The Influence of the Spatial Morphology of Township Streets on Summer Microclimate and Thermal Comfort

Abstract

Slow progress has been made on the study of thermal comfort studies in rural streets. The street construction lacks a corresponding theoretical basis, and the difference between city streets and township streets leads to the situation that the increased focus on improving the thermal comfort of city streets has not been effectively transferred to township construction. Therefore, this paper takes Huilongba Village as the research object, researching the mechanisms by which the spatial pattern of township streets influences the microclimate. This paper defines the spatial morphology of township streets by three indexes: the street aspect ratio, building density, and staggered arrangement of buildings. Additionally, it analyzes the microclimate influences of spatial morphology changes on township streets, verifies the validity of the ENVI-met model through field measurements, and designs a three-factor orthogonal experiment. With the help of software simulation, allowing for an investigation of the effects of indicators and their interactions on pedestrian thermal comfort, the optimal street spatial pattern construction scheme is proposed. The results show that the greater the density of street buildings, the more obvious the cooling effect and the better the comfort; in the staggered arrangement of buildings, the higher the high point of the building is to the south, the lower the overall temperature of the street and the better the cooling effect; and the larger the aspect ratio of the street, the better the cooling effect. Through orthogonal test and ANOVA, we can obtain the relationship between the contribution of each index to air temperature and the Universal Thermal Climate Index (UTCI) as street aspect ratio > building density > staggered building arrangement, and the overall thermal comfort of the street is the best when the aspect ratio of the street building is 1.5, the density of the building is 100%, and the south side of the building is higher. This study can provide a basis for rural street construction and thermal comfort retrofitting.

1. Introduction

With the acceleration of urbanization, the global climate issue has received widespread attention and has become a great challenge for mankind in the 21st century, with microclimate issues attracting increasing attention [1]. At the same time, the heat island effect and global warming are further aggravated, and the spatial morphology of township streets in many areas can no longer adapt to the development needs of modern life [2]. Studies show that by 2020, Earth’s average temperature will have increased by 1.1 °C compared to the end of the 19th century, and that the past decade (2011–2020) has been the hottest on record [3]. The above problems lead to the higher energy consumption of air-conditioning in summer and lower human thermal comfort, and even affect the health of individuals [4].

China has the largest number of townships in the world, especially in the western region; communes account for 42.4 per cent [5] and the relationship between spatial patterns and microclimate in townships is the most neglected. Studies have indicated that the poor thermal comfort of outdoor spaces in townships reduces the willingness of residents to engage in outdoor activities [6]. Under the basic national policy of rural revitalization, there is a great need to create more comfortable outdoor environments through the study of the rural microclimate [7].

Currently, microclimate research focuses on urban areas, with relatively little attention paid to rural areas [8], In the past decades, scholars have conducted a lot of research on urban street greening. The layout of vegetation and the coverage of vegetation are considered to be classic representatives of street greening [9,10], and research has shown that trees improve the thermal comfort near ground level by increasing the area of shading and reducing the influence of solar and ground radiation [11,12], thereby reducing near-surface temperatures and mitigating the heat island effect [13]. In addition, vegetation significantly reduces the daytime near-surface air temperature and wind speed and increases humidity [14,15]. Due to the limited amount of green space in the streets, adjusting the spatial morphological characteristics of the streets becomes the key to improving the thermal comfort of the streets [16,17]. Aspect ratios, building density, and building layout have been identified as the main spatial morphological features that affect thermal comfort in streets [18]. Adjusting the aspect ratio is commonly used to improve thermal comfort, and it has been documented that increasing the aspect ratio provides a larger area of shading at street level and reduces the impact of solar radiation on air temperature [19,20]. However, higher aspect ratios do not always result in better cooling, and when the aspect ratio is raised above 3.0, not only does the improvement in air temperature become less effective, but it also affects pedestrians’ perception of the street [21]. Building density is an indicator of the sparseness of a street’s one-sided interface [22]. Differences in the building density can affect the shading area of the street and the direction and speed of the hot air flowing through the street, changing the wind and heat environment of the street [23]. A research study has analyzed the comfort levels of three commonly used public spaces in townships [24,25]. Thermal comfort was found to be better in street spaces than in courtyard spaces and slightly lower than in square spaces, with the spatial form of the township having a greater impact on the thermal comfort [26]. In addition, the researcher found that in non-high-rise buildings, the advantages of the row layout, in terms of temperature and thermal comfort, were greater than those of the closed layout, using different building layouts in the simulation [27]. The building staggered arrangement is an important part of the building layout, reflecting the regularity of the street building ups and downs. The relevant literature shows that, through using an incremental or staggered approach to the building layout, you can use the air pressure difference formed by the differences in the heights of the different buildings to guide the movement of airflow and improve the thermal environment [28]. Furthermore, some scholars believe that changes in the way buildings are staggered will not only affect the spatial development of the city in the vertical direction, but also affect the zigzag pattern of the city skyline [29].

