Impact of Environmental Factors on Summer Thermal Comfort of Ribbon Waterfront Park in Hot Summer and Cold Winter Regions: A Case Study of Hefei

Abstract

Ribbon waterfront parks in hot summer and cold winter regions play a crucial role in microclimate regulation and thermal comfort enhancement due to the combined effects of water bodies and vegetation. This study focuses on ribbon waterfront parks in Hefei. This study investigates the influence of park environmental factors (e.g., plant community characteristics, spatial configuration of water bodies, and plaza layouts) on the summer thermal environment through field measurements and ENVI-met numerical simulations. Based on field studies and a literature review, five environmental factors were selected as test variables: water body direction (S), tree planting density and arrangement (A), square distribution form (B), square location (C), and pavement material (D). Using orthogonal testing, 64 different environmental scenarios under four distinct water body orientations were designed and simulated using ENVI-met (Version 5.6.1), followed by a quantitative analysis of the simulation results. The findings reveal that the interaction between water body orientation and prevailing wind direction significantly influences the cooling efficiency in both the upwind and downwind regions. In addition, through orthogonal testing, Range Analysis (RA), and analysis of variance (ANOVA), the order of magnitude of the effect of each experimental factor on the Universal Thermal Climate Index (UTCI) can be derived: density and form of tree planting (A) > pavement material (D) > location of the square in the park (C) > forms of distribution of squares in the park (B). Finally, this study suggests various environmental factor-setting schemes for ribbon waterfront parks that are tailored to distinct microclimatic requirements. It also provides design recommendations to improve thermal comfort in parks based on the orientation of different water bodies. Furthermore, it offers specific references and foundations for planning, designing, optimising, and renovating waterfront parks of similar scales.

1. Introduction

In the dual context of urban sprawl and climate change, the urban thermal environment has become a topic of increasing concern [1,2,3]. Microclimate is a key factor influencing the use of urban outdoor spaces, as people tend to prefer comfortable microclimate conditions when engaging in outdoor activities [4,5,6]. A well-designed urban outdoor environment encourages physical activity and social interactions, thereby enhancing space utilization and promoting daily urban vitality [7,8,9,10,11]. As a crucial component of urban outdoor spaces, waterfront parks possess unique natural water resources, and urban water bodies play a vital role in regulating the thermal environment, particularly in densely populated cities [12,13]. Green and blue spaces, such as parks and waterfronts, provide numerous physical and psychological benefits, including improving health, reducing stress, and encouraging physical activity [14,15,16]. The geographical characteristics and landscape features of waterfront parks make them distinct in terms of summer thermal comfort. The synergistic effect of green spaces and water bodies helps moderate high-temperature and high-humidity conditions during summer [17,18], thereby enhancing outdoor thermal comfort and improving urban liveability. Consequently, the microclimatic advantages and limitations of waterfront parks play a significant role in urban development.

Environmental factors in parks play a crucial role in regulating microclimates and optimising human thermal comfort [18,19,20,21,22]. Numerous studies have employed thermal environment measurements and numerical simulations to investigate the impact of environmental factors on park microclimates [23,24,25,26]. For instance, vegetation [15,27,28,29,30,31], water bodies [32,33,34,35], pavement albedo [36,37,38], and other environmental elements significantly influence the thermal environment of parks. Yan et al. [39] demonstrated that green spaces can effectively improve urban microclimates and mitigate the heat island effect. Chan et al. [40] found that the configuration of parks and buildings significantly affects thermal comfort in both summer and winter, with green spaces reducing perceived temperatures by up to 20% in shaded and vegetated areas. These findings suggest that urban areas should prioritise green spaces and reduce the use of heat-absorbing materials, such as asphalt and concrete, to enhance thermal comfort and promote sustainable urban environments. Ma et al. [41] discovered that the cooling effect of vegetation implies that increasing its coverage, encompassing trees and grasses, can directly alleviate heat stress. Furthermore, it can enhance individuals’ subjective perception by offering attractive landscapes. Additionally, reducing the area of hardened surfaces is another vital method to mitigate heat stress. Karimi et al. [36] found significant correlations between the sky view factor (SVF) and the physiologically equivalent temperature (PET) only during sunlight and sunset. Combining low-albedo pavements with trees that have broad canopies and tall trunks can create optimal thermal comfort conditions. The effect of albedo on thermal comfort was lower than that of vegetation cover. Chan et al. [42] found that increasing the number of trees or reservoirs enhanced thermal comfort; however, adding more grass did not result in any improvement. Li et al. [43] found that the complexity of vegetation structure affects the thermal environment, with multi-layered green spaces providing a more comfortable thermal environment than single-layer arrangements. Shi et al. [44] demonstrated through field surveys and ENVI-met simulations that a reduction of 1.0 in the Leaf Area Index (LAI) leads to a decrease in mean air temperature by 0.19–0.31 °C, likely due to increased ventilation flow. Chen et al. [45] concluded that elements such as water bodies, trees, and buildings can effectively regulate garden environments. They proposed strategies to optimise the microclimate and improve thermal comfort in urban green spaces in the subtropical monsoon climate zone south of the Yangtze River. Lin et al. [46] discovered that enhancing green cover and canopy density in parks significantly improves outdoor thermal comfort. However, Li et al. [43] also discovered that excessively dense planted forests may reduce the cooling efficiency of vegetation.

Several scholars have investigated the effects of water and vegetation on thermal comfort in waterfront spaces [17,44]. Fei et al. [33] explored the influence of water on the microclimate of waterfront spaces through questionnaires and field measurements, revealing that the cooling effect of water is intrinsically correlated with the distance of the measurement location from the centre of the water body. Jiang et al. [47] identified urban form factors affecting thermal comfort in waterfront spaces and analysed their impacts. Through field surveys of over 40 plant communities, Li et al. [48] found that community structure is the primary factor influencing the microclimatic effects of broadleaf communities, while the distance from roads and water bodies mainly affects the microclimatic performance of coniferous forest communities. Lim et al. [49] found through their research that small-scale water features can effectively improve thermal comfort in community parks, but they must be integrated into broader urban cooling strategies to maximise their impact. Jin et al. [50] analysed the impact of water bodies on the summer microclimate of the district and applied ENVI-met software to simulate the microclimate of urban residential districts with different water body configurations, and concluded that the optimal water body configuration is essential for regulating the microclimate of residential districts in summer. Xu et al. [51] investigated the mechanisms by which urban morphological parameters, such as building density, vegetation cover, volume ratio, and sky view factor, affect air temperature in the Wuhan waterfront area. Song et al. [32] analysed the key factors affecting waterfront areas’ thermal and humid environment based on the computational fluid dynamics (CFD) method with water effect and factor analysis (building type, building volume ratio, embankment height, distance from water bodies and landscaping). The results show that the geometry of ventilation corridors in waterfront neighbourhoods is highly dependent on the scale of influence of the thermal and humid environment. Specifically, temperatures in areas with low volume ratios and large ventilation corridors are more susceptible to the influence of water bodies.

This study focuses on the waterfront park along the Nanfei River in Hefei as the research object. It selects five research factors: the direction of the water body, the density and form of tree planting, the distribution form of the square, the location of the square within the park, and the material of the pavement. By employing numerical simulation and orthogonal testing, the study investigates the effects of these five factors on the microclimate and thermal comfort during summer. The analysis is conducted from the perspectives of the water body, vegetation, pavement, and square space, utilizing analysis of variance (ANOVA) and Range Analysis (RA). The optimal combinations of these factors are derived, providing valuable guidelines for designing thermal environments in waterfront parks located in hot summer and cold winter regions.

2. Methodology

2.1. Study Area and Data Collection

Hefei is located at 117°17′ E and 31°52′ N, in a region characterised by hot summers and mild winters, which falls under the subtropical monsoon climate. The high-temperature season typically lasts from June to August, with average monthly temperatures ranging from 25 °C to 36 °C and peak temperatures reaching between 36 °C and 38 °C. During the summer, the prevailing wind direction is from the southeast. Hefei has a well-developed water system, bounded by the Jianghuai Watershed, with the Huaihe River system to the north of the ridge and the Yangtze River system to the south of the ridge, in which the Nanfang River in the Yangtze River system runs north and south through the urban area of Hefei (Figure 1), with a length of 70 km and a watershed area of 1700 square kilometres, and it is the main river channel connecting the urban area with the western half of Chaohu Lake, and the Nanfang River is an ecological corridor within the urban area of Hefei. The waterfront parks along the river are important outdoor activity spaces for citizens’ summer activities, among which the ribbon waterfront parks are the most common forms of waterfront spaces. The ribbon waterfront parks along the Nanfei River in the main urban area of Hefei are selected as the research object, and Figure 1 is the location map of the research area and the distribution map of the waterfront parks researched in the present study.

In this study, a typical sunny summer day (12 August 2023) was selected as the test day. Measurement points were arranged in accordance with the Specification for Surface Meteorological Observation (GB/T 35221-2017) [52]. Microclimate monitoring was conducted hourly from 06:00 to 21:00, during which air temperature, relative humidity, and wind speed were recorded. The instruments used included the Testo 635 thermo-hygrometer, Testo 410 anemometer, and FLIR E6 infrared camera. The range and accuracy of these instruments complied with the ISO 7726 standard [53]. The instrumentation and accuracy of the microclimate tests are detailed in

Table 1. Microclimate measurement instruments.

