Influence of the thermal environment of urban sidewalks under green shading from a human scale

ABSTRACT

Urban footpaths are the main outdoor activity spaces for pedestrians. It is traditionally believed that the air temperature of a pedestrian walkway under tree shade is lower. However, from the perspective of vertical air temperature profiles, there is little difference in air temperatures between a shaded pedestrian walkway and one under unshaded sunlight. This paper adopts a human-centred approach to investigate the three-dimensional spatial distribution of air temperatures in a pedestrian walkway in Taipei City under tree shade. It analyzes the air temperature reduction caused by attenuation of ground reflected long-wave radiation due to shade, and the effect of shade on environmental ventilation. It was found that air temperatures varied along pedestrian walkway height and direction. The measured data shows that the main difference in air temperature between shaded areas and areas exposed to sunlight is in the 0–0.5 m height range, and there is little difference in air temperature between the 0.5 m-2 m height range.Higher shades led to poorer ventilation and were more likely to cause local overheating. This suggests the need to assess urban pedestrian thermal comfort from a three-dimensional perspective, considering not just planar shades but also vertical shading and ventilation.

KEYWORDS:

• Urban street canyons
• pedestrian walkways
• thermal images
• spatial dynamic probes
• Kriging

1. Introduction

The rapid expansion of cities has led to a decrease in available land, with most of the usable land being allocated for construction purposes, resulting in a yearly reduction of green spaces. The comfortable spaces needed by humans are being compressed, and the urban heat island effect is becoming more severe (Halder et al., 2022; Mahdavi Estalkhsari et al., 2022). Addressing this issue, relevant studies have utilised remote sensing data, microclimatic measurements, and questionnaire surveys across various urban and climatic contexts to thoroughly investigate the seasonal, diurnal, and spatial variations in heat stress and thermal comfort (Karimi & Mohammad, 2022; Karimi et al., 2022, 2023, 2023).In recent years, urban design trends have not only emphasised the creation of more humanised and high-quality urban spaces but have also increasingly focused on starting from the perceptions and experiences of residents, adopting a bottom-up approach to explore and achieve these goals. This trend is reflected in various aspects, including acoustic comfort, sociocultural integration, development of ecological parks, and the design of urban and rural environments with emotional care (Arrar et al., 2022; Bliankinshyein et al., 2021; Hao & Kozlov, 2022; Sheikh & Mitchell, 2018; Tai, 2020). Through these studies, we see the necessity of considering residents’ direct experiences and perceptions as crucial references in urban space design. Moreover, it emphasises the alterations in the physical environmental spaces within urban contexts, with associated studies increasingly concentrating on small-scale spatial physical elements that bear a closer relation to the everyday experiences of urban dwellers, such as the green view index, form, and permeability rates of streets. The study by Jingxuan et al. (2021) explores urban spatial networks from a broad perspective, assessing the application of spatial network survey techniques across various scales, with a particular emphasis on the importance of human-scale research. This research reveals the potential for understanding and improving residents’ quality of life, highlighting the necessity of considering individual daily experiences and behaviours in urban planning and design. By delving into the human scale, Jingxuan et al. (2021) offer new insights for fostering more humanised, inclusive, and sustainable urban environments. Building on this foundation, the study by Ye et al. (2021) further demonstrates the application of the human scale in researching the impact of urban form on street temperatures. Through the lens of the human scale, Ye et al. (2021) investigate the positive correlations between urban morphology, such as the ratio of green landscape to street, facade proportions, and shading effects, elucidating how these micro-design elements work together to optimise urban thermal comfort conditions, thereby directly affecting residents’ comfort and health. Similarly, the research by Karimi et al. (2020) emphasises the importance of considering personal perception from a human-scale and micro-perspective in assessing thermal comfort in urban parks. By analysing the impact of environmental parameters on thermal perception, Karimi et al. (2020) reveal the joint effect of specific micro-design elements, such as low-albedo pavements and trees with wide crowns and tall trunks, further affirming the significance of considering the human scale in urban design. Zhu et al. (2024) explore the impact of small-scale physical elements, specifically street trees, on urban experiences by quantifying their contribution to the Green View Index (GVI). Their study demonstrates how street trees enhance visual and thermal comfort in urban spaces, emphasising the importance of green infrastructure in urban planning for improved aesthetic and ecological outcomes. This research highlights the pivotal role of green view ratios and street morphologies in enriching urban living environments. Thus, elucidating the impact patterns of arborisation on the thermal ambience of urban pedestrian thoroughfares through a human-scale perspective is more consonant with the urgency of contemporary research. Deciphering these impact patterns can illuminate strategies to optimise pedestrian path thermal environments and provide robust theoretical bases to inform future pedestrian path planning and design.

