A Comparative Study of Outdoor Thermal Comfort in Centralized Traditional Organic and Modern Standardized Rural Settlements

Abstract

Global warming has significantly intensified the risks of summer heatwaves, making outdoor thermal comfort during extreme heat periods a critical research focus. Under centralized rural village reconstruction policies, traditional settlements are being replaced by regularized modern communities characterized by new materials and standardized layouts. However, the impact of these morphological transitions on the micro-scale thermal environment remains under-researched, with a notable lack of comparative perspectives between traditional organic and modern standardized typologies. This study identifies six representative zones based on spatial configuration. By integrating UAV photogrammetry (Pix4Dmapper v4.5), AutoCAD 2019, and QGIS (v3.22), morphological characteristics were quantified, followed by microclimate simulations using ENVI-met v5.9. The results reveal that while peak daytime Physiological Equivalent Temperature (PET) in the standardized zones (49.2–51.8 °C) is slightly lower than in traditional zones (53.5–55.2 °C), a phenomenon of thermal homogenization emerges in the former. Specifically, values in standardized zones are highly concentrated around the median (53.5 °C), contributing to a significant upward trend in the minimum PET values, with nearly all sampling points exceeding 47.0 °C. Quantitative analysis identifies green coverage and perviousness as primary cooling drivers, while spatial openness and imperviousness promote thermal homogenization. In contrast, traditional zones retain critical cool refuges due to their spatial heterogeneity. This research provides an empirical foundation and quantitative reference for understanding the thermal performance differences across different rural spatial typologies. The findings offer insights for planners to optimize street layouts and shading strategies, ultimately mitigating heat stress and fostering climate-resilient modern countryside development.

1. Introduction

The combination of global climate change and extreme weather has greatly increased the risk of summer heatwaves [1]. According to the IPCC’s sixth assessment report, extreme heat has become a common climate disaster [2], directly threatening human health and causing heat-related diseases [3,4]. While most studies focus on urban areas [5], high-density concentrated rural settlements are widely found in northern China [6]. These settlements have compact buildings and narrow street canyons rather than scattered layouts, showing heat accumulation similar to cities [7]. Since the permanent rural population mostly consists of vulnerable groups (such as the elderly and children), whose ability to adapt to extreme weather is weaker than that of urban residents [8], studying the rural thermal environment has demonstrated a clear necessity for public health.

Driven by the Rural Revitalization policy, many villages have changed from traditional organic villages to modern standardized communities for better living conditions [9]. This process has led to major changes in the rural physical environment [10]. In terms of geometry, traditional layouts were replaced by regular grids [11,12], which changed the Sky View Factor (SVF), building heights, and Aspect ratio (H/W). At the same time, the underlying surfaces changed from soil or natural materials to modern concrete and pavements that drastically change the thermal storage capacity of settlements [13,14], leading to a large increase in the Impervious Surface Fraction (ISF). While reconstruction has improved the infrastructure [15] and building quality [16], it is still not clear how this “regular” spatial change affects rural microclimates.

Spatial morphology is the core variable that controls radiation gain and heat diffusion. Its influence mechanism mainly involves two dimensions, namely, radiation shading and heat dissipation obstruction [17,18]. To standardize the description of morphological effects on microclimate, Stewart and Oke [19] proposed the Local Climate Zone (LCZ) system, classifying surfaces into 17 types based on building density, height, and materials. Currently, the LCZ system is widely used at meso-scales, often combined with satellite data to study Land Surface Temperature (LST). For example, research in Xi’an, China, has used LCZ to show that Surface Urban Heat Island Intensity (SUHII) peaks at noon [20], while a study in Guangzhou showed that Pervious Surface Fraction (PSF) was the dominant factor affecting high-temperature risks, followed by parameters like Roughness Element Height (REH) [21]. Micro-scale studies focus on street canyon morphology, mainly using field measurements [22] and high-resolution simulations [23] (such as ENVI-met) to explore how physical parameters contribute to human thermal comfort [24,25,26]. Key morphological parameters include H/W [27], building orientation, and SVF, which directly affect Mean Radiant Temperature (T_mrt) and wind speed by controlling shadow coverage [28]. Research in Malaysia showed that specific asymmetrical H/W ratios can effectively reduce heat island effects [29], while a study in Nanjing emphasized the role of block morphology in regulating solar radiation distribution [30]. Besides geometry, the thermal properties of materials (such as albedo and heat capacity) and greenery are also critical. High-albedo materials have been proven to reduce building heat loads [31], while high vegetation coverage (green infrastructure) enhances cooling through shading and transpiration [32].

To quantify these impacts on humans, researchers use various indicators based on the energy balance. Fanger [33] established the Predicted Mean Vote (PMV) model to quantify subjective perception, which was later expanded by Nikolopoulou et al. [34] through the thermal adaptation theory to include psychological adjustments. For physical modeling, Gagge [35] introduced the Standard Effective Temperature (SET) based on a two-node model, while Jendritzky et al. [36,37] developed the Universal Thermal Climate Index (UTCI) to improve simulation accuracy for extreme environments. Among these, the Physiological Equivalent Temperature (PET) developed by Höppe [38,39] has become the most widely applied metric in outdoor studies. Its output in degrees Celsius (°C) facilitates direct comparison with air temperature, making it a standard tool in extensive empirical research across various climate zones [40,41,42,43].

