Synergistic Landscape Design Strategies to Renew Thermal Environment: A Case Study of a Cfa-Climate Urban Community in Central Komatsu City, Japan

Abstract

An effective community landscape design consistently impacts thermally comfortable outdoor conditions and climate adaptation. Therefore, constructing sustainable communities requires a resilience assessment of existing built environments for optimal design mechanisms, especially the renewal of thermally resilient communities in densely populated cities. However, the current community only involves green space design and lacks synergistic landscape design for renewing the central community. The main contribution of this study is that it reveals a three-level optimization method to validate the Synergistic Landscape Design Strategies (SLDS) (i.e., planting, green building envelope, water body, and urban trees) for renewing urban communities. A typical Japanese community in central Komatsu City was selected to illustrate the simulation-based design strategies. The microclimate model ENVI-met reproduces communities involving 38 case implementations to evaluate the physiologically equivalent temperature (PET) and microclimate condition as a measure of the thermal environments in humid subtropical climates. The simulation results indicated that the single-family buildings and real estate flats were adapted to the summer thermal mitigation strategy of water bodies and green roofs (W). In small-scale and large-scale models, the mean PET was lowered by 1.4–5.0 °C (0.9–2.3 °C), and the cooling effect reduced mean air temperature by 0.4–2.3 °C (0.5–0.8 °C) and improved humidification by 3.7–15.2% (3.7–5.3%). The successful SLDS provides precise alternatives for realizing Sustainable Development Goals (SDGs) in the renewal of urban communities.

1. Introduction

1.1. Background

With climate change and heat waves worldwide, urban heat islands (UHI) deteriorate the urban thermal environment and microclimate conditions, which increases the potential health risks of inhabitants. Balancing the energy consumption of buildings and urban resilience has become a critical challenge for sustainable urban planning [1]. In the Revised Integrated Energy Statistics Report (2019), the Agency for Natural Resources and Energy (ANRE) stated that global warming has increased the mean air temperature in Japan by 1 °C, and deaths due to heatstroke are increasing among the elderly population. The literature about community-related microclimate and thermal environment focuses on energy consumption and air quality [2,3,4,5,6,7], while less attention has been paid to the synergistic landscape design to reduce thermal stress. The mitigation strategies in subtropical cities have focused on green space layout and tree design in microclimatic simulation for thermal comfort improvement [6,7,8,9,10], ignoring mechanisms verification and comparative studies in synergistic landscape strategies.

1.2. Synergistic Landscape Design Strategies (SLDS) for Improving Thermal Environment

Synergistic Landscape Design Strategies (SLDS) are important approaches to urban cooling that integrate design elements in vegetation, water conditions, albedo of materials, and urban form. Previous mitigation strategies in the thermal environment have shown that SLDS can reduce cooling energy demand and greenhouse gas emissions [4,11,12,13,14]. Planting design in the SLDS involves complex factors such as vegetation structure, layout, tree species, type, and number on the microclimate and thermal environment [6,7]. Multi-layered vegetation structures have been validated for cooling and humidifying effects on thermal environments, especially in summer, where air temperature (T_a) is reduced by 2.14–5.15 °C and relative humidity (RH) is increased by 6.21–8.30% [15]. Optimal vegetation configuration coupled with the building form of the block, studies have shown that vegetation in low-rise building blocks has a high cooling potential [16]. Research on improving thermal environments found that increasing the albedo of urban fabric material and developing green roofs can deal with subtropical UHI in Osaka, Japan [8]. A study in a residential area of Nanjing [5] found that increasing green quantity improves outdoor thermal comfort by using a mean radiant temperature (T_mrt) and the mean voting index (PMV) as the evaluation criteria. Moreover, the vertical cooling performance (T_a and T_mrt) of the residential environment benefits from urban trees [10].

In other studies on cooling community strategies of SDLS, green building envelopes (GBE) retrofits play an important role in the thermal stress attenuation of residential buildings [4,17]. Green roofs reduce thermal radiation flux through evapotranspiration to increase surface albedo and reduce shortwave radiation [18,19,20]. Knaus and Haase [19] have shown that green roofs improve the temperature of residential roofs during the day and reduce physiologically equivalent temperatures (PET) by an estimated 9 k. Furthermore, some studies have indicated that green facade systems can be used in different climates with significant cooling and humidification effects [21,22,23,24,25,26,27]. In contrast, installing a single green facade has limitations in mitigating the ambient temperature [28,29]. The SLDS integrated planting design [6,30,31,32], urban trees [6,7,10], GBE design [33,34], and water body design [35] significantly contribute to balancing thermal comfort. Feitosa and Wilkinson [36] verified that green roofs and walls contribute to better thermal insulation of the building envelope and effectively reduce the thermal index. Lai et al. [37] found that PET was reduced by 13.0 K and 4.6 K in vegetation and water bodies design. However, synergistic landscape interventions ignore a mix of more than three design strategies to alleviate urban climate change and UHI-induced thermal perturbation.

1.3. Microclimate Simulation in the Urban Community

Microclimate simulation can reproduce specific urban spaces on a system of physical environments to simulate urban surface-plant-air interaction by computational fluid dynamics (CFD) models [38,39,40]. Most microclimate studies examined CFD-integrated thermal indices to investigate the influencing factors and strategies to mitigate the thermal environment. The ENVI-met V5.0.1 simulation software is applicable for simultaneous cross-simulations and comparative studies of microclimatic conditions and thermal environments by adopting SLDS [41,42,43,44,45]. In Egypt, ENVI-met 4.4.5 simulated canyons in urban residential communities and showed that hybrid landscape design scenarios can improve summer thermal comfort by reducing PET and T_a [46]. In a previous ENVI-met study on Nanjing City, the quantity of trees effectively relieves the heat stress of a residential district [5]. Another ENVI-met study in a residential district in Tabriz, Iran, validated that tree coverage and species can improve the cooling effect (reduction of T_a, T_mrt, and PET) [32]. To date, the major challenges of these simulation models are related to inadequate modeling of a synergistic landscape strategy optimization mechanism and limited support for the practical design of multifactorial improvements with parameter complexity [47].

1.4. Motivation

As mentioned in previous studies, there are some research gaps in the existing studies. First, it is unclear how to develop an optimization mechanism to assess the thermal and microclimatic environments in humid subtropical cities to renew the urban communities. Second, there is limited research on microclimate simulation for SLDS integrating the design of planting, GBE, water bodies, and urban trees [15,33,48]. Third, some research is scarce on applying SLDS consistent with the multifactorial improvements in building form, landscape design, climatic characteristics, scale change, and time mitigation [5,8,10]. Therefore, an optimization method is proposed, coupling a renewal experience and microclimate simulation mechanism. The aim is to comprehensively assess the potential for summer mitigation thermal performance and cooling effect on urban communities.

To address these gaps, the detailed purposes and novelty of this study are as follows:

(i)

To develop a three-level optimization method to systematically verify the SLDS alternatives in a pre-renew stage of the central city for both summer outdoor thermal comfort and microclimate improvement.

(ii)

To apply to quantify which the SLDS alternatives can affect the small-scale and the large-scale optimization community cooling demand and the summer thermal comfort more effectively multifactor improvements strategy following the optimal criteria (i.e., improvement of T_a, RH, WS, T_mrt, and PET).