The pedestrian walkway is the primary outdoor space for pedestrians. Recent studies have progressively unveiled the significance of pedestrian walkways in urban planning, particularly in terms of their impact on the safety, health, and social interaction of urban residents. Initially, pedestrian walkways provide city dwellers with a safe walking environment, distancing them from the threats posed by motor vehicles. The research by Lui et al. (2022) delved into pedestrian movements around public buildings’ walkways, examining how pedestrians interact with surrounding facilities. This study underscored the role of pedestrian walkways in enhancing the interactivity and safety of urban spaces. Furthermore, the design and layout of pedestrian walkways directly influence the health and well-being of urban residents. Abdul Rahim et al. (2023) assessed the walkability of pedestrian walkways in the Royal Town of Klang, offering recommendations for improving pedestrian environments, pathway networks, pedestrian infrastructure, and maintenance. This highlights the importance of pedestrian walkways in promoting healthy walking habits. Additionally, pedestrian walkways facilitate social interaction and economic development within cities. Sanchez et al. (2022) described the design and construction process of the Mary Elmes Bridge along the river in Cork City Centre, illustrating how pedestrian walkways and bridges serve as vital links connecting different urban areas, thus fostering social interaction and economic activities. Lastly, given the challenges of global warming and urbanisation, pedestrian walkways play an increasingly crucial role in enhancing urban thermal comfort and addressing extreme climate conditions. Todd (2023) introduced the Cool Walkshed Index (CWI) as an assessment tool for pedestrian thermal comfort in tropical cities, emphasising the importance of considering pedestrian walkways in tropical city planning. Following the discussion on the importance of pedestrian walkways to the urban experience, it is noteworthy that planting street trees is considered one of the most crucial measures for improving the urban thermal environment (Rahman et al., 2020). Studies have demonstrated that on streets with varying levels of shade, those with more abundant shade can mitigate heat stress and enhance human thermal comfort (Ren et al., 2022). However, the effect of shade on temperature is likely to be more modest in the vertical direction compared to horizontally. Where temperatures are high, perimeter trees can shelter pedestrians from extreme heat (Lanza & Durand, 2021). In practical applications, the strategy of augmenting vegetation to mitigate urban temperatures falls short of addressing the complexity of urban thermal dynamics. A comprehensive evaluation of the greening effects on thermal comfort within urban street canyons necessitates an integrated consideration of critical external variables, including the Sky View Factor (SVF), surface albedo, the selection of tree species, and wind velocity. These factors synergistically contribute to the regulation of the microclimate and the enhancement of thermal conditions in urban environments (Mohammad et al., 2021; Narimani et al., 2022).Therefore, accurately quantifying and analysing the vertical impacts of street trees on ground and air temperatures is imperative for the judicious allocation of shade resources and mitigating the urban thermal environment.

The microclimate of pedestrian walkways in urban street valleys is influenced by the interaction of vertically orientated thermal environments. In order to understand the effect of vegetation shading, the surface temperature of the valley environment needs to be obtained. Studies that have hitherto been conducted have primarily relied on simulations (Elbardisy et al., 2022), which should be supplemented with real measurements. The Internet of Things (IoT) is used in this study to gather a lot of data on movement and location, dynamic and static, time and space, which is then integrated into a three-dimensional model of shade variation in an urban heat island. A ‘Dynamic probe’¹ device is used to collect data on temperature, humidity, pressure, air quality, and wind speed at each altitude. The surface temperatures of the ground and objects are measured using temperature measuring equipment with infrared cameras for the thermal environment of the pedestrian pathway in the urban valley.

This study finally wants to discuss the following issues through actual measurement and combining Kriging in GIS:

1. To analyse the influence of tree shade on the thermal environment of the vertical space of urban pedestrian walkways through actual detection.

2. To analyse the change pattern of thermal radiation reflection attenuation on pedestrian walkways at the human scale.

3. To analyse the impact of tree planting shading rate on the thermal environment in the surrounding space.

In the thermal environment measurement of the pedestrian path in the urban valley, a multi-lens FLIR ONE PRO infrared camera (real scene lens + thermal imaging lens) and an infrared thermometer are used to detect the surface temperature of the ground surface and objects. Based on actual measurements and overall data acquisition, the study analyzes the three-dimensional changes of plant shade and sun exposure, the resulting ground surface temperature reflection attenuation, and the slight cold air source generated below the shade that produces thermal environment changes.

1.1. Scale of urban thermal environment research

In subtropical cities, the urban thermal effect is mainly affected by human activities and the temperature in urban areas is usually higher than that in adjacent rural areas. This phenomenon was proposed by Howard (Howard, 1818) in the 19th century. So far, research on the urban thermal environment has been carried out on four scales: urban scale, block scale, street scale, and human scale. The multi-scale approach is crucial for comprehensively capturing the complexity of Urban Heat Island (UHI) effects and their impact on residents’ quality of life. For instance, Chen et al. (2020) analysed the regional urbanisation process, average surface temperature change characteristics, and the relationship between urban expansion, urban size, and UHI in the Beijing-Tianjin-Hebei urban agglomeration. Their findings revealed that an increase in urban size is likely to face a greater UHI effect (Chen et al., 2020). Additionally, Vogel and Afshari (2020) conducted mesoscale simulations of Berlin’s urban microclimate using the Weather Research and Forecasting (WRF) numerical weather prediction platform. By comparing different physical schemes and methods for estimating UHI intensity, they further emphasised the importance of interventions at the block and street scales for improving thermal comfort (Vogel & Afshari, 2020). Studies on outdoor thermal comfort in the historic squares of Seville and Madrid, Spain, provided optimal recommendations for visiting these heritage sites. By considering the impacts of urban planning and design at different scales, mitigation measures can be more effectively implemented to enhance the thermal comfort of residents and tourists, and reduce the negative impacts of UHI on individuals and the urban environment (Karimi & Mohammad, 2022). These studies demonstrate the significance of a multi-scale research approach for understanding and mitigating urban thermal environment issues, promoting urban sustainability. By considering the impacts of urban planning and design at different scales, mitigation measures can be more effectively implemented, improving residents’ thermal comfort and reducing the negative impacts of UHI on the urban environment. All relevant literature will be presented in Table 1.