Measurement Items	Instrument Model	Range	Accuracy
Wind speed (Va)	Testo 410-1 Compact Vane Anemometer	0.4~20 m/s	±0.03 m/s + 5% reading ±(0.2 m/s + 2% of measured value)
Air temperature (Ta)	Testo 635-1 Thermo-Hygrometer with Air Moisture	−200~1370 °C	±0.3 °C (−60~60 °C) ±(0.2 °C + 0.5% of measured value)
Relative humidity (Rh)	Testo 635-1 Thermo-Hygrometer with Air Moisture	0~100%Rh	±2%
Surface temperature (Ts)	FLIR E6 Commercial Thermal Imaging Camera	−20 °C~550 °C	±2 °C or ±2% of reading

. To minimise systematic errors, all instruments were calibrated prior to the test. Additionally, to ensure data reliability, the measured data were compared with records from the Hefei Meteorological Station (Station ID: 58321). The Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) were calculated to evaluate deviations and assess data consistency.

2.2. ENVI-Met Simulation and Model Validation

In urban microclimate research, ENVI-met is one of the most widely used dynamic simulation tools [26,54,55,56,57,58]. Numerous case studies have validated its reliability, and it has been extensively applied to investigate the influence of various factors on urban microclimates, including vegetation [22,59,60], urban morphology [61,62], and ground surface albedo [36,63], and their effects on human thermal comfort [64,65]. This study employed ENVI-met to simulate the microclimate of a ribbon waterfront park. Compared to other simulation tools, ENVI-met demonstrates broader applicability [54,66]. However, previous studies have identified persistent limitations in model parameterization, mesh independence, and boundary condition handling, which can affect the accuracy and applicability of simulation results. These limitations necessitate further validation and optimisation [67]. For instance, Crank et al. [68] conducted a systematic evaluation of ENVI-met’s microscale model in urban thermal mitigation analysis. They examined the model’s ability to reproduce vertical mixing, heat transport, and building energy exchange, identifying limitations in handling complex geometries and actual building heights. Additionally, ENVI-met’s simplified parameterization of vegetation transpiration and water surface evaporation processes may subtly influence thermal environment simulations at heights of 0.5 to 2 m, corresponding to the human activity zone. To address these limitations, this study applied isometric stratification when configuring the vertical grid. The bottom grid was subdivided into five sub-cells, each measuring 0.4 m, to enhance the simulation accuracy of surface microclimate variations. Despite these limitations, ENVI-met remains widely recognized as an effective tool for simulating thermal environments at the park scale [69].

According to Schlünzen and Vuckovic et al. [70,71], model calibration and uncertainty analysis are essential for evaluating the accuracy of simulation results before applying ENVI-met [72]. In ENVI-met simulation studies, most researchers have validated their models using field measurements [73]. Detommaso et al. [74] reviewed multiple ENVI-met simulation studies on outdoor thermal comfort assessment and found that most incorporated field measurements, with air temperature and relative humidity serving as the primary calibration metrics for ENVI-met models [75,76]. Therefore, to evaluate the accuracy of model predictions, this study simulated the existing conditions of the waterfront park using ENVI-met and compared the simulated data with observed meteorological parameters. Following previous studies on model calibration error metrics [75,77,78], this study selected Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and the Coefficient of Determination (R²) to evaluate model accuracy. MAE and RMSE quantify the magnitude of errors, with RMSE being more sensitive to larger errors. MAPE measures the relative proportion of errors, making it suitable for comparing datasets of different scales. R² assesses the goodness-of-fit, indicating the model’s ability to explain variations in observed data and measuring the agreement between predicted and actual values. Additionally, the 95% confidence interval was used to assess the statistical uncertainty of error means, reflecting the model’s stability.

$$MAE={\displaystyle \frac{1}{n}}\sum _{i=1}^n\left|x_i-x_i^{\prime }\right|$$

(1)

$$MAPE={\displaystyle \frac{1}{n}}\sum _{i=1}^n{\displaystyle \frac{\left|x_i-x_i^{\prime }\right|}{x_i}}\times 100\mathrm{\%}$$

(2)

$$RMSE=\sqrt{{\displaystyle \frac{\sum _{i=1}^n{\left(x_i-x_i^{\prime }\right)}^2}{n}}}$$

(3)

In the formula, $x_i$ is the measured value; $x_i^{\prime }$ is the simulated value; and $n$ is the number of measurements.

Additionally, to further assess the predictive performance of the ENVI-met model, regression analyses were conducted on simulated and observed values. As shown in Figure 2, 95% confidence bands and 95% prediction bands were employed to quantify the uncertainty associated with model predictions. The 95% confidence band represents the precision of the estimated relationship between simulated and observed values, defining the expected range of the true regression line. In contrast, the 95% prediction band delineates the expected range of future observations while accounting for model variability and residual uncertainty. The regression analyses indicate that the R² values for air temperature and relative humidity are 0.94 and 0.93, respectively, confirming the high reliability of the ENVI-met model in predicting microclimatic trends and accurately capturing observed data variations. Further error analyses confirm the model’s high accuracy in simulating both temperature and relative humidity. For temperature simulation, the MAE is 0.52 °C, the MAPE is 1.65%, the RMSE is 0.63 °C, and the 95% confidence interval for mean error is [−0.70, 0.33] °C, indicating a slight underestimation bias in the model. For relative humidity simulation, the MAE is 1.56%, the MAPE is 2.34%, and the RMSE is 1.88%, with a 95% confidence interval of [−0.22, 1.62], indicating a balanced error distribution with no apparent systematic bias. Overall, the ENVI-met model’s error indicators fall within acceptable limits, meeting the accuracy requirements for microclimate simulations. The model effectively captures real-world microclimate variations, demonstrating strong applicability in urban microclimate research and environmental assessment.

2.3. Research Framework

The research framework of this study is illustrated in Figure 3 and consists of the following steps:

• Identification of key factors and experimental design: Through field research and preliminary single-factor experiments, the main environmental factors affecting the waterfront park in Hefei City and their respective levels were identified. Based on these findings, orthogonal experiments were designed to determine the experimental scenarios.
• Numerical simulation and analysis: The three-dimensional urban microclimate simulation software ENVI-met was employed to conduct numerical simulations. These simulations were used to study the degree and order of influence of different factors on summer microclimate and thermal comfort, ultimately deriving the optimal combination scenario.
• Scenario development and optimised design recommendations: Different scenarios for setting environmental factors in ribbon waterfront parks were proposed, tailored to different microclimatic demands. Additionally, optimised design suggestions were made to enhance thermal comfort in ribbon waterfront parks with varying water body orientations.

Figure 3. Research framework.

Figure 3. Research framework.

2.4. Simulation Scenario Design

2.4.1. Model Construction

This study focuses on the typical city of Hefei, located in a region characterised by hot summers and cold winters, to analyse and research the ribbon waterfront park. By summarising several waterfront parks in Hefei, a typical model of the Hefei ribbon waterfront park was constructed. Statistical analysis reveals that the average length-to-width ratio of the ribbon park is 6:1. The Nanfei River flows through the city, averaging 50 m in width, while the parks on both sides of the river also average 50 m in width. Most surrounding buildings are multistorey structures arranged in rows, with a north–south orientation. The water body direction and the dominant wind direction are parallel to the building orientation, aligning with the river. The typical ribbon waterfront park model consists of two ribbon parks (each 300 m in length and 50 m in width) distributed on both sides of the river. On the long side of the park, city roads and residential areas are located, while a bridge connects the two sides of the park area on the short side. A 3 m wide walkway runs along the riverfront, creating a 3 m elevation difference with the park. The proportion of land allocated to garden paths and paved areas is 25%. The plant structure follows a tree–shrub–grass composite pattern, and the park includes a designated activity square. Analysis of data extracted from satellite maps indicates that the canopy coverage rate of the waterfront park ranges from 40% to 90%, with an average canopy coverage rate of 61.93%. According to the Specification for Landscaping Engineering Projects GB55014-2021 [79], for parks with a land area of less than 2 hectares, the proportion of land used for garden paths and paved areas should be between 10% and 30%.