Recently, scholars have focused on the thermal characteristics of traditional dwellings and settlements [44]. Simulations by Xin et al. [45] showed that in rural settlements in Shanxi, as artificial surfaces increased, the PET index rose by 27.43% during the day and 34.03% at night. Research by Henna et al. [46] in different Indian climate zones showed that vernacular settlements often have higher resilience to climate change compared to modern reconstructions. Gado et al. [47] conducted field measurements on the ventilation of traditional villages in the Dakhla region, demonstrating that the unique spatial design strategies of vernacular settlements provide an effective passive response to hot climatic conditions. However, most existing rural studies focus on cross-sectional comparisons of different regions or individual dwellings at single time points. There remains a critical lack of comparative research that explores the distinct thermal characteristics of different settlement types. More importantly, how standardized layouts impact spatial thermal patterns—specifically the stability of heat distribution and the shrinking of effective cool refuges—remains a significant gap in current rural microclimate literature.

This study investigates the distinct microclimate characteristics of different rural spatial typologies by comparing the thermal performance of traditional organic and standardized settlements under varying morphological parameters. The study selects Dabudi (traditional organic) and Hangtou (modern standardized) as typical cases. Representative zones were identified based on spatial layout, and a systematic comparative analysis was conducted using field measurements, weather station data, and high-resolution ENVI-met simulations. To accurately analyze the quantitative relationship between morphology and PET, this study refers to the LCZ classification logic and develops a morphological characteristic scale for small-scale rural spaces in Section 2.2. By selecting and quantifying key indicators. Building on this, the study focuses on deeply exploring spatial thermal patterns, especially the shrinking of effective cool refuges. Through a combined analysis of PET spatial distribution maps and box plots, the study discusses the fundamental differences in thermal pattern stability between modern standardized and traditional organic zones.

The main contributions of this study include

• Provides a typological comparison of rural thermal characteristics between centralized traditional organic and modern standardized layouts.
• Developing a morphological characteristic scale and a framework to explain the quantitative relationship between micro-morphology and PET in rural areas.
• Identifying the risk of thermal homogenization and the shrinking of effective cool refuges caused by standardized reconstruction.

These results provide empirical data and a quantitative reference for understanding thermal environment changes in centralized rural areas. The developed morphological scale can be used as a tool to evaluate micro-climate performance in similar villages. Furthermore, by identifying the factors that influence thermal comfort, this study offers practical design suggestions for planners—such as optimizing street layouts and shading—to reduce heat stress and improve the climate resilience of modern rural settlements.

2. Methods

2.1. Sites and Case Identification

2.1.1. Geography, Climate, and Reconstruction Policy

Lingyang Town in Juxian County, Shandong Province, was selected as the research site due to its dual representativeness in climatic stress and policy-driven spatial transformation. Located approximately 2 km southeast of the county center (Figure 1a), the region experiences a typical temperate monsoon climate where intense solar radiation and high humidity create severe summer heat stress. According to historical meteorological data, daily maximum temperatures in July and August frequently exceed 30.0 °C, with extreme peaks reaching 38.0 °C, posing significant health risks for outdoor rural activities. Moreover, the implementation of the Centralized Reconstruction Policy—characterized by shared responsibilities between the local government and residents—has resulted in a unique spatial juxtaposition within the town. This policy context provides a “living laboratory” where traditional organic villages and modern standardized communities coexist, facilitating a direct comparative analysis. Specifically, Dabudi (Zone D) (Figure 1b) was selected as a typical traditional organic village, while Hangtou (Zone H) (Figure 1c) represents a modern standardized community, with reconstruction beginning in 2005 and reaching full completion in 2015, ensuring the settlement had a mature 10-year operational history at the time of study.

Dabudi maintains its organic spatial characteristics, featuring a low-density expansion pattern and a high PSF. Buildings are predominantly spontaneous single-story structures made of brick, wood, or earth. Due to the lack of unified planning, street widths are irregular, and paths are winding, creating diverse street canyons with effective self-shading. Surfaces consist mainly of rammed earth, gravel, and permeable bricks, which have low heat capacity and high water permeability. Vegetation is scattered but efficient; many mature deciduous trees (predominantly poplars with a 5–8 year maturity cycle) provide natural shade in summer, effectively blocking solar radiation.

In contrast, Hangtou reflects the standardized features of modern reconstruction, using a rigid grid layout. High-density spaces and hardened surfaces have significantly reshaped the microclimate. Some buildings have increased to three stories, leading to a significant rise in REH. The spatial layout has changed from irregular courtyards to highly regular residential blocks. Materials with high heat capacity, such as asphalt, cement, and concrete, have replaced traditional permeable materials. This has caused a sharp increase in the ISF. Furthermore, the greenery pattern primarily consists of mature peach trees (with a 6–10 year maturity cycle) and decorative shrubs.