(iii)

To investigate the effect application of urban trees in the microclimate simulation ENVI-met of sustainable Japanese communities project and resilience assessment in the Cfa-climate by comparing analysis if the hypotheses do not reach the multifactorial improvements.

For these purposes, typical Japanese urban communities in central Komatsu City are selected as a case study for renewal planning for the Sustainable Development Goals. The potential SLDS in the pre-renewal stages are systemically verified on a small and large scale. The summer thermal comfort and community cooling demand prior to and after the application mechanism of the SLDS interventions are compared. The thermal sensation range of PET, specifically for humid subtropical cities, as detailed in Section 2.5, is applied to analyze the thermal environment improvement in this study. Finally, ideal SLDS are presented, considering thermal comfort and microclimate effects, which can serve as a practical guideline and method for urban designers.

2. Materials and Methods

This study combined field measurements with ENVI-met simulations in SLDS to assess the microclimatic conditions and thermal comfort in the urban community. The workflow can be divided into four steps (Figure 1): (1) Select three Japanese communities in Komatsu City as sample sites; (2) Investigate and monitor local meteorological data via weather stations to simulate the base model (L1 AC, L1 HC, and Ga BC); (3) Use numerical simulations and ArcGIS 10.8 created an accurate 3D environment model for microclimate simulation of the communities; (4) Validate the base model and develop the new SLDS simulations for comparative studies in renewal mechanism, as detailed in Section 3.

2.1. Study Area

Komatsu City is located at the center of the southwest Kaga Plain, and it has 106,292 inhabitants as of 2020 (Figure 2). This study area was identified as a future city for the Sustainable Development Goals (SDGs), which seek synergistic development with multiple goals in the areas of energy, climate change, and health. This study area was selected due to the Komatsu City Renewal and Maintenance Plan project, which implemented a specific plan to renew the city to a sustainable and resilient urban form starting in 2020 [49].

The local community was represented at the small scale by the single-family community (HC) of Doihara No. 1 (34.41° N, 136.45° E) and the real estate flat community (AC) at 1-chōme-178 Hinodemachi (36.40° N, 136.45° E) (Figure 2). The large scale was represented by the mixed cluster community (BC) of Yokaichimachi (36.24° N, 136.26° E). Japanese-style houses were present in the HC area with a civic square in the center, and the AC area formed an asymmetrical canyon with station hotels. The BC area was formed by the single-family (H type) and real estate flat buildings (A type). Moreover, the HC area is approximately 6536 m^2, and the tallest low-rise building is 12 m. The AC area had an overall area of 14,950 m² around the high-rise building in 30 m. The BC had an applied area of approximately 46,200 m², with the tallest building being 24 m. Among the landscape designs in the sample area, the HC and AC areas are populated with a mixture of trees, shrubs, and grasses, while the BC area is planted with a private garden landscape. A common feature is the distribution of functional spaces with parking lots, parks, and plazas within the urban communities.

2.2. Climate Condition

Komatsu City is located in the humid subtropical climate zone (Cfa) of Köppen’s climate classification system [50] (Figure 3a). Summer in Komatsu City is studied in this paper, which is generally characterized by heat, high temperature, and rain humidity, with high mean air temperatures of 30 °C. The hottest times are between 12:00 and 16:00; the hot season lasts for 2.8 months, from June to September [51]. In the GISS of global surface temperature analysis for 2019–2021 in Japan, the surface temperatures increased by 1–2 °C in August [52] (Figure 3b). Local weather data are provided in real-time by the Japan Meteorological Agency (JMA). The agency shows that air temperatures in August may reach a maximum of 38 °C based on the maximum air temperatures of the last three years [53] (Figure 3c). August is the hottest month based on the annual mean air temperature in Komatsu City (Figure 3d). Inhabitants are exposed to high air temperatures for long periods, which can potentially lead to health risks associated with discomfort, especially in elderly communities [54]. So, selecting August as a typical simulation period allows people to provide timely feedback on discomfort sensations during the summer months. The measured meteorological data also provided evidence to simulate an effective SLDS in urban communities.

2.3. Meteorological Monitoring

The meteorological data were monitored by the Komatsu City Weather Station and Komatsu Airport Station [51,53]. Weather stations monitored microclimate parameters, including T_a, RH, wind direction (WD), and wind speed (WS). Based on the climate data for 2021, August was the hottest month, with a daily maximum T_a of up to 35 °C (Figure 3c). In the visible meteorological data for evaluating thermal comfort in specific environments, the thermal perception of the inhabitants was most pronounced in terms of discomfort. Therefore, this study chose the monitoring date of 22 August 2021 as a typical summer day, representing extremely hot days for examining the SLDS in a Cfa-climate urban community. The monitoring time is setting 24 h at 1 h intervals. Recorded meteorological parameters are presented in

Table 1. Meteorological parameters (T_a, RH, WD, and WS) on 22 August 2021.

Time	Ta (°C)	RH (%)	WD (°)	WS (m/s)
00:00	26.5	74.0	112.5	0.8
01:00	28.7	62.0	180.0	3.2
02:00	27.1	70.0	247.5	1.1
03:00	28.5	66.0	180.0	3.0
04:00	28.3	66.0	180.0	2.9
05:00	27.6	66.0	180.0	1.2
06:00	26.9	70.0	135.0	1.6
07:00	29.3	62.0	180.0	2.0
08:00	29.7	58.0	202.5	2.4
09:00	31.4	58.0	180.0	2.9
10:00	32.5	52.0	180.0	4.6
11:00	33.0	55.0	180.0	4.4
12:00	34.1	52.0	202.5	3.5
13:00	33.6	55.0	202.5	4.2
14:00	32.2	59.0	202.5	3.3
15:00	31.8	62.0	202.5	3.7
16:00	31.3	62.0	202.5	2.1
17:00	28.2	79.0	202.5	2.2
18:00	27.2	84.0	202.5	2.0
19:00	27.0	84.0	202.5	0.7
20:00	27.0	84.0	225.0	0.3
21:00	27.0	84.0	157.5	0.3
22:00	25.7	94.0	270.0	1.2
23:00	25.8	89.0	135.0	1.2

.

2.4. ENVI-Met and ArcGIS

Urban microclimate reveals the behavior of microclimate in urban areas and between buildings, which helps researchers better understand the impact on the outdoor thermal environment and human health in a specific urban environment [55]. The ENVI-met model has been used for the numerical simulation of real scenes and a comparative study of strategic cases that affect thermal comfort and microclimate regulation. The ENVI-met software is widely used for multiscale microclimate simulations in humid subtropical climates [8,45,48]. This software is a grid-based model with fine resolution (0.5–10 m) based on the standard κ–ε turbulence model and Reynolds Averaged Navier-Stokes (RANS) equations [10,45]. ENVI-met V5.0.1 converted the vegetation model from geometric blocks to actual trunk and leaf shapes. The contour of the 3D tree model is more accurate and extends to new tree species in the DB database. The operation mode was selected in the ENVI-guide sub-tool, including full force and simple force [56]. The Bio-met in ENVI-met is used to evaluate the thermal index (PET). In this study, the simple force mode was selected, and the simulated height was set to a pedestrian height of 1.8 m. The X, Y, and Z gride dimensions in three sample models (HC, AC, and BC areas) based on the scale of the study areas were set to 38 × 43 × 15, 63 × 69 × 45, and 160 × 75 × 36; and the grid size is 2 × 2 × 2 m. The fundamental parameter settings in the ENVI-met models and the setting of the wall material are shown in Figure 4 and Table 1 and

Table 2. Fundamental parameter settings in the ENVI-met.

