Study Roadmap Selection Based on the Thermal Comfort of Street Trees in Summer: A Case Study from a University Campus in China

Abstract

The intensification of the urban heat island effect, characterized by persistent high temperatures in Chinese cities during summer, has led to notable shifts in urban residents’ activity patterns and travel preferences. Given that street trees, as fundamental components of urban road networks, have significant interaction with residents, it is imperative to investigate their thermal comfort impact. This study aims to enhance the comfortable summer travel experience for urban dwellers. Fujian Agriculture and Forestry University (FAFU) was selected as the case study site, with eight street tree species identified as measurement points. The summer solstice (21 June 2023) served as the representative weather condition. Through monitoring temperature and humidity, the study explored the correlation between street tree species, their characteristic factors, and thermal comfort. Utilizing ENVI-met and ArcGIS, the thermal comfort of campus travel routes was assessed, leading to the development of a summer travel guide based on thermal comfort considerations. The research novelty lies in applying a combined ENVI-met 5.0.2 and ArcGIS 10.8 software approach for modelling and visualizing the microclimate, which enables a more precise analysis of the thermal comfort variations of different campus paths, thus improving the accuracy and applicability of the results in urban planning. The findings reveal several points. (1) Different street trees possess varying capacities to enhance human comfort, with Falcataria falcata and Mangifera indica exhibiting the strongest cooling and humidifying effects, whereas Bauhinia purpurea and Amygdalus persica perform the poorest. Additionally, the research confirms ENVI-met’s scientific accuracy and practicality for microclimate studies. (2) The contribution of street trees to the comfort of campus road travel is primarily determined by the Sky View Factor (SVF), which negatively correlates with cooling and humidifying intensity and positively with thermal comfort. (3) During midday, travel comfort conditions on campus roads are better. Based on the thermal comfort assessment, a summer roadmap was created for the campus. In this case, the campus roads indicated by road A are considered the best travel routes in summer, and the roads indicated by roads B and C are considered alternatives for travelling. This practical application demonstrates how theoretical research results can be translated into practical tools for daily commuting and urban planning. It provides data references and empirical cases for the scientific optimization and enhancement of urban roads.

1. Introduction

Meteorological standards often define heatwaves as periods where the maximum temperature surpasses the average maximum temperature by 5 °C or more for at least five consecutive days [1]. Additionally, urban areas experiencing temperatures exceeding 35 °C (95°F) for prolonged periods are considered to experience severe heat conditions [2]. The increasing prevalence of intense heat during summer has necessitated a more holistic and integrated approach to addressing this issue. One effective strategy that has been identified involves enhancing the microclimate through urban greening initiatives, particularly by focusing on the strategic selection of tree species for roadways [3]. Recent studies contribute to a detailed understanding of these dynamics. Karimi et al. highlighted the new developments and challenges in mitigating urban heat islands, emphasizing the need to focus on vegetation and also consider broader aspects of urban planning and material selection [4,5]. Mohammad et al. examined the combined effects of albedo and vegetation in urban street canyons, demonstrating that the reflective properties of surrounding surfaces can either enhance or reduce the cooling benefits of tree species. This method integrates greenery with appropriate urban materials to maximize pedestrian thermal comfort [6]. This approach has been recognized as an effective means of improving the thermophysical properties of the urban road network’s spatial environment. Klemm et al. conducted a comprehensive study involving meteorological measurements and surveys across nine streets in the Netherlands, revealing that a 10% increase in the coverage of large trees could reduce the average radiant temperature by 1 Kelvin [7]. However, it is important to note that the extrapolation of these findings might be limited by the specific urban and climatic contexts of the Netherlands, which may not directly apply to regions with different environmental conditions or urban fabrics; similarly, Souch observed that street trees are more effective in moderating temperature and humidity than simply expanding green spaces, with tall trees possessing high canopy coverage offering the greatest cooling effects [8]. While insightful, Souch’s methodology may not fully account for the varying maintenance needs of different tree species or the long-term sustainability of these green spaces, which could influence the feasibility of these interventions in different urban settings. Building upon these findings, Xu Min and Ma Xiuzhi, among others, delved into the impact of different tree species on microclimate modulation, identifying Platanus × acerifolia [9] and Salix babylonica L. [10] as having the most significant cooling impact. However, their research was predominantly conducted in urban settings within East Asia, which might introduce regional biases related to specific atmospheric conditions and urban planning norms that are not universally applicable. This indicates that the effectiveness of strategies based on thermal comfort can vary significantly depending on local geographic and climatic conditions. Studies such as those by Karim [11] and Dokhanian [12] emphasize the importance of considering specific urban configurations and regional climate characteristics in microclimate research, suggesting that strategies effective in one geographical or climatic context may not be directly applicable in another. Given this, the current study aims to explore how street tree greening affects thermal comfort within the climatic backdrop of the subtropical humid, warm regions, providing a detailed understanding. This approach helps us understand effective urban greening strategies and underscores the necessity of effectively tailoring interventions to specific locations to improve the urban thermal environment [13].

