2025 Volume 13 Issue 2 Pages 219-234
Rapid urbanisation has significantly changed the landscape of urban environments. More specifically, land use in cities has increasingly shifted towards built-up areas for urban infrastructure, housing, transportation, and industry. This has substantially altered the microclimate of cities, with one of the most noticeable impacts being the rise in urban temperatures, which negatively affects the quality of life and health of urban dwellers. As such, there is an urgent need to address these microclimate issues. Planting vegetation is one potential solution. Although urban green open spaces (GOS) are effective at mitigating microclimate impacts, they are often favoured for their aesthetic value rather than their ecological benefits. As such, the present study conducted computational fluid dynamics (CFD) simulations to monitor the urban microclimate in tree-filled and treeless areas, specifically, the air temperature distribution, to evaluate the efficiency of trees in decreasing temperatures. The findings indicate that tree-filled are cooler than treeless areas. More specifically, approximately 24.1% of the tree-filled area recorded temperatures between 28.1-29°C, while 36.5% of the treeless area recorded temperatures between 29.1-30°C. Therefore, trees effectively decreased the temperature by 1.6°C on average. These findings provide valuable insights that stakeholders can use to mitigate the urban heat island (UHI) effect.
Over the past 25 years, the global population has increased by 2.1 billion, with the majority residing in urban areas. There has also been a significant increase in the rate of urbanisation in Asia, especially in Indonesia. In 2011, the country’s urbanisation rate increased by 43.3% and reached 50% in 2021 (Unctad, 2022), which has led to major changes in the landscape of urban environments. The land use in these areas have also changed, with an increase in built-up areas for infrastructure, settlements, transportation, and industry (Imran, Hossain, et al., 2021; Tutuko, Bonifacius, et al., 2022; Tutuko, Subagijo, et al., 2018)
The rate of urbanisation and rapid population growth have increased the urgency of addressing changes in the microclimate (Toparlar, Blocken, et al., 2017; Yang, Wang, et al., 2023). Of the many changes in a microclimate, a change in air temperature is the main change that urban dwellers can feel (Imran, Hossain, et al., 2021; Zeren Cetin, Varol, et al., 2023).
Changes in air temperature directly affect the quality of life of urban dwellers. These higher temperatures not only cause physical discomfort, but impact the well-being of urban dwellers and increase the likelihood of health risks, such as heat stress and respiratory disorders (Marinić, 2023; Wullenkord and Ojala, 2023; Zeren Cetin, Varol, et al., 2023). Furthermore, the resulting decline in air quality due to vehicular and industrial emissions further increases the air temperature, increasing the need for a more strategic approach with which to address microclimate changes in urban environments.
Establishing green open spaces (GOS) in urban environments is one such solution. A direct correlation has been established between higher vegetation and lower concentrations of air pollutants (Aini and Shen, 2019; Gromke and Blocken, 2015; Janhäll, 2015; Li, Lu, et al., 2016; Morakinyo and Lam, 2016). Vegetation has also been found to affect the distribution of air temperature. Urban vegetation functions as a natural cooler that absorbs heat via evapotranspiration and provides shaded areas that lower the temperature in its surrounding areas (Imran, Hossain, et al., 2021; Zeren Cetin, Varol, et al., 2023).
However, the ecological function of GOS in urban areas are in decline (Esbah, Cook, et al., 2009). When the existing GOS were established, they did not prioritise the existence, characteristics, and planting patterns of vegetation, especially trees. However, the planting patterns and characteristics of these trees impact microclimate changes (Aini and Shen, 2020b, 2020a; Aini, Shen, et al., 2023; Gromke and Blocken, 2015; Morakinyo and Lam, 2016). Therefore, the present study examined the efficacy of GOS in controlling the microclimate, specifically, the air temperature, in urban environments. Extant studies on this subject generally conducted direct measurements in the field. However, this method fails to fully visualise the overall air temperature distribution in an urban environment.
As such, the present study used computational fluid dynamics (CFD) simulations to better visualise the overall air temperature distribution in an urban environment. Computational fluid dynamics (CFD) simulations have emerged as an important tool in the understanding of the dynamics of urban microclimates. It is a field of engineering that utilises numerical and computational methods to model and analyse the behaviour of fluid flows, be it gasses or liquids, and heat transfer in various conditions (Chung, 2002; Sayma, 2009). Mathematically modelling fluid flows and heat transfer in an urban environment yields a detailed visualisation of the temperature distributions in such areas. Computational fluid dynamics (CFD) simulations have been used to accurately predict urban microclimates, identify microclimate problem areas, and evaluate the impact of climate adaptation measures, such as urban greening (Toparlar, Blocken, et al., 2015, 2017).