Sample Site/Setting		HC	AC	BC
Time setting	Total simulation time (h)	24
	Output time interval (min)	60
	Start of simulation date and time	On 22 August 2021; at 00:00
Create model	Model dimensions (X × Y × Z) (m)	38 × 43 × 15	63 × 69 × 45	160 × 75 × 36
	Grid size (m)	2 × 2 × 2	2 × 2 × 2	2 × 2 × 2
	Telescoping factor (%)	6%	-	-
	Start telescoping after height (m)	7	-	-
	Number of nested grids	5	-	-
Meteorological parameters	Minimum–maximum air temperature (°C)	25.7–34.1
	Minimum–maximum relative humidity in 2 m (%)	52–94
	Wind speed measured at 10 m height (m/s)	2.3
	Wind direction (°)	189
	Roughness length at the measurement site	0.01
Soil	Soil layer (0–200 cm) humidity (%) and temperature (°C)	65, 20
		70, 20
		75, 19
Cloud	Fraction of cloud (x/8)	6
Force mode	Simple force/Full force	Simple force

. Figure 5 shows that the urban building land is used for residential and commercial purposes, so Komatsu City forms a highly built-up central area. The central urban area is covered with low and medium vegetation and distributed the UHI phenomenon in ArcGIS analysis for Komatsu City.

Airflow field turbulence in ENVI-met is given by RANS equations, as shown in Equation (1) [10]:

$$\begin{gathered} \frac{\partial u}{\partial t}+u_i\frac{\partial u}{\partial }=-\frac{\partial u}{\partial x}+K_m\left(\frac{{\partial }^{2}u}{\partial {x_i}^2}\right)+f\left(V-V_g\right)-S_u \\ \frac{\partial u}{\partial t}+u_i\frac{\partial u}{\partial }=-\frac{\partial u}{\partial x}+K_m\left(\frac{{\partial }^{2}u}{\partial {x_i}^2}\right)+f\left(u-V_g\right)-S_V \\ \frac{\partial w}{\partial t}+u_i\frac{\partial u}{\partial x_i}=-\frac{\partial p’}{\partial z}+K_m\left(\frac{{\partial }^{2}w}{\partial {x_i}^2}\right)+g\left(\frac{\theta \left(z\right)}{\partial {x_i}^2}\right)-S_w \end{gathered}$$

(1)

$$\frac{\partial u}{\partial x}+\frac{\partial v}{\partial y}+\frac{\partial w}{\partial y}=0$$

(2)

where f = 10⁴ sce⁻¹ is the Coriolis parameter, p’ is the local pressure perturbation, θ is the potential temperature at level z, and the reference temperature θ_ref should be recognized as average mesoscale conditions and are provided by a one-dimensional model running parallel to the primary model. Equation (2) is used to keep the model mass conserving for each time step.

T_a and RH in ENVI-met is given by advection-diffusion equations, as shown in Equations (3) and (4) [57]:

$$\frac{\partial \theta }{\partial t}+u_i\frac{\partial \theta }{\partial {x_i}^{ }}=K_h\left(\frac{{\partial }^{2}u}{\partial {x_i}^2}\right)+Q_h$$

(3)

$$\frac{\partial \theta }{\partial t}+u_i\frac{\partial \theta }{\partial {x_i}^{ }}=K_q\left(\frac{{\partial }^{2}q}{\partial {x_i}^2}\right)+Q_q$$

(4)

Similar to the momentum equations, Q_h and Q_q are used to link heat and vapor exchange at the plant surface with the atmospheric model.

2.5. The Human Thermal Comfort Index

Urban thermal environments have been assessed based on meteorological elements. PET is a general thermobioclimatic index used to characterize all climate zones. PET was developed by Hoppe in the 1990s based on the Munich Personal Energy Balance Model, which is the most widely used index for assessing outdoor thermal environments in humid subtropical climates [58]. This index considers all thermal bioclimatic factors using the Munich Energy-balance Model for Individuals (MENI) for individuals. The model assumes that the body reaches thermal equilibrium, where the equation of state Equation (5) [59].

M + W + R + C + ED + E_VC + E_SE + S = 0

(5)

In Equation (5), M is the metabolic rate, W is the physical output, R is the net radiation of the physical body, C is the convective heat flow, ED is the skin diffusion of the latent heat flow, E_VC is the total latent heat flow, E_SE is the heat flow from sweat evaporation, and S is the heat that cools the body to store heat. The individual heat flow in Equation (5) is directly affected by the following microclimate parameters: T_a, RH, WS, and T_mrt. Moreover, the ENVI-met calculates the PET in Bio-met by personal setting (default male, summer clothing).

The evaluation criteria of PET were based on the classification of thermal perception in previous studies of humid subtropical cities in Taiwan and Kyoto [58,60]. Specifically, very cold (<14 °C), cold (14–18 °C), cool (18–22 °C), slightly cool (22–26 °C), comfortable (26–30 °C), slightly warm (30–34 °C), warm (34–38 °C), hot (38–42 °C), and very hot (>42 °C).

3. Synergistic Landscape Design Strategies (SLDS) in Japanese Communities

Based on the current case, the SLDS (planting design, GBE design, water body design, and urban trees) were compared in terms of their performance in improving the thermal environment. This study predicts the effect of thermal comfort improvement by vegetation structure (mixed in trees-grass-shrubs, no plant, trees and grass, shrubs and grass), number of urban trees, and planting layout (current layout, hard pavement layout, enclosed layout, and semi-enclosed layout) in the design of small-scale models (HC and AC areas). This SLDS establishes a mechanism divided into four steps (from a to d) to find an optimal strategy for improving thermal environment renewal. The details are as follows: (a) Testing the planting design (“L”) to improve the thermal environment on a small scale. The vegetation structure and layout are integrated into various tree numbers representing: “L1”—mixed structure in the current layout, “L2”—no plant in hard pavement layout, “L3”—trees and grass in an enclosed layout, and “L4” shrub and grass structure in a semi-enclosed layout. (b) Adopting the GBE design was divided into an “R” strategy for the green roof, an “F” strategy for the green facade and the green roof, and a “W” strategy for water bodies and green roofs and the GBE design (“R, F, W”) integrated planting design code cases in R1–4, F1–4, and W1–4. (c) Obtaining the output of the SLDS optimization simulated in small-scale communities. (d) Identifying optimization strategies for BC large-scale models based on small-scale results, particularly the thermal mitigation of urban tree types and numbers (Gb, Gc, Gd, Ge, and Gf) combined with pavement material. The community design cases for synergizing landscape design strategies are listed in

Table 3. All design cases using the Synergistic Landscape design strategies (SLDS) were conducted in three sample communities (HC, AC, and BC areas) (Note: “◎”indicates that SLDS cross-codes form different renewal cases).