Table 1. Collation of relevant literature.

Author	Location	Methods	Key Findings
Zhang et al. (2022)	Urban greenways in Singapore	Computer vision analysis	Different recreation activities clustered at different times related to environment
Fan et al. (2021)	Residential areas in Beijing	Field measurements	Vegetation and water bodies had cooling/humidifying effects; landscape metrics correlated with microclimate
Zhu et al. (2022)	Streets in Hong Kong	Field investigation	H/W ratio key factor affecting thermal environment
Xiong et al. (2022)	Streets in Chongqing	Field measurements	High spatiotemporal variability of street thermal environments
Yang et al. (2020)	Blocks in Nanjing	Simulation analysis	Complex effects of urban form on thermal environment
Triyuly et al. (2021)	Wetland area in Palembang	Field measurements	Surface changes affect urban thermal environment
Ma et al. (2021)	Commercial street in Southern China	Numerical simulation	Micro-scale thermal environment assessment
Fan et al. (2021)	Residential areas in Beijing	Field measurements	Vegetation and water bodies had cooling/humidifying effects; landscape metrics correlated with microclimate
Gao et al. (2019)	Urban areas in Xi’an	Satellite data inversion	Enhanced urban heat island during 2000–2016
Zoran et al. (2019)	Bucharest	Satellite data	Urban growth affected land surface radiative properties
Zhang et al. (2019)	Urban parks in Gold Coast	Field measurements	Internal and external factors affected diurnal temperature
Li et al. (2019)	Urban green spaces in Wuhan	Statistical analysis	Green space design and planning for thermal comfort
Deng et al. (2022)	Urban areas in Wuhan	Field measurements	Cooling effect of river and railway station at night. Higher temperature in man-made vs. natural areas.
Kim and Brown (2023)	Urban street canyons	Field measurements, modelling	Factors like SVF, FAR, BCR impact street canyon thermal environment. Developed MHI model to estimate micro-scale urban thermal environment.
Li et al. (2021)	Urban areas in Fuzhou	Street view data, satellite imagery	Differences in distribution of green visibility and vegetation coverage; Proposed strategies to improve street greenery quality
Zhang et al. (2020)	Urban areas in Taipei	Satellite data	There was an obvious reverse variation trend between the UHI and FVC
Chang, Wang, et al. (2020)	Urban street canyons	Field measurements	The lower visible green index is, the higher the temperature of infrared thermograph
Chen et al. (2020)	Beijing-Tianjin-Hebei Urban Agglomeration	Case Study	Urban scale and expansion correlate with increased UHI effect.
Vogel and Afshari (2020)	Berlin, Germany	Climate Modeling	Neighborhood and street-level interventions crucial for thermal comfort improvement.
Karimi and Mohammad (2022)	Sevilla and Madrid, Spain	Microclimatic measurement, questionnaire survey, Envi-met simulation	Recreation activities and tourist satisfaction vary with environmental thermal comfort and time.

The preponderance of metropolitan-scale inquiries have examined geographical expanses exceeding 10 km, harnessing field measurements or satellite remote sensing, with investigations principally concentrated on elucidating the variability patterns and underpinning mechanisms of the heat island phenomenon, etc. Urbanisation processes, land surface cover types, and urban morphologies have been identified as critical determinants influencing thermal ambiances. Specifically, Gao et al. (2019) selected Xi’an City as a case study, leveraging multi-temporal remote sensing imagery to derive surface temperature data and analyse the interrelations between urban expansion and heat island transformations. Zoran et al. (2019) chose the Bucharest metro region, applying satellite remote sensing to acquire surface parameters and probe the impacts of urbanisation on surface thermal environments. Triyuly et al. (2021) examined Palm Beach, utilising remote sensing to discern surface parameters and research the effects of urbanisation on surface thermal environments. Yang et al. (2020) proposed thermal environment optimisation strategies based on numerical simulations of Nanjing’s urban morphologies. Chang, Wang, et al. (2020) harnessed satellite remote sensing to analyse the relationships between heat island intensities and green cover rates in Taipei metro from 1990 to 2018, finding inverse correlations between green cover rates and urban heat island effects.

The block scale examines 1.0–10 km expanses of inner urban areas, harnessing GIS, numerical simulations, field measurements, and other investigative techniques to correlate the nature of urban soils and morphologies, urban textures, and thermal ambiences, subsequently proposing strategies to mitigate heat island effects based on these findings, such as optimising urban morphologies and rationally configuring greenspaces. Li et al. (2019) conducted an inquiry on urban core neighbourhoods, constructing a model to optimise the layout and configuration of greenspace. Zhang et al. (2019) analysed the thermal environments of 18 parks via field monitoring, finding that augmented tree cover can effectively reduce temperatures. Fan et al. (2021) probed the impacts of seven distinct surface types on air temperature and relative humidity through field observations, proposing countermeasures for greenspace planning and design. Zhu et al. (2022) analysed the effects of building aspect ratios and view factors in Hong Kong, eliciting greater influences on air temperatures from the former, while Deng et al. (2022) discerned significant moderating effects on surrounding thermal environments from the Yangtze River and railway station in Wuhan.