Research has established that microclimatic elements, including air temperature, humidity, wind speed, and solar radiation, significantly influence people’s thermal perceptions [14]. Notably, air temperature is identified as the primary determinant of outdoor thermal comfort [15], while the role of plant shading in modulating solar radiation also critically contributes to thermal comfort levels [16]. The evaluation of thermal comfort depends on the specific geographical location, climate type, and meteorological conditions of the study site. Thermal comfort assessment methodologies are broadly categorized into two types; namely, empirical models, which are based on subjective human perceptions and represented by indicators such as the wet bulb globe temperature and mean radiant temperature [17], and mechanistic models, which rely on the human body’s heat balance for objective evaluation, exemplified by indices like the Predictive Mean Vote (PMV) [18], Standard Effective Temperature (SET) [19], Universal Thermal Climate Index (UTCI) [20], and Physiological Equivalent Temperature (PET) [21]. China’s vast expanse and varied topography result in a diversity of climates, influenced by latitude, elevation, terrain, and solar radiation. In regions characterized by hot and humid climates, temperature and humidity emerge as principal meteorological determinants for thermal comfort research. Commonly utilized human comfort indices include the temperature–humidity index (THI) [22], Physiological Equivalent Temperature (PET) [23], and Predictive Mean Vote (PMV) [24]. It is important to note that PET assessments are typically conducted in controlled indoor environments devoid of wind and solar radiation influences [23], and the PMV index may exhibit considerable inaccuracies when applied to outdoor thermal comfort evaluations [18]. The THI, due to its simplicity of calculation, established thermal regime foundation, and widespread application, is particularly favored for studies in hot and humid locales. Research conducted by Haiyan Yan and Qianqian Liu, Dyah and Bambang, and Huilin Yi utilizing the THI in Beihai, Yogyakarta City, and Guangzhou, respectively, demonstrated a pronounced correlation between microclimate variables and thermal comfort levels, along with notable variations in thermoneutral temperatures across different geographic settings [25,26,27]. Current research methodologies, including fixed-point measurements [28,29] and questionnaire surveys [30], are frequently employed to assess the impact of landscape elements such as vegetation, water features, and pavements on thermal comfort. These methods are used to formulate enhancement strategies for urban parks [31], residential green spaces [31], community streets [32], and other public areas in relation to thermal comfort [33]. However, the process of gathering data through fixed-point measurements imposes stringent demands on the precision of measuring instruments and the comprehensiveness of indicators. Additionally, it demands significant time to monitor the real-time variations in the microclimate across different campus components. Conversely, the use of subjective questionnaires often leads to less intuitive evaluation outcomes and fails to capture the spatial distribution of the thermal environment comprehensively along campus pathways [34]. Consequently, numerical simulation and ArcGIS have become more appropriate tools for the analysis of urban spatial environmental performance and the evaluation of outdoor conditions [35]. For instance, Nazarian et al. demonstrated the importance of numerical modeling in investigating thermal comfort under complex urban conditions by focusing on urban streets as their study subject [36]. Building on this foundation, D. Antoniadis and colleagues utilized ENVI-met to assess various landscaping interventions aimed at enhancing outdoor human thermal comfort on campuses [37]. In light of these insights, this paper proposes utilizing ENVI-met and ArcGIS software to conduct a regional continuity analysis of thermal comfort on campus roads. This approach aims to ensure comprehensive coverage across the area, thereby contributing valuable insights and practical applications for improving pedestrian travel comfort. There exists a significant relationship between university campus roads and urban roads. An optimal campus road environment can greatly enhance the travel comfort of students and faculty, thereby boosting learning efficiency. This study uses Fujian Agriculture and Forestry University (FAFU) in Fujian, China, as a case study. It focuses on analyzing the spatial and temporal variations in the thermal comfort offered by campus roadway trees and examining how their characteristics affect this comfort. Additionally, this research employed ENVI-met and ArcGIS as a method to analyze the thermal comfort of urban campuses. After validating the model’s accuracy, a visual and regionally continuous evaluation of thermal comfort along campus roads was conducted—providing a practical tool for a summer travel guide. This offers valuable insights for urban planners, landscape architects, and policymakers involved in urban development and climate adaptation strategies. Not only is this study applicable to the specific context of Fujian Agriculture and Forestry University, but it can also be applied to similar urban environments globally, making a significant contribution to the existing body of knowledge.

2. Methods

2.1. Study Sites

Fuzhou City exhibits a subtropical monsoon climate, marked by high humidity, averaging 74% annually, and persistent high temperatures during summer, which can reach up to 40 °C. This has earned it the distinction of being one of China’s ‘four furnaces’ [38]. The present study is conducted at Fujian Agriculture and Forestry University’s Jinshan Campus in Fuzhou City for several compelling reasons. (1) The campus is home to over 30,000 teachers and students, whose daily routines and academic pursuits are significantly affected by the intense and humid summer conditions, necessitating urgent measures to alleviate the heat in the campus road environment. (2) The variety and abundance of street trees on the campus reflect the main types found in Fuzhou City; these trees are diverse, well-established, and effectively managed, making them ideal for this study on thermal comfort. (3) As a distinct social entity with comprehensive landscape elements and clear regional boundaries, the campus provides a valuable microcosm for research that can be applied to broader urban settings, offering useful insights for future road design.