SDLS	Code	Case	Tree Number (Tn)	Tree Number (Tn)	Planting Design				Adding GBE Design			Adding Urban Tree
			HC/BC	AC	L1	L2	L3	L4	R	F	W	Ga	Gb	Gc	Gd	Ge	Gf
Current cases/ Planting design cases	L1–4	L1	34	54	◎
		L2	0	0		◎
		L3	32	76			◎
		L4	0	0				◎
Renewal small-scale cases	R1–4	R1	34	54	◎				◎
		R2	0	0		◎			◎
		R3	32	76			◎		◎
		R4	0	0				◎	◎
	F1–4	F1	34	54	◎					◎
		F2	0	0		◎				◎
		F3	32	76			◎			◎
		F4	0	0				◎		◎
	W1–4	W1	34	54	◎						◎
		W2	0	0		◎					◎
		W3	32	76			◎				◎
		W4	0	0				◎			◎
Current cases/ Renewal large-scale cases	W-Ga-f	Ga	126		◎							◎
		W-Gb	126		◎						◎		◎
		W-Gc	208		◎						◎			◎
		W-Gd	208		◎						◎				◎
		W-Ge	208		◎						◎					◎
		W-Gf	208		◎						◎						◎

. Based on the current cases (L1 HC, L1 AC, and Ga BC), 35 design cases were established to investigate the performance of the new SLDS to optimize the thermal environment in the three sample communities. Among them, the small scale involves 16 SLDS cases for comparative study, including L1–4, R1–4, F1–4, and W1–4 cases. It crosses two sample communities with various building forms (AC and HC), forming 32 cases of small-scale alternatives. The large-scale model is based on small-scale optimal SLDS for six cases (i.e., Ga, W-Gb, W-Gc, W-Gd, W-Ge, W-Gf). The performance of comparative strategies for thermal environment improvement is verified through hierarchical SLDS and the thermal benefits of deleting optimal design decisions for renewing sustainable communities in a homogeneous building context (Figure 6).

3.1. Planting Design (L) at a Small-Scale

At the small scale of the HC and AC models, the vegetation configuration was based on the layout of the existing plaza and parking lot for perimeter planting, with the tree types being mainly deciduous in combination with a small number of evergreens and conifers. Typical evergreen trees in the local area are Camphorwood and Cyclobalanopsis glauca, coniferous trees are planted pine, and deciduous trees are mainly planted with Cornus florida, Acer rubrum, and Zelkova serrata. The renewal planting design involved vegetation structure, landscape layout, and tree numbers in horizontal and vertical dimensions formed by integrating the reconfiguration cases (L1–4) shown in Figure 7.

3.2. Green Roof (R)

Green roofs respond to urban environmental challenges as a multifunctional greening strategy. Green roofs have five basic layers: Vegetation, growing substrate, drainage layer, root barrier, and waterproof membrane [61]. The cooling performance of green roofs depends on the roof material and vegetation characteristics in the context of the urban microclimate [62]. This study systematically investigated the degree of thermal mitigation provided by extensive green roofs based on planting designs (R1–4) at a small scale.

3.3. Green Facade and Green Roof (F)

Green facade components cover the green facade using climbing vegetation [63]. Green facades improve the urban microclimate by absorbing, evaporating, and insulating solar radiation to comply with a climate-adaptative design. In the ENVI-met microclimate model, green roofs and facades were classified as default greenings with or without air gaps. This study used default greenings with air gaps to model the GBE designs based on planting designs (F1–4) in HC and AC areas. The planting thickness (0.3 m) and substrate thickness (15 cm) were specified, with the vegetation layer consisting of 25 cm of grass for the green roof and ivy for the green facade.

3.4. Water Body and Green Roof (W)

Traditional community buildings in Japan have been maintained as front and back yards, with a courtyard landscape in the center. The design of water bodies in ENVI-met was via discharge fountains at the street green space of the study area to present traditional garden design techniques such as chōzubachi, bamboo bucket fountains, spiral sprays, and rain chains, including the main community activity areas and backyards of the communities. Water bodies and green roofs as SLDS are considered for validating the effect on a small scale (W1–4).

3.5. Large-Scale Synergistic Landscape Design Strategies (W-Ga-f)

Based on the evaluation of small-scale SLDS, an optimal strategy was applied to the BC area to investigate the mitigation performance and cooling effect on UHI at a large scale (W-Ga-f). For the Leaf Area Density Index (LAD) values of typical local tree species, an increase of 82 urban trees is used for placement in the large-scale cases (W-Gc-f); the corresponding case codes are Ga, W-Gb, W-Gc, W-Gd, W-Ge, and W-Gf. Ga represents the current case in the large-scale model, while W-Gb optimizes the “W” strategy based on the Ga case. W-Gc-e is the introduction of coniferous (conic, small trunk, dense, medium, 15 m), deciduous (middle American sweet gum), and evergreen (Camphorwood was set to change LAD by coniferous tree) types as the main urban trees to replacement based on W-Gb. W-Gf is a renewal of the granite pavement (single tone) (0100GS) based on the W-Gc case (Figure 8).

4. Results

4.1. ENVI-Met Model Validation

A precise ENVI-met model can be validated by fitting measured results against simulated results. Statistical evaluation metrics are used to validate simulation results for microclimate parameters. Typical evaluation metrics include the coefficient of determination (R²) and root-mean-square error (RMSE) [64]. The reliability model in the ENVI-met related to the evaluation criteria for the simulation and monitor parameters: R² close to 1, RMSE close to 0 [45]. Following previous studies, we focus on validating the T_a and RH combined with the evaluation metrics of R² in this study [45,46]. On-site monitoring of T_a and RH was performed on 22 August 2021, from 00:00 to 24:00 (Table 1). The monitored and simulated meteorological data were used to validate the current cases in the ENVI-met models for local sample communities.

For validating T_a and RH in three sample communities, R² in T_a was 0.87, 0.88, and 0.81; R² in RH was 0.87, 0.85, and 0.85 (Figure 9). The evaluation metrics performed reasonably, with R² close to 1. In this regard, the ENVI-met models in current cases were valid and reliable for compliance with the criteria in R². Based on validated current cases, this study develops the SLDS to renew models and improve the local thermal environment.

4.2. Microclimate Conditions in Renewal Strategies

4.2.1. Air Temperature (T_a)

In Figure 10, the maximum T_a simulated in the current cases was up to 33 °C; the minimum simulated T_a was about 26.9 °C. For all cases, see microclimate results in the Appendix A, the planting design cases (L1, L2, L3, and L4) are similar under the T_n, the optimal reduction of 0.2 °C in the HC area (

Table A1. Mean values of the simulation microclimate condition (at 1.8 m height) for the Synergistic Landscape Design Strategies (SLDS) (from a to c) on a small scale.