Street-scale thermal ambiances examine streetscapes spanning 0.01-1 km using numerical simulations and in-situ measurements. Initial inquiries investigated the impacts of street space morphologies on urban thermal environments by analysing the energy balances of street surfaces and air, developing numerical models of urban street energy balances (Nunez & Oke, 1977). Contemporary studies aim to probe microclimates and propose enhanced methods or implements to ameliorate thermal environments from measurement or simulation perspectives. Zhang et al. (2020) applied image recognition techniques to discern relationships between street greenery visibility and thermal environments, eliciting inverse correlations where greater visibility corresponded with lower temperatures. Ma et al. (2021) harnessed numerical simulations to assess pedestrian street thermal comfort, furnishing quantitative bases to optimise microclimates. Xiong et al. (2022) conducted on-site observations of major Chongqing streets, discerning substantial spatiotemporal heterogeneities in thermal environments at the street scale, with temperature variations approaching 9.4°C. Kim and Brown (2023) constructed streetscape environment models accounting for micro-island effects, enhancing simulation fidelity.

In recent years, the concept of urban form at the human scale has received widespread attention, which is defined as the urban form that people can see, touch, and feel intimately related to the human body. The human-centred scale pays more attention to the bottom-up feeling and experience relative to the space field in which people reside in.

The scholarly works of Jacobs (1961) and Lefebvre (1974) underscore the criticality of urban planning and design that prioritise human needs, behaviours, and well-being in enhancing urban form and microclimate. Jacobs, in her 1961 work, underscored the significance of green spaces and pedestrian-friendly environments in mitigating urban heat and improving air quality, advocating for mixed-use development, vibrant street life, and organic urban growth. Lefebvre, through his 1974 exposition on ‘The Right to the City’ and theories on the social production of space, introduced a more democratic and inclusive approach to urban planning. He emphasised the importance of public space design in catering to the diverse needs of communities, including the creation of more comfortable microclimates.With the development of information technology and urban development, human scale is more and more important and focuses on the direction of microclimate improvement in high-density cities, and it is found that most of the researches focus on this aspect of the percentage and area of vegetation in the degree of depression. M. Li et al. (2021) explored correlations between green visibility and coverage across Fuzhou neighbourhoods via a human-scale approach combining Tencent Street View and satellite imagery. Zhang et al. (2022) leveraged computer vision to examine impacts of physical environments on recreational activities along Singaporean urban greenways. Different activities were found to concentrate at distinct times corresponding to specific environmental characteristics, informing greenway design strategies and community participation in maintenance. Fan et al. (2021) used field measurements to analyse seven typical Beijing settlement surface types, finding vegetation and water bodies with high/medium depression effects significantly reduced temperatures and increased humidity. Pavements exhibited the highest temperatures and lowest humidity. Correlations between landscape patterns and microclimates varied seasonally, but patch percentage and area were key influences. This provided landscape design strategies to improve micro-scale thermal ambiances. Synthesizing studies reveals a predominant focus on green shade cover distributions and air temperature impacts in human-scale urban space microclimate research. Thus, this study will measure vertical pedestrian walkway spaces in Taipei using in-situ techniques to explore tree/plant shading effects on thermal radiation decay, emphasising the indispensability of arborisation shading rates in thermal environment regulation.

1.2. Formulation of the street valley sensible heat balance

Tee plantings offer particular potential for cooling urban microclimates and providing other ecosystem services, and they can be integrated into dense urban street networks. Richards & Edwards (2017) found that 13% of the diffuse and direct solar radiation is shaded every year through a study of the street thermal environment in Singapore, and more than 70% of the shading effect is caused by tree crowns. Hirabayashi et al. (2018) evaluated the ways in which the Tokyo Metropolitan Government is planning to plant canopies of existing trees to expand shade and reduce air temperature and solar radiation.

Under the urban street canyon scale, the height of buildings on both sides of the street is limited, the length of the street is assumed to be infinite, and the surrounding environment is similar. In this case, it can be ignored that the long and short-wave radiation is absorbed by the air, and advective heat transfer cannot be performed. Therefore, the air temperature inside the street is the result of the interaction of sensible heat transfer between the upper atmosphere, the ground and the wall.

The energy balance relation of the road surface in street canyon area can be represented as:

(1) $Q^{\ast }=Q_{\mathit{HG}}+Q_{\mathit{EG}}+\mathrm{\Delta }{Q}_{\mathit{SG}}$(1)

In the formula, Q* is the net radiation heat transfer of the ground, Q_HG is the sensible heat transfer between the ground and street air, Q_GH is the latent heat exchange heat between the ground and the street air, ${ }_{\mathit{QSG}}$ is the heat storage on the ground, and the meanings of other parameters are the same as above.

The street thermal environment system is composed of air, buildings and ground surface, and its energy balance relation is:

(2) $Q^{\ast }=Q_H+Q_E+\mathrm{\Delta }{Q}_S+\mathrm{\Delta }{Q}_A$(2)

where is the net radiation, Q_H is the sensible heat flux, Q_E is the latent heat flux, ${ }_{\mathit{QS}}$ is the net storage heat flux, and ${ }_{\mathit{QA}}$ is the net heat convection.