The road network at Fujian Agriculture and Forestry University is extensive, with walking being the predominant form of transportation for faculty and students. The university’s primary thoroughfares, including Jinshan Road, Jiangle Road, Tongle Road, and Xia’an Road, link various campus areas (Figure 1). These roads cater to the daily commuting needs of the university community and assist in managing the influx of urban traffic from outside the campus. This study focuses on the peak noontime hours—12:00, 13:00, 14:00, and 15:00—to explore and document how comfort levels on campus roads fluctuate throughout this period.

2.2. Sample Site Selection

Based on commonly utilized urban greening street tree species in Fuzhou City [39], eight types of trees were selected within the campus for study, namely Michelia × alba, Syzygium hainanense, Amygdalus persica, Magnolia grandiflora, Falcataria falcata, Litchi chinensis, Mangifera indica, and Bauhinia purpurea (Figure 2), labeled Measurement Points 1 through 8. Additionally, an open space was designated as a control group, marked as A0, to facilitate discussion on the temperature and humidity effects of different tree species. The selection of these species adhered to specific criteria: (1) the street trees are prevalent throughout the urban area of Fuzhou and are maintained in stable condition with regular management; (2) the physical appearance and botanical attributes of the species clearly reflect their classification; (3) the environmental conditions on either side of the campus roads are uniform, with no obstruction from other plants or buildings, similar underlying surfaces and topography, and consistent road widths and tree spacing.

2.3. Field Measurement

In this paper, the summer solstice was selected for field measurements. The summer solstice usually occurs on June 21 in the northern hemisphere. It represents the day with the longest sunshine hours [40], with distinct microclimatic characteristics (temperature, humidity, wind speed), and is also considered to be the warmest day of the year in China [41]. This period is crucial for studying the microclimatic effects of vegetation and water bodies to effectively mitigate the urban heat island effect. Meanwhile, choosing the summer solstice for climate and comfort assessment has ensured that the study results are robust and valid under the widest possible range of conditions [42].

In the center of the sample plots, 1.5 m from the ground, an anemometer (Kestrel 5500) was placed to monitor 8 sample plots employing the manual movement detection method. Handheld measuring instruments were used to measure and record the detailed data, and then quickly rushed to the next measurement point. The data collected included air temperature and relative humidity. In the experiment, efforts were made to ensure that the interval between switching measurement points was the same. The actual measurement time was 21 June 2022 (summer solstice); the weather was sunny and had no wind. The measurement time was from 12:00 am to 15:00, with an interval of 1 h. The measured data were used to analyze the temporal variation patterns of different roadside trees and verify the ENVI-met model’s accuracy. In addition, the Sky View Factor (SVF) of the campus road was obtained by setting up a 1.6 m high fisheye camera, and then the photos were imported into Rayman and processed to obtain the SVF values. The results are shown in Figure 3.

2.4. ENVI-Met Simulation

2.4.1. Model Validation Parameter Settings

ENVI-met is a microenvironmental numerical simulation software developed by Prof. Michael’s team at the Institute of Geography, University of Bochum, Germany, in 1998 [43]. Its primary function is to calculate and simulate the interactions between “plant-building surface-air-soil” and present the results as two-dimensional or three-dimensional images [44]. In this study, we used ENVI-met version 5.0.2, which includes five sections, namely, Space, Database Manager, ENVI-guide, ENVI-core, and Leonardo, corresponding to five main functions, namely, scenario modeling, model parameter management, 3D plant modeling, simulation calculation, and result analysis. The accuracy of ENVI-met simulations heavily depends on the correctness of boundary conditions and initial parameter settings [45]. Any misestimation in these inputs can lead to significant deviations in the output, affecting the reliability of the simulation results [46]. Therefore, prior to running the model, the study area must be modeled (Figure 4), and only after the simulation data have been calibrated with actual measurements can further simulation studies be conducted [47]. This study obtained meteorological data from the Chinese website (https://rp5.ru/) on 21 June 2022, which hosts weather data for 241 countries, to provide initial meteorological parameters for the ENVI-met simulation. Based on the size of the simulated site, the grid size for ENVI-met modeling was set to 260 × 314 × 30. The grid cell dimensions were set as dx = 4, dy = 4, dz = 3. The simulation period was from 5:00 to 19:00, with data from 8:00 to 18:00 selected for study. To ensure consistency between the simulated and actual environmental conditions, the initial parameters such as plot size, meteorological parameters, surrounding environment, vegetation condition, and timing in the model were set to match those of the actual site. Details of other model validation parameters are shown in

Table 1. Model validation parameter settings.

