by a small decrease in air temperature due to its small canopy area. The apparent temperature at some spaces having low SVF is even lower than that of open spaces (Lai et al. 2019), which represents the dominant influence of shading over the effects of reduced wind speed. Likewise, in open areas which were considered a suitable option for tree placement due to the absence of building shadowing effect, Bauhinia × blakeana (TC-2), Azadirachta indica (TC-4), and Peltophorum pterocarpum (TC-5), having large LAD, large crown width, and large trunk height, were most effective in improving thermal comfort. This was due to their better ventilation capacity and shading as observed from monitoring points located between the 2 vegetation zone (P15 to P17). This observation corresponded to other previous studies (Morakinyo et al. 2017; Tan et al. 2017; Morakinyo et al. 2020) and suggests the planting of tall trees having large LAD irrespective of the urban density inside the studied microclimate.

Figure S2.
Figure S2.

Results at discrete monitoring locations for species morphological characteristics. (A) air temperature; (B) velocity; (C) apparent temperature.

Literature Cited

  1. ↵
    1. Akbari H,
    2. Cartalis C,
    3. Kolokotsa D,
    4. Muscio A,
    5. Pisello AL,
    6. Rossi F,
    7. Santamouris M,
    8. Synnefa A,
    9. Wong NH,
    10. Zinzi M.
    2016. Local climate change and urban heat island mitigation techniques—The state of the art. Journal of Civil Engineering and Management. 22(1):1–16. https://doi.org/10.3846/13923730.2015.1111934
    OpenUrl
  2. ↵
    ANSYS. 2016. ANSYS FLUENT: Fluid simulation software. Version 16.2. Cannonsburg (PA, USA): ANSYS, Inc. https://www.ansys.com/products/fluids/ansys-fluent
    1. Antoniou N,
    2. Montazeri H,
    3. Neophytou M,
    4. Blocken B.
    2019. CFD simulation of urban microclimate: Validation using high-resolution field measurements. Science of the Total Environment. 695:133743. https://doi.org/10.1016/j.scitotenv.2019.133743
    OpenUrl
    1. Aslam AQ,
    2. Ahmad SR,
    3. Ahmad I,
    4. Hussain Y,
    5. Hussain MS.
    2017. Vulnerability and impact assessment of extreme climatic event: A case study of southern Punjab, Pakistan. Science of the Total Environment. 580:468–481. https://doi.org/10.1016/j.scitotenv.2016.11.155
    OpenUrl
  5. ↵
    1. Belgacem H,
    2. Leduc T,
    3. Musy M.
    2019. Towards a QGIS-based graph carrier of urban information and spotting wind behavior at the pedestrian level. In: 10th International Conference on Urban Climate/14th Symposium on the Urban Environment; 2018 August; New York, United States. https://hal.science/hal-01977551
  6. ↵
    1. Berry R,
    2. Livesley SJ,
    3. Aye L.
    2013. Tree canopy shade impacts on solar irradiance received by building walls and their surface temperature. Building and Environment. 69:91–100. https://doi.org/10.1016/j.buildenv.2013.07.009
    OpenUrl
  7. ↵
    1. Blocken B.
    2015. Computational Fluid Dynamics for urban physics: Importance, scales, possibilities, limitations and ten tips and tricks towards accurate and reliable simulations. Building and Environment. 91:219–245. https://doi.org/10.1016/j.buildenv.2015.02.015
    OpenUrl
  8. ↵
    1. Franke J,
    2. Hellsten A,
    3. Schlünzen H,
    4. Carissimo B.
    2010. The Best Practice Guideline for the CFD simulation of flows in the urban environment: An outcome of COST 732. The 5th International Symposium on Computational Wind Engineering (CWE 2010); 2010 May 23–27; Chapel Hill (NC, USA). p. 1–10.
  9. ↵
    1. Green SR.
    1992. Modelling turbulent air flow in a stand of widely-spaced trees. Journal Computational Fluid Dynamics and its Applications. 5:294–312. https://cir.nii.ac.jp/crid/1570291225289659904
    OpenUrl
  10. ↵
    1. Gromke C,
    2. Blocken B,
    3. Janssen W,
    4. Merema B,
    5. van Hooff T,
    6. Timmermans H.