Sensible heat is the turbulent heat exchange between the atmosphere and the surface of an object or thermal system caused by temperature changes. The change of sensible heat release mainly depends on factors such as net radiation and the thermal condition of the underlying surface. There are two main factors affecting sensible heat release: one is the difference between the surface temperature and the air temperature; the other is the flow velocity of the air. The higher the wind speed, the stronger the surface air turbulence and the higher the sensible heat release. For the heat transfer process of sensible heat from the ground surface to the air, there are many sensible heat transfer equations. Related studies have also simplified many formulas empirically, as such, the formula for calculating thermal flux (Qin & Hiller, 2014) is:

(3) $\mathrm{H}=h_c\left(T_s-T_a\right)$(3)

where $\mathrm{H}$ is the sensible heat flux, $T_s$ is the ground surface temperature, $T_a$ is the air temperature, $h_c$ is the air turbulence coefficient, which can be obtained by the following empirical formula (Bentz, 2000):

(4) $h_c=\left\{\begin{array}{c}5.6\times 4.0\mathit{v}\mathit{v}<5\\ 7.2\times v^{0.78}\mathit{v}\ge 5\end{array}\right.$(4)

If the wind speed at the corresponding height cannot be obtained, the wind speed conversion can be carried out by using the exponential law of the sub-profile (Atkinson, 1981).

(5) $\bar{u}={\bar{u}}_1{\left[\frac{z}{z_1}\right]}^{\mathit{m}}$(5)

Among them: ${\bar{u}}_1$ is the average wind speed at the known height $z_1$,$\bar{u}$ is the average wind speed at the desired height $\mathit{z}$, and $\mathit{m}$ is the stability parameter, which refers to the degree of stability of the atmosphere.

In the case of the pedestrian walkway research area environment, the shady place shades part of the solar radiation and starts to affect the sensible heat flux, and the shaded place will generate a new cold air source to generate sensible heat convection. In this mode, the air temperature will also be related to Atmospheric humidity, atmospheric pressure, and wind speed have a certain correlation. Through the measurement of the above environmental factors, the air temperature change model formed by the ground reflection attenuation phenomenon caused by trees can be explored.

2. Methods

2.1. Scope of study area

The human-scale three-dimensional space with the street length as the research distance can be seen as a rectangular spatial area 3 metres wide and 2.5 metres high. In order to study the urban thermal environment in a subtropical region, a suitable pedestrian walkway with typical characteristics needs to be selected. For this study, the third section of a pedestrian walkway, which is the main arterial road in Taipei City running east to west with a street width of approximately 40 m, was selected. The pedestrian walkways and tree planting on both sides of the road provided an ideal research site. The data were collected along a linear pedestrian path with each detection point having an exposure and shade control group to explore the influence of the shade effect. The selected research area is depicted in Figure 1 below.

Figure 1. Schematic diagram of research area and equipment.

2.2. Data acquisition

2.2.1. Surface temperature data

In order to obtain the surface temperature data of the study area, this study first used the FLIR ONE PRO thermal imager lens and combined it with a mobile phone to capture real-scene infrared image data of pedestrian-scale pedestrian walkways. The use of FLIR ONE PRO thermal imager lens has the advantages of good imaging effect, portability, and a wide range of applicable mobile phone types, and it can use FLIR Tools with software to facilitate post-data analysis. The software interface is shown in Figure 2. Additionally, the infrared thermometer is used to obtain point data collection in a specific area. The infrared thermometer has the characteristics of high precision and easy portability. (The relevant parameters of FLIR ONE PRO thermal imager and infrared thermometer are shown in Tables 2 and 3)

Figure 2. The real scene image and thermal environment image of the pedestrian survey process.

Table 2. Technical indicators of FLIR ONE PRO thermal imager.

Attribute	Parameter
Thermal pixel size	12 μm
Thermal resolution	19200pixel（160 × 120）
Thermal sensitivity	70mK
Object temperature range	−20℃-400℃
Operating temperature	0 ℃ −35 ℃, battery charging temperature 0 ℃ −30 ℃
Non-operating temperature	−20℃-60℃
Size (width×height×depth)	68 × 34 × 14 mm
Weight（contain battery）	36.5 g
Long wave range	8–14 μm
Visible light resolution	1440 × 1080
HFOV/VFOV	50°±1°/43°±1°
Frame frequency	8.7 Hz
Focus	Fixed 15 cm – infinitely far
Precision	±3 ℃ or ±5% of the reading, the percentage difference between the typical ambient temperature and the scene temperature. When the temperature is between 15 ℃ −35 ℃ and the scene temperature is between 5 ℃ −120 ℃, it can be applied after 60 seconds of startup.
Capture mode	video、static images、delay
Video and still image display	save as1440 × 1080
File format	static images – Radiation measurement jpeg picture video – MPEG-4（file format MP4）
Point temperature measurement	turn on/turn off: resolution ratio0.1℃(0.1℉)
Adjustable MSX distance	0.3 m – afocal

Table 3. Technical specifications of UT380 infrared thermometer.

Attribute	Parameter
weight	147.5 g
temperature range	−50 ℃ to 380 ℃
Display accuracy	±2%
D:S	12:1
emissivity	0.1–1.00 (adjustable)
Spectral response and reaction time	（8–14）μm &500 ms
display resolution	0.1°C
operating temperature	0–50℃

2.2.2. Thermal data of vertical change

To obtain thermal environment data at the pedestrian scale, this study will employ a mobile dynamic probe sensor device for data collection. The device is equipped with sensors at heights of 0.5 m, 1 m, 1.5 m, and 2.0 m to measure the air temperature at different heights, while a thermometer will be used to measure the surface temperature. Data collection will be conducted for the entire pedestrian walkway area in the research area, and the measured positions need to be measured separately at the shaded area of each street tree and the sunlight exposure area next to it. The obtained data will form a basis for comparative analysis.