Therefore, the present study used CFD to simulate the distribution of air temperature in an urban environment to obtain a more complete picture of how GOS decreased the air temperature in the examined area. The findings yielded insights that could facilitate better decision-making where GOS planning is concerned. It also offers policymakers valuable information that could be used to design better strategies with which to mitigate the impact of urban heat islands (UHI), especially in the face of increasing urbanisation.
A quantitative approach, namely, CFD simulations, was used to analyse the impact of trees on the air temperature distribution in two cases: Case 1 – a tree-filled area, and Case 2 – a treeless area. This facilitated a detailed assessment of the cooling effect that trees had by visualising the spatial distribution of the air temperature under different environmental conditions. Figure 1 presents a clear overview of the workflow of the present study.
The research was conducted in Malang City, Indonesia – a central business district (CBD) with a sizable GOS that is frequented by many urban dwellers. The vegetation here comprises three strata: (1) stratum E, which comprises ground-covering vegetation; (2) stratum D, which comprises trees measuring 1-4 metres tall; and (3) stratum C, which comprises trees measuring 4-20 metres tall. This vegetation grows within the GOS and along the sides of several roads.
The buildings surrounding the GOS are commercial and service buildings, offices, and places of worship of varying heights and sizes, with the tallest standing 15 metres tall. Figure 2 depicts the research location on a map of Malang City.
Three-dimensional (3D) Modelling of the Research LocationThe three-dimensional (3D) modelling phase is a crucial step in any simulation. The research location and the GOS in its centre had to be accurately depicted in 3D. The 3D objects constructed include the buildings, roads, open spaces, and vegetation. Two 3D case studies were built: (1) one with trees in the GOS (Case 1) and (2) the other without trees in the GOS (Case 2) (Figure 3). Both these cases had to be accurately modelled to compare the microclimate of both conditions and determine the temperature reduction efficiency of trees in open spaces.
Autodesk® CFD was used to conduct the simulations. The CFD simulation was divided into several key stages. The first stage involved identifying and assigning the properties of the objects’ materials in the 3D model, which were the buildings, roads, open spaces, and vegetation. It is critical to accurately select the objects’ materials as it impacts how each object will react to solar heating. The second stage involved defining the domain size, while the third stage involved setting the boundary conditions, which ensured that the simulated environment reflected the actual physical boundaries of the research location.
The mesh sizing was determined in the fourth stage. The research location was segregated into a fine computational grid or mesh, where the calculations were performed. The final stage involved solving the CFD model., where the software iteratively solved heat transfer equations and determined the temperature distribution at the research location. The results of these simulations were then analysed to compare the temperature distributions in Cases 1 and 2, which quantified the role that GOS play in improving the urban microclimate and mitigating the UHI effect.
Assigning Objects’ Materials in the Three-dimensional (3D) ModelThe materials of the objects in the 3D model, such as buildings, roads, open spaces, and trees, were assigned. This step was crucial as the material chosen would influence how the object interacts with environmental factors, such as heat and airflow. The material selected for the buildings was brick, asphalt for the road, grass for the open space, and the trees were hardwood. These materials ensured that the simulation produced results that closely resembled the real-world condition of the research location. Figure 4 depicts the configuration of the materials used in the 3D model of the research location.
The domain size was defined to ensure that the simulated environment reflected the actual physical boundaries of the research location. In CFD simulations, the domain size must be based on the spatial complexity of the urban layout, especially in areas with varying building heights or vegetation, to accurately capture the thermal effects of vegetation and buildings. Various domain sizes have been used to simulate urban environments in the past. However, it is recommended that the domain maintains a minimum distance of 5-15 times the height of the main object from the object itself to the inlet, outlet, and lateral boundaries, to ensure that the airflow can develop freely and without interference from the boundaries of the domain (Franke, Hellsten, et al., 2010; Tominaga, Mochida, et al., 2008; Toparlar, Blocken, et al., 2015, 2017). The research location was simulated with a vertical and horizontal distance of 5H (Figure 5). This facilitated capturing the temperature distribution and thermal effects more comprehensively as well as represented the real-world conditions more accurately.
