Optimizing Public Space Quality in High-Density Old Districts of Asian Megacities: Thermal Environment Analysis of Shenzhen’s Urban Fringe

Abstract

High density old districts at the urban fringes of Asian megacities, such as Shenzhen, face significant thermal challenges due to dense building clusters, limited airflow, and heat retention. This study adopts an integrated approach combining Phoenics wind simulation, geographic information system (GIS) modeling, and spatial prototype analysis to assess and optimize the wind and thermal environments in these urban areas. It investigates how spatial configurations, including building density, height distribution, orientation, and green space integration, influence wind flow and thermal comfort. The results demonstrate that optimized spatial arrangements, including reduced building density, height adjustments, and strategic landscape design, improve ventilation and temperature regulation. Comparative analyses of different spatial prototypes reveal that radial configurations effectively channel external winds into the urban core, enhancing internal airflow, whereas rectangular layouts create wind shadows that hinder ventilation. Adjustments to building façades and vertical arrangements further mitigate pedestrian-level heat accumulation. Interventions in public spaces, including green roofs and vertical greening, offer cooling benefits and mitigate urban heat island effects. This study underscores the importance of aligning urban design with natural wind flow and offers a framework for sustainable landscape and architectural strategies in high-density, heat-prone environments. The findings offer valuable insights for urban planners and policymakers seeking sustainable development in similar megacities.

1. Introduction

The latest World Cities Report by UN-Habitat [1] highlights that billions of urban residents may face an additional temperature increase of at least 0.5 °C by 2040, which poses a growing challenge to urban livability. This trend is particularly concerning in rapidly urbanizing cities, such as Shenzhen, a representative megacity in Asia, where high-density old districts on the urban fringe experience significant thermal discomfort.

As a typical product of uneven urban spatial development, urban villages exhibit prominent non-standardized spatial attributes: irregular spatial layout, aging infrastructure, and mixed-function environmental patterns together constitute their core spatial contradictions [2]. In terms of the external environment, these areas are often closely surrounded by new urban areas with high-density development, and their wind environment quality and thermal comfort are obviously degraded by the continuous influence of the urban heat island effect [3]. It is noteworthy that the existing building groups, public spaces, and vegetation configurations are mostly formed based on historical development needs, and this traditional spatial organization pattern is difficult to effectively adapt to the current increasingly severe microclimate change challenges in the context of high-density cities [4].

Furthermore, as architectural entities with profound historical roots [5], urban villages have evolved gradually from naturally grown settlements to integrated components within the broader urban planning framework. The dynamic evolution of their spatial patterns—encompassing building arrangements, road networks, and public spaces—has given rise to complex and fragmented spatial layouts. The spatial structure of these areas serves as a critical determinant of their internal wind environment, with distinct spatial configurations generating unique wind patterns and thermal characteristics. Thus, the systematic analysis and feature extraction of spatial archetypal wind environment features can effectively refine the classification system of wind environments in old urban areas, thereby enhancing the relevance of planning and design strategies.

From the perspective of research content evolution, the field of wind environment simulation has exhibited a pronounced trajectory toward methodological refinement, with notable advancements particularly observed in the study of wind environments and microclimatic characteristics of specific spatial typologies [6]. In the early stage, wind environment simulation research mostly focused on the overall scale of urban climate analysis, with the wind and heat environment as the core, providing theoretical support for regional planning, ventilation corridor construction [7], greenway system design, and urban surface temperature change prediction [8]. With the deepening of the research, wind environment simulation was gradually applied to smaller-scale space types and has covered diverse scenarios such as urban areas [9,10], street spaces [11,12,13], monolithic buildings [14], vegetation configurations [15], and the assessment of extreme climate scenarios [16,17]. Meanwhile, microclimate, as a key factor affecting human thermal comfort [18], pushes the research scope to contract further and the analysis object to be more specific. Areas such as small-volume urban villages [19], historic districts [20], and streets and alleys of traditional settlements [21,22] all present differentiated wind environment characteristics due to their unique spatial patterns. Additionally, scholars such as Shirzadi have focused on the typical spatial type of urban intersections, systematically exploring optimization pathways and characterization methods for their wind environment simulation techniques [23].

From the dimension of research methodology, the existing wind environment research is mainly divided into two types of methodological paths: one is classification modeling based on spatial type modules, and the other is targeted optimization for a single case. The two methods have their own advantages and disadvantages: classification research is mostly based on key parameters such as building density, orientation, layout, height, etc. [24], and the spatial prototype model is constructed through a subjective induction method, which has a certain degree of universality, but its over-qualitative treatment makes it difficult to cope with the complex and changing spatial patterns in the actual project [25]. In contrast, case studies, rooted in in situ analysis, provide specific optimization suggestions but often lack generalizability due to limited commonality extraction [19,26]. To address these limitations, some scholars have effectively improved the completeness and applicability of their research results by increasing the categorization dimensions [27] or introducing a practical feedback mechanism [28]. For example, Fang H et al. expanded the scope of sampling sites and classified urban village spaces by combining the three dimensions of household density, enclosure degree, and dispersion, systematically exploring the relationship between different spaces and wind environment indicators, which is both empirical and methodological [29].

From the perspective of research object expansion, wind environment research has gradually shifted from independent analysis of single-layout building or vegetation elements to multi-factor coupling and holistic analysis. It is worth noting that some studies have adopted the conditional qualification method to categorize and simulate single elements [29], but it is easy to ignore the potential influence of other spatial factors. In fact, the layout and morphology of buildings, streets, and vegetation, as the core elements of the spatial structure of the site [30,31], show synergistic effects of their attributes in shaping the wind environment and thermal comfort. Although single-element studies can reveal the relationship mechanism between specific variables and wind environment characteristics more clearly, their explanatory power in complex sites is susceptible to multi-factor interference; while multi-factor classification and holistic analysis face the challenge of cross-interference of variables, their conclusions are closer to the actual spatial context, and their theoretical explanatory power and practical guidance significance are more prominent. For example, Hu et al. comprehensively considered the three dimensions of neighborhoods, streets, and buildings in their research [28]; Zhu et al. developed a multi-scale simulation and analysis framework integrating macro, meso, and micro scales from the perspectives of wind, thermal, and acoustic environments [32].