The Raspberry Pi is a dynamic detection sensor that is small in size and has low power consumption, making it a convenient mobile space detector. To obtain more detailed parameters of the space environment, air temperature and GPS sensors are included in the installation. The equipment installation facilities, equipment parameters, and software design are based on previous work using Raspberry Pi devices (Chen et al., 2021) The software component of the device is modularised into four parts: the Firebase cloud storage module, space sensor module, Raspberry Pi module, and 4 G LTE transmission module. This modularisation makes it easier to replace and maintain the device. The research uses Python as the development language, and Linux as the operating system for the Raspberry Pi. Relevant drivers are written to connect the sensor component to the Raspberry Pi, and a SQLite database is built to automatically store sensor data every second, which is then transmitted to the cloud storage module through 4 G. (The parameters related to Raspberry Pi are shown in Table 4).

Table 4. Parameters of raspberry pi related devices.

Name	Model	Major Function	Parameters and Characteristics
Raspberry Pi	Raspberry Pi 3 Model B	Data processing, collection, and bidirectional transmission with the cloud	Four core small single crystal computer with fast speed and low price.
Temperature, humidity, atmospheric pressure sensor	CJMCU-280E BME280	Collection of temperature, humidity, and atmospheric pressure data	1. Temperature: Accuracy ±0.5 ℃ (at 25 ℃). 2. Humidity: Response time of 1 second, accuracy ±3%. 3. Atmospheric pressure: The relative accuracy is ±12hPa, equivalent to a resolution of 1.5 cm at an altitude difference of ±1m.
GPS satellite positioning module	ATGM336H	Collect coordinate point information	Support BDS/GPS/GLONASS multi-mode hybrid positioning, with positioning accuracy less than 2 metres.

2.2.3. Timeline for research survey

The survey was conducted on 16 June 2021, near the Taipei weather station. The temperature on the day ranged from 28.9°C to 35.5°C, which is typical of the summer solstice and avoids the one-week period before and after the rainy season. The collection time corresponds to a period when the thermal environment in Taipei is most significant, i.e. between 13:00 and 14:00. This period is characterised by the angle of direct sunlight at noon, which is in line with the research requirements. To collect thermal environment data, the survey was divided into two parts: the collection of thermal image data of the entire block and the collection of vertical space thermal environment data by setting spatial points (see Figure 3).

Figure 3. Distribution map of spatial detection points in pedestrian walkway.

3. Results

3.1. Vegetation coverage and shading

Figure 4 shows the distribution of trees in the research section. The green coverage rate of each road section was calculated using the formula: (projected area of vertical vegetation/individual street area) × 100%. The green coverage rate on the north side of the pedestrian area of a pedestrian walkway was found to be 54.90%, while the south side had a rate of 23.16%. Overall, the green coverage rate for the entire road section was 25.36%. Similarly, the shading rate was calculated using the formula: (projected area of vegetation/individual street area) × 100%. The shading rate on the north side of the pedestrian area of a pedestrian walkway was 65.66%, while the south side had a rate of 27.69%. Overall, the shading rate for the entire road section was 30.33%.

Figure 4. Distribution map of tree planting in pedestrian walkway.

3.2. Analysis of the surface thermal environment in the study area

Surface thermal environment data in the research area is collected using a FLIR ONE PRO thermal imaging lens attached to a mobile phone. Thermal images of the surface temperature are taken to gather accurate data for further analysis. The relevant shooting data is shown in Table 5.

Table 5. Data collection table of thermal image-related data in the study area.

Road name	Road section length	Collection route	Collection quantity
The north side of the third section of a pedestrian walkway	360 m	east to west	121 sheets
The south side of the third section of a pedestrian walkway	360 m	west to east	125 sheets

This study utilised an average sampling of 10 thermal imaging images of road sections on both sides of the research area to serve as the data source. The selected images were subjected to data mining using the FLIR Tools software to obtain the temperature data of each image data. The collected data of road sections in the research area were divided into two groups. The first set includes the maximum temperature, minimum temperature, and average temperature of the thermal surface of the overall space of the sampled thermal imaging data, while the second set comprises the maximum temperature, minimum temperature, and average temperature of the thermal surface of the pedestrian street surface of the image data. Using the summary of the collected data, a table was compiled, which displays the summary of the average thermal surface temperature of each research area (Table 6). The average temperature of the street valley and the pedestrian street valley on the south side of a pedestrian walkway are both higher than those on the north side of the pedestrian walkway, with a temperature difference of 2.2°C and 2.8°C, respectively. The south side of a pedestrian walkway registers as the highest temperature area in the street valley of each research section.

Table 6. Summary of average hot surface temperature in each study area.