Based on the above studies, the research on the spatial morphological characteristics of streets focuses on city streets, and there is little mention of township streets, resulting in the fact that the excessive focus on thermal comfort improvements in city streets has not been effectively transferred to township construction; unreasonable construction on township streets creates far greater thermal comfort problems than on city streets. In addition, the past papers mostly studied the influence of each factor on thermal comfort individually, and fewer articles have examined the strengths and weaknesses of the effects of each factor in the presence of multiple factors. Under China’s basic national policy of “revitalization of the countryside”, the construction of modernized townships is a necessary way forward, and the optimization of the thermal comfort of the streets in the townships should also be inevitable. Therefore, this paper takes the township roads in Beibei District, Chongqing as an example, adopts the combination of on-site measurement and ENVI-met simulation, uses software simulation to study the influence of three street spatial morphology indexes, namely, building staggered arrangement, building density, and aspect ratio, on the street thermal environment of the township, and uses the orthogonal test and analysis of variance (ANOVA) to study the contribution of the building staggered arrangement, building density, and aspect ratio to the microclimate of the township. We also provide improvement strategies based on the results of the study, and we hope that this will provide ideas for comfort-retrofitting traditional township streets in our region.

2. Methodology

2.1. Validation of the Actual Model

2.1.1. Study Site

Located in western China, Chongqing is situated at 29°49′ north latitude and 106°24′ east longitude. It is the place where the two rivers, Jialing River and Yangtze River, converge, and with a three-dimensional transportation network of waterways and airways, it is an important transportation hub in China, a tourist leisure center, and a financial service center in the southwest region. Chongqing has a hot and humid subtropical monsoon climate with a long summer season (i.e., May to September), which is hot and humid with prevailing south-west winds. In the hot and humid summer, the maximum daily temperature can reach 40 °C and relative humidity is often more than 90%, with one of China’s four major fire pits called [30].

In this study, the township roads in Beibei District, Chongqing Municipality were selected as the research object, as shown in Figure 1. Beibei District belongs to one of the main urban areas of Chongqing, but it is located at the edge of the urban area, belonging to the urban area and the township development junction. Most of the townships in the region along the highway are established, the buildings on both sides of the road are relatively old, the overall layout is neat, the adjoining buildings have a spacing, and some of the buildings are larger than the difference between most of them, and have an open courtyard. Building heights are approximately 12 m, street widths are approximately 14 m, the aspect ratio is approximately 0.85, the building density is approximately 87%, and a little planting exists along the road, a layout that is common in western townships.

2.1.2. On-Site Measurement

The weather at the study site has been followed for a long time before the field measurements; the day that is closest to the average values of temperature, daily comparisons, relative humidity, and solar radiation illuminance of the hottest month of a typical meteorological year in Beibei District has been selected as a typical meteorological day of the summer season, i.e., the date of the field measurements. The field measurement date was 22 May 2024, a typical summer meteorological day in Chongqing; field measurements were taken on the village streets and field measurements were collected to validate the simulation model. The first image on the left side of Figure 1 shows a section of the street area in Huilongba village; this street is representative of the township streets in the southwest area and was selected for validation; such streets are characterized by a street aspect ratio between 0.5 and 1, mostly north-south oriented streets, low green coverage, roads made of asphalt, and houses and front yards made of concrete.