Parameter Name	Parameter Name	Parameter Values
Grid Settings	Model dimensions /Size of grid cell in meter	260 × 314 × 30/4 × 4 × 3
Model Location	Base settings	Fujian Agriculture and Forestry University. (26.08° N, 119.23° E)
	Microscale roughness length of surface (m)	0.01
Time and date	Start date	21 June 2022
	Start time	5:00 am
	Total simulation time	14
Meteorological data	Specific humidity in 2500 m (g/kg)	7
	Wind direction	135 degrees (south-east)
	windspeed (m/s)	2.5
	temperature range	17–28
Soil Section	Upper layer (0–20 cm)	65 °C/50%RH
	Middle layer (20–50 cm)	70 °C/50%RH
	Deep layer (50–200 cm)	75 °C/50%RH

.

2.4.2. Validation of Model Validity

The methods used in this paper to validate the model accuracy include Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) (Table 2).

This study utilized ENVI-met 5.0.2 to compare actual measurements with simulation results using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) to assess deviations between observed and simulated values. The results indicate that, through the comparative analysis of average measured and simulated values of various microclimatic elements during summer (Figure 5), both simulation and actual measurements show consistent trends and high levels of fit. The errors between daily average simulated and measured temperatures ranged from 0.35 °C to 1.43 °C; daily average simulated and measured humidity errors were ≤2%; and daily average simulated and measured wind speeds had errors of <0.07 m/s. For the study area in summer, the RMSE values for temperature, humidity, and wind speed were 0.87 °C, 1.01%, and 0.05 m/s, respectively; the MAE values were 0.82 °C, 0.82 °C, and 0.04 m/s, respectively; and the MAPE values were 2.58%, 1.18%, and 9.22%, respectively (

Table 3. Model accuracy value.

Accuracy Value	Temperature/°C	Humidity/%	Wind Speed/m/s
RSME	0.87	1.01	0.05
MAE	0.82	0.82	0.44
MAPE	2.58	1.18	9.22

). The acceptable error ranges for the model were RMSE 0 to 1.5; MAE: 0 to 1.5; MAPE ≤ 10%, indicating that the model is highly accurate and the experiment is effective, allowing further experimental simulations to be conducted [48].

Table 2. Model accuracy validation equation.

Number	Formula	Meaning	Symbol and Its Representation	Reference
1	$RMSE=\sqrt{\frac{{\sum }_{i=1}^n{\left(x_i-x_i^{\prime }\right)}^2}{n}}$	Evaluate the accuracy of the model, compare the error between the measured and simulated values, and reflect the applicability of the model.	xi denotes the simulated value; xi’ denotes the measured value; n denotes the number of tests.	[49]
2	$MAE=\frac{1}{n}\sum \left|x_i-x_i^{\prime }\right|$
3	$MAPE=\frac{1}{n}{\displaystyle \sum _{i=1}^n\frac{\left|x_i-x_i^{\prime }\right|}{xi}}\times 100\%$

Table 2. Model accuracy validation equation.

Number	Formula	Meaning	Symbol and Its Representation	Reference
1	$RMSE=\sqrt{\frac{{\sum }_{i=1}^n{\left(x_i-x_i^{\prime }\right)}^2}{n}}$	Evaluate the accuracy of the model, compare the error between the measured and simulated values, and reflect the applicability of the model.	xi denotes the simulated value; xi’ denotes the measured value; n denotes the number of tests.	[49]
2	$MAE=\frac{1}{n}\sum \left|x_i-x_i^{\prime }\right|$
3	$MAPE=\frac{1}{n}{\displaystyle \sum _{i=1}^n\frac{\left|x_i-x_i^{\prime }\right|}{xi}}\times 100\%$

2.4.3. Parameters of Street Tree Models

The ENV-met plant database contains rich plant models defined by different leaf area density (LAD) and root area density (RAD). The parameters include plant height, crown spread, plant leaf type (evergreen or deciduous), CO2 fixation type, leaf reflectance, root depth, root diameter, and root area density. This paper is based on a field study of campus street trees, with eight street tree species containing Michelia × alba, Syzygium hainanense, Amygdalus persica, Magnolia grandiflora, Falcataria falcata, Litchi chinensis, Mangifera indica, Bauhinia purpurea. The tall tree plant models that need to be used were selected from the plant database. The approximation modules were selected for the plant modules not found in ENV-met to replace them with the relevant plant parameters and related attributes, as shown in

Table 4. Attributes related to plant models.

Plant Models	Code	MA	Plant Models	Code	SH
	Name	Michelia × alba		Name	Syzygium hainanense
	Plant height/m	15		Plant height/m	9
	Crown width/m	6		Crown width/m	6
	Height below branch/m	6		Height below branch/m	3
Plant models	Code	AP	Plant models	Code	MG
	Name	Amygdalus persica		Name	Magnolia grandiflora
	Plant height/m	3		Plant height/m	15
	Crown width/m	4		Crown width/m	7
	Height below branch/m	0.8		Height below branch/m	3
Plant models	Code	FF	Plant models	Code	LC
	Name	Falcataria falcata		Name	Litchi chinensis
	Plant height/m	20		Plant height/m	16
	Crown width/m	15		Crown width/m	8
	Height below branch/m	10		Height below branch/m	6
Plant models	Code	MI	Plant models	Code	BP
	Name	Mangifera indica		Name	Bauhinia purpurea
	Plant height/m	16		Plant height/m	10
	Crown width/m	9		Crown width/m	5
	Height below branch/m	6		Height below branch/m	4

.