In terms of methods and tools, CFD (computational fluid dynamics) simulation technology is a numerical modeling method for analyzing wind environments [6,33]. Compared to traditional wind tunnel experiments, it has a more significant advantage in terms of convenience and intuitiveness and is widely used in urban planning and architectural design. The current mainstream tools include ANSYS Fluent 2022 R2, Phoenics 2022, Ladybug 1.5.0, and ClimateStudio 1.9.x. Among these, ANSYS Fluent is often used for mechanical engineering simulations, while Phoenics is more suitable for analyzing the wind environment outside buildings and the wind flow in urban canyons [34], and it supports modeling complex buildings. It can also be used in conjunction with ArcGIS [1], remote sensing [35], and spatial syntax analysis [34], significantly enhancing the ability to assess multiple scales of space.

Therefore, this study centers on high-density urban villages, taking Shenzhen as a case study, and employs Phoenics wind simulation technology to derive regional summer wind environment simulation results. A comprehensive qualitative research approach is proposed, encompassing three hierarchical levels: spatial prototype generalization, comparative analysis of wind environment characteristics, and classification-based strategy formulation. This study provides theoretical and practical value for understanding how to use spatial configurations to improve the thermal comfort and sustainability of similarly structured urban areas and offers targeted intervention pathways for promoting sustainable design, enhancing livability, and strengthening climate resilience in high-density urban village environments.

2. Wind Environment Characteristics of Urban Villages in Shenzhen

2.1. Macroclimatic Characteristics in Shenzhen

As one of the four central cities in the Guangdong–Hong Kong–Macao Greater Bay Area and with a longitude ranging from 113.53° E to 114.52° E and a latitude ranging from 22.25° N to 22.75° N, Shenzhen is characterized by an important geographic location and numerous ports of entry. Since the 1980s, its accelerated industrialization and urbanization have made it a symbol of China’s reform and opening up. By 2019, Shenzhen’s built-up area had reached 927.96 square kilometers, with a resident population of 13,438,800 and an average daily population of 21,830,000, with a large portion of the population located in Bao’an District [36]. However, given this high-density population, heat generated by human activities, such as urban construction and associated atmospheric pollution, has intensified the deterioration of the wind environment and enhanced the heat island effect in urban areas [37].

Shenzhen has a subtropical oceanic monsoon climate with abundant precipitation; its rainy season occurs from April to September each year, and its total annual rainfall is approximately 2000 mm. In summer, under the influence of subtropical high-pressure systems, hot weather is common, and the hottest month is generally July, with the highest temperature in history reaching 38.7 °C; therefore, the area is highly susceptible to the effects of extremely hot weather. Under the influence of monsoons, southeasterly winds prevail in Shenzhen in summer and are often influenced by low-pressure and tropical cyclone systems, with high temperatures and heavy rainfall (Figure 1).

2.2. Study Area and Wind Environment Characteristics in Urban Villages

2.2.1. Study Area

The study area is located in the Xixiang Street region of Bao’an District, Shenzhen (Figure 2), and is strategically situated within a regional ventilation corridor that remains largely unaffected by mountain-induced shading. This district originated as a settlement with villages and townships on the basis of historical urban thermal comfort needs rather than natural urban development patterns. Since urban expansion accelerated in 1978, suburban construction has gradually enveloped the district, and the district reached 100% urbanization in 2000; however, older buildings within the core area have been largely preserved.

The spatial organization of the area includes an increase in building density and age from the periphery toward the interior. Field studies and data collection indicate that most buildings were constructed between 1990 and 2010; these buildings predominantly range from mid-rise structures (20–60 m) to low-rise buildings (under 20 m), and high-rise structures are present at the lowest frequency. The majority of buildings are residential, with limited education, recreation, and healthcare facilities; public green spaces; and water systems. The current spatial and environmental configuration of urban villages is characterized by high-density development, inadequate ventilation, and limited landscape quality; therefore, urban villages face unique challenges in achieving wind and thermal comfort. Poor wind circulation can intensify air pollution and heat island effects because of high population density, whereas inadequate thermal comfort can lead to decreased quality of life and increased energy consumption for cooling and heating due to the consequent need for effective temperature regulation. In this study, wind environment simulation technology and urban morphology classification are used to analyze the wind environment types in the study area; then, the underlying mechanisms are identified, and adaptive optimization strategies are proposed from both architectural and landscape perspectives.

2.2.2. Wind Environment Characteristics in Urban Villages in Shenzhen

The compact architecture and narrow streets common in Shenzhen’s urban villages create a complex and constrained wind environment, hindering natural ventilation and limiting the reflective and absorptive capacity of surfaces and vegetation. While some coastal areas in Shenzhen exhibit favorable ventilation conditions that enhance urban airflow, the specific density and spatial configuration of urban villages hinder the efficacy of these ventilation corridors. High ground friction caused by clusters of densely packed buildings further decreases the wind speed.

At the interfaces between urban villages and adjacent high-rise buildings, abrupt height differences often induce localized strong airflows between buildings. Conversely, in zones with more gradual height variations, airflow rarely reaches ground level, which creates a “roof effect” that traps heat at the pedestrian level and intensifies thermal discomfort. Addressing these challenges requires tailored interventions to improve both the microclimate and overall wind comfort within these unique high-density urban spaces.

3. Materials and Methods

In this study, computational fluid dynamics (CFD) technology is employed to analyze the wind environment of urban villages in Shenzhen. CFD is chosen for its ability to solve fluid dynamics problems numerically and its successful applications in various fields. The methodology was selected with the aim of providing data support for the design of transformation strategies by simulating and evaluating the wind environment. Data collection and processing, model data preparation, and detailed analysis methods were undertaken to ensure accurate and reliable simulation results (Figure 3).