In order to accurately study the effect of changes in the spatial pattern of streets on the thermal comfort of villagers, the measurement time was chosen to be during the villagers’ daytime activities, i.e., the data measurements were continuously conducted from 8:00 to 20:00, and the equipment was adjusted before the measurements to minimize the errors; the day of the measurements was cloudy, with the maximum sun at noon, and a slight wind. Measurement items were air temperature (AT) and relative humidity (RH), with air temperature and relative humidity data measured by the thermal temperature and a humidity meter (model: THM-01; air temperature accuracy: ±0.3 °C; relative humidity accuracy: ±2.0%). Using the air temperature and relative humidity data to verify the reliability and stability of the ENVI-met (V5.6.1) software is a commonly used method of outdoor thermal environment research [31]. Because the street area is small, only two real points were selected; in order to test the accuracy of numerical simulation in different environments, the open concrete pavement next to the road was selected as point 1, and the pavement at the base of the tree was selected as point 2, as shown in Figure 2. In order to monitor the thermal environment at the pedestrian level, all measurements were made at 1.5 m above the ground, and the temperature and humidity sensors are protected from solar radiation by a protective cover. Measurements were taken every half hour, with a single measurement lasting 5 min, and the average value over the 5-min period was taken as the measurement value to ensure the stability of the measurement results.

2.1.3. Actual Model Simulation Scenarios

ENVI-met is a dynamic numerical simulation software designed by German scholars Bruse et al. in the 1990s based on the relevant theories of fluid dynamics, thermodynamics, and urban meteorology to study the impact of urban microclimate on thermal comfort, and it has been widely used in urban environmental design and the assessment of the impact of urban planning programs on air temperature, relative humidity, and solar radiation in the region. The ENVI-met model can be used to simulate the regional microclimate, and the simulated data can be obtained by the software, which can then be used to study the effects of greenery, building spatial morphology, materials, etc., on thermal comfort at spatial scales of 0.5 m and above, and time scales of 1–5 s. The ENVI-met model can be used to simulate the regional microclimate, and the simulated data can be obtained by the software.

This study is based on ENVI-met V5.6.1 and modelled based on the actual situation of the street as well as the pictures taken by the unmanned aerial vehicle as the base map. The size of the model area is 180 m × 120 m × 30 m, the cell grid size is 1 m × 1 m × 1 m, the upper boundary of the model is set to be greater than twice the height of the tallest building, and 3–5 layers of grids are nested in the boundary of the model to ensure the accuracy of the model in the ENVI- met simulation process. Based on the field research data and according to the modelling approach of ENVI-met, the measured model was established as shown in Figure 2.

According to the selected measurement date, the simulation time is set as 8:00–20:00 on 22 May 2024. Based on the geographic and climatic conditions of the study area, the geographic information, meteorological parameters, and simulation control parameters are set up; in this case, the initial values of macro-meteorological parameters are quoted from the nearest meteorological station to the measurement location, i.e., the meteorological station in Beibei District. Because the simulation is prone to large numerical errors at the beginning of the simulation, in order to reduce the impact of unstable data on the experiment, the simulation start time was carried out 3 h earlier, i.e., it started at 05:00 and continued until the end of 20:00. The simulation length was 15 h, and finally, the temperature and humidity simulation data from 8:00 to 20:00 were extracted and analyzed with the measured data by extracting the measured temperature and humidity data of each set point and calculating the simulated value of temperature and humidity data of each set point through the software and adding the measured and simulated values into the formula for error calculation. If the calculated error meets the requirements, the ENVI-met simulation method of the reliability and stability of the needs of this paper can be satisfied.

2.2. Optimization of Model Design Solutions

This paper investigates the spatial morphology of streets on microclimate influences, and through preliminary research, it is found that the composition of the streets in Beibei District is relatively homogeneous. The spatial morphology skeleton consists of a main highway running through a building complex and a parallel layout of buildings on both sides, with the main body presenting the Chinese character “three” layout, i.e., the main highway and the buildings on the left and right sides are in three parallel lines, there are small alleys between buildings, and the overall layout is in the shape of a fish-bone. In this case, the comprehensive modeling of street spatial morphology in Beibei District can be achieved by using aspect ratio (A), building density (D), and the staggered arrangement of buildings (S). In order to more accurately study the influence of street spatial morphology characteristics on thermal comfort, we extracted a simplified model of township street morphology to highlight the morphological characteristics, controlled the secondary influencing factors, and established different street spatial morphology models for simulation studies by changing the street height-to-width ratio, the staggered arrangement of buildings, and the density of buildings, taking into account the spatial morphology indicators of the streets only.