2.5. Spatial Statistical Tools

2.5.1. ArcGIS

ArcGIS is a desktop geographic information system platform that integrates numerous functions including spatial data display, editing, querying, statistics, report generation, spatial analysis, and advanced mapping. As an emerging cross-disciplinary technology, ArcGIS has been widely applied in urban planning [50], route selection for tours [51], and regional climate mapping [52], among others. This study uses ENVI-met simulated data, utilizing ArcGIS 10.8 to perform a series of operations such as buffering, raster calculation, clipping, and reclassification to create visual maps. The reasons include the following: (1) Compatibility and Interoperability: The ability of ArcGIS to interoperate with other software and systems (such as ENVI-met) is crucial. Although not seamless, the compatibility of ArcGIS with formats and data used by ENVI-met enables a workflow that can maintain data integrity and facilitate efficient analysis; (2) Handling Large Datasets: ArcGIS is well-equipped to manage large datasets, which is crucial for handling the extensive and detailed data output from ENVI-met simulations. These simulations can generate a significant volume of data, especially when modeling the microclimate across different times and conditions for a large urban area. ArcGIS’s capacity to handle and process large spatial datasets without significant performance degradation makes it a suitable choice for this purpose; (3) Complex Geoprocessing Tasks: ArcGIS offers a comprehensive suite of geoprocessing tools that are essential for analyzing and manipulating spatial data. This includes tasks such as raster calculations, spatial overlay, buffering, and reclassification. These tools are necessary for integrating and analyzing ENVI-met outputs, such as temperature and humidity distributions, vegetation effects, and their spatial impacts on urban microclimates; (4) Spatial results visualization: ArcGIS integrates geographic information with data to create map visualizations, effectively displaying the spatial characteristics and distribution patterns of the data, which facilitates the discovery of spatial correlations and impacts.

2.5.2. Combination of ArcGIS and ENVI-Met

ArcGIS is a powerful tool for spatial analysis and data integration, and it is highly compatible with ENVI-met, making it widely used in the field of microclimate research for visualization studies [53]. Currently, there are some potential limitations or challenges in using ArcGIS for data analysis, such as limitations in data interoperability or potential biases in the interpretation of spatial patterns. These issues are related to data formats, types, spatial coordinate systems, and spatial resolutions [54]. ENVI-met software is well-suited to address these issues. The data required for this study are all sourced from ENVI-met 5.0.2, ensuring consistency in data format, type, spatial coordinate system, and spatial resolution, which in turn enhances the scientific accuracy of spatial analysis. Furthermore, by converting ENVI-met simulation data from .csv file format to .xlsx file format, it is possible to perform spatial analysis and the visualization mapping of urban climate distribution with ArcGIS [55].

2.6. Data Analysis

In this paper, the intensity of cooling and humidification is calculated from the difference between the measured and control data (

Table 5. Data calculation formula.

Equations	Formulas	Definition	Symbols and Their Meanings	Reference
1	ΔT = t0 − t	Cooling effect	ΔT denotes the temperature drop; t0 denotes the control temperature; t denotes the measurement point temperature.	[58]
2	ΔH = h − h0	Humidifying strength	ΔH denotes incremental humidity; h0 denotes control group degree; h denotes measurement point humidity.	[49]
3	$THI=\left( 1. 8\mathrm{t}+32\right)-0. 55\ast ( 1-\mathrm{f}) \ast ( 1. 8\ast \mathrm{t}-26)$	Evaluation of human comfort in different meteorological conditions	THI refers to temperature and humidity number, t is the average temperature, and f is the relative humidity.	[59]

). Thermal comfort aims to elucidate the degree of the influence of meteorological factors such as air temperature, relative humidity, and wind speed on human function. The human body has a surface skin temperature of 33 °C and maintains an internal temperature of approximately 37 °C, with an optimal external ambient perceived temperature of 26 °C. When the external environment is too high, it affects the heat alternation between the human body and the outside world. The paper chose the temperature and humidity index (THI) as the calculation standard for thermal comfort. The temperature and humidity index is a combination of two parameters, temperature and humidity, to respond to the exchange of heat between the human body and the outside world, and it is the first index for evaluating the feeling of thermal comfort [22]. Many studies apply the temperature–humidity index (THI) to hot and humid areas in China, which is widely used and can accurately reflect the level of thermal comfort [56,57]. The degree of classification of THI comfort levels is shown in

Table 6. Temperature and Humidity Index (THI) vs. Thermal Comfort Levels.

THI Value Range	Thermal Comfort Assessment
<21.1	Comfortable
21.1–23.8	Comparatively comfortable
23.9–26.6	Normal
26.7–29.4	Less comfortable
≥29.5	Uncomfortable

. The formula of the temperature–humidity index is shown in Table 5 (Equation (3)).