2.4.2. Design Factors and Levels

Based on thermal environment test research, pre-experiments, and previous studies [13,21,80], this study selects five variables as research factors: water body direction, density and form of tree planting [22,81], forms of distribution of squares in parks, location of the square in the park, and pavement material. The influence of these five factors on summer microclimate and thermal comfort is explored through comparative analyses, focusing on the perspectives of the water body, plants, and pavement. The experimental factor levels were set as shown in Table 2. The direction of the water body is set in four orientations. The simulation results of various scenarios under different water body directions are compared and analysed to derive the optimal microclimate and thermal comfort scenarios, further exploring the waterfront park space. The specific factors are selected as follows:

• Factor S: Water Body Direction
  The direction of the water body is determined according to the actual flow of the Nanfei River through Hefei City, in relation to the dominant wind direction. Four conditions are set: east–west direction, north–south direction, perpendicular to the wind direction, and parallel to the wind direction (Figure 4).
• Factor A: Density and Form of Tree Planting
  The planting density and form are adjusted by varying the planting spacing, based on the intersection, tangency, and separation relationships between tree canopies. Three levels are set:
  • High-density uniform planting (planting spacing 8 m × 8 m);
  • Low-density rows and columns planting (rows with 16 m spacing and 8 m spacing between trees);
  • Medium-density uniform planting (planting spacing 10 m × 10 m).
  These correspond to the generation of sparsely spaced planted landscape spaces, dense forest spaces, top-covered spaces, and semi-open spaces, with varying tree canopy coverage (Figure 5a).
• Factor B: Forms of Distribution of Squares in the Park
  The research found that the square activity space in the waterfront park is primarily divided into a large centralised square and decentralised small-area activity squares. In accordance with the norms, the proportion of the park’s pavement area is set to 25%. The total paving area remains unchanged and is arranged in two forms: centralised (a single large square) and decentralised (two smaller squares) (Figure 5c).
• Factor C: Location of the Square in the Park
  The research found that the location of the square (near the water body, centrally in the park, or near the road) affects the influence of the water body on the square’s activity space. Additionally, the orientation of the square (facing the city road or the water body) creates semi-open or enclosed spaces, which significantly impact the wind environment. Therefore, three levels are set in the model with reference to the square: near the water body, in the middle, and near the road location (Figure 5b).
• Factor D: Pavement Material
  Based on field research, four materials with different albedo values were selected [59]: red bricks (albedo of 0.3), coloured asphalt (albedo of 0.4), light-coloured concrete (albedo of 0.5), and light-coloured granite (albedo of 0.6).

Table 2. Orthogonal factor setting.

Level	Factors
	S	A	B	C	D
1	East–west direction	High density, uniform planting (Planting spacing (8 m × 8 m))	Centralised	Near the water body	Red brick
2	North–south direction	Low-density, determinant planting (Row spacing 16 m, plant spacing 8 m)	Decentralised	Centring	Coloured asphalt
3	Perpendicular to the prevailing wind direction	Medium density, uniform planting (Planting spacing 10 m × 10 m)	-	Near the road	Light-coloured concrete
4	Parallel to the prevailing wind direction	-	-	-	Light-coloured granite

Table 2. Orthogonal factor setting.

Level	Factors
	S	A	B	C	D
1	East–west direction	High density, uniform planting (Planting spacing (8 m × 8 m))	Centralised	Near the water body	Red brick
2	North–south direction	Low-density, determinant planting (Row spacing 16 m, plant spacing 8 m)	Decentralised	Centring	Coloured asphalt
3	Perpendicular to the prevailing wind direction	Medium density, uniform planting (Planting spacing 10 m × 10 m)	-	Near the road	Light-coloured concrete
4	Parallel to the prevailing wind direction	-	-	-	Light-coloured granite

2.4.3. Experimental Design

In this study, an orthogonal experiment was conducted to optimise the microclimate of the waterfront park by simulating it under different combinations of four factors and four levels of operating conditions using ENVI-met software. If a full simulation were carried out, 4⁴ (256) simulations would be required, which would consume significant humanpower, material resources, and time. The orthogonal experimental design is an incomplete experimental design that reduces the number of simulations by conducting multifactor and multilevel analyses according to an orthogonal table, which is simple and easy to operate [81].

The direction of the water body, density and form of tree planting, forms of square distribution in parks, location of the square in the park, and pavement material were designated as factors S, A, B, C, and D, respectively. Since the direction of the water body (factor S) significantly affects the distribution of the overall thermal environment in the park, this study conducted orthogonal tests under different water body directions to compare their effects laterally. The effects of other factors were analysed separately under each water body direction. Test factors A, B, C, and D have 3, 2, 3, and 4 levels, respectively, resulting in a total degree of freedom of the test: f = 4 × (4 − 1) = 12. Therefore, an orthogonal table with more than 12 degrees of freedom was required. Using SPSS (Version 27.0.1)’s orthogonal design function, a relatively close-to-standard orthogonal table, L16 (⁴), was generated (L denotes the orthogonal table, and 16 represents the number of level combinations). Based on the study’s requirements, only 16 simulations were needed to draw conclusions, satisfying the degrees of freedom requirements. The proposed level method was used for processing and simulation operations, as shown in

Table 3. Orthogonal experiment.

Case	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
A	2	2	1	3	1	1	1	3	3	2	2	3	1	2	3	2
B	1	2	1	2	1	2	2	1	1	2	1	2	1	2	2	1
C	2	3	2	1	1	1	3	2	1	2	3	3	2	3	1	3
D	4	3	1	1	4	2	3	4	2	3	2	1	3	4	2	1

. The 16 waterfront park simulation scenarios were simulated under four different water body directions relative to the dominant wind direction, resulting in a total of 64 test simulations.

2.5. ENVI-Met Simulation Parameter Settings and Model Construction

The ENVI-met simulation parameter settings are shown in

Table 4. ENVI-met simulation parameter settings.

Category	Set Item	Parameter Value
Grid data	Grid size	2 m × 2 m × 2 m
	The number of grids	150 × 210 × 30
Simulation time	Date	August 12, 2023
	Period	5:00~21:00
	Duration	16 h
Boundary and initial conditions	Full forcing	Typical annual meteorological data of Hefei weather station
	Air temperature	26.3~36.7 °C
	Wind speed at 10 m altitude	2.0 m/s
	Wind direction	135° (southeast)

. The input meteorological data for this study were obtained from the typical daily meteorological station data provided by the National Weather Science Data Center (https://data.cma.cn/, accessed on 15 September 2023), with the station number 58321, latitude 31.78°, longitude 117.3°, and an elevation of 27 m. The simulation date was 12 August 2023, and the total simulation duration was 16 h. The dominant wind direction was 135° (south-southeast), at a 10 m altitude. The simulation start time was 05:00, with a dominant wind direction of 135° (south-southeast) and a wind speed of 2.0 m/s at 10 m altitude. The initial air temperature was 26.3 °C, and other surface parameter inputs were based on the Design Code for Urban Roads. In the plant model setup, common tree species in Hefei parks, such as magnolia, camphor, ginkgo, and luan, were selected. These trees typically have heights of 12–20 m and crown widths of 8–15 m. Models with similar morphology and sizes to these trees were chosen from the ENVI-met botanical library, with tree heights of 15 m and spherical crowns of 10 m width. The systematic model was used for shrubs and ground covers. The pavement materials included both artificial and natural surfaces. The research variables comprised floor tiles (KK), coloured asphalt (AR), light-coloured concrete (PL), and light-coloured granite (G2). Other materials included asphalt pavement (ST), grey cement pavement (PG), and soil (00). Specific parameters are detailed in

Table 5. Model material parameters.

Material	Red Bricks	Coloured Asphalt	Light Coloured Concrete	Light-Coloured Granite	Asphalt Pavement	Grey Cement Pavement	Soil
Code	KK	AR	PL	G2	ST	PG	00
Albedo	0.3	0.4	0.5	0.6	0.2	0.9	0.2
Legend

.

Based on the level parameters of the above factors, the ENVI-met model was established (Figure 6 provides an example of Scenario 5). Red markers in the figure indicate monitoring points (sensors), which were arranged to account for different offshore distances (0 m, 15 m, 30 m, 50 m, and above the water body) and various types of outdoor spaces (shaded green areas, squares, boulevards, and riverfront walkways). Control groups with no water body and no greenery were established under each of the four water body direction layouts. The floor plans of the 16 orthogonal test scenarios are shown in Figure 7.

2.6. Selection of Thermal Comfort Evaluation Indexes

Currently, the commonly used outdoor thermal comfort models include the Universal Thermal Climate Index (UTCI) [82,83,84], the physiologically equivalent temperature (PET) [85,86,87,88,89,90], the Predicted Mean Vote (PMV) [91,92,93,94], and the Standard Effective Temperature for Outdoors (OUT_SET*) [95,96,97]. The theoretical basis of these evaluation indexes and a comparative analysis of their influencing factors are presented in

Table 6. Comparison of thermal comfort evaluation indexes.

Thermal Comfort Model	Ta	RH	Va	Tmrt	Tsk	Ms	Tc	Icl	HMR
UTCI	+	+	+	+	+	+	+	+	+
PET	+	+	+	+	+	−	+	+	+
PMV	+	+	+	+	−	−	−	+	+
OUT_SET*	+	+	+	+	+	−	+	+	+

Note: “+” represents factors considered, “−” represents factors not considered.

[98,99]. Compared with other thermal comfort models, the UTCI encompasses a more comprehensive set of parameters, providing a more accurate assessment of human thermal comfort across various global climate conditions [100,101]. Therefore, the Universal Thermal Climate Index (UTCI) was selected as the index for evaluating human thermal comfort in this study.

3. Results and Discussion

According to the subjective thermal comfort questionnaire, residents in the study area were more sensitive to air temperature and wind speed and less sensitive to relative humidity. Therefore, air temperature, wind speed, and their differences compared to the control group were selected as microclimate evaluation indicators, while the Universal Thermal Climate Index (UTCI) was used for thermal comfort evaluation. The time of day with the highest temperature and the most prominent thermal comfort issue—14:00—was selected to analyse the significance, influence ranking, primary and secondary relationships, and the optimal combination of factors affecting microclimate and thermal comfort.