We change the aspect ratio by increasing and decreasing the width of the street while keeping the building height constant; we change the building density by changing the proportion of buildings to the total length of the street while keeping the aspect ratio and the building density constant; and we change the building staggering by changing the location of the high points of the buildings on the street as well as the undulation pattern while keeping the aspect ratio and the building density constant. Based on the field research data, the parameters of aspect ratio (A), building density (D), and staggered building arrangement (S) are set as shown in

Table 1. Orthogonal factor setting.

Project	Factors
	A	S	D
1	0.5	The highest on the north side	100%
2	1	The highest on the south side	87%
3	1.5	The highest in the central region	75%

. In order to reflect the improvement effect of street spatial morphology, we will establish the original street model (D2) with an aspect ratio of 0.85, a consistent building height, and a building density of 87% as a comparison based on the actual street situation.

Simulation studies under the thermal environment conditions created by meteorological parameters on a typical weather day to propose optimized design solutions is a common methodology used in current outdoor thermal environment studies [32]. In this paper, the optimization simulation study is based on the meteorological parameters of typical summer weather days in Beibei District, Chongqing, to explore the impact of various indicators on the summer thermal environment of the street.

2.3. Data Analysis Methods

1. Model validation was performed using ENVI-met software and measuring instruments, and using three indicators, the root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination of the fit (R²) to test the accuracy of the measured and simulated data. R² is used to measure the linear correlation between the simulated and measured values; values closer to 1 indicate that the correlation between the two is higher. MAPE is defined as the sum of the absolute difference between the simulated and measured values, and RMSE is defined as the square root of the sum of the squares of the differences between the simulated and measured values, which reflects the size of the error between the predicted values and the actual values. The simulation has a RMSE and MAPE are less than 10%, indicating that the simulation has a high accuracy; the closer the value is to 0, the smaller the error between the simulated value and the measured value, and the more accurate the model is.

$$\mathrm{M}\mathrm{A}\mathrm{P}\mathrm{E}={\displaystyle \frac{{\sum }_{i=1}^N|S_i-O_i|}{N{\times O}_i}}\times 100\mathrm{\%}$$

(1)

$$\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}=\sqrt{{\displaystyle \frac{{\sum }_{i=1}^N{(S_i-O_i)}^2}{N}}}$$

(2)

$$R^2={\displaystyle \frac{{\sum }_{i=1}^N{(S_i-\overline{O})}^2}{{\sum }_{i=1}^N{(O_i-\overline{O})}^2}}$$

(3)

Note: N is the total amount of data, $S_i$ is the simulated value, $O_i$ is the measured value, and $\overline{O}$ is the mean of the measured values.

2. Orthogonal experimental design: In order to study the influence of the contribution of each factor to the microclimate, the street microclimate of each orthogonal test item was simulated by ENVI-met software using a three-factor, three-level orthogonal test with three-factor analysis of variance (ANOVA). If a full simulation is carried out, 3³ (27) simulations are required, which will consume a lot of manpower, material, and time. An orthogonal experimental design, according to the orthogonal table for design and analysis, can reduce the number of simulations and is easy to operate. It is a kind of incomplete experimental design, which combines the contents of the study, and only nine simulations are needed to draw conclusions. The staggered building arrangement method, aspect ratio, and building density were designed as factors S, A, and D, respectively, and each factor was designed as three levels. Orthogonal design analysis was carried out using SPSSAU (V24.0) software, and orthogonal experimental tables were designed as shown in Table 1 and

Table 2. Orthogonal combination setting.

Number	A	S	D	Composition
1	1	1	1	A1S1D1
2	1	2	3	A1S2D3
3	1	3	2	A1S3D2
4	2	1	3	A2S1D3
5	2	2	2	A2S2D2
6	2	3	1	A2S3D1
7	3	1	2	A3S1D2
8	3	2	1	A3S2D1
9	3	3	3	A3S3D3

below.