3.1. Computational Fluid Dynamics (CFD) Technology

In 1933, British scientist George Templeton advanced the field of CFD by employing numerical methods to solve two-dimensional partial differential equations for viscous fluids. CFD encompasses a broad spectrum of computational techniques and fluid dynamics principles. Tools such as Phoenics and Fluent, among others [38], are commonly used in the field. Currently, the application of CFD technology is rapidly expanding into various domains, such as indoor and outdoor ventilation, urban planning, building fire prevention, and pollutant dispersion, and remarkable success across these fields has been attained.

In this work, we select the wind environment of urban villages in Shenzhen as the research object and use Phoenics 2019 (CHAM Ltd., London, UK) and ArcGIS 10.8 (Esri, Redlands, CA, USA) to analyze the typical spatial wind environment and provide data support for the design of transformation strategies.

3.2. Data Sources and Processing

3.2.1. Climate Data

This study is based on the wind direction statistics from June to August over the past four years, obtained from the World Weather Online platform (https://www.worldweatheronline.com/, accessed on 12 February 2024) and the Meteorological Bureau of Shenzhen Municipality (https://weather.sz.gov.cn, accessed on 20 February 2024). The analysis indicates that the prevailing summer wind direction in the study area is southeast, with an average wind speed of approximately 2.6 m/s and an average temperature of around 29 °C.

3.2.2. Model Data

The site model data were obtained from BIGEMAP 30.0.0.0 and imported into the ArcGIS platform, where they were verified and refined using high-resolution satellite imagery. Based on this, spatial models were constructed with the aid of 3D modeling software such as Blender 4.3.0 and SketchUp 2022 in preparation for subsequent wind environment simulations using the Phoenics platform.

3.3. Analysis Methods and Procedures

3.3.1. Area Modeling

To ensure a certain range of simulation and to prevent the boundary from being too small and having an effect on the interior, the simulation domain was carefully selected to be five times the length and width of the selected site, and the height of the simulation domain was set to be approximately three times the height of the tallest building in the area. This approach helps mitigate the impact of the boundary conditions on the internal wind environment simulation results.

3.3.2. Regional Wind Environment Formulation

Based on the previously obtained wind environment data, the site-specific wind environment model was constructed using the Wind Attributes module in the simulation software (accessed via Object > New Object > Wind > Attributes). The parameters were configured as follows: the external temperature was set to 29 °C, representing the average summer temperature of the site; the wind speed was set to 2.6 m/s; and the wind direction was defined as S–S–E, corresponding to the prevailing southeast-southerly wind. The profile type was selected as power law, considering the flat terrain in Bao’an District, Shenzhen. The vertical direction was set as Z.

For surface roughness, the effective roughness height was set to Parkland, reflecting the urban context of the site. To ensure unimpeded airflow, Include Open Sky was enabled (Yes), while External Radiative Link was also enabled (Yes), with Texternal adjusted to 0. The Ground Plane was disabled (No), and the WAWP option was enabled (Yes). The reference height was set at 1.5 m, representing the standard pedestrian level for assessing wind comfort [39].

3.3.3. Grid Division

Grid division in the simulation area affects the calculation accuracy, and a greater density of grid cells in the study region correlates with a higher accuracy of the simulation results. In this work, a high-density grid is established in the main study domain and near the ground (a = 1.5 m), a low-density grid approach is used outside the simulation area and is extremely close to and far from the ground (a < 1.5 m or a > 1.5 m), and a gradual change in the division method is adopted to ensure a certain degree of accuracy while increasing the computational speed given the actual wind environment in the study area [40].

3.4. Classification of Wind Environment Suitability

Different factors affect the evaluation of the wind environment in urban villages, and the reasonable selection of various factors affecting the wind environment is conducive to the continuous improvement of wind environment evaluation systems. On the basis of relevant studies, wind speed is classified into 5 levels according to human perception and thermal comfort levels (

Table 1. Wind environment suitability levels.

Level	Wind Speed (m/s)	Suitability Class
1	2.20 < a ≤ 2.82	Very suitable
2	1.88 < a ≤ 2.20; 2.82 < a ≤ 3.45	Appropriate
3	0.63 < a ≤ 1.88; 3.45 < a ≤ 4.39	Generally suitable
4	0.00 ≤ a ≤ 0.63; 4.39 < a ≤ 5.02	Unsuitable

).

4. Results

Based on the established research methodology (Figure 3), the wind environment simulation results for the site were obtained (Figure 4a). The spatial types of the site were systematically categorized (Figure 4b), encompassing typical features of urban villages such as building forms, public space configurations, and vegetation layouts. In addition, the “Image–Analysis–Record Measurement” function in Photoshop 2022 was employed to conduct a quantitative analysis of the simulation results, measuring the area and corresponding proportion of each wind speed category (

Table 2. Results of quantitative statistics on the wind environment of the site.

No.	Wind Speed (m/s)	Area (m2)	Suitability Class	Level	Percentage (%)
A	0.00–0.31	254,321.83	Unsuitable	4	25.49
B	0.31–0.63	201,098.09	Unsuitable	4	20.16
C	0.63–0.94	177,548.83	Generally suitable	3	17.81
D	0.94–1.25	99,046.22	Generally suitable	3	9.93
E	1.25–1.57	72,432.73	Generally suitable	3	7.26
F	1.57–1.88	62,783.2	Generally suitable	3	6.29
G	1.88–2.20	60,045.57	Appropriate	2	6.02
H	2.20–2.51	42,600.91	Very suitable	1	4.27
I	2.51–2.82	16,580.95	Very suitable	1	1.66
J	2.82–3.14	4301.22	Appropriate	2	0.43
K	2.82–3.45	1915.15	Appropriate	2	0.19
L	3.45–3.76	1408.42	Generally suitable	3	0.14
M	3.76–4.08	1600.48	Generally suitable	3	0.16
N	4.08–4.39	1830.51	Generally suitable	3	0.18
O	4.39–4.70	118.27	Unsuitable	4	0.01
P	4.70–5.02	0	Unsuitable	4	0
Public space	—	997,632.38	—	—	100
Total site	—	2,072,889.54	—	—	—

Note: Wind speeds of 1–15 correspond to the data distributions for each respective wind speed category. Total site denotes the entire study area, while public space refers to the study area excluding building footprints. Percentage (%) indicates the proportion of the area within each wind speed category relative to the total public space area.

and Figure 5).