2.7. Technical Lines

The technical roadmap of this study is illustrated in Figure 6 and mainly involves three aspects: field measurements, ENVI-met numerical simulations, and ArcGIS visualization. The details are as follows:

(1) Field measurements: Field measurements are used to record temperature and humidity data and the Sky View Factor (SVF) to analyze the spatiotemporal variation patterns and related characteristics of typical street trees and their impact on thermal comfort;

(2) ENVI-met numerical simulations: After accuracy validation, ENVI-met numerical simulation software is employed to simulate the temperature and humidity across the entire campus; the specific steps include downloading the research site’s jpg file from https://www.google.com/maps accessed on 21 June 2022, converting this data into a bmp format base map using Photoshop 2020 software, and then creating a campus model in the ENVI-met modeling module; the simulation period is set from 5:00 AM to 7:00 PM, with simulation data from 8:00 AM to 6:00 PM used for model validation; the data from 12:00 PM to 3:00 PM are specifically used for thermal comfort assessment;

(3) ArcGIS visualization: ArcGIS tools are utilized to calculate the Thermal Heat Index (THI) and generate visual maps; the steps include exporting the simulation data to a csv file and converting it into an Excel spreadsheet, then importing it into GIS 10.8 software to set the XY coordinate system; the data from 12:00 PM to 3:00 PM are rasterized and reclassified, followed by an overlay analysis based on weights, ultimately determining the optimal travel routes on the campus (Figure 6).

3. Results

3.1. Influence of Street Tree Species on Thermal Comfort

ANOVA one-way analysis of variance (ANOVA) was conducted to analyze the microclimatic characteristic factors (air temperature, relative humidity, cooling effect and humidifying effect) of the tree types at each time. The results showed that the street tree types were highly significant (F = 11.21, p = 0.000; F = 51.69, p = 0.000; F = 50.66, p = 0.000; F = 8.91, p = 0.000) in terms of air temperature, relative humidity, cooling effect, and the humidifying effect of the campus roadway space at different times of the midday. It shows that street trees have a significant role in improving the thermal environment for pedestrians.

As shown in Figure 7, during the midday hours in summer, the cooling effect of street tree species was ranked from highest to lowest as follows: Falcataria falcata > Mangifera indica > Litchi chinensis > Magnolia grandiflora > Michelia × alba > Syzygium hainanense > Bauhinia purpurea > Amygdalus persica; and the humidifying effect was ranked as Falcataria falcata > Mangifera indica > Litchi chinensis > Michelia × alba > Syzygium hainanense > Magnolia grandiflora > Bauhinia purpurea > Amygdalus persica. Notably, Falcataria falcata demonstrated the strongest cooling and humidifying effects, followed by the Mangifera indica and Litchi chinensis, and the weakest cooling and humidifying effect was shown by the Amygdalus persica. We hypothesized that the large canopy area of Falcataria falcata, Mangifera indica, and Litchi chinensis species types creates a more extensive shade cover, which can withstand more solar radiation, and thus, reduce the spatial temperature and humidity. This is consistent with the findings of Liu Zhenwei [60] and Dimoudi et al [61].

3.2. Influence of Street Tree Characteristics on Thermal Comfort

The comfort of campus road travel is primarily influenced by the degree of shade provided by the street trees lining both sides of the road. This shading is dependent on the extent of tree cover, which in turn relates to the size and shape of the tree canopy and the distribution of the leaf area. Trees that provide inadequate shade allow greater solar radiation penetration, adversely affecting pedestrian thermal comfort [62]. The Sky Visibility Factor (SVF) serves as a reliable indicator to quantify the degree of tree shading, replacing measures such as the Leaf Area Index (LAI) and Projected Area Index (PAI), and has been utilized in related studies [63,64]. The SVF of street trees correlates with cooling and humidifying intensities, as well as thermal comfort. The results, depicted in Figure 8, demonstrate a strong linear relationship between the SVF and the cooling intensity, humidifying intensity, and thermal comfort of street trees along campus roads (R² = 0.8534; R² = 0.9086; R² = 0.9004). These findings suggest that a larger shaded area from street trees enhances cooling and humidifying effects, as well as the overall thermal comfort. The respective linear relationships are expressed as y = −1.86 x + 2.15, y = −4.47 x + 7.919, and y = 0.97 x + 27.79, These results are in alignment with the research conducted by Zhang Zheng et al. [65].

3.3. Evaluation of Campus Traveling Comfort

The fixed-point measurement data make it difficult to comprehensively reflect the overall road comfort changes on campus. ENVI-met software can build a campus model to simulate the thermal comfort changes of road travel on the whole campus, which has been used in related studies [66,67]. Due to the lower temperature in the morning and evening, human comfort is better. Therefore, this paper mainly selects the simulation data of four moments, 12:00 pm, 13:00, 14:00, and 15:00, for analysis.