3. Thermal comfort indicators: A number of biometeorological indicators have been developed to describe human thermal comfort by linking local microclimatic conditions to human thermal sensations. Such indicators are usually based on steady-state models, which are based on the assumption that people’s exposure to the surrounding climate environment allows them to reach thermal equilibrium over time, and they provide numerical solutions to the energy balance equations that control temperature regulation. Since the human body heat load is always under change in the real outdoor environment, the steady state human heat balance model cannot accurately describe the process. In 1990, under the initiative of the president of the International Society of Biometeorology, scholars from 23 countries, after comparing many models, chose to establish the Universal Thermal Climate Index (UTCI) based on the Multi-node Human Physiology and Comfort Model (MNHPCM). The UTCI fully considers the heat exchange and thermal adaptation process between the human body and the outdoor environment, and it can accurately evaluate slight changes in the external environment under a variety of climatic zones and meteorological conditions. The UTCI generally categorizes the human body’s thermal comfort level into 10 levels, as shown in

Table 3. UTCI rating scale.

UTCI Class Name	UTCI Temperature Range
Extremely hot and uncomfortable	UTCI ≥ 46 °C
Very hot and uncomfortable	46 °C > UTCI ≥ 38 °C
Hot and uncomfortable	38 °C > UTCI ≥ 32 °C
Hotter and uncomfortable	32 °C > UTCI ≥ 26 °C
Comfortable	26 °C > UTCI ≥ 9 °C
Cold	9 °C > UTCI ≥ 0 °C
Colder and uncomfortable	0 °C > UTCI ≥ −13 °C
Cold and uncomfortable	−13 °C > UTCI ≥ −27 °C
Very cold and uncomfortable	−27 °C > UTCI ≥ −40 °C
Extremely cold and uncomfortable	UTCI < −40 °C

below [33].

3. Results

3.1. Validation of the Accuracy of the Simulation Method

Comparing the measured values of air temperature and relative humidity at the measurement point every hour with the simulated values of the software, as shown in Figure 3, and qualitatively evaluating the degree of coincidence of the curves, it can be seen that the trend of the curves is basically the same, and in comparison, the measured results, due to the influence of various real-life factors, such as crowd activities, have a higher overall value and stronger fluctuation than the simulated results. The linear fitting of the simulated values to the measured values shows that the fitted coefficients of determination, R², for air temperature and relative humidity at measurement point one are 0.951 and 0.858, respectively; the fitted coefficients of determination, R², for air temperature and relative humidity at measurement point two are 0.891 and 0.771, respectively, and the validity of the simulation results can be seen from the values of the coefficients of determination, R², at the two measurement points. The results of theRMES and MAPE calculations are shown in

Table 4. Coefficient values for simulation accuracy judgement.

High Degree	Projects	Designation	Measurement Point I	Measurement Point II
1.5 m	Air temperature	RMSE (%)	1.51%	1%
		MAPE (%)	4.26%	2.81%
	Relative humidity	RMSE (%)	4.16%	4.87%
		MAPE (%)	6.88%	8.30%

. There will be differences between the measured values and the simulated values when comparing but the overall error is within the acceptable range, and the errors are all in line with the corresponding value standards, indicating that the ENVI-met software constructs a physical model, which can be a better response to the actual microclimate of the street, and the results are valid.

3.2. Effects of Building Density on Microclimate

3.2.1. Air Temperature

The cooling curve is shown in Figure 4a. From the figure, we can obtain, with the reduction of building density, that the average temperature of the street has a trend to rise. The cooling effect of scenario D1 and D3 with the change of time are showing a trend of first, rise and then, fall about 15:00 to reach the maximum value. The maximum decrease in temperature is 0.17 °C for the D1 scenario and 0.47 °C for the D3 scenario, indicating that the change in building density affects the shading area of the street in a certain period of time, which, in turn, affects the temperature of the street. The lower the density of the building, the higher the temperature is, and this pattern is most obvious at 10:00 and near 15:00. In addition, we found that in the 11:00–13:00 time period, the temperature difference curve will be out of the reverse downward trend; this is due to the direct sun. The building density of the change in the shading gap will be reduced to a minimum, at this time, due to the southwesterly wind direction. The reduction in the building density will produce a larger wind channel, which is conducive to the entry of the lateral winds and the cooling effect on the street. The lower the density, the better the cooling effect.