The overall analysis indicates that the wind environment at the site is generally poor, with the area occupied by higher wind speeds decreasing correspondingly. Integrating the suitability classification results reveals that the largest proportion, 45.7%, corresponds to unsuitable areas, followed by relatively suitable areas at 41.8%, while suitable and highly suitable areas account for only 6.6% and 5.9%, respectively. These findings demonstrate the generally poor wind conditions in urban villages. Moreover, comparison of wind speed and occupied area shows that the extent of unsuitable areas associated with low wind speeds far exceeds that related to high wind speeds. This pattern is closely linked to the unique architectural layout of the densely built old city and its complex public space network. Therefore, systematically categorizing spatial prototypes based on wind environment analysis is crucial for understanding wind environment characteristics and guiding microclimate improvements in old urban districts.

Urban spatial morphology is a product of urban spatial interactions and serves as the foundation for deconstructing the spatial characteristics of high-density urban villages. This field of study encompasses morphology analysis, environmental behavior, and political economy research, all of which contribute to a comprehensive understanding of urban form and function [41]. Therefore, building on the aforementioned categories, this study further investigates the relationship between spatial prototypes and the wind environment through a finer spatial classification and analysis of their corresponding wind characteristics.

Overall, among the three types under similar density conditions, the belt extension type generally exhibited better spatial performance in urban villages, whereas the low-density point grouping type presented the most favorable wind conditions. With respect to public space prototypes, wind conditions varied among types, with staggered alleys presenting the poorest ventilation performance, whereas single-sided open parks offered the most suitable wind environment. For architectural space prototypes, building elevation configurations posed certain challenges for wind optimization, whereas vertical space contributed to the formation of unique microclimates in urban villages. The staggered and overlapping features of the fifth elevation provided diverse possibilities for improving local wind environments and thermal comfort. In terms of plant relationship prototypes, moderate vegetation density and height had the potential to regulate microclimatic conditions and support the development of favorable wind environments.

4.1. Building Group Prototypes

According to the different spatial characteristics of building group spaces with different development modes, building group prototypes can be categorized into three forms: encircling diffusion, belt extension, and point arrangement types (

Table 3. Building cluster prototypes and wind environment characteristics in urban villages.

Type 1 Encircling Diffusion		Type 2 Belt Extension		Type 3 Point Arrangement
01 Fan diffusion	Limited circulation in downwind areas; enhanced airflow along streets and in pockets.	01 Riverfront ribbon cluster	Perpendicular buildings block airflow; minimal ventilation within clusters.	01 High-density spot formation	Minimal wind; airflow limited to access points.
02 Perimeter block	Overall wind speed remains low, with slight improvements observed along the site boundaries.	02 Major belt	Wind directed along roads; restricted interior airflow.	02 Low-density point grouping	Enhanced wind reflection and circulation on the windward side.
03 Semi-enclosed perimeter layout	The wind environment around the open side shows notable improvement.	03 Secondary strip belt	Overall low ventilation; slight improvement in street-facing pockets.	03 Strip point groupings	Narrow gaps limit internal airflow.

Note: Dark blue denotes low-wind areas, red–yellow denotes high-wind areas, and green denotes areas with comfortable wind environments; the proposed dominant wind direction is southeast; the site selected is along Xixiang Street in Bao’an District, Shenzhen City, China.

).

4.1.1. Encircling Diffusion Type

Rapid urban expansion presents a pattern of spatial diffusion in urban villages characterized by clusters that radiate outward in a dispersed configuration. This encircling diffusion typically originates from the central areas of urban villages, where building groups form semi-enclosed, outward-spreading clusters. Around natural resources such as lakes, green spaces, and adjacent settlements, the spatial layout extends in a pattern that reflects the characteristics of the surrounding environment. These clusters generally exhibit high overall building density with low-rise, tightly packed structures, and the buildings in these clusters are typically relatively old. Consequently, this spatial arrangement restricts natural airflow, which results in a limited wind environment and confined air circulation within these areas.

4.1.2. Belt Extension Type

Urban belts are often located around main roads, rivers, and other linear features. As development occurs on both sides of these belts, planning often varies, with different building layouts and corridors. Thus, the ventilation direction in belt zones often misaligns with the ventilation corridors in neighboring urban areas, which results in highly variable and often limited ventilation conditions.

4.1.3. Point Arrangement Type

Point arrangements are mostly located within center-periphery road networks and are caused by initial urban construction development. The overall building volume is small, the building arrangement is regular, and the building density is gradually reduced from the inside to the outside of these areas. The road width, building spacing, and building height increase accordingly, but the overall changes in these factors are not large. The density of these spaces is lower than that in central urban areas. Notably, in point arrangement areas, there are gray spaces that are well ventilated, such as street corners and courtyards, and the overall ventilation conditions are generally sufficient.

4.1.4. In-Depth Wind Analysis of Building Group Prototypes

Following the above wind environment analysis of different building layout types, the “Record Measurement” method was further applied to quantify the proportion of areas under various wind speed classes for each spatial configuration (Figure 6). The results indicate that for most spatial types, the wind environment is predominantly characterized by grades A–C, with wind speeds largely concentrated in the 0.00–0.94 m/s range. The major belt (Figure 6e) and low-density point grouping (Figure 6h) exhibit a relatively higher proportion of areas with wind speeds exceeding 0.94 m/s, suggesting more favorable wind conditions resulting from their downwind-oriented strip layouts and lower building density. In contrast, strip point grouping (Figure 6i) demonstrates poor wind performance, with 95.09% of its area experiencing wind speeds below 0.31 m/s and the remaining areas also falling within suboptimal wind conditions.

4.1.5. Architectural Space Prototypes

Architectural spaces in urban villages can be classified into three primary spatial forms on the basis of their structural characteristics: architectural façade type, fifth elevation type, and vertical space type. Each prototype uniquely influences wind flow and thermal comfort in these high-density areas (

Table 4. Architectural spatial archetypes and images of urban villages.