2.1.2. Representative Zones and Geometric Characteristics

Following the LCZ classification framework proposed by Stewart and Oke [19], this study selected building height, building density, and underlying surface properties as core indicators for zone identification. These parameters are widely recognized in microclimate research as the primary drivers of radiation exchange and airflow patterns [20]. Specifically, building height and density determine the urban canyon’s roughness and shading capacity [24], while surface properties—particularly the ratio of impervious to pervious materials—dictate the thermal storage and evaporative cooling potential of the settlement [21,32]. By integrating these established criteria with the specific morphological features of centralized rural settlements, six representative zones were identified in Hangtou and Dabudi (Figure 2).

Hangtou, representing the modern standardized community, shows significant spatial homogeneity and regular grid layouts.

• Zone 1 (H1) is defined as a multi-story high-density area, covering 23,904 m² (249 m × 96 m). It mainly consists of 9 m high three-story houses and 3 m high auxiliary courtyards. The area includes three north–south roads of varying widths and four east–west roads of uniform width.
• Zone 2 (H2) is a single-story high-density area with a uniform building height of 6 m. It covers an area of 15,372 m² (183 m × 84 m). The road network consists of two north–south roads of varying widths and four east–west roads of uniform width.
• Zone 3 (H3) maintains the same geometric dimensions, road structure, and building height as H2 but is identified as a single-story low-density area. The main difference is that H3 contains large open spaces covered with shrubs, which significantly reduces the proportion of hard impervious surfaces.

Dabudi, as a traditional organic village, features an organic morphology consisting of irregular road networks and houses of varying scales.

• Zone 1 (D1) is a single-story high-density area composed of 6 m high houses and 3 m high auxiliary rooms. It covers 6240 m² and includes three north–south and three east–west roads with significant imbalances in street width.
• Zone 2 (D2) is classified as a single-story low-density area with a building height of 6 m and a total area of approximately 13,200 m². It contains four north–south and six east–west roads. The road network is highly complex, with large width variations between and within roads, as well as several dead-end paths.
• Zone 3 (D3) is identified as a combination of a single-story high-density area and a public plaza. Due to the constraints of the main village road, it shows an irregular trapezoidal shape with a total area of 7425 m². As a core public gathering space, D3 includes a large activity plaza and other open spaces, with a road network of two north–south and three east–west roads.

2.2. Morphology Data Collection and Processing

The workflow for rural morphological data collection and processing in this study is divided into three phases: high-precision data acquisition, vector modeling, and indicator quantification.

Data acquisition and model construction: Unmanned Aerial Vehicle (UAV) tilt photography was used for data collection, with image processing conducted via Pix4Dmapper (Pix4D SA, Prilly, Switzerland). The resulting Ground Sampling Distance (GSD) was 2.24 cm, and a high-density 3D point cloud was exported using the WGS 84 coordinate system. In the initial processing stage, the point cloud was optimized by removing redundant points and outliers, then imported into AutoCAD 2019 (Autodesk, San Rafael, CA, USA) for detailed vector modeling. During modeling, spatial structures such as residential buildings, auxiliary structures, and courtyards were drawn in separate layers. Different underlying surface materials (e.g., asphalt, sand, and soil) and vegetation were also categorized. The final morphology model provided a consistent dataset for spatial analysis in QGIS v3.22 (Open Source Geospatial Foundation), geometric verification in SketchUp 2019 (Trimble Inc., Sunnyvale, CA, USA), and microclimate simulations in ENVI-met v5.9 (ENVI-met GmbH, Essen, Germany).

Construction of the morphological characteristic scale: Eight spatial indicators were selected to fully characterize the physical form of the study zones (

Table 1. Morphological characteristic indicators for rural communities.

Parameter	Definition	Unit
1. Sky View Factor (SVF)	The ratio of the visible sky hemisphere	-
2. Aspect Ratio (H/W)	Mean building height divided by street/space width	-
3. Building Surface Fraction (BSF)	Area proportion covered by buildings	-
4. Impervious Surface Fraction (ISF)	Proportion of paved/hard surfaces	-
5. Pervious Surface Fraction (PSF)	Proportion of soil/grass permeable surfaces	-
6. Green Coverage Ratio (GCR)	Proportion of vegetation surfaces	-
7. Roughness Element Height (REH)	Average height of dominant roughness elements	m
8. Tree Canopy Cover (TCC)	Horizontal projection of tree crowns over the ground area	-

). The raw data for SVF were simulated using ENVI-met, and the raster data were exported via the Leonardo module for secondary processing in MATLAB R2023b (MathWorks, Natick, MA, USA). To eliminate boundary effect errors during the calculations, MATLAB was used to remove outliers within the buffer zone of the model.

The average SVF for all valid grid points within the study unit was calculated as follows:

$$\overline{SVF}={\displaystyle \frac{1}{n}}\sum _{i=1}^n{SVF}_i$$

(1)

where ${SVF}_i$ is the value of the $i$-th valid grid point, and $n$ is the total number of grid points in the study zone.