3.2.2. UTCI Index

The UTCI cloud is shown in Figure 5 below, where changing the building density results in a greater change in the UTCI of the street environment, with a decrease in building density resulting in an increase in high-temperature areas of the street, and high-temperature areas at low building densities occupying a greater overall extent of the street compared to high building densities. The UTCI curves are shown in Figure 5d and it can be shown that as the building density decreases, the environmental UTCI curve shifts significantly upward and the street thermal comfort decreases. All three scenarios show a trend of increasing and then decreasing over time, reaching a maximum near 13:00, with the average UTCI value approaching 46 degrees. According to the trend of the UTCI curve, the D3 scenario with the highest building density enters the “very hot and uncomfortable” (38 °C < UTCI < 46 °C) zone later, leaves the “very hot and uncomfortable” zone earlier, and stays in the higher temperature zone for a shorter period of time. The D1 scenario, with its lower building density, is the earliest to enter and the latest to exit the “very hot and uncomfortable” zone, and has been in this zone for the longest period of time. Combining the three curves suggests that a reduction in building density would be detrimental to the travel experience of residents.

3.3. Effects of Staggered Building Arrangement on Microclimate

3.3.1. Air Temperature

The cooling curves are shown in Figure 4b, from which it can be seen that the average temperature near the ground level of the street is affected by the different staggered arrangements of buildings in the order of higher buildings on the south side of the street > higher buildings in the middle of the street > higher buildings on the north side of the street. As the staggered arrangement of buildings affects the shading area and concentration of shading on the street, which, in turn, affects the mean street temperature, the cooling effect of scenarios S1, S2, S3 with the change in time are showing a trend of first, rise and then, fall about 13:00 to reach the maximum value. The S1 scenario temperature will make the street temperature appear to rise, the S2 scenario temperature has a maximum reduction of 0.45 °C, and the S3 scenario temperature has a maximum reduction of 0.31 °C. This suggests that a higher building on the south side and a taller building in the center results in a better cooling effect due to the enhancement of the shading area compared to a building of the same height. As the taller buildings are on the north side, due to the angle of the sun, compared to buildings of the consistent height, the reduction of shading area will instead lead to an increase in temperature.

3.3.2. UTCI Index

The UTCI cloud map, shown in Figure 6 below, shows that changing the staggered building arrangement affects the UTCI of the street environment, with taller buildings on the north side of the street compared to the south side of the street, and the midday street minimum and maximum temperatures decreasing from 38.09 °C and 49.2 °C to 36.99 °C and 48.91 °C, respectively. As the highest point of the street building changes from the south side to the north side, the hotter areas of the street increase and thermal comfort decreases. The UTCI curves are shown in Figure 6d, and as the highest point of the building moves from north to south, the UTCI curves move significantly upward, with all three scenarios showing a trend of increasing and then, decreasing over time, with a maximum near 13:00–14:00 and a UTCI average approaching 46 degrees. As can be seen from the UTCI curves, the UTCI is ranked from high to low as higher on the south side of the building > higher on the center side of the building > higher on the north side of the building. The higher S2 scenario on the south side of the building enters the “very hot and uncomfortable” (38 °C < UTCI < 46 °C) zone later, leaves the “very hot and uncomfortable” zone earlier, and is at higher temperatures for a shorter period of time. The higher S1 scenario on the north side of the building is the earliest to enter and the latest to exit the “very hot and uncomfortable” zone, and it stays in this zone for the longest period of time. Combining the three curves suggests that building the high point to the south is friendlier to residential travel.

3.4. Effects of Aspect Ratio on Microclimate

3.4.1. Air Temperature

The cooling curve is shown in Figure 7a, from which it can be seen that the building’s ability to shade the street in the morning and evening hours decreases as the aspect ratio decreases. The mean temperature of the street has an increasing trend, and the cooling effect over time shows an increasing and then, decreasing trend. The maximum value was reached at about 15:00, in which scenario A1 had a warming trend to the street, with a maximum value of 0.51 °C, and scenarios A2 and A3 both had a cooling trend, with a maximum drop of 0.26 °C and 0.83 °C, respectively. It is obvious that the increase in aspect ratio has a significant effect on the thermal comfort of the street. In addition, the cooling curve tends to decrease at midday, reaching a low value around 12:00–13:00, which is attributed to the fact that the increase in the aspect ratio increases the shading area of the building on the street. In consequence, the average temperature of the street will be reduced, and in the midday, due to direct sunlight, the difference in shading area brought about by the difference in building height and aspect ratios will be extremely minimized, and, therefore, the temperature gap between the various scenarios and the original street will be narrowed down.