Type 1 Architectural Façade		Type 2 Fifth Elevation		Type 3 Vertical Space
01 Door and window structures		01 Low-rise street level		01 Architectural crevices
02 Complex materials		02 Mid-rise rooftop		02 Architectural corridors
03 Living facilities		03 Side terrace

Note: The site is located in the area of Xixiang Street in Bao’an District, Shenzhen City, China.

).

• Architectural Façade Type

The architectural façade type refers to the varied and complex external surfaces of buildings in urban villages. These façades typically feature a mix of older facilities, diverse door and window styles, and numerous balconies, all of which increase surface roughness. This roughness disrupts wind circulation, traps heat near ground level, and reduces ventilation efficiency. Additionally, the dense visual effect created by these varied façade elements contributes to a psychological perception of increased warmth, which intensifies thermal discomfort. For clarity and consistency, the term architectural façade type is used throughout the manuscript to uniformly describe these external building features and their impact on the microclimate.

• Fifth Elevation Type

In urban villages characterized by high building density and low floor area ratios, the fifth elevation type refers to building rooftops and provides open, unobstructed spaces with minimal shading from surrounding structures. However, these rooftops are often cluttered with equipment, which limits their potential to enhance ventilation. Effective design strategies for these fifth elevations can improve airflow and reduce heat accumulation.

• Vertical Space Type

The vertical space type describes the narrow gaps between closely spaced buildings, which form a network of ventilation seams. These confined spaces typically experience lower wind speeds than main roads or other open spaces but are crucial for internal air circulation within high-density areas. Enhancing the design and spacing of these vertical spaces can improve the airflow within these compact settings.

4.2. Public Space Prototypes

The different spatial characteristics of public spaces can be divided into ventilation components: parks, streets, and corner courtyards (

Table 5. Archetypes, characteristics, and forms of public spaces in urban villages.

Archetypes	Characteristics	Forms
Type 1 Parks	Enhanced internal airflow; shading limits climate impacts on surrounding areas	Low-density green space
		High-density yard areas
		Single-sided open parks
Type 2 Streets	Narrow, sheltered; dense facades restrict airflow	Staggered alleys
		Crossroad alleys
		Bilateral alleys
Type 3 Corner courtyards	Limited climate control; moderate airflow advantages	Internal courtyards
		Street-side courtyards
		Corner spaces

and

Table 6. Public space archetypes and wind environment characteristics in urban villages.

Type 1 Parks		Type 2 Streets		Type 3 Corner Courtyards
01 Low-density green space	(1) Overall well-ventilated environment. (2) Limited shading from surrounding buildings.	01 Staggered alleys	(1) Streets and alleys are folded. (2) Airflow is impeded.	01 Internal courtyards	(1) Sheltered by dense outer buildings. (2) Near-wind zone.
02 High-density yards areas	(1) High-density surrounding buildings. (2) Open spaces with high winds.	02 Crossroad alleyways	(1) Few shaded roads and with openings at crossroads. (2) Improved wind environment at local street corners.	02 Street-side courtyards	(1) Embedded within building zones. (2) Often aligned with the dominant wind direction. (3) Wind often guided through these areas.
03 Single-sided open parks	(1) Openings on the windward side. (2) An overall suitable wind environment.	03 Bilateral alleyways	(1) Unidirectional roads. (2) Dense and cluttered buildings on both sides. (3) Poor overall ventilation.	03 Corner Spaces	(1) Complicated wind environment. (2) Good overall wind environment. (3) High winds often form.

Note: Dark blue indicates low-wind areas, red to yellow denotes high-wind areas, and green denotes areas with comfortable wind environments; the proposed dominant wind direction is southeast; the study site is located in the area of Xixiang Street in Bao’an District, Shenzhen City, China.

).

4.2.1. Park Type

Fields and parks are generally central activity areas in urban villages and are located in inner urban zones. Features of this type also include irregular green spaces, lakes, sports fields, flat dams, and open spaces, among others, with buildings adjacent to the periphery and limited ecological elements. Owing to the relative openness of these urban areas, the wind flow conditions are generally satisfactory; thus, these areas meet the minimum criteria for effective ventilation and thermal comfort, which are typically defined by wind speeds of at least 1–2 meters per second at the pedestrian level. However, severe shading by buildings at the edges of these areas leads to a moderating effect on the regional climate; therefore, many of these areas cannot be effectively integrated with the local wind corridors, and their influence occurs primarily in downwind areas.

4.2.2. Street Type

Narrow streets and alleys are concentrated mainly in central urban areas; when these narrow streets and alleys are closer to the interior of urban areas, the road structures become more complex. These streets vary in width, and the layout on both sides of the streets can be variable, with limited areas for walking, landscaping, and parking in most cases. Moreover, gray space and transition areas are lacking, and most of these streets are directly adjacent to building façades. In general, street access is sheltered, with narrow passages, high congestion, and poor ventilation.

4.2.3. Corner Courtyard Type

Street corner courtyards are mostly located on both sides of roads, at interchanges, in front of houses, and in other places; they are distributed and scattered, covering a wide area. Ventilation is limited by the narrowness, irregularity, and poor continuity of these spaces. Therefore, this type of space plays a limited role in the regulation of the surrounding climate, but the air circulation here is generally greater than that in other interior urban areas.

4.2.4. In-Depth Wind Analysis of Public Space Prototype

The analysis of public spaces reveals distinct variations in wind conditions. Wind speeds in staggered alleys, bilateral alleyways, and internal courtyards are predominantly concentrated within the low wind speed categories A (0.00–0.31 m/s) and B (0.31–0.63 m/s). Single-sided open parks (Figure 7c) exhibit a relatively large proportion of area falling within the moderately suitable wind speed range G–K (1.88–3.45 m/s), although a small portion remains unsuitable due to excessively high wind speeds. Within single-sided open parks, the areas classified as highly suitable and suitable account for 5.05% and 13.23% of the total, respectively (Figure 7c). Additionally, parts of low-density green space (Figure 7a), crossroad alleyways (Figure 7e), and street-side courtyards (Figure 7h) also fall within the relatively suitable wind speed range.