The analysis of factors involved evaluating the advantages and disadvantages of each factor in the orthogonal experimental data using the method of extreme difference analysis. The optimal level of each factor was selected and combined to derive the theoretical optimal scenario. The average value of the simulated data at a specific level of a factor was recorded as k. For example, k1 represents the average value at level 1 of factor A. The optimal level and combination of factors were determined based on the magnitude of the average values across different water body directions. In the extreme difference analysis of air temperature and thermal comfort, a lower mean value indicates better performance, while in the analysis of wind speed, a higher mean value indicates better performance. The range value of each factor is denoted as R, and the influence of each factor on the simulation results was assessed by comparing the magnitude of the R values.

Temperatures throughout the day exhibited a typical diurnal pattern, as shown in Figure 8, with a trend of increasing and then decreasing. The lowest temperature (29.5 °C) occurred at 06:00, gradually rising as the sun rose and reaching a maximum of 35.2 °C at 14:00. Subsequently, temperatures began to decline, reaching 30.9 °C as sunlight weakened. In contrast, relative humidity showed a decreasing and then increasing trend throughout the day, peaking at 70% at 06:00, gradually decreasing to a minimum of 44.5% at 14:00, and then increasing to 61% during the night.

3.1. Comparative Analysis of Air Temperature

3.1.1. Cooling Effect of Different Water Body Directions and Offshore Distance

The average air temperature of the waterfront park at 14:00, the hottest moment in the simulation, was analysed under different water body directions. The difference in air temperature, calculated using simulation data from the control group, represents the cooling effect of different scenarios. A larger temperature difference indicates a better cooling effect.

As shown in Figure 9, the average air temperature in the waterfront park decreased under different water body orientations compared to the control group, with reductions ranging from 0.42 °C to 1 °C. The most significant cooling effect on both sides of the river was observed when the water body was aligned parallel to the wind direction. All downwind zones exhibited a cooling effect, while in upwind zones, except for the east–west orientation, air temperature decreased in the other three directions. The water body aligned parallel to the wind direction demonstrated the greatest cooling effect, with a temperature reduction of 1.38 °C. Conversely, in the east–west orientation, the air temperature in the upwind zone increased, likely due to tree-induced wind blockage, which restricted airflow into the park’s interior. This led to significant air stagnation, creating a stifling microclimate. These findings suggest that water bodies in waterfront parks exert a significant cooling effect on the surrounding environment while also enhancing local wind conditions [102,103,104]. Existing studies indicate that water bodies exhibit stronger cooling efficacy than tree-dominated green spaces, with a significantly greater cooling range [105]. Wind direction influences the spatial extent of microclimate modifications induced by water bodies [106] and affects the cooling performance of urban parks [107]. When the water body in a ribbon waterfront park is aligned parallel to the wind direction, the unobstructed wind gradient accelerates water surface transpiration, facilitating efficient heat dissipation from both the water body and the surrounding air. Consequently, the maximum daytime temperature difference between the waterfront area and its surroundings can reach 9 °C [108], aligning with Liu Jing’s study on the cooling effects of static water features in Guangzhou [109]. Jiang et al. [110] demonstrated that greenery corridors with a 20–25 m width aligned with the summer monsoon direction had more significant cooling effect ranges in Shanghai, based on comparisons of different spatial structures and waterfront greenery morphologies.

Therefore, when planning the location of waterfront parks, it is appropriate to choose the downwind region of the water body to establish a ribbon waterfront park. When the river flows parallel to the prevailing wind, constructing a ribbon waterfront park in the upwind region of the water body can enhance its cooling effect.

To analyse the relationship between the cooling effect of the water body in the waterfront park and the offshore distance, the microclimate data of several groups of monitoring points with different offshore distances were compared, and the offshore distances of each group of monitoring points were 0 m, 15 m, 30 m, 50 m, and above the water body. The data at 14:00 from Scenario 12 (water body direction perpendicular to the wind direction), which exhibited the best cooling effect, were used for analysis (Figure 10). Using the water body as the dividing line, where the downwind region is considered positive and the upwind region negative, the average air temperature values of the measurement points at various offshore distances are presented [111] (Figure 11). The air temperature in the water body’s downwind region is generally 0.84 °C lower than that in the upwind region, indicating a stronger cooling effect in the downwind region. Water bodies reduce air temperature through evaporative cooling and thermal convection mechanisms [112]. In the downwind region, the cooling effect induced by water surface evaporation propagates with the wind flow, leading to a significant reduction in air temperature [113]. When the water body is oriented perpendicular to the wind direction, the cooled air from water surface evaporation is carried by the wind, thereby enhancing the cooling effect, particularly in the downwind region. Conversely, in the upwind region, the cooling effect is less pronounced since the airflow remains unaffected by the water body [114,115].

The lowest air temperature, highest humidity, and strongest wind speed were observed in the downwind zone at the riverfront walkway, at 0 m from the riverbank, indicating that the cooling and humidifying effect of the water body was most pronounced at this location. The latent heat of evaporation facilitated ambient heat absorption, leading to a reduction in near-surface temperature. Simultaneously, the thermal pressure difference between the water surface and adjacent land induced local air circulation, thereby increasing wind speed in the waterfront area. Additionally, the synergistic effects of vegetation transpiration and water vapour diffusion further contributed to increased air humidity. As the distance from the riverbank increased, the curve flattened, and the cooling and humidifying effects of the water body gradually diminished beyond 15 m [116,117]. In the upwind region, the trend of the UTCI differs from that of air temperature. The UTCI value peaks above the water body and gradually decreases with distance from the shore, reaching its minimum at 30 m. This phenomenon is attributed to the absence of shade over the water surface, which is directly exposed to solar radiation, increasing the radiant temperature and consequently raising the UTCI value. At 30 m from the shore, the inner park area benefits from tree canopy shading, amplifying the water body’s cooling effect. Beyond 30 m offshore, the UTCI becomes more pronounced, indicating a decline in the water body’s role in enhancing thermal comfort. Wind speed and the UTCI significantly decrease within the 0–30 m offshore range, suggesting that the water body’s attenuation of wind speed is primarily limited to this zone.

3.1.2. Cooling Effect of Different Internal Environmental Factors

The ANOVA results (

Table 7. Analysis of variance of air temperature by test factors in different water directions.

Direction of Water Body		Air Temperature
		A	B	C	D
S1	F	39.3	0.207	3.073	24.79
	sig.	0.007	0.68	0.272	0.013
S2	F	37.113	13.043	3.636	4.489
	sig.	0.008	0.036	0.157	0.125
S3	F	0.524	1.154	1.331	0.503
	sig.	0.638	0.361	0.386	0.707
S4	F	52.706	0.007	0.995	23.778
	sig.	0.005	0.937	0.466	0.014

) indicate that the density and form of tree planting (A) significantly affect air temperature in the east–west direction, north–south direction, and parallel to the wind direction (Sig. < 0.01). However, no significant effect is observed when the water body is perpendicular to the wind direction (Sig. > 0.05). The square layout form (B) significantly influences temperature changes only when the water body is oriented north–south (Sig. < 0.05). The position of the square (C) does not significantly affect temperature changes (Sig. > 0.05). Pavement material (D) significantly impacts temperature changes in the east–west direction, perpendicular to the wind direction, and parallel to the wind direction (Sig. < 0.05).

According to the orthogonal test results (

Table 8. Results of orthogonal test for air temperature.

Direction of Water Body	Mean and Range	Test Factors
		A	B	C	D
S1	Ri	0.111	0.013	0.051	0.108
	Optimal combination	A1 + B1 + C1 + D4
S2	Ri	0.205	0.069	0.099	0.080
	Optimal combination	A1 + B1 + C2 + D1
S3	Ri	0.146	0.083	0.125	0.146
	Optimal combination	A3 + B2 + C3 + D1
S4	Ri	0.143	0.020	0.065	0.109
	Optimal combination	A1 + B1 + C1 + D4

), the ranking of factors influencing air temperature in the east–west direction, perpendicular to the wind direction, and parallel to the wind direction is RA > RD > RC > RB, indicating that tree planting density and form have the greatest impact, followed by pavement material, square position, and square layout form. In the north–south direction, the ranking shifts to RA > RC > RD > RB, meaning tree planting density and form remain the most influential, followed by square position, pavement material, and square layout form. Across all scenarios, tree planting density and form exert the most significant influence on air temperature. As Cameron et al. [118] and Shashua-Bar [119] found, shading accounts for 60% and 80% of the total cooling effect of trees, respectively, and optimising the distribution pattern of buildings and trees can significantly enhance shading effectiveness [116].

For the east–west, north–south, and parallel-to-wind directions, the average air temperature values for tree planting density and form levels k1, k3, and k2 (corresponding to canopy coverages of 80%, 70%, and 50%, respectively) follow the trend k1 < k3 < k2. This indicates a negative correlation between tree planting density and air temperature, with higher densities resulting in lower temperatures [117]. In the perpendicular-to-wind direction, the trend is k3 < k1 < k2, indicating a non-linear relationship between planting density and air temperature. The lowest temperatures occur at moderate planting densities, with uniform 10 m spacing yielding the lowest recorded temperatures. This is because the tangential arrangement of tree canopies facilitates air circulation, reducing air temperature in the perpendicular-to-wind scenario. Thus, a planting spacing and crown ratio of 1:1 are optimal, a finding consistent with the studies by Huang Y and Zhang L [26,75,120,121].