The H/W characterizes the geometric enclosure of street spaces. This study used the floor area-weighted method to calculate the average building height. QGIS software was used for centerline extraction (skeletonization) of the road polygons to calculate the average street width [48]. The formulas are as follows:

$$\overline{H}={\displaystyle \frac{{\sum }_{i=1}^n\left(H_i{A}_i\right)}{{\sum }_{i=1}^n{A}_i}}$$

(2)

$$\overline{W}={\displaystyle \frac{A_{road}}{L_{centerline}}}$$

(3)

$${\displaystyle \raisebox{1ex}{$H$}\!\left/ \!\raisebox{-1ex}{$W$}\right.}={\displaystyle \frac{\overline{H}}{\overline{W}}}$$

(4)

where $H_i$ and $A_i$ are the height (m) and footprint area (m²) of the $i$-th building, respectively, and $n$ is the total number of buildings in the zone. $A_{road}$ represents the projected area of road polygons (m²) within the study unit, and $L_{centerline}$ is the total length of the corresponding road centerlines (m).

Building Surface Fraction (BSF), ISF, PSF, and Green Coverage Ratio (GCR) were all extracted from the CAD model. These were quantified as the ratio of the horizontal projection area of each element to the total area of the zone, as shown in Equation (5). REH was calculated as the arithmetic average of all building heights (m) in the zone. The Tree Canopy Cover (TCC) [49] is defined as the ratio of the total tree canopy projection area to the zone area, as shown in Equation (6),

$${\lambda }_x={\displaystyle \frac{\sum A_x}{A_{total}}}$$

(5)

$$TCC={\displaystyle \frac{\sum A_{canopy,i}}{A_{total}}}$$

(6)

where ${\lambda }_x$ represents the fraction of a specific surface (buildings, hard pavement, soil, etc.), $A_x$ is the corresponding area (m²), $A_{total}$ is the total area of the study unit (m²), and $A_{canopy,i}$ is the canopy projection area (m²) of the $i$-th tree.

2.3. Model Configuration and Validation

This study uses ENVI-met v5.9 to conduct high-resolution numerical simulations of the outdoor thermal environment in the six representative zones. 1 August 2025, was selected as a typical summer calculation day, with a total simulation duration of 23 h. The boundary conditions used the Full Forcing mode, with meteorological driving data obtained from the Juxian ground weather station (No. 54936). Air temperature and humidity were measured at a height of 2 m, and wind speed was measured at 10 m. Detailed initial settings and boundary conditions are shown in

Table 2. ENVI-met parameters setting.

	Parameter	Value
Space	Location	Ju Xian
	Position	35.63 °E 118.90 °N
	Time zone	+8
	Model dimensions	H1: L = 382.0 m × D = 151.0 m × H = 25.0  m H2: L = 253.0 m × D = 142.0 m × H = 30.0  m H3: L = 274.0 m × D = 163.0 m × H = 30.0  m D1: L = 147.0 m × D = 86.0 m × H = 30.0  m D2: L = 213.0 m × D = 125.0 m × H = 25.0  m D3: L = 112.0 m × D = 152.0 m × H = 25.0  m
	Size of grid cell in meters	dx = 1 dy = 1 dz = 1
Date and time	Start date	2025.08.01
	Start time	0:00 a.m.
	Total simulation time	23 h
Meteorology	Boundary conditions	Full forcing
	Forcing data	Wind&Air temperature&Clouds&Humidity
	Measurement height	Air temperature and Relative humidity 2.0 m Wind speed and direction 10.0 m
	Roughness length	0.1 m

.

For spatial modeling, a uniform grid standard was used across the six zones. The spatial resolution was set to 1 m to ensure the capture of street-scale microclimate details. The roughness length was uniformly set to 0.1. Specific simulation domain sizes for each zone are detailed in Table 2.

To verify the accuracy of the simulation results, field measurements were conducted simultaneously for the simulation period. The measurement site was located at the center of a north–south street in Dabudi Village, which features typical rural street canyon characteristics. Measured parameters included air temperature (T_a), relative humidity (RH), globe temperature (T_g), and wind speed (V_a). All sensors were installed at a height of 1.5 m above the ground, representing human breathing height. The measurement site setup is shown in Figure 3, and instrument parameters are detailed in

Table 3. Parameters of field measurement instruments.

Device Type	Model	Parameter	Interval	Accuracy
Temperature/RH Logger	HOBO MX2301 and MX1101	Air temperature	5 min	0.01 °C
		Relative humidity	5 min	0.01%
Black-bulb Thermometer	HQZY-1	Black globe temperature	5 min	0.1 °C
Multifunction Meter	KIMO AMI310	Wind speed	1 h	0.1 m/s

.

The T_mrt in the street and courtyard spaces was calculated using the obtained T_a, T_g, and V_a data. According to ASHRAE 2013 [50], the calculation formula is as follows:

$$T_{mrt}={\left[{\left(T_g+273.15\right)}^4+{\displaystyle \frac{1.1\times {10}^8{V}_a^{0.6}}{\epsilon \times D^{0.4}}}\times \left(T_g-T_a\right)\right]}^{0.25}-273.15$$

(7)

where D is the globe diameter in meters, and ε is the globe emissivity.

Validation results (Figure 4) show that the simulated and measured values for T_a were highly consistent (R² = 0.92), indicating that the model accurately captured diurnal temperature fluctuations. The correlation for RH remained stable with R² = 0.87. Regarding the T_g, which is critical for thermal environment assessment, the simulation results showed strong predictive capability with an R² of 0.79. Based on these statistical indicators, the simulation scheme used in this study demonstrates high reliability.