4.3. Plant Relationship Prototypes

The spatial distribution of plants can be classified on the basis of their density in the urban plane and their vertical height (

Table 7. Planar density and vertical height of vegetation in urban areas and their respective impacts.

Type 1 Planar Density
01 Classification	02 Coverage	03 Example situation	04 Shading effect	05 Transpiration	06 Impact on the wind-heated environment
High density	Greater than 80%		Provides a large area of shade, effectively reducing the ground temperature.	Strong transpiration increases air humidity.	Regulating the temperature of the zone wind environment also has the potential to block the flow of natural air.
Medium density	50–80%		Provides an appropriate amount of shade while allowing some sunlight to reach the ground.	Transpiration is moderate and helps regulate local climate, but not as significantly as high-density vegetation.	It has a certain effect on guiding and changing the wind direction.
Low density	Below 50%		Provides less shade, with limited effect on reducing ground temperatures.	Transpiration is relatively weak, contributing minimally to increasing air humidity.	It has a small impact on wind speed but has a moderating effect on the comfort of the wind environment in the surrounding residential areas.
Type2 Vertical Height
01 Classification	02 Height	03 Example situation	04 Vegetation type	05 Function	06 Impact
High-level vegetation	Taller than 5 meters		Tall trees	Forms a cool canopy layer to provide extensive shade.	The canopy layer can significantly reduce the temperature of the area below, minimize heat accumulation, and provide a cool environment for people.
Middle-level vegetation	Between 1 and 5 meters		Medium-height shrubs and smaller trees	Provides shaded areas, blocking some direct sunlight from reaching the ground and buildings.	Reduce the heat load from direct sunlight, lower the temperature of the surrounding environment, and enhance the comfort of the shaded areas.
Low-level vegetation	Below 1 meter		Ground cover plants, herbaceous plants, and low shrubs	Provide ground coverage to reduce the amount of heat reflected from the ground.	By the blocking of direct sunlight, the surface temperature is reduced, and the transfer of heat to the surrounding environment is decreased, thereby improving thermal comfort.

). Reasonable planning of the horizontal density and vertical height of plants can effectively mitigate the UHI effect and improve residents’ quality of life. For example, a horizontal density that ensures at least 30% green cover in urban areas can significantly contribute to mitigating the UHI effect. Vertically, plants of varying heights can provide different levels of shade. Foliage shields buildings from solar radiation and wind [42]. The canopies formed by tall trees provide extensive shaded areas and help reduce the air temperature in these areas, whereas shorter shrubs and ground cover plants can decrease the heat absorbed from paved surfaces and thereby lower the overall temperature of the surrounding environment.

4.3.1. Planar Density

The distribution of vegetation with different planar densities creates shaded areas of varying sizes. These shaded areas reduce the direct sunlight on the ground and building surfaces to varying degrees; thus, the amount of heat absorbed by these surfaces is regulated, and the surface temperatures change. This temperature regulation is particularly important for mitigating the UHI effect since buildings and hard surfaces in cities absorb and reradiate a significant amount of heat. The presence of vegetation can effectively reduce the temperatures of these surfaces and create a cooler and more comfortable living environment for urban residents.

4.3.2. Vertical Height

Vegetation at different vertical heights can form a multilayered greening structure. Tall vegetation, such as trees, can provide large areas of shade, whereas low-level vegetation, such as shrubs and herbaceous plants, can fill in spaces and reduce ground-level heat radiation. This multilayered greening structure helps create a microclimate, regulates the local temperature and humidity, and enhances the thermal comfort of residents.

4.3.3. Comparative Analysis of Wind Environments

This study conducted separate wind simulation experiments on selected spatial prototypes featuring varying vegetation densities and vertical heights. For each prototype, comparative analyses were performed between vegetated and unvegetated scenarios to investigate the influence of vegetation characteristics on the wind environment (

Table 8. Impact of vegetation density and height on urban wind.

Type 1 Planar Density
High Density		Medium Density		Low Density
01 Layout features	04 Comparison	01 Layout features	04 Comparison	01 Layout features	04 Comparison
	(1) High vegetation density can significantly obstruct wind flow, thereby exacerbating the deterioration of the wind environment within settlements.		(1) Medium-density plantings have a limited impact on wind speed in relatively wide lanes. (2) In narrower areas, vegetation can obstruct wind flow. (3) In confined spaces, vegetation may partially alter the direction of wind flow.		(1) Low-density spot planting can help regulate wind speed and alter flow direction in specific areas. (2) It also contributes to mitigating high-velocity wind phenomena induced by building layout.
02 Vegetated		02 Vegetated		02 Vegetated
03 Non-vegetated		03 Non-vegetated		03 Non-vegetated
Type 2 Vertical Height
High-Level Vegetation		Middle-Level Vegetation		Low-Level Vegetation
01 Layout features	04 Comparison	01 Layout features	04 Comparison	01 Layout features	04 Comparison
	(1) The elevated vegetation zone along the lakefront exerts a slight decelerating effect on wind speeds across the roadway; however, its influence on the wind environment surrounding buildings located farther inland is minimal to negligible.		(1) Vegetation of medium height situated in relatively open areas exerts minimal influence on wind speed. (2) Conversely, vegetation surrounding the perimeter of adjacent buildings facilitates the formation of an extended area of reduced wind velocity on one side.		(1) Lower vegetation belts in the landscape can moderately enhance regional wind speeds.
02 Vegetated		02 Vegetated		02 Vegetated
03 Non-vegetated		03 Non-vegetated		03 Non-vegetated

Note: The six spatial prototypes are all derived from the site, so the single relationship between the different densities and heights of the vegetation layout and the wind environment of the current high-density old area is explored more here.

). The results indicate that, in terms of horizontal layout, excessively high vegetation density can significantly impede air circulation, whereas moderate vegetation density or isolated plants can effectively guide wind direction within confined spaces or mitigate elevated wind speeds caused by canyon effects. Regarding vertical structure, vegetation of medium height notably reduces wind speed along roadways, while lower shrub belts tend to uniformly increase wind speed in the same areas.