For the east–west, north–south, and parallel-to-wind directions, the mean temperature values for square layout forms follow k1 < k2, indicating that centralised layouts result in lower temperatures compared to decentralised layouts of the same area. In the perpendicular-to-wind direction, the trend reverses, with decentralised layouts yielding lower temperatures than centralised ones. For the east–west and parallel-to-wind directions, the mean temperature values for square positions follow k1 < k2 < k3, with squares near water bodies exhibiting the lowest temperatures and those near roads the highest. In the north–south direction, the trend is k2 < k1 < k3, with squares near roads recording the highest temperatures. In the perpendicular-to-wind direction, the trend is k3 < k1 < k2, with squares near roads showing the lowest temperatures and centrally located squares the highest. When a square is located near a water body, the cooling effect of the water helps moderate the surrounding air temperature. A water body not only lowers air temperature through evaporative cooling, but also transfers cool air via convection, leading to a significant temperature reduction [35,114]. Centrally located squares, though relatively distant from the water body, still experience some cooling effects, maintaining moderate temperatures. Squares adjacent to roads are typically affected by the urban heat island effect, particularly due to heat accumulation from roads and impervious surfaces. Areas near roads tend to exhibit higher heat radiation and heat storage, resulting in elevated temperatures. Cruz et al. [107] investigated the cooling effects of green and blue spaces (i.e., vegetation and water bodies) on urban microclimates using numerical simulations and demonstrated that water bodies effectively lower ambient air temperatures through evapotranspiration and convection. However, areas near impervious surfaces exhibit higher temperatures due to heat accumulation. Smith et al. [122] examined the impact of urban green spaces and surface albedo on urban temperatures across multiple U.S. cities. Their findings indicate that built surfaces, such as roads, accumulate more heat due to greater heat storage, whereas areas near water bodies benefit from enhanced cooling mechanisms. This finding further implies that areas near water bodies exhibit lower temperatures and greater thermal comfort compared to regions dominated by road infrastructure.

The mean value of the temperature corresponding to each level of the pavement material is k4 < k3 < k2 < k1, i.e., the temperature is sorted in the following manner: light-coloured granite < light-coloured concrete < coloured asphalt < red brick, and the albedos of k1, k2, k3, and k4 are 0.3, 0.4, 0.5, and 0.6, respectively. This indicates that the albedo of the pavement surface is negatively correlated with the air temperature, and that the higher the albedo of the pavement material, the higher the air temperature, and the higher the albedo of the air surface, the higher the air temperature. The higher the albedo of the material, the lower the air temperature. Surface albedo is a crucial parameter that describes the ground’s capacity to reflect solar radiation and is strongly associated with the urban thermal environment [122]. A high-albedo surface reflects more solar radiation, thereby reducing heat absorption, lowering local temperatures, and enhancing thermal comfort. Conversely, a low-albedo surface absorbs more heat, resulting in elevated local temperatures [123,124]. Liu et al. [125] demonstrated that the surface temperature of surrounding buildings was higher with asphalt pavement compared to turfgrass coverage. Lopez-Cabeza et al. [126] examined the impact of surface albedo on microclimate and thermal comfort in courtyards under hot summer conditions in the Mediterranean region. Their findings indicate that moderately increasing albedo can effectively reduce thermal environmental loads. Smith et al. [122] validated the combined influence of urban green spaces and albedo on surface temperature regulation across several U.S. cities, providing a quantitative framework for heat island effect management in cities with diverse climatic conditions.

In summary, it can be seen that the planting density and form of trees are the primary factors affecting air temperature, especially in the case of high planting density or moderate spacing, and the cooling effect is significant [127]. The albedo of the pavement, the layout of the square, and the location of the square in the park have a certain regulating effect on the air temperature under the conditions of different water directions, in which the high-albedo material and the design of the square close to the water body help to reduce the local air temperature [123]. At the same time, the mechanism of each factor varies under different wind directions, which indicates that the matching of wind environment and spatial layout should be considered comprehensively in the design to optimise the thermal comfort of the local microclimate [128].

3.1.3. Comparison of Scenarios and the Best Combination of Scenarios

When comparing the average temperature differences between each experimental scenario and the control group (cooling amplitude), a more significant difference indicates a stronger cooling effect, as illustrated in Figure 12. Regarding the direction of the water body, the better cooling effects in the east–west direction and parallel to the wind direction are found in Scenarios 5, 7, and 6; the better cooling effects in the north–south direction are in Scenarios 3, 5, and 7; while the better cooling effects in cases perpendicular to the wind direction are in Scenarios 12, 7, and 5.

For each test scenario at 14:00, a 1.5 m height temperature plane distribution analysis (Figure 13) reveals that in Scenario 12, the direction towards the water body, perpendicular to the wind direction, shows the largest cooling effect in the region. By analysing the temperature distribution at 14:00 across various scenarios, Scenario 5 emerges as the optimal choice. It features high-density uniform tree planting (8 m × 8 m), an impressive tree canopy shade rate of 82%, a square arrangement near the water body, and light-coloured granite pavement (albedo of 0.6). Scenario 5 demonstrates significant cooling effects in different water bodies and wind directions, mainly exhibiting the best performance when oriented east–west and parallel to the wind direction. When considering all scenarios, it shows both universality and stability.

The suboptimal scenarios are Scenarios 12, 3 and 7. Scenario 12 has the largest cooling area in the direction of the water body perpendicular to the wind direction, but its performance in other directions is not as good as that of Scenario 5, which is less applicable; Scenario 3 has the best cooling effect in the north–south direction (90°), but its performance in other cases is not as good as that of Scenario 5; Scenario 7 has a stable performance, and its cooling effect is second to that of Scenario 5, which belongs to the alternative scenarios.

The reason explanation and mechanism analysis are as follows: High-density uniform planting makes the park’s tree canopy shade rate as high as 87.6%, providing effective shading and reducing ground heating by solar radiation, thereby significantly lowering air temperature [81]. A centralised layout enhances the interaction between vegetation and water bodies, amplifying the cooling effect. When perpendicular to the wind direction, this layout concentrates cool airflow, reinforcing local cooling. Proximity to water bodies leverages evaporative cooling, vegetation shading, and air circulation to enhance overall cooling capacity. Light-coloured granite pavement (albedo of 0.6) reduces surface heat storage and mitigates the local heat island effect, complementing the cooling effects of vegetation and water bodies [59,129].

Scenario 5’s excellent cooling performance, adaptability to various water body directions and wind conditions, and stability make it the optimal choice. To further enhance microclimatic benefits and outdoor thermal comfort, it is recommended to prioritise high-density uniform planting (8 m spacing), centralised square layouts, and light-coloured, high-albedo paving materials.

By prioritising air temperature as the key optimisation objective, the ideal combination of factors (A1 + B1 + C1 + D1) achieves the best thermal environment. This includes high-density uniform tree planting (8 m spacing), a centralised square near the water body, and light-coloured granite pavement, with the water body aligned parallel to the prevailing wind direction (Figure 14).

3.2. Comparative Analysis of Wind Speed

3.2.1. Influence of Different Water Body Directions on Wind Speed at the Square

The average wind speed at the waterfront park at 14:00, the hottest moment in the simulation, was analysed under different water body directions (Figure 15). The results reveal that wind speed decreases significantly when the water body is perpendicular to the wind direction, as the water body acts as a barrier to incoming wind. This alignment facilitates the formation of a downwind channel, reducing wind resistance and significantly enhancing wind speed. Conversely, wind speed increases in other orientations, particularly when the water body is parallel to the wind direction. This alignment facilitates the formation of a downwind channel, reducing wind resistance and significantly enhancing wind speed. Xin Guo et al. [130] examined urban street intersections with varying greening layouts and found that vegetation can reduce ventilation at these intersections. Fang et al. [131] confirmed that ventilation corridors significantly improve environmental conditions and air circulation in Hefei City. Xu et al. [132] demonstrated through numerical simulations that optimising corridor design, particularly when aligned parallel to the dominant wind direction, can substantially increase local wind speed, enhancing natural ventilation and mitigating the local heat island effect.

3.2.2. The Influence of Internal Environmental Factors on Wind Speed

Internal environmental factors, particularly tree planting spacing, significantly influence wind speed in waterfront parks. Wider tree spacing facilitates air circulation and convective heat transfer, improving the microclimate. For instance, Scenario 4 (10 m tree spacing) exhibited higher wind speeds and better thermal comfort compared to Scenario 6 (8 m spacing).

ANOVA results (

Table 9. Analysis of variance of wind speed by test factors in different water directions.

Direction of Water Body		Test Factors
		A	B	C	D
S1	F	23.334	0	1.499	4.673
	sig.	0.015	0.997	0.354	0.119
S2	F	22.035	0.714	0.924	2.274
	sig.	0.016	0.46	0.487	0.259
S3	F	23.79	2.072	1.531	1.156
	sig.	0.014	0.246	0.348	0.454
S4	F	36.284	0.35	2.503	4.225
	sig.	0.008	0.596	0.229	0.134

) indicate that tree planting density and arrangement significantly affect wind speed in all four water body directions (Sig. < 0.05), while other factors do not (Sig. > 0.05). Among planting configurations, row planting (16 m × 8 m) produced the highest wind speeds, as it facilitates coherent wind channels and enhances air circulation. In contrast, high-density uniform planting obstructs airflow, reducing wind speed. Morakinyo and Lam [27] investigated the effects of tree configuration, planting pattern, and wind conditions on the microclimate of a street canyon through numerical simulations, and the results showed that the aspect ratio of the trees affects the degree of decrease in PET (physiologically equivalent temperature). When the planting pattern has a larger aspect ratio, it can effectively regulate wind speed and improve thermal comfort for pedestrians. This further validates the conclusion of this study that planting trees in rows helps to improve the wind environment. Although high-density planting may generate localised thermal convective circulation due to its cooling effect, its ventilation capacity remains limited, and overall wind speed tends to decrease. When the water body is parallel to the wind direction, tree planting density and form profoundly impact wind speed (Sig. < 0.01). This alignment creates a belt-shaped wind channel within the park, enhancing ventilation efficiency and optimising the local microclimate. A square layout also plays a crucial role; Wang et al. [133] demonstrated that enclosed plazas with higher height-to-width ratios reduce wind speed, improving wind comfort. Extreme deviation and ANOVA analyses of wind speed at a 1.5 m height (

Table 10. Results of orthogonal test for wind speed.