All the boundary conditions were defined as air temperature (Figure 6), which was based on data obtained from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) for 1:00 AM in Malang City, where the temperature reached 32°C. This temperature was used to define the boundary conditions in the CFD simulation.
Mesh SizingThe research location was divided into fine computational grids or meshes, where the calculations were performed on the 3D model. Mesh sizing is a critical aspect of a CFD simulation as it directly affects its accuracy and computational efficiency. It is important to systematically develop the mesh in CFD-based 3D simulations, particularly in domains that are complex (Dixon, Ertan Taskin, et al., 2011). Although finer mesh provides more accurate results, especially in areas with high gradients, they increase the amount of computational power required. Therefore, it is crucial to finetune the mesh refinement according to the computational resources available. Figure 7 shows the mesh sizing process for both cases.
The heat transfer equations and temperature distribution in the research location were calculated, which was crucial for understanding the temperature distribution patterns in urban environments. The Reynolds-averaged Navier-Stokes (RANS) method is often used in CFD-based urban temperature simulations as it can efficiently handle turbulent flows by averaging the effects of turbulence over time, which is relevant in complex urban environments.
Despite some limitations in handling highly complex geometries, the RANS is excellent for simulating turbulent flow in built environments as it requires less computational power than other methods, such as large eddy simulation (Tominaga, 2024). It also effectively simulates heat transfer in complex variations of walls, especially in cases where the temperature fluctuations can be simplified into a steady-state (Marineau, Schetz, et al., 2007).
Therefore, the RANS was used under steady-state conditions in Autodesk® CFD, which yielded a straightforward yet accurate simulation that was suitable for analysing the temperature distribution and the UHI phenomenon in urban environments.
In Case 2 (Figure 8), the highest air temperature was recorded on the roads (33-34°C), while the lowest was recorded in the open space in the centre (30-31.5°C). Therefore, the treeless areas experienced higher temperatures as they were directly exposed to solar heating.
In Case 1 (Figure 9), however, the air temperature recorded on the roads was lower, as indicated by the shift in colour from orange to yellow, indicating that the temperature was 31-32°C. Furthermore, the open space in the centre, which was green in Case 2, was now blue, indicating that the temperature was 29-30°C.
Effect of Vegetation on Air TemperatureAs seen in Figure 8 and Figure 9, the air temperature distribution patterns of both cases differed significantly, reflecting the influence of vegetation, particularly trees, on environmental temperatures and in enhancing thermal comfort.
The air temperature distribution was higher in Case 2. Figure 10 depicts the percentage of areas in Case 2 and which air temperature range they fell into. As seen, during the day, the average air temperature was 33°C and none (0%) of the areas experienced air temperatures <27-28°C. The air temperature in most areas was between 29.1-33.1°C. More specifically, the air temperature in 37.5% of the area was between 29.1-30°C, while 21.9% of the area experienced air temperatures between 32.1-33°C. Therefore, the air temperature in nearly 60% of the area was >30°C, with 3.1% of the area experiencing extreme air temperatures (>36°C) reaching. Therefore, the layout of Case 2 yielded minimal natural cooling, leading to significant heat accumulation.
The air temperature distribution was cooler in Case 1. Figure 11 depicts the percentage of areas in Case 1 and which air temperature range they fell into. As seen, the air temperature in 24.1% of the area was between 28.1-29°C, while 21% of the area experienced air temperatures between 29.1-30°C.
Although the air temperature in some areas were >30°C, more specifically 32.1-33°C (19.2%) and 33.1-34°C (12.1%), overall, the percentage of areas experiencing extreme air temperatures (>36°C) was smaller (2.2%). Therefore, vegetation significantly cools urban environments.
Figure 12 provides a detailed comparison of the air temperature distribution patterns of both cases, which differed significantly. The air temperature distribution patterns of Case 2 mainly fell within 29.1-33.1°C, while that of Case 1 was lower, with a larger proportion falling within 28.1-29°C and 29.1-30°C. Case 1, which featured tree-filled areas, experienced better cooling, as evidenced by the substantial amount of area (45%) experiencing air temperatures <30°C. Conversely, 80% of the areas in Case 2, which featured treeless areas, experienced air temperatures >30°C. Therefore, trees significantly lower the air temperature in urban environments and create more comfortable conditions.