3. Results

3.1. Morphological Characteristics

The quantitative analysis of morphological characteristics for the six zones reveals a significant transition in spatial structure from heterogeneous low-rise forms to homogeneous high-density configurations during the rural reconstruction process (

Table 4. Results of morphological characteristics for the six representative study zones.

Parameter	H1	H2	H3	D1	D2	D3
1. Sky View Factor (SVF)	0.40	0.42	0.48	0.47	0.48	0.59
2. Aspect Ratio (H/W)	1.16	0.63	0.54	0.57	0.79	0.52
3. Building Surface Fraction (BSF)	0.48	0.47	0.33	0.48	0.39	0.35
4. Impervious Surface Fraction (ISF)	0.39	0.38	0.42	0.19	0.12	0.30
5. Pervious Surface Fraction (PSF)	0.13	0.15	0.25	0.32	0.49	0.35
6. Green Coverage Ratio (GCR)	0.13	0.15	0.25	0.29	0.35	0.30
7. Roughness Element Height (REH)	9.55	5.21	5.70	4.56	4.71	4.81
8. Tree Canopy Cover (TCC)	0.13	0.16	0.15	0.12	0.08	0.16

). Regarding geometric parameters, the standardized zone H1, a typical multi-story high-density area, reached a REH of 9.55 m, which is significantly higher than other single-story zones. This increase in vertical dimension directly leads to H1 having the highest H/W and the lowest SVF, showing typical features of a closed street canyon. In contrast, although the traditional zone D3 also has high density, its SVF reaches the maximum value due to the influence of a large public plaza, providing the most open geometry for long-wave radiation emission.

The comparison of underlying surface composition highlights the reshaping of rural surface physical properties by reconstruction policies. Data shows that while the BSF of H1, H2, and D1 remains at similar high levels, there is a huge difference in the ISF. H zones show significant hardening characteristics, with values all above and reaching a peak in H3. In contrast, the number of traditional zones D1 and D2 is only 0.19 and 0.12, respectively. This difference is directly reflected in the PSF; the traditional zone D2 has a PSF as high as 0.49. This is mainly due to the retention of large areas of natural soil and unpaved courtyards, which typically have better water retention and cooling potential.

Regarding vegetation configuration, traditional zones are superior to standardized communities in overall green coverage, but the layout of shade trees follows a different logic. The GCR of D zones ranges from 0.29 to 0.35, generally higher than that of standardized zones. However, the TCC indicators for the H zones and D zones vary. Notably, although the traditional zone D2 has the highest GCR, its TCC is the lowest, indicating that the greenery in this area consists mostly of ground vegetation or scattered shrubs, lacking effective vertical shading. In contrast, while the overall amount of greenery in standardized zones is lower, the TCC shows better spatial consistency due to the standardized planting of street trees in the regular grid layout.

3.2. Temporal Thermal Comfort

Simulation results show that the diurnal evolution of PET is highly consistent across all study zones, reflecting a synchronous response of the rural microclimate to solar radiation intensity (Figure 5). Before 06:00 a.m., PET curves remain low and rise rapidly after sunrise. Around 08:00 a.m., a period of rapid transition begins; the maximum and minimum PET values in each zone start to diverge significantly, marking the point where building shadows and underlying surface materials begin to exert different regulatory effects. Between 10:00 a.m. and 3:00 p.m., PET remains in a high-level plateau, with maximum values fluctuating between 55.0 °C and 60.0 °C. After 4:00 p.m., as the solar altitude angle decreases, data divergence occurs again due to different shading logics. After 18:00, the curves gradually stabilize and drop below 35.0 °C, relieving the thermal stress.

The study areas experience extreme heat stress for long periods during a typical summer day, with both village types showing significant thermal environment degradation. During the daytime, the average PET in all zones exceeds 39.0 °C (the threshold for extreme heat stress) for 8 to 10 h. The peak average PET for each zone occurs between 12:00 p.m. and 2:00 p.m. Specifically, peak values in the modern standardized communities range from 49.2 °C to 51.8 °C, while those in the traditional organic zones are slightly higher, ranging from 53.5 °C to 55.2 °C. This indicates that despite distinct spatial morphologies, intense summer solar radiation poses severe thermal comfort challenges for rural outdoor environments.

As the SVF increases, the maximum average PET in each zone rises significantly. For example, in the traditional village, from the high-density D1 to D3 (which features a large plaza), the maximum average PET increases from 54.1 °C to 55.2 °C. This phenomenon is due to both increased short-wave radiation absorption associated with higher SVF and the properties of the underlying surfaces. Although the D zones have a higher PSF, open spaces without effective tree shading lead to longer durations of extreme heat stress.

A stacked analysis of daytime heat stress levels reveals that the study areas face extremely severe thermal challenges, with high heat stress occupying most of the daytime. Statistical data (Figure 6) show that between 4:00 a.m. and 8:00 p.m., the time proportion in the “Extreme Hot” range for all zones is close to or exceeds 50% of the total duration. Zone D1 performs the worst, with “Extreme Hot” conditions accounting for 57.6% of the time, while “Comfortable” or “Cool” periods are completely absent during the day. In addition to extreme heat, the “Hot” and “Warm” levels account for approximately 20% to 30%, meaning residents are exposed to varying degrees of thermal discomfort risks nearly all day when outdoors.