Simulation Runs in a Small-Scale	Code	Case	Mean Ta in HC (°C)	Mean RH in HC (%)	Mean WS in HC (m/s)	Mean Tmrt in HC (°C)	Mean Ta in AC (°C)	Mean RH in AC (%)	Mean WS in AC (m/s)	Mean Tmrt in AC (°C)
Current case	L1–4	L1	29.2	62.5	0.9	38.0	29.4	62.2	0.9	30.9
Planting design case		L2	29.1	62.3	1.1	35.2	29.4	62.6	1.9	31.4
		L3	29.0	63.5	0.7	33.1	29.5	62.8	1.7	32.1
		L4	29.1	62.5	1.1	35.0	29.5	62.9	1.7	31.5
Renewal small-scale cases	R1–4	R1	29.2	63.0	0.9	39.1	29.5	63.0	1.7	31.4
		R2	29.0	62.7	1.2	35.3	29.4	63.1	1.6	30.7
		R3	29.1	63.8	0.8	34.7	29.5	62.7	1.8	33.0
		R4	29.1	63.1	0.7	34.6	29.5	62.7	1.6	31.4
	F1–4	F1	29.1	63.3	0.6	37.3	29.5	62.6	1.6	33.5
		F2	29.1	63.8	1.1	35.2	29.5	62.8	1.9	32.3
		F3	29.0	64.5	1.2	34.4	29.5	62.9	1.7	32.3
		F4	29.1	63.1	1.2	35.2	29.4	62.7	0.9	34.2
	W1–4	W1	26.9	77.7	0.6	36.0	29.0	65.9	1.9	32.2
		W2	27.4	72.6	0.3	32.7	29.1	64.8	1.7	31.4
		W3	28.2	67.8	1.0	34.8	29.0	65.8	1.7	31.5
		W4	27.3	73.6	1.3	34.8	29.3	63.3	1.0	33.8

). In

Table A2. Microclimate conditions at 14:00 (at 1.8 m height) for the Synergistic Landscape Design Strategies (SLDS) (from a to c) on a small scale.

Simulation Runs in a Small-Scale	Code	Case	Mean Ta in HC (°C)	Mean RH in HC (%)	Mean WS in HC (m/s)	Mean Tmrt in HC (°C)	Mean Ta in AC (°C)	Mean RH in AC (%)	Mean WS in AC (m/s)	Mean Tmrt in AC (°C)
Current case	L1–4	L1	32.1	54.9	0.8	57.3	32.3	53.9	0.9	43.3
Planting design case		L2	31.8	54.4	1.1	54.9	32.8	52.5	1.9	46.3
		L3	31.8	56.3	0.7	50.4	33.0	52.2	1.7	46.5
		L4	31.8	55.0	1.1	54.3	33.0	52.2	1.6	46.3
Renewal small-scale cases	R1–4	R1	32.1	55.6	0.8	57.3	33.0	52.4	1.7	45.7
		R2	31.8	54.9	1.2	55.0	32.9	52.5	1.5	45.0
		R3	31.9	56.6	0.7	53.3	33.0	52.2	1.8	47.7
		R4	31.8	55.9	0.7	53.3	32.9	52.3	1.6	45.9
	F1–4	F1	31.9	56.3	0.6	55.8	32.9	54.2	1.5	47.6
		F2	32.2	54.9	1.1	54.4	32.9	52.9	1.9	46.9
		F3	31.8	57.5	1.2	52.6	33.0	52.4	1.7	46.4
		F4	31.8	56.1	1.1	54.2	32.7	53.3	1.0	46.0
	W1–4	W1	30.2	67.0	0.6	55.4	32.5	54.6	1.9	47.3
		W2	30.3	63.0	0.3	49.9	32.6	53.7	1.7	46.4
		W3	31.0	60.4	1.0	53.7	32.5	54.6	1.7	46.0
		W4	30.2	63.8	1.3	54.1	32.5	53.5	1.1	45.9

, T_a is reduced by 0.3 °C in the HC area and increased to 0.7 °C in the AC area. These results indicated that the combination of landscape layout and vegetation structure had a minor cooling effect in HC area under similar tree numbers [48]. Planting design involved mitigation strategies (“R, F, W”), and T_a significant reduction occurred in the “W” series of renewal cases (Figure 10). In the HC area, the W1 case had the lowest mean T_a of 2.3 °C compared to the L1 case. In the AC area, the W1 and W3 cases effectively reduce the heat transmission by up to 0.4 °C (Table A1). At 14:00, W1 and W4 cases were maximum reduced by 1.9 °C in the HC area (Table A2). Therefore, the cooling effect was more significantly observed in the HC area. In the large-scale strategies (“W-Ga-f”) (

Table A3. Mean values of the simulation microclimate condition (at 1.8 m height) for the Synergistic Landscape Design Strategies (SLDS) (from d) on a large scale.

Simulation Runs in a Large-Scale	Code	Case	Mean Ta in BC (°C)	Mean RH in BC (%)	Mean WS in BC (m/s)	Mean Tmrt in BC (°C)
Current case	W-Ga-f	Ga	29.1	60.8	0.4	29.6
Renewal large-scale cases		W-Gb	28.6	64.5	0.5	31.7
		W-Gc	28.3	66.1	0.3	28.7
		W-Gd	28.6	65.4	0.5	30.1
		W-Ge	28.5	64.9	0.4	32.0
		W-Gf	28.3	65.9	0.5	29.4

and

Table A4. Microclimate conditions at 14:00 (at 1.8 m height) for the Synergistic Landscape Design Strategies (SLDS) (from d) on a large scale.

Simulation Runs in a Large-Scale	Code	Case	Mean Ta in BC (°C)	Mean RH in BC (%)	Mean WS in BC (m/s)	Mean Tmrt in BC (°C)
Current case	W-Ga-f	Ga	31.7	53.4	0.4	43.6
Renewal large-scale cases		W-Gb	31.2	57.5	0.5	47.5
		W-Gc	30.6	60.1	0.3	41.9
		W-Gd	31.2	58.3	0.5	44.8
		W-Ge	31.0	58.5	0.4	47.3
		W-Gf	30.6	60.6	0.2	42.9

), the W-Gc and W-Gf cases effectively compared to the Ga case, decreased mean T_a by 0.8 °C and reduced the T_a by up to 1.1 °C at 14:00. Accordingly, the W-Gb case was only 0.5 °C lower than the Ga case at 14:00, further indicating that adding the urban trees positively affects cooling air temperatures than only used “W” strategy at large-scale (Table A4). The simulation results showed that the cooling effect of introducing quantitative trees with the same “W” strategy followed the order of coniferous trees > evergreen trees > deciduous trees.

4.2.2. Relative Humidity (RH)

In Table A1, the mean RH increased by up to 0.7–1.0% in the HC and AC areas. At 14:00, the L3 case was effectively increased by 1.4% in the HC area, in contrast to a 1.7% reduction of RH in the AC area. The simulation results of RH further prove that the planting design has a minor effect on humidification. However, the W-series cases were also highly effective in increasing humidification (Figure 10). The W1 case yielded the best humidification effect compared to the current cases in HC and AC areas, with a mean humidification increase of 3.7–15.2% and an improvement RH of 0.7–12.1% at 14:00 (Table A1 and Table A2). In Table A3 and Table A4, the mean RH increases by 5.3% and 7.2% at 14:00 in W-Gc and W-Gf cases. The large-scale simulated results indicated synergistic landscape design strategies can help regulate community humidification.

4.2.3. Wind Speed (WS)

In the microclimate simulation, the mean WS of the current case reached 0.9 m/s in the HC and AC areas. In comparison, the mean WS of the BC area with dense buildings was only 0.4 m/s (Table A1 and Table A3). The small-scale models simulated WS without vegetation obstruction (T_n = 0) can maximize the improvement of 1.0 m/s in the AC area (Table A1 and Table A2). The AC area formed an urban canyon, which realizes optimal summer ventilation in the absence of vegetation at the L2 case. Under the “W” strategy, the mean WS can maximize ventilation improvement up to 0.4–1.0 m/s in small-scale communities (Table A1). The well ventilation focused on the green roof and water body design (Figure 10). On this basis, the large-scale cases improve mean WS to 0.1 m/s, validated here to accommodate scale variations.