    2015. CFD analysis of transpirational cooling by vegetation: Case study for specific meteorological conditions during a heat wave in Arnhem, Netherlands. Building and Environment. 83:11–26. https://doi.org/10.1016/j.buildenv.2014.04.022
    OpenUrl
  11. ↵
    1. Jamei E,
    2. Rajagopalan P,
    3. Seyedmahmoudian M,
    4. Jamei Y.
    2016. Review on the impact of urban geometry and pedestrian level greening on outdoor thermal comfort. Renewable and Sustainable Energy Reviews. 54:1002–1017. https://doi.org/10.1016/j.rser.2015.10.104
    OpenUrl
    1. Kazmi SJH,
    2. Mehdi R,
    3. Arsalan MH.
    2008. Karachi: Environmental challenges of a mega city. New Delhi (India): Cambridge University Press. p. 23–36.
  13. ↵
    1. Kovats S,
    2. Akhtar R.
    2008. Climate, climate change and human health in Asian cities. Environment and Urbanization. 20(1): 165–175. https://doi.org/10.1177/0956247808089154
    OpenUrlCrossRef
  14. ↵
    1. Lai D,
    2. Liu W,
    3. Gan T,
    4. Liu K,
    5. Chen Q.
    2019. A review of mitigating strategies to improve the thermal environment and thermal comfort in urban outdoor spaces. Science of the Total Environment. 661:337–353. https://doi.org/10.1016/j.scitotenv.2019.01.062
    OpenUrl
  15. ↵
    1. Launder BE,
    2. Spalding DB.
    1974. The numerical computation of turbulent flows. Computer Methods in Applied Mechanics and Engineering. 3(2):269–289. https://doi.org/10.1016/0045-7825(74)90029-2
    OpenUrlCrossRef
  16. ↵
    1. Lin H,
    2. Xiao Y,
    3. Musso F,
    4. Lu Y.
    2019. Green façade effects on thermal environment in transitional space: Field measurement studies and computational fluid dynamics simulations. Sustainability. 11(20):5691. https://doi.org/10.3390/su11205691
    OpenUrl
  17. ↵
    1. Liu J,
    2. Chen JM,
    3. Black TA,
    4. Novak MD.
    1996. E-ε modelling of turbulent air flow downwind of a model forest edge. Boundary-Layer Meteorology. 77:21–44. https://doi.org/10.1007/BF00121857
    OpenUrl
  18. ↵
    1. Massetti L,
    2. Petralli M,
    3. Napoli M,
    4. Brandani G,
    5. Orlandini S,
    6. Pearlmutter D.
    2019. Effects of deciduous shade trees on surface temperature and pedestrian thermal stress during summer and autumn. International Journal of Biometeorology. 63:467–479. https://doi.org/10.1007/s00484-019-01678-1
    OpenUrl
  19. ↵
    1. Mazdiyasni O,
    2. Sadegh M,
    3. Chiang F,
    4. AghaKouchak A.
    2019. Heat wave intensity duration frequency curve: A multivariate approach for hazard and attribution analysis. Scientific Reports. 9:1–8. https://doi.org/10.1038/s41598-019-50643-w
    OpenUrl
    1. Montazeri H,
    2. Blocken B,
    3. Hensen JLM.
    2015. CFD analysis of the impact of physical parameters on evaporative cooling by a mist spray system. Applied Thermal Engineering. 75:608–622. https://doi.org/10.1016/j.applthermaleng.2014.09.078
    OpenUrl
  21. ↵
    1. Morakinyo TE,
    2. Kong L,
    3. Lau KK,
    4. Yuan C,
    5. Ng E.
    2017. A study on the impact of shadow-cast and tree species on in-canyon and neighborhood’s thermal comfort. Building and Environment. 115:1–7. https://doi.org/10.1016/j.buildenv.2017.01.005
    OpenUrl
  22. ↵
    1. Morakinyo TE,
    2. Lau KKL,
    3. Ren C,
    4. Ng E.
    2018. Performance of Hong Kong’s common trees species for outdoor temperature regulation, thermal comfort and energy saving. Building and Environment. 137:157–170. https://doi.org/10.1016/j.buildenv.2018.04.012
    OpenUrl
  23. ↵
    1. Morakinyo TE,
    2. Ouyang W,
    3. Lau KKL,
    4. Ren C,
    5. Ng E.
    2020. Right tree, right place (urban canyon): Tree species selection approach for optimum urban heat mitigation—Development and evaluation. Science of the Total Environment. 719:137461. https://doi.org/10.1016/j.scitotenv.2020.137461
    OpenUrl
  24. ↵
    National Parks Board. 2017. https://www.nparks.gov.sg
  25. ↵
    1. Nowak DJ.
    1996. Estimating leaf area and leaf biomass of open-grown deciduous urban trees. Forest Science. 42(4):504–507. https://doi.org/10.1093/forestscience/42.4.504