The thermal performance of the standardized community is slightly better than that of the traditional village, characterized by a reduction in the duration of extreme heat stress and an increase in low-stress periods. The “Extreme Hot” proportion in the H zones are generally lower than in the D zones. For instance, the extreme heat stress proportion in H2 drops to a minimum of 48.5%, which is about 9.1 percentage points lower than in D1. Meanwhile, the proportion of the “Slightly Warm” level increased significantly in the standardized community, reaching 24.2% in both H2 and H3, compared to only 9.1% in D1. This shift in distribution suggests that the regular grid layout and building spacing in the standardized community help mitigate excessive heat accumulation to some extent, providing residents with more time windows of relatively low thermal stress.

3.3. Spatial Thermal Patterns

The actual thermal sensation of residents depends not only on the duration of heat stress but also on the spatial intensity of heat exposure along their activity paths. Therefore, to analyze the distribution of spatial heat exposure, PET maps for the six zones at 08:00 a.m. and 12:00 p.m. are presented in Figure 7.

The study finds that the modern standardized community (Hangtou) exhibits significant thermal pattern stability, while the traditional organic village (Dabudi) shows obvious thermal pattern instability. In Hangtou, the high heat stress areas at 08:00 a.m. remain hotspots at 12:00 p.m. For example, in H1, the 08:00 a.m. PET map shows that heat stress is concentrated inside courtyards and at the southeast corners of intersections; by 12:00 p.m., although overall values increase, these locations remain the core areas with the highest PET. This stability is also observed in H2 and H3. In contrast, the spatial evolution of PET in Dabudi shows significant asymmetry. For instance, in Zone D1, an east–west road on the north side shows a uniform temperature distribution at 08:00 a.m., but by 12:00 p.m., drastic spatial divergence occurs. This suggests that the complex organic morphology of traditional zones produces more variable thermal responses under dynamic solar radiation.

Street geometry has a clear regulatory effect on PET. Increasing the width of north–south roads leads to higher heat stress, but there is a threshold for this growth. In H3, at 08:00 a.m., the PET of the wider north–south road is approximately 43.0 °C, while the narrower one is only 38.0 °C. However, in H2, once the road width reaches a certain point, PET values for different street widths tend to converge at 08:00 a.m. and 12:00 p.m., indicating that the negative impact of road width no longer increases significantly after a saturation point. For east–west roads, heat stress decreases as the width decreases, also showing a floor effect. In D1, at 12:00 p.m., the PET of the widest street is approximately 55.0 °C, the next is about 47.0 °C, and the narrowest drops to 37.0 °C. In the extremely narrow streets of D3, despite variations in width, PET remains stable below 40.0 °C.

The numerous narrow passages and green courtyards in traditional zones create a significant cool refuge area, which is crucial for regulating local extreme heat stress. At 12:00 p.m., when radiation is strongest, the PET in the narrow passages of D3 is below 37.0 °C, while the open plaza next to it exceeds 57.0 °C. This sharp temperature difference of 20.0 °C highlights the huge contribution of high shading proportions in traditional spaces to thermal comfort. Additionally, simulation results clearly show that courtyards with vegetation have significantly lower heat stress than paved courtyards. Specifically, in spaces with tall shrubs or trees, transpiration and vertical shading work together to effectively reduce heat accumulation. This finding emphasizes that maintaining narrow alley logic and introducing multi-level vegetation are key strategies for mitigating thermal environment degradation caused by modern reconstruction.

To describe the distribution and dispersion of the thermal environment within D zones and H zones, spatial PET boxplots were created (Figure 8).

Looking at the overall shape of the boxplots, the PET distribution in traditional zones is significantly wider, with whiskers extending far beyond those of the modern standardized zones. At 12:00 p.m., the range of PET extremes in traditional zones reaches nearly 15.0 °C. This high heterogeneity reflects the diverse microclimate characteristics within traditional zones [51]. In contrast, PET values in standardized zones show a strong tendency to concentrate around the median, with a tighter distribution representing thermal homogenization [52]. At 08:00 a.m. (Figure 8b), the span of the boxes and whiskers for traditional zones D2 and D3 is markedly larger than for H1 and H2. In particular, PET values in H1 are highly concentrated within a narrow range of 37.5 °C to 38.5 °C, making it the most spatially uniform unit. During the period of strongest radiation (Figure 8c), the PET distribution in D2 and D3 shows strong divergence, with the lower edge dropping to around 44.0 °C and the upper edge approaching 58.0 °C. In standardized zones, H2 and H3 show high similarity, with medians stable around 54.0 °C and highly compressed boxes, meaning there is very little difference in PET across most of the space.

The bottom of the distribution range in standardized zones remains at a high PET level, indicating a lack of cool refuges. Observations reveal that the lower edge of PET in standardized zones is significantly higher than in traditional zones. In traditional zones D2 and D3, relatively cool areas with PET between 44.0 °C and 47.0 °C are preserved due to narrow alley shading and local greenery. However, almost all sampling points in the standardized zones have a PET higher than 47.0 °C. This means that in modern standardized zones, residents can find almost no effective outdoor shaded refuges during periods of extreme heat.