3.4.2. UTCI Index

The UTCI cloud is shown in Figure 8, and changing the street aspect ratio can have a large effect on the street UTCI. As the aspect ratio increases from scenarios A1 to A3, the midday street minimum and maximum temperatures decrease from 37.09 °C and 48.38 °C to 36.93 °C and 46.84 °C, respectively, and as the aspect ratio increases, the high temperature area of the street decreases and the thermal comfort of the street decreases. The UTCI curves are shown in Figure 8d. As the aspect ratio decreases, the UTCI curves shift upward significantly, and all three scenarios show a tendency of increasing and then, decreasing with time, reaching a maximum value near 13:00–14:00. The temperature difference between the scenarios is small, near 12:00, fitting the previous analysis, and reaching a relatively large difference near 10:00 as well as 15:00, indicating that the effect of the aspect ratio on thermal comfort in the street peaks at this time. The A3 scenario with a larger aspect ratio enters the “very hot and uncomfortable” (38 °C < UTCI < 46 °C) zone later, leaves the “very hot and uncomfortable” zone earlier, and spends a shorter period at higher temperatures. The A1 scenario with a smaller aspect ratio was the first to enter and the last to exit the “very hot and uncomfortable” zone, and stayed in this zone for the longest period of time. Combining the three curves suggests that taller building aspect ratios are friendlier to residential transportation.

4. Optimized Design and Discussion

Microclimate simulations were conducted according to the orthogonal experimental design, and nine groups of time-dependent simulation data were obtained. Considering the range of time during the day when people are active, the air temperature and UTCI will be extracted from morning (8:00–10:00), midday (12:00–15:00), and evening (17:00–20:00) time intervals to analyze the effect of each spatial morphology index on thermal comfort.

The most disadvantageous scenario, the midday UTCI, was taken as the dependent variable, and the effect of each factor, A, D, and S, in UTCI was tested for significance by three-way analysis of variance (ANOVA), as shown in

Table 5. Effect of midday factors on UTCI.

Source of Variation	Type III Sum of Squares	Df	Mean Square	F	p
D	1.491	2	0.745	80.626	0.012 *
S	0.422	2	0.211	22.851	0.042 *
A	10.074	2	5.037	544.847	0.002 **
Residual	0.018	2	0.009

Note: * denotes p < 0.05, significant (two-sided); ** denotes p < 0.01, highly significant (two-sided).

. The primary effect of each factor exists and the significance is relatively powerful, although there are some differences in the significance of the effect between factors, including that the aspect ratio (A) has the most obvious effect towards the UTCI. From Table 5, we can get the degree of influence of each factor over UTCI as A > D > S. Through the post hoc multiple comparison analysis, the mean value of each water of the factor is obtained as shown in Figure 7b, from which it can be seen that A3, D1, and S3 have the smallest mean values inside their respective factors, respectively, so that when UTCI is used as the effect indicator, the best combination of spatial morphology indicators is A3D1S2. In the same way, using the midday air temperature as the dependent variable, the effect of each of the factors, A, D, and S, towards air temperature was tested for significance by three-way analysis of variance (ANOVA). As shown in

Table 6. Effect of midday factors on air temperatures.

Source of Variation	Type III Sum of Squares	Df	Mean Square	F	p
D	0.227	2	0.113	16.019	0.059
S	0.148	2	0.074	10.443	0.087
A	0.534	2	0.267	37.725	0.026 *
Residual	0.014	2	0.007

Note: * denotes p < 0.05, significant (two-sided).

, similar to the above, the significance of the aspect ratio is still more obvious, and the remaining two are slightly weaker. Based on the F-value, it can be obtained that the degree of influence of each factor on the air temperature is still in the rank order of A > D > S. Also, from Figure 7b of the average values of each water of the factors, the best combination of spatial morphology indicators is A3D1S2, confirmed by post hoc multiple comparisons analysis with air temperature as the influence indicator.

The contribution is obtained as the ratio of the sum of squares of each factor to the total sum of squares [34], calculated separately according to the three time periods of morning, midday, and evening. The results are shown in

Table 7. Contribution of factors to the effect of UTCI and air temperature.