    OpenUrl
  26. ↵
    1. Oke TR.
    1982. The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society. 108(455):1–24. https://doi.org/10.1002/qj.49710845502
    OpenUrlCrossRef
  27. ↵
    1. Oke TR,
    2. Mills G,
    3. Christen A,
    4. Voogt JA.
    2017. Urban climates. Cambridge (UK): Cambridge University Press. 526 p. https://doi.org/10.1017/9781139016476
    1. Park M,
    2. Hagishima A,
    3. Tanimoto J,
    4. Narita K.
    2012. Effect of urban vegetation on outdoor thermal environment: Field measurement at a scale model site. Building and Environment. 56:38–46. https://doi.org/10.1016/j.buildenv.2012.02.015
    OpenUrl
  29. ↵
    Pointwise. 2017. Mesh Generation Software for CFD. Version 18.0. http://www.pointwise.com/support/release-notes-V18R3.html
  30. ↵
    1. Qureshi S.
    2010. The fast growing megacity Karachi as a frontier of environmental challenges: Urbanization and contemporary urbanism issues. Journal of Geography and Regional Planning. 3(11):306–321. https://doi.org/10.5897/JGRP.9000139
    OpenUrl
  31. ↵
    1. Rahman MA,
    2. Hartmann C,
    3. Moser-Reischl A,
    4. von Strachwitz MF,
    5. Paeth H,
    6. Pretzsch H,
    7. Pauleit S,
    8. Rötzer T.
    2020. Tree cooling effects and human thermal comfort under contrasting species and sites. Agricultural and Forest Meteorology. 287:107947. https://doi.org/10.1016/j.agrformet.2020.107947
    OpenUrl
  32. ↵
    1. Richards PJ,
    2. Hoxey RP.
    1993. Appropriate boundary conditions for computational wind engineering models using k-ε turbulence model. Journal of Wind Engineering and Industrial Aerodynamics. 47-46:145–153. https://doi.org/10.1016/0167-6105(93)90124-7
    OpenUrl
  33. ↵
    1. Rizwan AM,
    2. Dennis LYC,
    3. Liu C.
    2008. A review on the generation, determination and mitigation of urban heat island. Journal of Environmental Sciences. 20(1):120–128. https://doi.org/10.1016/S1001-0742(08)60019-4
    OpenUrl
  34. ↵
    1. Sajjad SH,
    2. Blond N,
    3. Batool R,
    4. Shirazi S,
    5. Shakrullah K,
    6. Bhalli M.
    2015. Study of urban heat island of Karachi by using finite volume mesoscale model. Journal of Basic & Applied Sciences. 11:101–105. https://doi.org/10.6000/1927-5129.2015.11.13
    OpenUrl
  35. ↵
    1. Sajjad SH,
    2. Hussain B,
    3. Ahmed Khan M,
    4. Raza A,
    5. Zaman B,
    6. Ahmed I.
    2009. On rising temperature trends of Karachi in Pakistan. Climatic Change. 96:539–547. https://doi.org/10.1007/s10584-009-9598-y
    OpenUrl
  36. ↵
    1. Santamouris M,
    2. Cartalis C,
    3. Synnefa A,
    4. Kolokotsa D.
    2015. On the impact of urban heat island and global warming on the power demand and electricity consumption of buildings — A review. Energy and Buildings. 98:119–124. https://doi.org/10.1016/j.enbuild.2014.09.052
    OpenUrl
  37. ↵
    1. Santamouris M,
    2. Yun GY.
    2020. Recent development and research priorities on cool and super cool materials to mitigate urban heat island. Renewable Energy. 161:792–807. https://doi.org/10.1016/j.renene.2020.07.109
    OpenUrl
  38. ↵
    1. Sanz C.
    2003. A note on k-epsilon modelling of vegetation canopy air-flows. Boundary-Layer Meteorology. 108(1):191–197. https://doi.org/10.1023/A:1023066012766
    OpenUrl
  39. ↵
    1. Shashua-Bar L,
    2. Pearlmutter D,
    3. Erell E.
    2011. The influence of trees and grass on outdoor thermal comfort in a hot-arid environment. International Journal of Climatology. 31(10): 1498–1506. https://doi.org/10.1002/joc.2177
    OpenUrl
  40. ↵
    1. Shih T,
    2. Liou W,
    3. Shabbir A,
    4. Yang Z,
    5. Zhu J.
    1995. A new k-∈ eddy viscosity model for high reynolds number turbulent flows. Computers & Fluids. 24(3):227–238. https://doi.org/10.1016/0045-7930(94)00032-T
    OpenUrlCrossRef
  41. ↵
    1. Steadman RG.
    1984. A universal scale of apparent temperature. Journal of Applied Meteorology and Climatology. https://doi.org/10.1175/1520-0450(1984)023<1674:AUSOAT>2.0.CO;2