The thermal comfort evaluations of campus roads during midday hours (12:00 pm, 13:00, 14:00, 15:00) in summer are illustrated in Figure 9. Overall, the thermal comfort on campus roads is satisfactory. The distribution of thermal comfort varies across different times, with 13:00 experiencing the lowest level of comfort and 15:00 showing improved comfort levels. This variation suggests that campus thermal comfort fluctuates with changes in solar radiation, aligning with Thorsson’s findings [68]. The lowest comfort level at 13:00 contrasts with the results obtained by Chen Jashuo et al. [69], primarily due to the cooling and humidifying effects of campus street trees. These trees provide shade, effectively mitigating ground-level solar radiation and reducing near-ground heat. This confirms that street trees significantly enhance thermal comfort along campus travel routes. Furthermore, all campus roads have areas of low comfort, notably Taoshan Road in the Haan area, which consistently shows poor thermal comfort at all four times. This area, mainly a dormitory block, suffers from the building layout and sparse street tree coverage, leading to travel discomfort. This underscores the importance of roadway greening in enhancing thermal comfort, a finding corroborated by Lin et al. [70]. The northern sections of Jiangle Road and Xichen Road exhibited the best thermal comfort during these periods, attributed to the dominant presence of Falcataria falcata and Litchi chinensis. These trees, with their extensive canopy coverage, block a substantial amount of direct sunlight and, influenced by the downwind direction, provide an optimal travel environment for teachers and students. These observations are consistent with the research by Xue Sihan et al. [57].

Finally, we synthesized temperature and humidity data collected during four midday intervals and calculated their averages for layer overlay, which facilitated the creation of a summer campus roadmap tailored to the travel needs at various campus entrances and exits. As depicted in Figure 10, this comprehensive assessment yielded a map delineating optimal routes for summer travel. Route A, extending from the East to the West gate, offers the highest comfort due to the presence of tall trees, predominantly Mangifera indica and Litchi chinensis, with branching points at approximately 1.5–3 m high. These trees feature wide crowns and dense foliage, providing continuous, extensive shade, thus making it the most favorable route for summer. Road B, lined with Magnolia grandiflora, offers substantial shade and ranks as the secondary option for summer travel. Road C, flanked by trees with lower Sky View Factor (SVF) values, offers enhanced shade in certain sections, positioning it as a viable alternative route. Conversely, the route through the south gate suffers from lower comfort levels due to sparse tree coverage, necessitating an increase in street tree density for improved comfort. In summary, Figure 6 highlights that Roads A, B, and C are preferred for their higher comfort indices, whereas other campus roads are less recommended for summer travel.

4. Discussion

4.1. Street Trees and Campus Road Thermal Comfort

By selecting eight typical street trees along campus roads for onsite measurements and evaluating their cooling intensity, humidifying strength, and thermal comfort, we discovered that different types of trees vary in their effectiveness at improving thermal comfort. Notably, Falcataria falcata, Mangifera indica, and Litchi chinensis demonstrated superior cooling and humidifying effects, characterized by their broad canopies and extensive shade areas, which are particularly effective in enhancing the microclimatic conditions along pedestrian pathways. This underscores the importance of strategic tree selection in urban green spaces. Urban planners are advised to prioritize tall trees with high bifurcation points, such as Cinnamomum camphora, Magnolia grandiflora, and Mangifera indica to optimize thermal comfort. It is also beneficial to intersperse the planting of larger and smaller trees, such as Sophora xanthoantha C.Y. Ma, Ficus concinna, and Bombax ceiba L., which can quickly enhance the microclimatic environment for commuting while also satisfying people’s aesthetic demands for rich and diverse greenery [71]. Furthermore, the study confirmed a strong linear relationship between the Sky View Factor (SVF) and cooling intensity, humidifying strength, and thermal comfort. We found that street trees with a lower SVF often have higher leaf area indices, closure rates, and shading rates, indicating that SVF can effectively replace leaf area index, closure rate, and crown width as indicators for assessing the thermal comfort of street trees. This also suggests that SVF can effectively serve as a substitute for more complex vegetation parameters in urban planning. This is consistent with the research findings of Lau [72], Jendritzky [63] and Matzarakis [64]. It is important to note that different tree species exhibit varying degrees of adaptability to high temperatures and drought stress, which may become more pronounced as climate change progresses. Selecting species that not only provide effective shading and cooling but also adapt to changing climatic conditions becomes crucial. Future urban planning should consider the resilience characteristics of tree species to ensure long-term sustainability.