Direction of Water Body	Mean and Range	Test Factors
		A	B	C	D
S1	Ri	0.080	0.003	0.045	0.028
	Optimal combination	A2 + B2 + C23 + D3
S2	Ri	0.088	0.012	0.051	0.036
	Optimal combination	A2 + B2 + C3 + D4
S3.	Ri	0.058	0.007	0.014	0.014
	Optimal combination	A2 + B1 + C2 + D2
S4	Ri	0.114	0.005	0.068	0.038
	Optimal combination	A2 + B2 + C3 + D3

) reveal that the influencing factors, in descending order of significance, are tree planting density and form > location of the square in the park > pavement material > square layout form (RA > RC > RD > RB).

Trees influence wind dynamics through two primary mechanisms: (1) Physical obstruction and ventilation regulation: trees can block wind flow or create ventilation corridors through gaps between branches and leaves. (2) Temperature-induced local air circulation: trees lower ambient temperatures, generating a temperature differential that induces air movement. The relationship between planting density and wind speed follows a clear trend, with mean wind speeds ranked as k2 > k3 > k1. Row planting promotes higher wind speeds compared to uniform planting. In a uniform layout, greater planting spacing results in higher wind speeds, as it facilitates air circulation. Conversely, high-density planting obstructs airflow, reducing wind speed. These findings align with previous research. Kang et al. [134] used computational fluid dynamics (CFD) simulations to demonstrate that tree crown size, planting density, and spatial positioning significantly influence local wind speed and pedestrian comfort. Similarly, Gao et al. [135] found that optimised tree layouts enhance wind speed and regulate wind direction, improving ventilation efficiency. Chen et al. [136] revealed that strategic tree configurations stabilise wind fields and enhance outdoor environmental quality. Together, these studies highlight that tree planting strategies play a crucial role in shaping urban wind environments. By optimising planting density, spatial arrangement, and crown size, planners can enhance ventilation efficiency while maintaining pedestrian thermal comfort in urban public spaces.

3.2.3. Comparison of Scenarios and the Best Combination of Scenarios

A comparison of the average wind speed differences between each test scenario and the control group (growth rate) reveals that a larger difference indicates a stronger ventilation effect and wind speed regulation, as shown in Figure 16. Positive values denote enhanced wind speed (good ventilation), while negative values indicate reduced wind speed (wind-blocking effect). When the water body is perpendicular to the wind direction, wind speed decreases across all scenarios. High-density planting (e.g., Scenarios 5, 6, and 7) exacerbates this effect by forming dense forest belts that obstruct incoming wind, further reducing wind speed. In the east–west and north–south directions, row planting (Scenarios 1, 2, 10, 11, 12, and 16) slightly increases wind speed, particularly in medium-density planting (10 m × 10 m, Scenario 12), which demonstrates a moderate wind-enhancing effect. The most significant wind enhancement occurs when the water body is parallel to the wind direction, especially in row planting (Scenario 2), where the ribbon tree configuration forms a distinct ventilation corridor, maximising ventilation efficiency.

Figure 17 illustrates the wind speed distribution at a height of 1.5 m at 14:00 for each test scenario. The optimal scenario is Scenario 2, featuring medium-density row planting (48% coverage), a decentralised square near the road, and light-coloured concrete pavement. Row planting facilitates the formation of clear wind paths, particularly in the parallel wind direction, yielding the best ventilation effect.

Moderate planting density ensures air circulation while avoiding wind speed loss due to overly sparse planting. A suboptimal solution is Scenario 11, which employs row-type planting with 52% coverage. Although higher coverage provides some wind channeling, its ventilation effect is slightly inferior to Scenario 2. From the spatial distribution of wind speed, Scenario 2 exhibits the best ventilation effect at 14:00, with the largest high-wind-speed area and the smallest low-wind-speed area. This scenario’s row-type planting (16 m × 8 m) features appropriate tree spacing and an efficient air duct design, significantly improving ventilation compared to other scenarios.

Consequently, based on the simulation analysis, Scenario 2 (row planting, 16 m × 8 m) emerges as the optimal choice for enhancing the wind speed regulation in ribbon waterfront parks. Its design effectively achieves significant ventilation across various water directions, mainly exhibiting the strongest wind enhancement in the parallel wind direction. This contributes to improving the microclimate environment and the thermal comfort of visitors. This design is well suited for waterfront parks requiring marked improvements in ventilation.

Taking the wind environment as the primary optimisation objective, the best scenario combines the optimal levels. Its configuration, A2 + B1 + C3 + D3, occurs when the direction of the water body is parallel to the wind direction, utilizing medium-density row and column planting with a spacing of 8 × 16 m. The layout is centrally positioned and located close to the city road, with light-coloured concrete used for the pavement material, as illustrated in Figure 18.

3.3. Comparative Analysis of Thermal Comfort

3.3.1. Comparison of UTCI for Different Water Body Directions

The analysis of the average UTCI at the waterfront park at 14:00, the hottest point in the simulation results under various water body orientations (Figure 19), aligns with the air temperature findings in Section 2.1. The lowest UTCI occurs across all scenarios when the water body is parallel to the wind direction, while the highest UTCI is observed when the water body is perpendicular to the wind direction.

This phenomenon can be attributed to two primary mechanisms: (1) Evaporative cooling and thermal inertia: when the water body is parallel to the wind direction, unobstructed airflow enhances evaporative cooling, allowing cooler air to spread across the park and reducing the UTCI. Conversely, a perpendicular orientation disrupts airflow, weakening evaporative cooling and limiting its effectiveness in lowering the UTCI. (2) Wind-induced heat dissipation: a parallel water body orientation facilitates better air circulation, enhancing heat dissipation and maintaining lower UTCI values. These findings align with previous studies. Fei et al. [17] examined the combined effects of water bodies and greenery in waterfront spaces in cold regions of China. Their findings indicate that water bodies significantly lower the summer UTCI through thermal inertia and evaporative cooling, thereby alleviating thermal stress. Similarly, Du et al. [137] validated the importance of integrating water bodies in blue-green space planning for enhancing thermal environments and reducing UTCI values in a Shanghai-based case study. Additionally, Yu et al. [138] and Liu et al. [139] further corroborated the beneficial role of water bodies in optimising urban microclimates and enhancing thermal comfort.

Based on the distribution of the UTCI in the simulation results, the proportion of the park area with comfortable UTCI values (comfortable UTCI ratio) can be calculated to evaluate the spatial microclimate of the waterfront park. As shown in

Table 11. Relationship between UTCI and thermal sensation.

UTCI (°C) Range	Stress Category
above +46	extreme heat stress
+38 to +46	very strong heat stress
+32 to +38	strong heat stress
+26 to +32	moderate heat stress
+9 to +26	no thermal stress
+9 to 0	slight cold stress
0 to −13	moderate cold stress
−13 to −27	strong cold stress
−27 to −40	very strong cold stress
below −40	extreme cold stress
above +46	extreme heat stress

[140], a UTCI value of 38 °C marks the threshold between strong and very strong thermal stress.

Taking into account the actual activities of residents in the study area, the park is divided into comfortable and uncomfortable UTCI zones at 38 °C. The comfortable UTCI ratio for each scenario is shown in Figure 20. Data analysis indicates that a higher percentage of comfort zones typically correlates with lower UTCI conditions. For example, under a 0° wind direction, parallel-aligned trees (Scenario 5) exhibit lower temperatures, with a comfort zone percentage of 69.23%, whereas vertically aligned trees (Scenario 1) have higher temperatures, with a comfort zone percentage of only 54.26%. This trend suggests that lower temperatures increase the comfort zone, and scenarios with higher comfort zone percentages are usually accompanied by lower temperatures. Pearson’s correlation coefficient between the UTCI and the percentage of the comfort zone is −0.923, indicating a strong negative correlation. As the UTCI rises, the comfort zone percentage decreases, suggesting that higher temperatures reduce thermal comfort.

3.3.2. Effect of Internal Environmental Factors on Thermal Comfort

ANOVA results (

Table 12. Analysis of variance of UTC by test factors in different water directions.