Name	Maximum temperature ℃	Minimum temperature ℃	Average temperature ℃
The north side of a pedestrian walkway	56.9	34.9	41.1
The south side of a pedestrian walkway	61.3	30	43.3
Pedestrians on the north side of a pedestrian walkway	56.9	36.3	40.8
Pedestrians on the south side of a pedestrian walkway	55.8	36.8	43.6

4. Discussion

4.1. Temperature change at each elevation level of the research section

The temperature change map at each height and level of the third section of a pedestrian walkway (Figure 5) reveals that temperature change is more significant at the range of 0-100 cm, and the temperature change is less pronounced at the range of 100 cm-200 cm. To display temperature change more vividly, the Kriging interpolation method is employed to analyse the interval temperature change map. The surface temperature change map distinctly illustrates that the detection points in the shaded and sun-exposed areas show a sharp temperature contrast. The data demonstrates that the temperature of the sun-exposed areas is higher, whereas the temperature of the shaded areas is relatively lower. The third section of a pedestrian walkway’s detection point data shows that the maximum temperature of sun-exposed areas can reach 56°C, the minimum can reach 39°C, and the average can reach 47.2°C. In contrast, the highest temperature in the shaded area detection point in the third section of a pedestrian walkway was 45°C, the lowest was 33°C, and the average temperature reached 37.6°C. The values indicate a temperature difference of nearly 10°C, demonstrating that the shading effect of tree cover plays a crucial role in controlling surface temperature.

Figure 5. Temperature variation at different elevations in pedestrian walkway.

4.2. Vertical temperature variation at different elevations in the research area

Table 7 and Figure 6 show the vertical space data and temperature variation data of each measurement point in the research area. The vertical temperature difference between sun exposure and shade was noticeable. At surface level (elevation 0 m), the minimum temperature difference between sun exposure and shade was 8°C and the maximum difference was 15°C, with an average temperature difference of about 11.9°C. This indicates that the surface pavement under the sun exposure has remarkable heat storage and long-wave reflection. As per general theory and experience, the temperature of the surface is mostly influenced by the material, and the temperature decreases with an increase in distance from the ground. The temperature under the shade of trees is usually lower than that exposed to the sun. However, at an elevation of 0.5 m, the average temperature difference between sunlight and shade was about 1.1°C, with the smallest difference between sun and shade at 0.08–1.07°C in four vertical temperature changes. Another phenomenon observed was that the average temperature difference between sunlight and shade at an elevation of 1.5 m was about 0.6°C, with two vertical temperature changes in all measurement points, with the difference between sun and shade at 0.2–0.6°C. Moreover, there were two vertical temperature changes where the temperature under the shade was higher than the sun temperature, with the smallest difference at about 0.25–1.0°C. Therefore, higher distance from the ground does not necessarily mean lower temperature, and the temperature under the shade of trees is not always lower than the temperature exposed to the sun.

Figure 6. Line graph of vertical temperature change under sunlight and shade.

Table 7. Vertical temperature change between sunlight and shade.

H℃	0.0 m	0.5 m	1.0 m	1.5 m	2.0 m
BK1S	48.00	41.35	38.82	39.36	43.03
BK1	37.00	40.28	38.71	40.36	40.08
BK2S	45.00	40.08	42.26	41.62	43.07
BK2	33.00	39.84	40.81	40.98	41.47
BK3S	49.00	41.59	42.98	42.46	42.32
BK3	35.00	40.61	40.91	41.43	38.39
BK4S	50.00	40.76	39.01	38.72	36.70
BK4	40.00	39.52	38.27	38.11	35.98
BK5S	56.00	40.40	37.76	37.70	38.98
BK5	41.00	38.78	37.43	37.50	36.31
BK6S	47.00	40.65	39.40	38.70	36.69
BK6	34.00	40.57	38.54	38.47	36.48
BK7S	49.00	41.95	38.64	38.30	40.87
BK7	41.00	40.65	38.63	38.55	37.75
BK8S	49.00	44.65	43.73	42.71	39.68
BK8	37.50	42.25	40.86	40.59	38.62
BK9S	49.00	43.02	42.66	42.32	42.43
BK9	36.00	41.95	40.21	40.37	38.35

Based on the vertical spatial data of each detection point in the pedestrian walkway, we utilised the Kriging interpolation method in ArcGIS software to analyse and visualise the temperature change map of the vertical spatial data. Figure 7 and Figure 8 illustrate the spatial vertical temperature change on the north and south sides of pedestrian walkway, respectively.

Figure 7. Vertical spatial temperature change diagram of detection points on the north side of a pedestrian walkway.

Figure 8. Vertical spatial temperature change map of detection points on the south side of a pedestrian walkway.

The vertical spatial temperature change map of the detection points on the north side of a pedestrian walkway is depicted in Figure 7. The areas with higher temperatures are predominantly observed on both sides of the road, whereas the middle area displays a more pronounced cooling effect. The hot zone in the lower area is mainly the region exposed to sunlight. The local area appears to be relatively high and corresponds to the intersection area. The left and right sides of the region exhibit a cluster-like hot spot. This is an area with denser trees on the street, making it difficult for the wind to disperse. As a result, local hot clusters form due to the lack of air circulation.