As such, trees significantly reduce the air temperature of their surroundings by increasing the size of areas with cooler temperatures. Treeless areas experience more intense heat due to the absence of vegetation that can absorb heat and induce cooling via evapotranspiration. Therefore, vegetation, especially trees, is essential to create more comfortable microclimates, reduce extreme air temperatures, and enhance the quality of the environment.
Effectiveness of Air Temperature Reduction by TreesA separate analysis was conducted to examine the ability of trees to reduce the air temperature. The air temperature post-tree insertion (Case 1) and pre-tree insertion (Case 2) and were recorded at 20 points each (Figure 13).
As seen in Figure 14, trees significantly decreased the air temperature of their surroundings. The average air temperature in Case 2 was 29.0-33.9°C, while that of Case 1 was lower, at 28.3-31.9°C.
The air temperature significantly decreased at many of the sampled points. For instance, at Point 3, the addition of trees decreased the air temperature by 3.4°C, from 33.1 to 29.7°C. Meanwhile, at Point 6, the air temperature decreased by 2.9°C, from 32.6 to 29.7°C. At Point 16, the air temperature decreased by 2.4°C, from 30.8 to 28.4°C. However, the decrease in air temperature at some of the sampled points were minimal. For instance, at Point 19, the addition of trees decreased the air temperature by only 0.5°C, from 29.3 to 28.8°C. Meanwhile, at Point 10, the air temperature decreased by 0.6°C, from 29.9 to 29.3°C.
Therefore, trees may lower the air temperature by 0.4-3.4°C. However, this depends on several factors, such as the species of tree, the density of the vegetation, the size of tree-filled areas, and the condition of the surrounding environment. In the present study, the air temperature in tree-filled areas was 1.6°C lower on average. Therefore, vegetation, particularly trees, significantly decrease environmental temperatures, especially in urban environments that frequently experience temperature increases due to the UHI effect.
To validate the results of the present study, the Universal Thermal Climate Index (UTCI) was used to evaluate the recorded reductions in air temperature. The UTCI is a widely recognised measure that is used to assess the thermal comfort of humans in outdoor environments (Błazejczyk, Jendritzky, et al., 2013; Krüger and Di Napoli, 2022; Mahdavinejad, Shaeri, et al., 2024). It categorises thermal comfort into several levels (Figure 15). The temperatures recorded at the 20 points in Cases 1 and 2 were analysed within these comfort levels, which objectively demonstrated the ability of trees to effectively decrease temperatures.
Based on the UTCI (Figure 15), Case 2 experienced higher air temperatures, leading to varying levels of heat stress. The temperatures recorded in some areas fell into the Strong Heat Stress category, while that of the other areas were in the Moderate Heat Stress category. Conversely, temperatures recorded in Case 1 consistently fell within the Moderate Heat Stress category, with none falling in the Strong Heat Stress category.
These findings emphasise the crucial role that trees play in lowering the air temperature and reducing heat stress levels, which enhances the thermal comfort of humans. Furthermore, vegetation regulates temperature fluctuations by preventing the extreme heat conditions that often occur in urban environments that lack greenery. This is because, without trees, the surfaces and the surrounding air absorb more heat, which increases thermal discomfort. The temperature reductions observed at the 20 sampled points provide a strong foundation for microclimate management and should encourage governments and stakeholders to strategically plant trees in heat-prone areas to effectively mitigate temperature increases.
Furthermore, the results of the present study underscore the importance of integrating trees into urban environments to mitigate the UHI effect and improve thermal comfort. It is evident that incorporating vegetation into urban planning creates environments that are not only more liveable, but sustainable and thermally-comfortable, which would benefit the overall well-being of the community (Akbari, Pomerantz, et al., 2001).
Validation of Research ResultsThe findings of the present study empirically prove that trees significantly decrease the air temperature and reinforce the important role of urban vegetation in improving microclimates. The CFD simulations demonstrated that trees decrease the air temperature by 0.4-3.4°C and had an average cooling effect of 1.6°C across all 20 points. The air temperatures recorded in Case 1 were cooler and more stable, with air temperatures in 5% of the area <30°C. Conversely, the air temperatures recorded in Case 2 were higher during the day, with the air temperature in approximately 60% of the area between 29.1-33.1°C.