Direction of Water Body		Test Factors
		A	B	C	D
S1	F	8.553	0.202	0.147	1.27
	sig.	0.058	0.683	0.869	0.424
S2	F	38.747	0.486	0.305	1.48
	sig.	0.007	0.536	0.758	0.378
S3	F	10.107	0.407	0.444	0.718
	sig.	0.046	0.569	0.678	0.604
S4	F	25.065	0.004	0.773	0.983
	sig.	0.013	0.955	0.536	0.505

) indicate that tree planting density and form significantly impact UTCI values across all four water body orientations (Sig. < 0.05), with a particularly pronounced effect in the north–south direction (Sig. < 0.01). In contrast, other factors did not exhibit statistically significant effects on the UTCI (Sig. > 0.05). This strong correlation can be attributed to two key mechanisms: (1) Shading effect: higher tree canopy density reduces direct solar radiation, lowering air and surface temperatures. (2) Evapotranspiration cooling: trees release water vapour through transpiration, promoting latent heat dissipation and reducing local thermal stress.

These findings align with previous studies. Tree canopy cover exhibits a significant negative correlation with the UTCI [119,141]. Meili et al. [142] demonstrated that vegetation cover, transpiration efficiency, and leaf area index can lower the UTCI by up to 3 °C in tropical cities, with the synergistic effect of shading and evapotranspiration reducing the probability of heat stress by 23%. Similarly, Chen et al. [45] found that increasing tree cover from 18% to 70% can reduce the UTCI by approximately 3 °C. These results highlight the critical role of tree planting strategies in urban heat mitigation. By optimising tree density, planting form, and canopy coverage, urban planners can effectively reduce UTCI levels, enhancing outdoor thermal comfort and mitigating urban heat stress.

In the orthogonal test results (

Table 13. Results of orthogonal test for UTCI.

Direction of Water Body	Mean and Range	Test Factors
		A	B	C	D
S1	Ri	1.06	0.136	0.555	0.62
	Optimal combination	A1 + B1 + C1 + D4
S2	Ri	1.29	0.1	0.56	0.72
	Optimal combination	A1 + B1 + C1 + D1
S3	Ri	1.14	0.04	0.54	0.71
	Optimal combination	A1 + B2 + C1 + D1
S4	Ri	1.12	0.04	0.52	0.58
	Optimal combination	A1 + B1 + C1 + D1

), tree planting density and form have the greatest impact on UTCI values, followed by pavement materials, plaza location, and plaza layout (RA > RD > RC > RB). The mean UTCI values for different tree planting densities follow the order k1 < k3 < k2, where k1 represents high-density uniform planting (80% canopy coverage), k3 represents low-density row planting (70% coverage), and k2 represents medium-density uniform planting (50% coverage). This phenomenon can be attributed to the shading and evapotranspiration effects of vegetation, which reduce the UTCI by absorbing and scattering long-wave radiation, forming a localised “umbrella effect”. Higher planting density leads to lower UTCI values due to increased canopy cover, which provides shading and evapotranspiration cooling. This aligns with the findings of Brinda Deevi [143] and Shata, R.O. [144].

The UTCI values for different plaza layouts follow the pattern k1 < k2, indicating that a centralised layout provides better thermal comfort than a decentralised layout when the plaza area remains constant. However, when the plaza is perpendicular to the wind direction, the UTCI values exhibit the order k1 < k3 < k2, suggesting that decentralised layouts offer better thermal comfort in this scenario. Plaza location also significantly affects UTCI values. When the water body is oriented east–west and parallel to the wind direction, the mean UTCI values follow the order k1 < k2 < k3, indicating that plazas near water bodies experience the lowest UTCI values and the best thermal comfort, while those adjacent to roads exhibit the highest UTCI values and the poorest thermal comfort. Conversely, when the plaza is perpendicular to the wind direction, UTCI values exhibit the order k1 < k3 < k2, suggesting that proximity to roads enhances thermal comfort, while centralised plaza locations experience the highest UTCI values and the poorest comfort levels. The influence of pavement materials on UTCI values is also significant, with the ranking k4 < k1 < k2 < k3. Light-coloured granite exhibits the lowest UTCI values, followed by red brick, coloured asphalt, and light-coloured concrete. This pattern remains consistent across other directional analyses. This finding confirms that high-albedo paving materials effectively mitigate the urban heat island effect and enhance human thermal comfort [145,146].

These findings align with previous research. Wang et al. [147] demonstrated that a combination of green infrastructure and high-reflectivity materials can significantly lower the UTCI and reduce building energy consumption. Similarly, Sedaghat and Sharif [148] simulated various urban heat island (UHI) mitigation strategies, revealing that integrating high-albedo materials and vegetation can reduce cooling loads by up to 29%. These results highlight the importance of strategic urban planning in mitigating heat stress and enhancing thermal comfort. By optimising tree planting density, plaza layouts, and pavement materials, urban planners can effectively regulate UTCI values and create more thermally comfortable outdoor environments.

3.3.3. Comparison of Scenarios and the Best Combination of Scenarios

As illustrated in Figure 21, Scenario 5 exhibits the lowest UTCI values across all four scenarios, accompanied by the highest proportion of comfortable UTCI zones. The UTCI values are minimised when the water body aligns parallel to the wind direction, as this orientation enhances air convection and heat dissipation, thereby reducing ambient temperatures and improving thermal comfort conditions [105,110]. Across the four water body orientations, Scenarios 5, 6, 7, and 3 demonstrate lower UTCI values when aligned east–west, perpendicular, or parallel to the wind. Under the four water body directions, the comfort UTCI accounts for a more significant proportion of the east–west direction in Scenario 5, 1, and 6; of the north–south direction in Scenario 5, 3, and 6; perpendicular to the wind in Scenarios 5, 7, and 6; and parallel to the wind in Scenario 5, 7, and 3. Collectively, Scenario 5 not only records the lowest UTCI values but also encompasses the largest comfortable UTCI area, aligning with the optimal scenario identified in Section 2.1 for minimal temperature, thereby substantiating its efficacy in thermal comfort optimisation [149].

Figure 22 illustrates the more comfortable scenarios (3, 5, 6, and 7), all of which are characterised by a high density of trees. This indicates that the higher the planting density, the greater the canopy coverage and the more significant the shading effect and transpiration, thus effectively reducing the ambient temperature and improving thermal comfort. Conversely, Scenario 2, characterised by low-density tree planting, exhibits the highest UTCI values and the smallest comfortable UTCI area, attributable to reduced canopy coverage and diminished radiation shading, culminating in elevated ground and air temperatures and markedly reduced thermal comfort [116,150,151].

The scenario with the best thermal comfort is obtained by combining the optimal levels of each. It is combined as A1 + B1 + C1 + D1, consistent with the lowest air temperature scenario, described in Section 3.1 (Figure 14).

4. Conclusions

This study is based on an ENVI-met numerical simulation of the thermal environment in ribbon waterfront parks in Hefei’s primary urban area. It evaluates the influence of environmental factors on thermal comfort. The results indicate that water bodies in waterfront parks not only provide a significant cooling effect but also enhance wind conditions [102,103,104]. Various factors influence the thermal comfort of waterfront parks, and their influence, in descending order, is as follows: density and form of tree planting (A) > pavement material (D) > location of the square in the park (C) > distribution forms of squares in the park (B).

4.1. Influence of Environmental Factors on Thermal Comfort

The cooling efficiency of water bodies is influenced by multiple factors, including width, depth, flow rate, and spatial relationship with the dominant wind direction [152,153,154]. In ribbon waterfront parks, the interaction between water bodies and the dominant wind direction directly affects both the thermal environment and pedestrian thermal comfort. The study found that when the water body is aligned parallel to the dominant wind direction, the cooling effect is strongest, with a temperature reduction of up to 1.38 °C.

In a ribbon waterfront park, the interaction between water bodies and the dominant wind direction directly affects both the thermal environment of the park and the thermal comfort of pedestrians. The study found that when the water body is aligned parallel to the dominant wind direction, the cooling effect is at its strongest, with a temperature reduction of up to 1.38 °C. This conclusion aligns with the research of Liu et al. [109], who demonstrated that the orientation of water bodies plays a crucial role in regulating the thermal environment of waterfront spaces, with a linear layout parallel to the wind direction offering the best cooling and humidifying effects. However, when the wind direction forms a large angle with the water body (close to perpendicular), the wind shear effect weakens, leading to reduced evaporative cooling efficiency and potential lateral heat loss from the water body. Consequently, the spatial configuration of waterfront parks should dynamically align with the prevailing wind direction to maximise cooling efficiency [110,155]. This finding has significant implications for the location and layout of urban waterfront parks. During urban planning, priority should be given to areas where the wind direction aligns with the water body to maximise cooling effects.

In terms of internal environmental factors, this study identified the following ranking of elements impacting thermal comfort, listed in order of decreasing influence: tree planting density and form (A) > paving materials (D) > park plaza location (C) > plaza distribution form (B).

Among these, tree shading significantly reduces the UTCI [142]; however, excessively dense vegetation may obstruct airflow, causing localised heat stagnation and reducing thermal comfort [119]. This observation is consistent with Zhang et al.’s [75] findings on the impact of urban greening density on thermal comfort. Additionally, this study found that moderate-density tree planting (e.g., 8 m × 16 m) effectively enhances wind speed and improves thermal comfort [156,157]. Furthermore, paving material selection plays a crucial role in modifying the thermal environment. High-albedo surfaces reflect more solar radiation, reduce heat absorption, lower local temperatures, and improve thermal comfort, aligning with Taleghani’s findings [158]. Therefore, in waterfront park design, an optimal allocation of vegetation and water bodies is essential for enhancing thermal comfort.