4.2. Street Trees and Urban Planning

The insights gained from this study are crucial for the practical application of urban planning and landscape design and the development of related policies. By demonstrating the direct benefits of typical street tree species on thermal comfort, this research provides a rationale for municipal policies that advocate for the strategic planting of beneficial tree species in urban areas. Local governments and urban planners can use these findings to draft guidelines that not only enhance pedestrian comfort but also contribute to broader sustainable development goals, such as energy conservation and heat island mitigation. Future planning practices should consider their adaptability to local conditions and maintenance capabilities, including considerations of management, ecological balance, and the economic and social costs [73]. According to our research, species such as Falcataria falcata and Mangifera indica, which have shown superior cooling and humidifying effects, should be considered for planting in areas requiring extensive shading (e.g., playgrounds, sports activity areas, rest areas, and pathways). This clustering not only provides continuous shade but also creates a cool microclimate that can significantly enhance the comfort of park users. Due to their rapid growth rate, regular management, pruning, and maintenance are necessary to maintain their shape and ensure safety. In terms of ecological balance, we emphasize the selection of local species wherever possible to minimize the ecological footprint and enhance the resilience of urban green spaces, considering their adaptability to local conditions and maintenance capabilities. Regarding economic and social costs, it is emphasized that urban planners need to consider long-term costs associated with watering, pruning, pest control, and potential alternatives. We suggest that municipal budgets take these costs into account to ensure the sustainability of urban greening projects.

4.3. Campus Road Thermal Comfort Assessment

We focused on using actual measured data to generate more reliable, spatially continuous data in order to better estimate the microclimatic spatial distribution within the campus environment. We also employed geographic mapping methods to delineate the optimal travel routes for summer. This study utilizes sophisticated geographic mapping techniques to transform complex and abstract climate data into clear, concise color maps, thus enhancing the scientific foundation for urban design and planning based on thermal comfort. This approach is in line with the research expectations of Yuan S et al. [52]. Centered around Fujian Agriculture and Forestry University, this study offers detailed insights into how typical street trees contribute to thermal comfort in a subtropical urban environment. While these results are directly applicable to similar educational campus settings or urban areas within comparable climatic zones, applying these findings to a broader urban context necessitates adjustments in research methodologies and considerations. As this study has shown, the effectiveness of street trees in modifying microclimatic conditions is often influenced by local regional climate conditions, soil types, and urban planning standards. Therefore, to enhance the applicability of these findings, future research should incorporate a more diverse range of tree species. This could improve the generalizability of the results and include comparative analyses across different climatic zones and urban structures, aiding in understanding how variable environmental, biological, and urban characteristics impact the role of street trees in thermal comfort.

4.4. Applicability of Combining ENVI-Met with ArcGIS

Although ArcGIS effectively integrates various data types and sources, the accuracy of the output largely depends on the precision of the input data. Inaccuracies in spatial data, resolution mismatches between different data layers, and errors in data processing steps can all lead to inaccuracies in the final visualizations and analyses. However, data simulated by ENVI-met feature consistent resolution and uniform spatial coordinates, which, to some extent, ensure the scientific validity and usability of the research results. Consequently, urban planners and landscape designers can utilize the methods developed in this study, such as the spatial analysis of microclimate simulations combining ENVI-met and ArcGIS, to assess the suitability of various tree species for specific urban environments beyond educational campuses. By adapting these tools to encompass urban residential, commercial, and industrial areas, it becomes possible to more effectively realize the potential for creating more livable and cooler urban environments through strategic tree planting.

5. Conclusions

Through empirical measurements and the simulation modeling of campus street trees, we derived several key conclusions, which are as follows:

(1) Street trees markedly enhance the thermal comfort of campus pedestrians, with different species exhibiting varied efficacies in improving comfort levels; the ranking of species based on their cooling effects is as follows: Falcataria falcata, Mangifera indica, Litchi chinensis, Magnolia grandiflora, Michelia × alba, Syzygium hainanense, Bauhinia purpurea, and Amygdalus persica; similarly, their effectiveness in humidification follows this order: Falcataria falcata, Mangifera indica, Litchi chinensis, Michelia × alba, Syzygium hainanense, Magnolia grandiflora, Bauhinia purpurea, and Amygdalus persica;

(2) The contribution of street trees to the comfort of campus roadway travel is predominantly influenced by sky view factors (SVF), which exhibit a negative correlation with cooling and humidification intensity and a positive correlation with thermal comfort;

(3) In landscape microclimate research, ENVI-met numerical simulation software is considered a crucial tool for urban landscape microclimate design [63]; this study reaffirms the usability of ENVI-met software in campus microclimate research and integrates ArcGIS technology in the planning and design of campus roads, resulting in a heat comfort-based summer travel guide map that enhances the comprehensiveness, detail, and practicality of campus road studies; moreover, as urban street vitality and the night market economy continue to expand, travel comfort is becoming an increasingly important factor affecting urban vitality and transportation convenience; the road travel route selection developed in this study offers significant reference value, and with ongoing improvements in data collection, the method could be extended to a broader scope and is anticipated to be applied in larger urban contexts;

Additionally, the comfort evaluation model proposed in this article requires further refinement: (1) The ENVI-met model does not account for the impact of vegetation structures beyond street trees on road comfort, nor assess how variations in input parameters, such as tree density and leaf area index, might influence the outcomes of thermal comfort assessments; future research should address these limitations; (2) ENVI-met is constrained by computer hardware and the finite resolution of the grid system, posing challenges in achieving optimal accuracy; (3) the current study is limited to the campus setting during midday in summer and lacks examples from different locations and times; future studies should establish more detailed in-situ models at various types of research sites to render the results more scientific and comprehensive.