Much like the present study, extant studies have, similarly, reported that vegetation affects the air temperature. For instance, the RayMan Pro® simulations that Gherraz, Guechi, et al. (2018) conducted revealed that vegetation decreased the radiation temperature in open spaces from 52.9 to 42.9°C. The direct measurements and questionnaires that Elnabawi and Hamza (2020) used, on the other hand, concluded that shading devices, specifically trees, were best at providing thermal comfort at an urban scale as trees reduced the air temperature and physiological equivalent temperature by 2.3–2.4°C.
Meanwhile, Bowler, Buyung-Ali, et al. (2010) systematic literature review of urban greening interventions concluded that, on average, parks with trees are 0.94°C cooler during the day. Schwaab, Meier, et al. (2021), on the other hand, analysed land cover data as well as land surface temperatures (LST) from high-resolution satellites and concluded that the temperature in tree-filled urban areas were 2-4 times lower than treeless areas.
The CFD simulations of the present study, similarly, revealed that the air temperature in tree-filled areas (Case 1) were lower than than that of treeless areas (Case 2). The air temperature in 24.1% of the areas in Case 1 was lower (28.1-29°C), while the air temperature in this range was only 1.6% in Case 2. Moreover, the CFD simulations showed a consistent pattern, where Case 2 had hotter air temperature distributions, with more than 40% of the areas experiencing temperatures >32°C. The air temperature distribution in Case 1, on the other hand, tended to be in the lower range.
The results of the present study, which were obtained via CFD simulations, indicate that vegetation, particularly trees, plays a crucial role in lowering air temperatures and enhancing thermal comfort in urban environments. More specifically, as seen in Case 2, the air temperature was higher during the day, especially on the roads and in open spaces that lacked vegetation. Furthermore, approximately 60% of the total area in Case 2 experienced temperatures between 29.1-33.1°C. The temperature distribution in Case 1, on the other hand, was significantly cooler, with around 45% of the total area experiencing temperatures <30°C, indicating that trees effectively reduced heat accumulation. Apart from that, the ability of trees to decrease the air temperature was further analysed by recording the air temperature at 20 points each in Cases 1 and 2, which revealed an average air temperature reduction of 1.6°C. At some of the points, the difference in temperature was as much as 3.4°C, while the smallest reduction was 0.5°C. This could be attributed to the density of the trees and other factors in the surrounding environment. Nevertheless, the average temperature reduction of 1.6°C provides strong evidence that trees are an effective strategy for microclimate mitigation.
The findings of the present study provide a scientific basis for urban climate adaptation strategies. It highlights that trees are a natural cooling mechanism that can be leveraged in urban planning, environmental management, and policy-making to mitigate the UHI effect and enhance outdoor thermal comfort. It also provides valuable insights that governments, urban planners, landscape architects, and environmental policymakers can use to develop more climate-responsive urban designs, optimise green infrastructure strategies, and create healthier and more sustainable urban environments. Furthermore, urban environments can effectively manage microclimates, decrease their reliance on artificial cooling systems, and improve the overall quality of their environment by incorporating vegetation into their urban spaces.
However, it is noteworthy that the present study only considered one type of vegetation—hardwood trees—and did not examine other types of vegetation, such as shrubs, grass, or trees of varying canopy structures. As such, future studies may explore the impact of various types of vegetation and vegetation planting patterns on air temperature distribution as it would yield more comprehensive data with which to develop better urban greening and heat mitigation strategies for outdoor spaces.
Conceptualization, RW; methodology, RW, PT, and NA; software, SDC NA; investigation, AS and NA; resources, PT and AGS; data curation, RW, NA, and CDW; writing—original draft preparation, RW and NA; writing—review and editing, SDC and NA; supervision, NA. All authors have read and agreed to the published version of the manuscript.
The authors declare no conflicts of interest in the publication of this paper.
This study was funded by the Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia, whose support we gratefully acknowledge. We also extend our appreciation to the members of the Laboratory of Urban and Human Settlements, Department of Architecture, Faculty of Engineering, University of Merdeka Malang, as well as to all individuals who contributed to the Housing and Human Settlement project, particularly by providing supporting data.
This research was funded by the Ministry of Education and Culture-Research and Technology, Republic of Indonesia(109/E5/PG.02.00.PL/2024).