This study systematically assessed the combined effects of vegetation, water bodies, and plazas on thermal comfort and the thermal environment. The findings underscore the critical role of wind–water interaction in microclimate optimisation, a factor previously underexplored in existing studies. This research provides valuable insights for the scientific planning of urban waterfront parks, offering an evidence-based approach for optimising thermal comfort in urban spaces.

4.2. Microclimate Optimisation Strategies for Ribbon Waterfront Parks

Microclimate optimisation in ribbon waterfront parks is guided by the dual principles of “functional adaptation” and “wind direction responsiveness”, establishing a comprehensive design framework. Given the diversity of activities—ranging from static recreation to dynamic movement—each requiring specific microclimate conditions, a strategic management approach is essential. Furthermore, the spatial arrangement of water bodies relative to prevailing wind directions (perpendicular, parallel, or oblique) significantly affects cooling distribution.

This study employs orthogonal experiments and ENVI-met simulations to elucidate the mechanisms of microclimate regulation and thermal comfort, providing a quantifiable foundation for design optimisation. Two strategies are proposed: proactive planning for new parks and adaptive enhancement for existing ones. For new parks, proactive planning prioritises environmental factors such as tree planting density, plaza layout, and water body orientation based on functional requirements and user needs to pre-regulate thermal comfort. For existing parks, adaptive enhancement focuses on spatial reconfiguration and element optimisation, including ventilation corridors, plaza adjustments, high-reflectance paving replacement, and vegetation optimisation.

These strategies—“proactive design” and “adaptive renewal”—provide systematic solutions for achieving sustainable thermal comfort and enhancing environmental efficiency. Based on the above principles, we propose specific strategies to optimise the microclimate of the waterfront park from the following two aspects.

4.2.1. Tailored Environmental Design for Waterfront Parks Based on Activity Needs

Environmental design must be orientated to the crowd’s needs, flexible adjustment, and different demand-oriented focuses on the thermal environment. There are differences, and, thus, the setting of environmental factors is also different. For long-stay activity spaces, more attention should be paid to shade microclimate and thermal comfort regulation, while wind environment and air circulation should be optimised for slow walking trails. The following is a specific analysis of the design suggestions for waterfront parks under different goal directions:

(1)

Long-stay activity spaces: The primary focus in the construction of this type of park space is the optimisation of thermal comfort in activity areas. Based on the study’s findings, the optimal scenario is A1 + B1 + C1 + D4, characterised by a high-density uniform tree layout with 8 m spacing, a centralised plaza positioned near the water body, light-coloured granite hard pavement, and a water body orientation parallel to the wind direction. Under this combination of conditions, the thermal comfort performance is the best. The combination of extensive tree canopy coverage and high-albedo pavement enhances the park’s thermal comfort, aligning with the findings of Wang [147] and Sedaghat [148].

(2)

Waterfront walkways and running tracks: When the design objectives focus on a waterfront walkway and a healthy running track [15,159], optimising the wind environment becomes the primary consideration. The optimal scenario is A2 + B1 + C3 + D3, featuring medium-density row-and-column tree planting with 8 m × 16 m spacing, a centralised plaza near the city road, light-coloured concrete pavement, and a water body orientation parallel to the wind direction. In this layout, the row-and-column vegetation planting forms a ventilation corridor, while the plaza’s proximity to the road enhances park connectivity, facilitating airflow and improving the wind environment. These findings are consistent with those of Fang et al. [131] and Xu et al. [132]. Additionally, the results of [159] indicate that the primary spatial morphological factors influencing the microclimate and thermal comfort of waterfront walkways in summer include average tree height, average tree spacing, and the ratio of shadow-producing side tree height to walkway width.

4.2.2. Recommendations for Optimal Design of Ribbon Waterfront Parks with Different Water Body Directions

Based on the preceding analysis, the planning and design of the ribbon waterfront park should prioritise the relationship between the water body’s location and the dominant wind direction to maximise the water body’s optimisation effect on the thermal and wind environment. Specifically, when the direction of the water body in the construction site is at an angle to the wind direction, the cooling effect of the downwind region of the water body is better than that of the upwind region, so the park should be arranged in the downwind region as a priority. Conversely, when the water body aligns parallel to the wind direction, parks on both sides benefit equally from enhanced cooling effects. The optimisation of the microclimate of the existing ribbon waterfront park is based on the direction of the water body and the wind and thermal environment of the park, which should be adjusted and optimised. The following are recommendations for optimising the design of ribbon waterfront parks for different water body orientations:

(1)

When the water body is in the east–west direction (S1), i.e., the ribbon waterfront park is distributed in the north and south banks of the water body, the optimal scenario is A1 + B1 + C1 + D4; the vegetation and the pavement are the factors that affect the thermal comfort the most. The planting density of trees should be improved as much as possible to form a high canopy coverage and provide enough tree shade. The paving materials should be selected with high albedo (e.g., light-coloured granite and light-coloured concrete) [59]. The waterfront walkways should be combined with trees for shade, and squares should be set up in areas close to the water body to facilitate the cooling effect of the water body. A study [160] showed that park visitors tend to seek out waterside spaces for activities and to escape the heat. Stocco et al. [161] found that light-coloured gravel paving (albedo 0.55) maintains a shortened daytime heat stress duration through numerical simulation studies.

(2)

When the water body is in the north–south direction (S2), i.e., the ribbon waterfront park is distributed on the east and west sides of the water body, the optimal scenario is A1 + B1 + C2 + D1. The form of vegetation and square layout are the most important factors affecting thermal comfort, and a higher planting density can help to improve thermal comfort. A centralised square is better than a decentralised square in terms of thermal comfort. A centralised square can form more obvious ventilation paths and improve air circulation and heat dissipation, while trees should be planted around the square for people to rest in the shade. This is in line with the findings of Stocco et al. [161] in an arid region, who found through CFD simulations that increasing vegetative shading coverage synergistically with porous, permeable paving improves thermal comfort, and that centralised layouts significantly outperform decentralised layouts.

(3)

When the water body is perpendicular to the dominant wind direction (S3), the optimal scenario is A3 + B2 + C3 + D1. The planting density of trees should be moderate; a planting density that is too high is not conducive to airflow into the park when the park is oriented perpendicular to the wind direction. A moderate planting density will easily lead to the air circulation inside the park. A canopy coverage that is too high is also not conducive to the improvement of thermal comfort, also due to the wind resistance of the foliage. This is consistent with the results of a study in Fuzhou, which showed that for every 10 per cent increase in tree cover, the maximum reduction in pedestrian height-averaged air temperature was about 0.26 °C; after the tree cover exceeded 50 per cent, the value of the improvement in the thermal benefits from tree planting decreased [44]. In this case, you can connect urban roads and water bodies to set up activity squares and increase the area of the air inlet.

(4)

When the water body is aligned with the prevailing wind direction (S4), the optimal solution is A1 + B1 + C1 + D4. Planners can effectively enhance thermal comfort by increasing the planting density of trees, adopting a row-type planting layout, and utilizing high-albedo pavement. The dense planting of trees improves the shading effect, while the row layout, which aligns with the wind direction, ensures canopy coverage and prevents branches and leaves from blocking the wind [108,110]. Improve the material of the pavement, select the material with high albedo, and reduce the heat transfer from the pavement to the activity space. The layout of the square (centralised or decentralised) has less impact on thermal comfort in this case, and the arrangement, close to the road or in the middle, is conducive to the optimisation of the wind environment [132]; moreover, the waterfront walkway in both the upwind and downwind zones can fully benefit from the cooling effect of the water body.

4.3. Research Contributions and Future Directions

Through orthogonal tests, numerical simulation, and extreme difference analysis, this study investigates the characteristics of water bodies, squares, and plant communities in Hefei City, examining their primary and secondary roles in influencing temperature, wind speed, and thermal comfort in ribbon waterfront parks, as well as identifying optimal scenario combinations. Based on the findings, the study proposes the optimal scenario for enhancing thermal comfort in ribbon waterfront parks and provides design recommendations, offering a new perspective for microclimate research in such spaces. The study’s results can inform microclimate research in waterfront parks and serve as a reference during the planning, design, or renovation stages to support decision-making. Future research on waterfront parks should explore the comprehensive influence of complex environmental factors, such as surrounding building layouts [40,110], park shade facilities [162], street configurations [28,163,164,165,166,167], and river patterns [50,157], to provide comprehensive theoretical references and data support for the development of urban waterfront park spaces.

This study is geographically constrained, as all samples are derived from Hefei, a city characterised by a hot summer and cold winter climate. To apply the ENVI-met model to other climatic zones or regions outside China, localised adjustments to input parameters, climatic conditions, and vegetation characteristics are essential. Furthermore, field data validation is required to account for regional variations in temperature dynamics, urban morphology, wind patterns, and green space configurations, ensuring the model’s accuracy and reliability across diverse environmental contexts. To broaden the study’s applicability beyond China [168,169], future research should incorporate comparative analyses across distinct climatic regions, such as Mediterranean [170,171,172], arid [169], tropical [173], or humid subtropical zones [25], where differences in vegetation types, built environments, and microclimate interactions may yield varying thermal comfort outcomes. Such comparative investigations would not only enhance the generalizability of the findings but also contribute to a more globally adaptable framework for waterfront park design.