The vertical spatial temperature change map of the detection points on the south side of a pedestrian walkway is presented in Figure 8. It is observed that the low-temperature effect is concentrated below the height of 1.0 m, while the overall temperature of the south side of a pedestrian walkway is lower than that of the north side of the road. The temperature on both sides of the road shows a similar variation pattern with height. Specifically, the exposed area surface has a higher cooling effect from the surface to 0.5 m to 1.0 m height, whereas the temperature rises significantly from the surface to 0.5 m under the shade of trees, and the height from 0.5 m to 1.0 m has a relatively high cooling effect. From 1.0 cm to 2.0 m, the overall temperature has a certain continuous cooling, but there is a warming phenomenon in the local area at the height of 1.5 m to 2.0 m. The areas that cause local heat clusters are mainly distributed near some intersection areas, where a large area is exposed to the sun resulting in higher temperature. On the north side of the road, there are left and right heat clusters formed in the same area, while on the south side, the trees are denser, which affects the ventilation effect, and heat clusters are formed in different areas. Therefore, it can be inferred that intersection areas and densely planted areas have a greater impact on the air temperature.

4.3. Relationship between green coverage rate and spatial temperature variation pattern

The pedestrian walkway’s vertical spatial temperature change was analysed using the Kriging interpolation method in the ArcGIS software. The overall data of each road section was used to investigate the spatial temperature change at the pedestrian scale. The lowest temperature area in the entire block was found to be in the shade of the pedestrian walkway. This suggests that the shade effect of trees plays a role in cooling the space temperature. The temperature change from the surface to the height of 1.0 m is relatively significant, while the change from 1.0 m to 2.0 m is relatively mild. The overall temperature drops slightly from the height of 1.0 m-1.5 m, followed by a temperature decrease trend from the height of 1.5 m-2.0 m. However, some local temperature rises instead of decreasing.

To better understand the temperature variation trend of the road section with height, a broken line diagram of the spatial temperature change trend was drawn using the vertical spatial data of the research section, as shown in Figure 9. The average value of the height section of the research area was used to draw the diagram. The shaded area has lower temperatures than the sun-exposed area at all heights in each section of the road, indicating that the shade of trees has a cooling effect. As mentioned above, the temperature change is significant only between 0 m-1.0 m. There is a continuous temperature drop trend between 1.0 m-2.0 m, but the cooling effect is not significant, and the temperature difference from the sun-exposed place is not obvious.

Figure 9. Schematic diagram of broken line of vertical spatial temperature change trend under different modes of the research road section.

The presence of hotspots in certain areas of the pedestrian walkway can be attributed to two factors. Firstly, locations near the intersection on both sides of the road or in the middle of the road are exposed to a large-scale sun-exposed area. Secondly, the corner of the road and dense areas create regions where the wind cannot blow, leading to the formation of hotspots. The density of tree planting has a positive correlation with the green coverage rate and shading rate of each road section. Therefore, the analysis highlights the following characteristics: First, the air temperature at various heights in the shade area is slightly lower than the air temperature at the same height under the surrounding sunlight exposure. Second, areas with low green coverage and coverage of pedestrian walkways (i.e. intersection areas) are prone to forming hot spots. Third, areas with uniform tree planting density, i.e. areas with moderate green coverage in the road section, have better cooling effects.

In the past, studies have mostly proposed that increasing vegetation cover and increasing the number of shade trees has some positive effects on reducing regional temperatures and improving human comfort (Lanza & Durand, 2021; Ren et al., 2022), but the data in this study showed that in the vertical spatial direction, the difference between air temperatures in the open and in the shade is not very large, and that locally excessive tree planting densities instead produce localised regional heat clusters, although this may be partly due to localised poor ventilation, for example. From the measured data of this study, it seems that blindly increasing the number of trees and plants to achieve the cooling effect is not a wise behaviour, but an appropriate amount of scientific dispersal of the number of trees and plants set the site will produce better cooling purposes.

5. Conclusions

Specifically, the study investigates the green coverage rate of pedestrian walkways and its influence on the thermal environment. Based on the actual measurement and analysis results of pedestrian walkways, the following characteristics are displayed. The shading of non-densely planted street trees has limited effect on mitigating the thermal environment temperature of sidewalks. The phenomenon pattern of air temperature attenuation from ground up varies at different heights on sparsely planted pedestrian walkways in urban areas. Compared to the vertical variation amplitude, the degree of attenuation in sunlight exposure is greater than under tree shade. When the air temperature is at a height of 1.0 m, the shade and sunlight exposure areas exhibit the same temperature. However, there is little difference in temperature between the shaded area and the sunny area at a height of 1.5 m-2 m. The shading effect of planting will cause different ground reflection temperature attenuation modes in shaded places and exposed places, which will affect the temperature of pedestrian walkways. Sparse planting of green cover will cause vertical changes in air temperature, while dense planting of green cover will cause the phenomenon of poor wind flow. Therefore, an appropriate tree planting density can produce a better cooling effect.

This study proposed to analyse and infer microclimate pedestrian areas on a human scale, which provided new ideas and research methods for urban heat island research. Of course, research also has certain limitations, and in the future, further research will be conducted on more time periods and timelines, and the achievements made will be supplemented and revised. Subsequent research will delve deeper into the impact of tree shading on neighbourhoods and explore the influencing factors of air temperature changes at different heights. Provide application methodologies for walkable cities and thermal comfort environments, as well as methods for improving urban design under the influence of climate change.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. The dynamic probe used self-designed and assembled environmental sensors for this research, and assembled them with Raspberry Pi to collect environmental sensing data. The sensor is suspended above the mobile detection device of 0.5 m, 1.0 m, 1.5 m, 2.0 m, and records the environmental data of the detection point per second, including air temperature, humidity, atmospheric pressure, air quality, wind speed and GPS position。.