Projects	UTCI 8:00–10:00	UTCI 12:00–15:00	UTCI 17:00–20:00	AT 8:00–10:00	AT 12:00–15:00	AT 17:00–20:00
A	83.20%	84%	68%	42.10%	58.70%	64%
D	14.50%	12.40%	28.50%	30%	25%	22.90%
S	2.30%	3.60%	2.50%	27.90%	16.30%	13.10%

below, and it can be seen that among the various spatial morphology indicators, the staggered arrangement of buildings contributes to the UTCI in the morning, midday, and evening with a low percentage; the contribution of the aspect ratio to the UTCI was greater in the morning versus the midday, and slightly lower in the evening. Building density contributes slightly less to the UTCI in the morning as well as the midday, and increases in the evening with a contribution of 28.5%. In contrast to the former indicator, the contribution of building density as well as building arrangement to air temperature is increased, and the contribution of the aspect ratio to air temperature is reduced to a minimum of less than 50%, much lower than the percentage of contribution of this indicator to UTCI, this suggests that the effect on air temperature is multifaceted, with each indicator having a large impact.

In summary, the influence of the spatial morphology indicators of street buildings on the thermal comfort, in order of contribution, is aspect ratio (A) > building density (D) > building staggered arrangement (S). This indicates that the aspect ratio has the greatest influence on the thermal comfort of the street, and more attention should be paid to the improvement of the aspect ratio when optimizing the design of the street, but the influence of building density and staggered arrangement of buildings on the thermal comfort should not be ignored, especially when the aim is to reduce air temperatures. Through the post hoc multiple comparison analysis, the optimal scenario is A3D1S2, when the aspect ratio is 1.5, the building density is 100%, and the staggered arrangement of buildings is higher on the south side; the pedestrian thermal comfort is optimal in this case.

5. Conclusions

This paper explores how to optimize the thermal environment of township streets from the perspective of the spatial morphology of streets. A method combining on-site research and ENVI-met numerical simulation was adopted to analyze the influence of different street spatial morphology features on the thermal comfort. Since actual street conditions occur as a combination of features, this article utilizes orthogonal tests and ANOVA to explore the optimal solutions for different combinations of spatial morphology features of streets. It was found that the aspect ratio, building density, and staggered arrangement of buildings have significant influences on the near-surface microclimate of streets, and the effect of the aspect ratio is relatively strong whether the UTCI or air temperature is taken as the influenced quantity, which indicates that aspect ratio is still important in the optimization of spatial morphology of streets in the future. However, from the study of the contribution of each spatial morphology indicator to the street microclimate influences, it was found that when air temperature was used as the influenced quantity and the UTCI was used as the influencing quantity, there was a significant decrease in the contribution of the aspect ratio to the street microclimate, and a significant increase in the contribution of the staggered building arrangement, which may be due to the greater sensitivity of the UTCI to the changes in the meteorological parameters. When the air temperature is used as the affected quantity, all three indicators have a greater impact on the street as a climate, which also proves that the building density and the staggered arrangement of buildings, as the object of study, are reasonable, and should be taken into account when optimizing the design of the street, especially when the purpose is to reduce the air temperature.

The experimental results show that when the street is oriented in the north-south direction, increasing the building density and aspect ratio, and choosing the higher southern side of the staggered building arrangement, can reduce the time in the “very hot and uncomfortable” zone and improve the comfort of the residents. In the further analysis of the interaction of the factors, the orthogonal test results show that the contribution of each factor to the thermal comfort is in the order of aspect ratio A > building density D > building staggered arrangement S. Through the analysis of multiple comparisons after the fact, the optimal spatial morphology of the street is A3D1S2.

City streets and township streets in the green coverage, building aspect ratio, building staggered arrangement of regularity has a large gap, which shows that the two rules cannot be generalized. The research content of this paper can provide a certain basis for the construction of township streets. This study also has some limitations. The research model is a common street morphology in the townships of Southwest China, where the spatial morphology of the street consists of a main highway running through the building complex and parallel layouts of buildings on both sides, with the main body showing a Chinese character “three” layout. This kind of area is generally concentrated in the townships in the plains of Southwest China; thus, the research rule is more applicable to the plains in a hot and humid climate. For the streets in areas with dense buildings and large aspect ratios, the research rule can only provide a reference, but it cannot guarantee the applicability. In the future, the microclimate influences of high building densities and different orientations on township streets will be further studied.