4.2.4. Mean Radiation Temperature (T_mrt)

The AC and BC areas have the lowest simulated T_mrt for the current case. The shadows of the high-rise buildings reduce the reflections (Figure 10). In contrast, the HC area has the highest mean T_mrt, reaching a maximum of 38.0 °C in the L1 case (Table A1). With the intervention in the L3 case, the mean T_mrt decreased by 4.9 °C in the HC area. In comparison, it increased by 1.2 °C in the AC area. At 14:00, it also reduced by 6.9 °C in HC and increased by 3.2 °C in AC (Table A2). The thermal radiation temperature in the planting design fails to synergize the two community building forms to optimize the thermal environment. In Table A1, the W2 case decreased the mean T_mrt by 5.3 °C in the HC area, while the R2 case decreased the mean T_mrt by 0.2 °C in the AC area. Compared to the Ga case, the W-Gc case was able to reduce the mean T_mrt up to 0.9 °C (Table A3). Introducing more coniferous trees based on the “W” strategy has a positive effect on thermal radiation resistance under a fixed number of urban trees.

4.3. Mitigation Performance in Community Thermal Environments

The thermal environment is improved by reducing the thermal index and mitigation time. In Figure 11, renewal cases with a mean PET below 38–42 °C (in a hot state) are predominantly found in the “W” serie simulations, with an overall mitigation time of 18–24 h, except for the W-Gd case.

4.3.1. Planting Design Effects on Sample Communities

Table 4. Daily mean PET (a) and hourly PET (at 14:00) distribution (b) relative to current case (L1) for the small-scale community simulations with planting design (L1–4).

(a)
Simulation runs in a small-scale	Code	Case	Mean PET in HC (°C)	Mean deviation relative to L1 (ΔPET) in HC (°C)	Mean PET in AC (°C)	Mean deviation relative to L1 (ΔPET) in AC (°C)
Current case	L1–4	L1	32.1	-	29.2	-
Planting design cases		L2	32.1	0.0	30.5	1.3
		L3	30.4	−1.7	28.7	−0.5
		L4	29.7	−2.4	28.1	−1.1
(b)
Simulation runs in a small-scale	Code	Case	Mean PET in HC (°C)	Mean deviation relative to L1 (ΔPET) in HC (°C)	Mean PET in AC (°C)	Mean deviation relative to L1 (ΔPET) in AC (°C)
Current case	L1–4	L1	43.0	-	37.9	-
Planting design cases		L2	41.4	−1.6	39.9	2.0
		L3	41.1	−1.9	37.5	−0.4
		L4	40.5	−2.5	37.5	−0.4

a shows that the mean PET is 29.7–32.1 °C in the HC area and 28.1–30.5 °C in the AC area. The HC area was significantly observed with PET values of 40.5–43.0 °C at 14:00 (in the hot and very hot state). In the AC area, the PET was 37.5–39.9 °C at 14:00 (in a hot state) (Table 4b). The thermal comfort was poorer in the HC area than in the AC area, and extremely poor weather conditions were more likely to lower the PET to 2.4 °C (Table 4a). However, the thermal comfort performance failed to significantly alleviate the heat stress conditions on a small scale. On thermal maps in Figure 12, the overall level distribution still keeps hot and very hot states.

4.3.2. Green Building Envelope (GBE) Design Effects on Small-Scale Communities

The renewal cases (R1–4) compared the L1 case in the HC and AC areas. The mean PET can be reduced by 0.2–3.7 °C. R2 and R3 cases can lower the mean PET by 1.0–5.5 °C at 14:00. In the F1–4 cases, the maximum reduction in mean PET was 0.8–2.4 °C. The F3 and F4 cases reduced PET by 1.4–2.9 °C at 14:00. In the W1–4 cases, the mean PET was reduced by up to 1.4–5.0 °C, and the hot hour (at 14:00) was also reduced by 0.8–5.1 °C (

Table 5. Daily mean PET (a) and hourly PET (at 14:00) distribution (b) relative to current case (L1) for the small-scale community simulations with GBE design cases (R1–4, F1–4, W1–4) based planting design (L1–4).

(a)
Simulation runs in a small-scale	Code	Case	Mean PET in HC (°C)	Mean deviation relative to L1 (ΔPET) in HC (°C)	Mean PET in AC (°C)	Mean deviation relative to L1 (ΔPET) in AC (°C)
Renewal of the small-scale cases	R1–4	R1	32.4	0.3	28.2	−1.0
		R2	30.0	−2.1	28.0	−0.2
		R3	28.4	−3.7	28.5	−0.8
		R4	29.6	−2.5	28.2	−1.0
	F1–4	F1	31.9	−0.2	29.3	0.1
		F2	29.8	−2.2	28.4	−0.8
		F3	29.9	−2.2	29.4	0.2
		F4	29.7	−2.4	28.4	−0.8
	W1–4	W1	29.9	−2.1	29.0	−0.2
		W2	29.0	−3.1	27.8	−1.4
		W3	27.1	−5.0	29.1	−0.1
		W4	28.6	−3.5	27.9	−1.3
(b)
Simulation runs in a small-scale	Code	Case	Mean PET in HC (°C)	Mean deviation relative to L1 (ΔPET) in HC (°C)	Mean PET in AC (°C)	Mean deviation relative to L1 (ΔPET) in AC (°C)
Renewal of the small-scale cases	R1–4	R1	43.4	0.4	37.4	−0.5
		R2	41.1	−1.9	36.9	−1.0
		R3	37.5	−5.5	37.6	−0.3
		R4	40.4	−2.6	37.5	−0.4
	F1–4	F1	42.9	−0.1	38.4	0.5
		F2	40.9	−2.1	37.6	−0.3
		F3	40.1	−2.9	38.3	0.4
		F4	40.4	−2.6	36.5	−1.4
	W1–4	W1	41.4	−1.6	38.1	0.2
		W2	40.4	−2.6	37.1	−0.8
		W3	37.9	−5.1	37.7	−0.2
		W4	39.5	−3.5	37.2	−0.7

). In Figure 13, the mitigation thermal comfort of the HC area was better than that of the AC area. Compared to the planted design cases (L1–4) (Figure 12), the W2 and W3 cases reduce the thermal index significantly to improve the thermal environment (ΔPET = 1.4–5.0 °C) (Figure 13 and Table 5). This result indicates that the water body and green roof realized a multiplicative effect on the thermal environment and the microclimate condition based on the planting design (Figure 10 and Figure 13). At 14:00, the HC area varied from very hot (PET > 42 °C) to warm (PET = 34–38 °C) in W3 and R3 cases, and the AC area stayed to warm level in R1–4 and W2–4 cases (Figure 13 and Table 5b). The GBE design strategies in 92% of cases provided for adjusting the thermal environment to a comfortable state (PET = 26–30 °C) (Table 5a). So the introduced water body and green roof effectively reduce the risk of summer exposure and facilitate thermoregulation.

4.3.3. Urban Tree Effects on Large-Scale Community

In

Table 6. Daily mean PET (a) and hourly PET (at 14:00) distribution (b) relative to current case (Ga) for the large-scale community simulations under urban tree effect (W-Ga-f) based water body and green roof design.