5. Discussion

5.1. Planning and Renovation Strategies

The form of the urban layout plays a crucial role in influencing the wind environment. The layout should be designed with the goal of “overall first, then local,” and the transformations are divided into three levels: the community, residential units, and building units. This three-pronged approach is used to reduce wind loss, guide winds into windless areas, reduce the temperature of the building surfaces, and adjust the building layout, building height, building density, and building form to increase the comfort level of the wind environment and improve the internal thermal environment in urban areas (

Table 9. Specific strategies and regulatory roles in the planning and rehabilitation of high-density urban villages.

Levels of Strategy	Strategic Approaches	Specific Practices	Thermal Comfort Improvement
01 Neighborhood Level	(1) Rationally use terrain.	Expand the ventilation corridors in the windward direction, leveraging terrain and natural features.	Mitigate the overall UHI effect in the area.
	(2) Demolish old and outdated buildings.	Demolish old buildings and building features located in potential wind corridors.
	(3) Expand the width of the ventilation corridors.	Expand the width of the ventilation corridors in the direction of the prevailing wind.
	(4) Adjust the building density.	Adjust the high-density, low-rise layout to a high-rise, low-density layout.
02 Residential Unit Level	(1) Optimize the building layout structure.	Modify the row layout for buildings that border open areas and stagger the layout for adjacent buildings.	Enhance the heat dissipation in residential units. Improve the internal thermal environment of buildings.
	(2) Optimize the building layout orientation.	Reasonably design the building layout orientation according to the dominant wind direction.
	(3) Increase the size and number of public spaces.	Remove some buildings in high-density areas and increase point-based open spaces.
03 Building Unit Level	(1) Optimize the building materials.	Optimize the surface of the building materials to reduce the surface roughness.	Reduce temperatures on the building surface.
	(2) Optimize the building form.	Adopt more streamlined designs.
	(3) Enhance the building ventilation.	Install ventilation on elevated floors to guide wind flows through the buildings.
	(4) Optimize the building design.	Ensure that the building podium overlaps with the main body of the building.
	(5) Increase greenery on the building surfaces.	Increase the amount of greenery on rooftops or facades of buildings.

Note: Strategy illustrations were created by the author.

).

In this context, the main methods and measures are as follows: (1) community-wide renovation—removing obstacles, increasing the width of ventilation corridors, and providing reasonable access to certain point spaces; (2) residential unit renovation—reducing the building density, adjusting the building layout, and adopting stepped treatment at the top of buildings; and (3) single-unit building renovation—establishing buildings of appropriate heights, reasonably utilizing rooftops, optimizing the material used for façades, and streamlining corner design. These factors contribute to improved ventilation, reduced surface layer temperature of buildings, limited wind speed reductions, and the attainment of a comfortable wind environment.

The supplementary and contingency measures mainly include (1) optimizing the ventilation paths, (2) reasonably using the topography to further guide the wind flow direction, (3) establishing greening to supplement the bare areas between buildings, and (4) adopting an urban structure that maximizes wind comfort and uses a vignette-based structure to adjust the wind pressure.

5.2. Landscape Intervention Measures

Blue–green spaces refer to spaces composed of water bodies and vegetation provided by natural ecosystems [43]; the construction of these spaces is key to developing urban ecological systems and is highly important for improving the thermal environment in urban villages. The adjustment from the overall pattern to specific neighborhood layouts is represented in the construction of a three-dimensional grid of ecological corridors at the macro level and in the optimization of the spatial configurations at the micro level to fully exploit the cooling benefits of plants and water bodies (

Table 10. Specific landscape intervention strategies and moderating effects in high-density urban villages.

Main Retrofit Strategies	Specific Measures	Wind Environment Regulation	Thermal Comfort Improvement
1. Regional planning level
Optimizing ecological nodes	(1) Restore environmental quality at damaged ecological nodes.	(1) Optimize the local wind environment.	(1) Isolating heat islands, promoting temperature exchange between the internal spaces and localized “cold islands” in the peripheral areas, and improving the overall thermal environment within the urban areas.
	(2) Establish key ecological nodes such as eco-parks, nature reserves, and nature conservation buffer zones.
Constructing ecological corridors	(1) Optimize the layout of the green spaces in the direction of the unventilated corridors and increase the vegetation coverage to the greatest extent.	(2) Ensure the directional continuity of the ventilation corridors to support cold air circulation.
	(2) Improve river channels, integrate shoreline resources, and increase the width of greenways.
Build an ecological network	(1) Enhance the areas of ecological corridors and the transitions between ecological corridors.	(3) Form an overall ventilation network.
	(2) Enhance the penetration and connection of ecological corridors and the transitions between corridors.
	(3) Strengthen the green infrastructure and ecological isolation zones.
2. Neighborhood improvement level
Community parks	(1) Support reasonable, dense planting.	(1) Increase the areas of shaded surfaces. (2) Increase the water content of the air.	(1) Utilize the shaded surfaces of plants and buildings to enhance thermal comfort. (2) Utilize the evaporation of water vapor to lower the ambient temperature.
	(2) Set up pavilions, umbrellas, and rest areas and plant tall shade trees.
	(3) Conduct artificial cooling with water from sprinkler irrigation.
Waterfront green spaces	(1) Utilize terrain to guide wind direction.	(1) Direct winds to increase wind flows.	(1) Reduce the area of sunlight absorption, reduce the temperature of water surfaces, and better utilize thermal regulation by water bodies.
	(2) Establish tall plants along rivers.
	(3) Build ecological green islands.
Street spaces	(1) Incorporate green concepts into the design of street layouts; ecological planting in public areas.	(1) Direct winds to create neighborhood cold corridors. (2) Utilize the moderating effect of plants to create localized microclimates.	(1) Enhance the thermal comfort of the travel corridors and recreational areas.
	(2) Green the facades of buildings and plant trees along streets.
	(3) Avoid bare and impermeable land areas.
	(4) Adopt specific layouts for leisure, sport, and fitness facilities, surrounded by plants.
	(5) Reasonably arrange bus stops and rest areas.
Pocket green spaces	(1) Establish a regular, block-based layout of small green spaces.	(1) Create localized islands. (2) Regulate the circulation of cold air.	(1) Reduce building temperatures. (2) Build cool spaces.
	(2) Plant tall shade trees.
	(3) Add water features.
	(4) Install cooling pavement.
Rooftop green spaces	(1) Roof gardens.
	(2) Rooftop farms.
	(3) Eco-paving.