3.4. Quantitative Thermal Drivers

To further investigate the quantitative relationship between spatial morphology and thermal comfort, a Pearson correlation analysis and linear regression were performed across eight morphological indicators for the six representative zones. The morphological data for these eight indicators were derived from Table 4. The PET data selected for analysis include the maximum value of the area-averaged PET throughout the day, as well as the spatial maximum and minimum PET values at the specific time step when the maximum area-averaged PET occurs.

As shown in

Table 5. Pearson correlation coefficients (r) between morphological indicators and PET metrics.

PET Metrics	SVF	H/W	BSF	ISF	PSF	GCR	REH	TCC
Peak average PET	0.81	−0.63	−0.39	−0.74	0.85	0.91	−0.80	−0.26
Max PET at peak	0.77	−0.65	−0.71	−0.02	0.39	0.42	−0.61	0.36
Min PET at peak	0.17	0.01	0.45	−0.82	0.49	0.43	−0.33	−0.51

, several indicators exhibit statistically significant correlations with PET intensity. GCR shows the strongest positive correlation with the peak average PET (r = 0.91), confirming that greenery is the most dominant factor in mitigating overall heat stress. SVF (r = 0.81) and PSF (r = 0.85) also display strong positive links to the thermal baseline. Notably, ISF exhibits a strong negative correlation with the minimum PET (r = −0.82), quantitatively proving that the reduction in permeable surfaces in standardized layouts directly leads to the loss of micro-scale cool refuges.

In the linear regression analysis, the eight morphological indicators were further categorized into three dimensions: Shape, Surface Properties, and Greening Efficiency. The PET data used for this analysis corresponds to the peak-averaged PET (Figure 9).

The results demonstrate that GCR and PSF are the most significant positive predictors of thermal mitigation, while SVF and ISF act as the primary drivers of elevated heat stress and thermal homogenization in standardized rural settlements. Specifically, regarding Shape indicators, SVF yields an R² of 0.66, indicating that open spatial configurations in standardized zones significantly amplify radiation absorption. In contrast, H/W shows a moderate negative impact (R² = 0.39). Surface Properties results indicate that PSF is a reliable predictor for thermal mitigation (R² = 0.71), whereas the high ISF in standardized zones drives the trend of “thermal homogenization”. The sharp contrast in slopes between ISF (−14.04) and PSF (14.51) highlights the trade-off between modern pavement efficiency and microclimate resilience. Finally, for Greening Efficiency, GCR remains the most influential cooling driver (R² = 0.82). While the normalized REH shows a moderate negative correlation (R² = 0.63), its impact is secondary to the spatial extent of greenery, suggesting that canopy coverage is more critical than building height in these rural contexts.

4. Discussion

4.1. Potential Risks of Homogenized Thermal Environments

Centralized rural reconstruction achieves significant spatial justice through standardized spatial patterns and shows clear advantages in improving infrastructure and reducing extreme peak temperatures. From a social equity perspective, the uniform building specifications, regular grid layouts, and wide transportation networks in the standardized zones ensure equal access to living conditions, public services, and transportation convenience for all residents. Simulation results reflect this justice in thermal performance: due to standardized building spacing and uniform street trees, the maximum PET values (49.2–51.8 °C) are significantly lower than those in traditional zones (53.5–55.2 °C). This indicates that modern planning effectively suppresses local extreme hotspots, providing a physical environment with overall lower heat stress for all residents.

However, this justice based on a homogenized space comes at the cost of losing spatial thermal resilience, exposing the vulnerability of modern planning to climate change. The results reveal a clear trade-off: while regular grids slightly lower the ”upper limit“ of heat stress, they simultaneously raise the ”baseline“ by eliminating micro-scale cool spots. Although traditional zones face severe heat stress in some open areas, their complex geometry (such as narrow alleys with high H/W ratios) and fragmented green spaces create a series of micro-scale cool refuges. This high heterogeneity provides a large spatial temperature span (up to 15.0 °C), allowing villagers to adapt their behavior by choosing the shade. In contrast, the homogenized layout of standardized areas flattens microclimate gradients. This loss of resilience means that while residents are in an environment with a lower average value, they lose the opportunity to find cooler spaces for refuge during extreme periods.

In contrast, residents in standardized areas are exposed to continuous heat stress risks caused by homogenized space. The highly concentrated PET distribution indicates that wide streets, uniform building heights, and extensive hard surfaces have eliminated microclimate differences. This uniformity causes the community to lose the multi-level cooling barriers typically found in traditional spaces. When the thermal environment is nearly identical across the entire community, residents as a whole are trapped in a state of high heat stress with no relief. This situation suggests that while standardized layouts provide a form of spatial justice, they may create systemic environmental risks during extreme weather.

To mitigate the negative impacts of this homogenized degradation, several optimization paths should be considered. First, there should be a shift from “grid-based” to “climate-sensitive” layouts [53], suggesting the introduction of asymmetrical street designs in reconstruction planning and the purposeful retention of narrow spaces to create permanent shaded corridors [54,55]. Second, the introduction of softening surfaces and vertical greenery should be strengthened [56,57]. Given the high degree of surface hardening in standardized areas, low-cost cooling strategies such as permeable pavements and climbing plants on building facades should be promoted [58]. Finally, planners need to perform a trade-off analysis of landscape features: while ensuring vehicle access and fire safety, they should find a balance for walking comfort through local spatial contraction or shading structures [59]. Through these strategies, rural reconstruction can maintain standardized justice while rebuilding microclimate diversity and climate resilience.