(a)
Simulation runs in a large-scale	Code	Case	Mean PET in BC (°C)	Mean deviation relative to Ga (ΔPET) in BC (°C)
Current case	W-Ga-f	Ga	30.4	-
Renewal of large-scale cases		W-Gb	29.5	−0.9
		W-Gc	28.1	−2.3
		W-Gd	30.5	0.1
		W-Ge	28.2	−2.2
		W-Gf	29.0	−1.5
(b)
Simulation runs in a large-scale	Code	Case	Mean PET inBC (°C)	Mean deviation relative to Ga (ΔPET) in BC (°C)
Current case	W-Ga-f	Ga	40.1	-
Renewal of large-scale cases		W-Gb	38.6	−1.5
		W-Gc	36.4	−3.7
		W-Gd	39.3	−0.8
		W-Ge	35.4	−4.7
		W-Gf	37.8	−2.3

a, the W-Gc and W-Ge cases effectively reduce the mean PET by 2.2–2.3 °C compared to the Ga case. However, W-Gb was only reduced by 0.9 °C. The W-Ge case reduced PET by 4.7 °C but only by a reduction of 1.5 °C in the W-Gb case at 14:00. The results indicate that adding coniferous or evergreen trees combined with the “W” strategy resulted in multiple effects on mitigating the thermal environment (Figure 11 and Figure 14). Adding water bodies and green roofs provides less thermal stress mitigation for the large-scale community. Still, it introduces a proportionate number of coniferous or evergreen trees that can adapt to the thermal conditions from hot to warm at 14:00 (Table 6b). Compared to the results of Table 5 and Table 6, this research found that the difficulty of thermal mitigation is greater on a large scale than on a small scale. The nature-based case is to reintroduce urban trees, where the mitigation thermal index performance (ΔPET) is in the order of coniferous trees > evergreen trees > deciduous trees (Table 6a). With the synergistic GBE, urban trees, and pavement material, the mean PET was changed from slightly warm (PET = 30–34 °C) to a comfortable standard (Table 6a). Thus, holistic blue-green infrastructure contributes to the climatic resilience and sustainability of Komatsu’s traditional residential and flat houses.

5. Discussion

Synergistic landscape design strategies were used to regulate microclimate and thermal stress in Japanese communities. This study validated that sample communities are more suitable for SDLS adopting water bodies and green roofs to improve daytime and nighttime thermal regulation. In addition, more coniferous or evergreen trees as urban trees in large-scale communities capture the overall building form, landscaping, climatic characteristics, scale, and time mitigation. Simultaneously, the synergy of more than three landscape design strategies is more conducive to the renewal plan of the central urban community.

5.1. Implications for the Planting Design

The simulation results in this paper show that planting design has a small effect on improving the thermal environment in Japanese communities under the SLDS condition (a). This result is confirmed in some subtropical cities [48,65], especially changing landscape layout and vegetation structure under similar tree numbers. In this study, L2 and L4 cases result in similar microclimate conditions under T_n = 0. The reduction of the thermal index is strongly affected in the L4 case (ΔPET = −2.5 °C) (Table 4). The reduction of thermal radiation in the neighborhood was mainly concentrated in the HC area by planting trees and grass in an enclosed layout (L3) (Table A1). This finding further indicated that urban trees reflect and absorb via irradiance, obtaining transpiration and shade cooling for the local microclimate. However, thermal comfort was still in the hot level state at 14:00. Compared to previous studies on community design in ENVI-met [41,42,44,45], this study found that the planting design fails to realize improvements in scale effect and the mitigation time for the HC and AC building form (Figure 10 and Figure 12). Therefore, considering the planting design level alone will not achieve resilient community regeneration. Three or more strategies are sufficient to realize a wide range of environmental parameter improvements, especially the SLDS conditions (c) and (d) have significant positive effects (Figure 11).

5.2. Advantages of Green Building Envelope (GBE) Design

The GBE designs show that introducing water bodies and green roofs (W) has a strong cooling effect and leads to a more comfortable thermal environment (Figure 11 and Table 5). Within this strategy, 85% of the alternatives resulted in thermal mitigation during the summer day and night (Figure 11). Moreover, the community thermal conditions in the sample area were moderated to warm levels at 14:00, with an optimal PET reduction of 0.8–5.1 °C (Figure 13 and Table 5b). According to Lalošević et al. [66], green roofs at the pedestrian level achieve a maximum cooling effect of 0.9–1.6 °C when combined with high albedo cooling materials. However, in the present study, the water bodies and green roofs effectively reduced T_a to 0.4–2.3 °C and increased RH to 3.7–15.2% (Table A1). These results indicated that water body dispersion obtained the greatest cooling effect [14,67]. Previous studies on Cfa-climate design strategies [48,58,62,65,66] via one or two-level optimization methods, this study seeks a better thermal contribution under these three-level strategies (planting design, green roofs, and water body design) to mitigate the thermal environment producing positive cooling and humidifying effects.

Mitigation strategies in the green roof (R) at the pedestrian level significantly reduced the PET values in the HC area. In contrast, the AC area had a negative effect (Table 5). The AC area is modeled for building heights above 30 m; therefore, the effect of green roofs on air temperature at the pedestrian level becomes negligible for building heights of 10 m above [19]. There is very low microclimate regulation at pedestrian heights (Table A1), and these results are mostly influenced by the elevation between vertical buildings and green roofs [13,68,69]. The mitigation strategy of the green roof and green facade (F) is the least effective in cooling and improving thermal comfort at the pedestrian level (Table 5b), with 40% of the alternatives failing to provide thermal regulation during the day and night (Figure 11). The primary explanation for this result is that the time-spatial distribution during the night provides a passive warming effect by suppressing longwave radiation emitted from building facades and attenuating vegetation evaporation [13,43,70]. The HC area provides better cooling performance than the AC area in small-scale simulation with a green facade and green roof, validating the advantage of thermal mitigation for low-rise communities (Table A1 and Table A2). This result is consistent with a previous study [4].

5.3. Urban Tree Effect on Sample Communities

Urban tree effects produced different effects on thermal performance due to wind speed, wind direction, and shade [13]. The L2 case has a maximum wind speed that improves thermal comfort in hot time, whereas the L3 and L4 cases significantly reduce the thermal index due to crown shade (Figure 12 and Table 4). The tree-planting design includes key attributes such as the number, type, layout, and arrangement of the trees [13,32,44,48]. In similar tree coverage and the GBE design, the cooling effect and thermal mitigation depend on planting design and building form in renewal cases such as R3, F3, and W3 cases on a small scale (Table 5 and Table A1). The results suggest that the number of urban trees can provide more shade and thermal benefits, especially in an enclosed layout (a rectangular pattern) [48,65]. In this study of the renewal cases (L2, R2, F2, and W2) without plant trees, great wind speed has a lower thermal index among the green building envelope designs, especially with the green roofs and the water body design. Compared to previous studies that have improved colling potential by increasing the number of trees [6,7], this study focuses on summer thermal mitigation synergizing multifaceted optimization of environmental criteria, with coniferous trees and evergreen trees implementing a co-optimized large-scale community design for microclimatic conditions (T_a, RH, WS, T_mrt) and thermal comfort (PET) in T_n = 208 (Table 6 and Table A3). Previous studies on a higher tree coverage ratio have shown that it can lower T_a and PET and raise RH in the daytime [32]. Still, this study has effectively multifactor improvements (T_a, RH, WS, T_mrt, and PET) via green roof and water body design (W) via introducing coniferous or evergreen trees in sample communities can mitigate summer thermal comfort for all day, especially at large-scale community (Figure 10 and Figure 11).