  42. ↵
    1. Teshnehdel S,
    2. Akbari H,
    3. Di Giuseppe E,
    4. Brown RD.
    2020. Effect of tree cover and tree species on microclimate and pedestrian comfort in a residential district in Iran. Building and Environment. 178:106899. https://doi.org/10.1016/j.buildenv.2020.106899
    OpenUrl
  43. ↵
    1. Tominaga Y,
    2. Mochida A,
    3. Yoshie R,
    4. Kataoka H,
    5. Nozu T,
    6. Yoshikawa M,
    7. Shirasawa T.
    2008. AIJ guidelines for practical applications of CFD to pedestrian wind environment around buildings. Journal of Wind Engineering and Industrial Aerodynamics. 96(10-11):1749–1761. https://doi.org/10.1016/j.jweia.2008.02.058
    OpenUrlCrossRef
  44. ↵
    1. Toparlar Y,
    2. Blocken B,
    3. Maiheu B,
    4. van Heijst GJF.
    2018. The effect of an urban park on the microclimate in its vicinity: A case study for Antwerp, Belgium. International Journal of Climatology. 38(S1):e303–e322. https://doi.org/10.1002/joc.5371
    OpenUrl
  45. ↵
    1. Wan Z.
    2015. MODIS land surface temperature and emissivity. Version MOD11. Santa Barbara (CA, USA): Institutue for Computational Earth System Science. https://modis.gsfc.nasa.gov/data/dataprod/mod11.php
  46. ↵
    1. Yan C,
    2. Guo Q,
    3. Li H,
    4. Li L,
    5. Qiu GY.
    2020. Quantifying the cooling effect of urban vegetation by mobile traverse method: A local-scale urban heat island study in a subtropical megacity. Building and Environment. 169:106541. https://doi.org/10.1016/j.buildenv.2019.106541
    OpenUrl
  47. ↵
    1. Yang L,
    2. Liu X,
    3. Qian F.
    2020. Research on water thermal effect on surrounding environment in summer. Energy and Buildings. 207:109613. https://doi.org/10.1016/j.enbuild.2019.109613
    OpenUrl
  48. ↵
    1. Zeeshan M,
    2. Ali Z.
    2022a. Heat stress mitigation in urban streets having hot humid climatic conditions: Strategies and performance results from a real scale retrofitting project. Science and Technology for the Built Environment. 28(4):513–526. https://doi.org/10.1080/23744731.2022.2040322
    OpenUrl
  49. ↵
    1. Zeeshan M,
    2. Ali Z.
    2022b. The potential of cool materials towards improving thermal comfort conditions inside real-urban hot-humid microclimate. 13(1):1–17. https://doi.org/10.1177/09754253221083206
  50. ↵
    1. Zeeshan M,
    2. Ali Z,
    3. Sajid M,
    4. Ali M,
    5. Usman M.
    2022a. Modelling the cooling effectiveness of street trees with actual canopy drag and real transpiration rate under representative climatic conditions. Journal of Building Performance Simulation. https://doi.org/10.1080/19401493.2022.2080865
  51. ↵
    1. Zeeshan M,
    2. Ali Z,
    3. Ud Din E.
    2022b. Thermal performance prediction of street trees inside isolated open spaces—Evaluations from real scale retrofitting project. Journal of Building Performance Simulation. 16(4):381–397. https://doi.org/10.1080/19401493.2022.2038270
  52. ↵
    1. Zeng F,
    2. Lei C,
    3. Liu J,
    4. Niu J,
    5. Gao N.
    2020. CFD simulation of the drag effect of urban trees : Source term modification method revisited at the tree scale. Sustainable Cities and Society. 56:102079. https://doi.org/10.1016/j.scs.2020.102079
  53. ↵
    1. Zhang M,
    2. Bae W,
    3. Kim J.
    2019. The effects of the layouts of vegetation and wind flow in an apartment housing complex to mitigate outdoor microclimate air temperature. Sustainability. 11(11):3081. https://doi.org/10.3390/su11113081
Back to top

In this issue

Print
Download PDF
Email Article
Citation Tools
Green Infrastructure with Actual Canopy Parameterization: A Simulation Study for Heat Stress Mitigation in a Hot-Humid Urban Environment
Muhammad Zeeshan, Zaib Ali, Qasim A. Ranjha
Arboriculture & Urban Forestry (AUF) Sep 2023, 49 (5) 247-267; DOI: 10.48044/jauf.2023.016
del.icio.us logo Twitter logo Facebook logo Mendeley logo
Bookmark this article

Jump to section

Keywords