4.2. Research Limitations and Future Study Directions

Although this study reveals the significant impact of spatial morphology on thermal comfort in rural reconstruction, there are several limitations that should be addressed in future research.

First, regarding the time scale, this study focuses only on the microclimate performance during a typical extreme heat day in summer. While summer heat stress is a major challenge for rural living environments under climate change, the influence of spatial morphology on winter wind protection and thermal comfort during transitional seasons is also important. Future research should include long-term monitoring and simulation across different seasons to fully evaluate the year-round microclimate sustainability of centralized reconstruction.

Second, the geographical sample of this study is relatively limited, focusing on typical villages in a specific climate zone. Although the two selected settlements have representative spatial features, traditional zones in other climate regions (such as cold or hot-humid zones) have unique morphological traits and microclimate adaptation strategies [60]. Therefore, the applicability of the ”thermal homogenization“ conclusion to diverse climatic contexts requires further verification to avoid over-generalization. Furthermore, the limited sample size of six representative zones restricts the statistical generalizability of the quantitative regression results, which should be interpreted as exploratory indicators of general trends rather than universal predictive outcomes. Future work could expand the sample size to explore the relationship between spatial justice and climate resilience across different climatic backgrounds.

Finally, this study emphasizes the objective simulation of the physical environment and does not fully consider the behavioral adaptation and subjective perception [61] of residents. Rural residents have developed unique habits for using microclimates through long-term agricultural life. Their behavioral adjustments, such as changing activity paths or timing, may significantly reduce their exposure to high heat at the physical level [62]. Future studies should introduce social surveys and physiological monitoring data [63] to build a comprehensive evaluation framework covering the “physical environment—physiological response—behavioral adjustment” process. This will provide better support for creating more human-centered rural landscapes.

Furthermore, while the single-point validation demonstrates high statistical reliability, the lack of multi-point simultaneous measurements is a limitation. Consequently, spatial PET differences between zones should be interpreted as relative comparative tendencies rather than absolute values. Additionally, this study focuses on daytime outdoor performance and does not account for potential nighttime heat island effects. In standardized zones, lower SVF and impervious materials may delay nighttime cooling, which is a critical risk for vulnerable groups who lack air conditioning. Future research will incorporate 24 h monitoring to address these complexities.

5. Conclusions

Under centralized reconstruction policies, rural settlements in North China are transitioning from heterogeneous, low-rise organic forms to homogeneous, high-density standardized configurations. While these modern communities improve infrastructure, their impact on the micro-scale thermal environment remains a critical concern for sustainable rural development. Despite the widespread implementation of these policies, there is a lack of quantitative research comparing the thermal characteristics of different settlement typologies. Most studies ignore the spatial patterns of heat and the specific risk of “thermal homogenization,” leaving the disappearance of traditional “cool refuges” under-researched.

To address this, the present study established an 8-parameter morphological indicator system—covering geometric configuration (e.g., SVF, REH, H/W) and surface properties (e.g., ISF, PSF)—to capture the physical transition of rural spaces. Using UAV photogrammetry and high-resolution ENVI-met simulations, a typologies comparison was conducted between traditional organic villages and modern standardized communities.

The quantitative analysis of morphological parameters and the subsequent microclimate simulations yielded several key findings:

• Standardized communities exhibit significant vertical growth and surface hardening, where the REH of 9.55 m is nearly double that of traditional villages (4.5–5.7 m), while ISF consistently exceed 0.38.
• All study areas experience extreme heat stress (PET > 39 °C) for 8 to 10 h daily, and although standardized zones show slightly lower peak values (49.2–51.8 °C) than traditional ones (53.5–55.2 °C), they provide a longer window of non-extreme stress.
• A core finding is the risk of thermal homogenization in standardized areas, which leads to a loss of spatial thermal resilience. Although peak temperatures are slightly mitigated, the overall thermal environment degrades as the ”lower limit“ of the temperature range is raised. This trade-off means that residents in standardized communities are deprived of the shaded refuges essential for behavioral adaptation during extreme heat.
• Standardized reconstruction leads to a significant upward shift in the lower thermal limit, as traditional villages retain local cool spots with PET as low as 44.0 °C, while the PET at the majority of sampling points in standardized communities exceed 47.0 °C, indicating a systemic shift toward the reduction in shaded refuges. Specifically, the results establish a quantitative link confirming that greening coverage (GCR, R² = 0.82) and surface perviousness (PSF, R² = 0.71) are the most significant positive predictors for thermal mitigation.

Future rural settlement planning must move beyond standardized justice to prioritize climate resilience. Practical strategies include adopting asymmetrical street designs to create permanent shaded corridors, increasing pervious surface ratios, and introducing vertical greenery. By reintroducing spatial heterogeneity into standardized planning, the loss of microclimate regulatory resilience can be offset, effectively reducing the risk of continuous heat exposure for residents during extreme heat periods.