5.4. Scale Adaptation for Microclimate and Thermal Comfort Mitigation

5.4.1. Small-Scale Community Models

A strong synergistic cooling effect exists between green and blue infrastructure (GBI) [13]. Especially in small-scale models, T_a reduction reached a maximum of 2.3 °C in green roof and water body design. In contrast, a single green roof under intervention is less effective for cooling than the former at pedestrian heights (Table A1). The main reason is that water bodies have a strong microclimatic impact, especially the mitigation of summer temperatures and the openness of water bodies to increase arboricultural shading effects and to promote natural ventilation, where small-scale modeling ventilation is significant and responsive to scale requirements. The optimal performance of the planting design varies with the characteristics of the building form [65]. It forms an area of high-rise building canyons with weaker heat stress than the single-family building areas of the HC, where the building blocks are heat sources by conductive and convective heat fluxes to reduce the urban surface heat stress (Figure 13 and Figure 14) [71]. In the renewal strategy of the small-scale model, the best results for the optimization of microclimate conditions and thermal comfort at the pedestrian level were in the following order: W > R > F > L (Table 5 and Table A1). This study found that the GBE design contributed to improving the thermal environment more than the small-scale planting design with similar numbers of trees in terms of cooling and humidifying effects (Figure 10). In addition, the thermal comfort of the Japanese community in both scales was simultaneously optimized from very hot and hot levels to a warm condition at 14:00 (Figure 13 and Figure 14). Referencing Zhang et al. [15], this study found that in subtropical urban parks with field measurements results of 15 small-scale vegetation communities, the cooling and humidification effect of the multi-layer vegetation structure (trees, shrubs, and grasses) is significant. Accordingly, this simulation experiment (W1–4 cases) found that the cooling and humidification effect is optimal under the mixed structure planting (W1 case) (Table A1). In contrast to Rui et al. [5], an increased green quantity results in a smaller improvement in the thermal environment. The SLDS in the present study was used to obtain a more significant thermal mitigation effect on a small scale, such as a mitigation time that can optimally reach 24 h. The optimal SLDS (“W”) satisfies multifactorial improvements (T_a, RH, WS, T_mrt, and PET) (Table A1). Notably, the predominance of high-rise buildings in the AC area has a disadvantage in reducing thermal radiant temperatures than in the HC area. Similar to a previous study [16], the vegetation had the lowest impact on the community’s mean T_mrt, and the low-rise community had a better cooling effect on the microclimate than the high-rise community.

5.4.2. Large-Scale Community Models

The large-scale models re-simulated the best optimization strategy “W” from the small-scale models, which showed an optimal reduction in ΔPET of 4.7 °C, and the thermal comfort level is assessed at a warm level at 14:00 (Table 6b). These results verified the effectiveness of small- and large-scale models in implementing the “W” renewal plan, especially regarding climate resilience. What’s more, introducing tree types and numbers (Gb, Gc, Gd, Ge, Gf) (T_n = 208) based on the “W” strategy, the large-scale model thermal comfort states (ΔPET) are in order of improvement: Gc > Gf > Ge > Gb > Gd (Table 6). Therefore, coniferous trees in the Japanese community are more effective in reducing PET than other tree types. The cooling effect of trees by reducing mean T_a on the thermal environment in the order Gc ≥ Gf > Ge > Gb ≥ Gd (Table A3). Moreover, the coniferous tree is beneficial in reducing T_mrt and affecting PET levels in the W-Gc and W-Gf cases. According to Xiao and Yuizono [48], in the ENVI-met simulation for summer, planting more evergreen trees results in a more comfortable environment than planting deciduous trees, which is consistent with this finding. Compared to other studies that have improved cooling potential by urban tree effects [6,7,48], this study’s contribution validated that urban tree strategies (Gc, Ge, and Gf) could mitigate thermal comfort during the day and night (Figure 11). Accordingly, the best strategy to promote the regeneration of the Komatsu community is to validate multiple thermal mitigation to capture urban physical features (i.e., scale, climatic characteristics, cooling, humidification, and time mitigation). The introduction of optimal strategies allows a more climate-responsive architectural design of the community for renewal, adapting to the summer characteristics of the Cfa climate.

5.5. Limitations

Here, this study selected two typical building types of Japanese communities in a hot-humid climate context and used the combination of numerical simulation and thermal index assessment to select a synergistic landscape design strategy. However, some limitations and areas for improvement in the simulation method and our outcomes are listed as follows. (i) In the synergistic landscape design strategy, implementations focused on natural solutions to mitigate the thermal environment regarding planting design with structure, layout, tree numbers, green roofs/facades, and water body design. The pavement material was only replaced in the W-Gf case, which lessened the consideration of the cooling ground material. Simultaneously, this model of the urban community focuses on improving the thermal environment in the summer and lacks an assessment of the overall renewal of the area in all seasons. (ii) In the ENVI-met simulation, no explicit microclimate sensitivity parameters were given for native trees and green roofs/facades. Instead, the default arbor types settings and LAD values were used in ENVI-met V5.0.1, which has some limitations on the model accuracy. Unlike other studies, most low LAD trees were replaced with different tree species. Among them, the practice of substituting coniferous trees for evergreen ones in this study is relatively less in reference cases [72]. (iii) There are some limitations in modeling water body design using fountain settings in ENVI-met, and most of the simulations investigating water body design replace the model with a fountain model, lacking specific parameter settings and iterations of model algorithms.

6. Conclusions

Synergistic landscape design strategies were proposed and assessed thermal environments in densely populated Japanese communities to guide anticipatory design decisions and guidance for urban renewal and resilience. Summer thermal comfort was analyzed using the PET thermal index and microclimatic parameters. ENVI-met V5 reproduced the current and alternative cases to simulate and apply the new design at a small and large scale. The community-based design strategies in a mechanism were systematically examined and compared. A three-level optimization method demonstrates its effectiveness. Key findings include:

(1)

Applying a systematic SDLS mechanism to the urban community created an optimal strategy using a water body and green roof at two scales, which proved effective in microclimate and thermal improvement in summer during the day and night, optimizing outdoor comfort all day in 85% of cases.

(2)

There are synergistic landscape design strategies that result from integrated water design, green roofs, and urban trees (i.e., three-level optimization method) on Japanese building-specific community forms to achieve better adaptive design for building form, landscape design, climatic characteristics, scale, and time mitigation that reduces occupant health risks in summer, especially improves thermal comfort from a very hot to warm level at 14:00.

(3)

Regarding urban trees, adding evergreen or coniferous trees is more conducive to reducing the urban heat island phenomenon in a large-scale community. The optimal strategy “W” will be applied to the Komatsu City Renewal and Maintenance Plan project.

These planning strategies can guide a decision-making reference for landscape architects who design communities for the SDGs, including specific cases and other Japanese community areas characterized by Cfa-climates. Furthermore, our simulation results can provide design support for urban community planning and design to confirm the optimization of the thermal environment in hot summer, including the selection of urban trees.