Note: Strategy illustrations were created by the author.

).

In this context, the main approaches and measures are as follows: (1) at the regional planning level, optimizing the native ecology at given points, strengthening the connections between the points, and establishing a regional ecological network through eco-acupuncture need to be emphasized, and (2) at the neighborhood level, the areas of vegetation and water bodies need to be increased, the layout should be rationalized, the role of natural elements in the regulation of the climate needs to be fully understood, and the local microclimatic environment needs to be artificially improved by applying surface shading and artificial humidification methods.

The supplementary and contingency measures mainly include the following: (1) introduce interactive water features in public spaces for regular cooling; (2) establish a multilevel greening system to create clear wind paths, change the wind direction, and provide shade using the joint effects of plants at different levels; (3) use permeable and low-reflectivity sustainable paving materials to create resilient spaces; (4) establish green belts to guide wind flows between high-density built-up areas and natural areas to promote airflow and circulation; and (5) improve nighttime cooling systems by installing automatic nighttime sprinkler systems in neighborhoods and public spaces to promote evaporation for the mitigation of heat accumulation during the day.

5.3. Specific Associations Between Spatial Prototypes and Wind Environments

The classification of spatial prototypes used as the main framework in this study effectively reveals the relationships between specific layout forms and wind environments and aligns well with existing generalized classifications [29] and comparative studies [28]. Previous research on building classification has focused on differences in thermal environments across varying building densities and heights [35] but has rarely provided detailed categorizations on the basis of spatial morphology, such as street width or block size. Moreover, urban space and microclimate studies have shown that in addition to the planar layout of residential clusters, building attributes—including orientation, arrangement, massing, openness, and shading—significantly influence local wind conditions [28], which is consistent with the building-type classifications in this study of urban villages. Moreover, vegetation, as a critical element of urban residual spaces, plays a vital role in modulating the wind environment. Some studies have demonstrated that a rational vegetation configuration is key to improving airflow, and an optimized green space layout can alleviate the population’s experiential demands [1]. Given the complex living environment in urban villages, enhancing green spaces can provide more comfortable conditions, and the landscape vegetation strategies discussed herein offer relevant guidance.

5.4. Limitations and Future Research Directions

In this study, simulated data based on the climatic and urban conditions of Shenzhen’s high-density villages are used; because of this, the generalizability of this method is likely limited to this climate type and may not be applicable to other regions with different climatic or urban characteristics. While the simulation provides valuable insights into urban wind dynamics, empirical studies involving real-world data would be beneficial for validating and refining these strategies across longer timescales and in varied environmental contexts. Importantly, key thermal factors, such as the sky view factor, albedo, and surface heat storage, were not considered in the current simulations, which limits the scope of the thermal modeling. Future researchers could modify and test this model in diverse climatic zones by incorporating additional factors that may influence thermal comfort to assess its broader applicability and explore how local environmental factors may influence the effectiveness of spatial and landscape interventions.

Moreover, in this study, advanced simulation tools capable of capturing more complex interactions between urban structures and dynamic microclimates were not used, and while GIS and PHOENICS were applied to simulate the urban wind environment, their use presents specific limitations. The GIS-based modeling process considered only buildings and excluded vegetation, which can significantly affect airflow in high-density settings. Furthermore, the building height and location data were not field verified, which may introduce spatial inaccuracies. These issues may lead to localized deviations in the CFD simulation results, especially in narrow alleyways or compact street canyons. The development and use of more sophisticated modeling technologies could enable a finer-grained understanding of airflow and heat distribution patterns in dense urban areas. Additional research could also examine the socioeconomic dimensions of implementing these interventions and evaluate both the feasibility and the potential impact of the proposed landscape and design strategies within various urban contexts.

When this study’s methods and findings are applied to other urban contexts, in addition to local climatic differences, it is necessary to classify and analyze the site layout and building morphology. Urban typologies such as historic districts and mountainous communities have distinct spatial and wind environment characteristics. Therefore, generalization requires the integration of climate factors with these typologies to develop adaptive, context-specific design strategies.

6. Conclusions

This study focused on optimizing the quality of public space in high-density old districts at the urban fringe of Asian megacities, taking Shenzhen as a representative case. Through the integration of wind simulation, geographic information systems (GIS), and spatial prototype analysis, we explored how variations in spatial configuration affect microclimatic conditions and urban livability, with the aim to offer applicable strategies for similar compact urban environments.

The research yielded several key findings:

• Spatial configurations—including building density, height distribution, orientation, and green space integration—are essential factors that influence thermal comfort and wind environment performance.
• Reducing the building density, adjusting the height distribution, and embedding vegetation improved ventilation and thermal comfort. Among prototypes, belt extension, low-density point grouping, and single-sided open parks were most effective for wind environment optimization.
• Radial spatial layouts can effectively channel prevailing winds into dense urban interiors, thereby increasing airflow and reducing heat accumulation.
• Green infrastructure elements, such as green roofs and vertical greening, significantly alleviate the urban heat island (UHI) effect.
• The combined application of GIS and wind simulation tools at early planning stages provides effective support for spatial decision-making and environmental performance evaluation.

Overall, this study provides a transferable qualitative framework that informs spatial planning and public space improvement in high-density urban villages. Aligning spatial form optimization with environmental mechanisms, such as wind flow and heat regulation, contributes to adaptive urban design strategies aimed at improving both thermal resilience and the overall quality of urban life. The results offer theoretical guidance and practical reference for urban renewal in rapidly urbanizing contexts, especially across Asia.