Relevance of Ground and Wall Albedo for Outdoor Thermal Comfort in Tropical Savanna Climates: Evidence from Parametric Simulations

Abstract

High-albedo ground and wall materials are promoted to mitigate heat stress in tropical climates, yet conflicting evidence driven by climatic and metric variability make their impact on Outdoor Thermal Comfort (OTC) unclear. This study employed parametric simulations to assess how ground and wall albedo affect OTC, measured via the Universal Thermal Climate Index (UTCI) in typical urban canyons. Using ENVI-met, we tested ground albedo (0.2–0.8) and wall albedo (0.05–0.90) with emissivity fixed at 0.9. Findings reveal that ground albedo had a minimal impact on the UTCI (mean amplitude 0.44 °C), while wall albedo reduced the UTCI by up to 2.80 °C, prioritizing wall material selection for heat mitigation. It was also found that the increase in ground albedo offsets the cooling potential of high-albedo walls. Furthermore, differences in the impact under shaded and unshaded areas were observed. These results question assumptions of universal high-albedo benefits, recommending case-specific simulations in urban design.

1. Introduction

1.1. Urban Heat and Thermal Comfort in Tropical Areas

Tropical areas, encompassing rainforest, monsoon, and savanna climates, face challenges in achieving pedestrian thermal comfort due to high temperatures, humidity, and intense solar radiation [1]. These thermal conditions can result in elevated levels of heat stress, adversely affecting human health, well-being, and productivity and limiting the quality of outdoor spaces and their usability, particularly in public areas where people engage in social, recreational, and economic activities [2]. Tropical savanna climates, covering 17.5% of the global population and characterized by intense solar exposure and rapid urbanization, are particularly critical yet understudied [3,4]. The Urban Heat Island (UHI) effect, which refers to the phenomenon of urban areas being significantly warmer than their rural surroundings due to human activities and dense infrastructure (e.g., buildings, roads, and other surfaces), exacerbates this problem [5,6] and, as such, has become a key consideration in sustainable urban studies, design, and planning, to improve urban livability [7].

1.2. Influence of Surface Materials on Thermal Comfort

In urban environments, surface materials play a significant role in influencing outdoor thermal comfort [6,8,9]. Materials are characterized by their albedo, the fraction of shortwave solar radiation reflected (0–1), and emissivity, the efficiency of longwave radiation emission, among other properties [6]. Studies have shown that materials such as concrete, asphalt, glass, and high-albedo surfaces are key factors in shaping the thermal environment [10,11]. These materials interact with solar radiation by absorbing or reflecting it, impacting local temperatures and pedestrian comfort [12,13]. Ground albedo has been a focal point in several investigations of how urban surfaces affect the outdoor thermal environment. Some authors like Xinjie Huang et al. explored the use of retro-reflective materials on urban surfaces. They found that retro-reflective surfaces could reduce surface temperatures by as much as 20 °C and air temperature by 2.6 °C, significantly improving pedestrian thermal comfort [14]. This demonstrates the potential of high-albedo materials in mitigating the Urban Heat Island effect and enhancing thermal comfort in outdoor spaces. However, Erell et al. [15] cautioned that high-albedo pavements increase reflected radiation, potentially elevating pedestrian heat stress, highlighting the need for sustainable material selection to balance cooling and radiant loads. Other studies examined the application of reflective coatings on urban surfaces in tropical settings. They highlighted how reflective paints and high-albedo pavements could reduce heat absorption, thereby improving thermal comfort in public spaces [16,17,18]. In contrast to ground materials, wall albedo has received less attention in outdoor-focused studies but is equally important in the study of pedestrian thermal comfort. Elisavet Tsekeri and colleagues (Technical University of Crete) researched the impact of highly reflective materials on building facades. Their study found that the application of cool materials to building envelopes could reduce cooling energy needs by up to 30%, illustrating the significant role building walls can play in both energy efficiency and thermal comfort [19]. This suggests wall albedo’s potential for sustainable urban design, though its interaction with ground albedo remains underexplored, necessitating integrated approaches to optimize livability [12,17,19]. Studies have shown that roof material albedo can affect thermal conditions. Specifically, the choice of highly reflective roofs can lower indoor air temperatures and reduce building cooling energy needs [20,21,22], but apart from building-scale studies (focused on indoor conditions and energy saving), urban-scale studies have shown a less significant impact of roof albedo on street-level heat mitigation; the impact is quite limited unless extensively implemented on a large scale [23,24], which is why it will not be discussed further in this paper (rather focused on pedestrian-level thermal conditions). Some studies in tropical climates have further emphasized the importance of material albedo in improving thermal comfort. For instance, the potential of reflective coatings on walls to reduce heat transmission and improve outdoor comfort in hot, humid environments was discussed by Bulbaai and Halman [16]. The potential of combining reflective materials with the optimal building orientation, as a passive design strategy, was also discussed, concluding that the careful selection of materials, in combination with smart design choices, can substantially mitigate thermal discomfort in urban areas [25].

1.3. Current Research Limitations and Study Objectives

The existing literature, for the most part, highlights the significance of both ground and wall albedo in improving pedestrian thermal comfort in tropical urban settings. However, despite the importance of albedo, these studies also indicate that the effectiveness of high-albedo materials is context-dependent, and further research is needed on the matter [26]. Although a significant body of research has investigated the role of urban surface albedo in mitigating urban heat, there remains a lack of consensus regarding the net impact of high-albedo materials on the outdoor thermal environment in general and in tropical areas in particular. The general understanding in the literature is that high-albedo materials reflect more solar radiation, thus reducing heat absorption and minimizing urban heat stress in summer [27,28]. Conversely, materials with lower albedo absorb more solar radiation, which they re-emit as longwave radiation, contributing to increased heat retention and exacerbating the Urban Heat Island (UHI) effect [29]. However, some studies have highlighted the complexities of this relationship, noting that local climate conditions and hydrological feedback can offset the radiative effects of surface albedo, leading to differing results across studies [15,30]. Myhre and Myhre [31] demonstrated that albedo changes in tropical regions have stronger radiative forcing due to high solar insolation, reducing the surface absorption of solar radiation and lowering surface temperatures. This decrease in surface heating and longwave emission directly mitigates radiant heat exposure for pedestrians, influencing the Universal Thermal Climate Index (UTCI) and underscoring the need for sustainable urban planning to optimize surface materials for pedestrian-level thermal comfort.

This complexity is particularly pronounced in tropical regions, where the impact of albedo changes can vary significantly compared to higher latitudes. Research by Myhre and Myhre (2003) [31] and others has shown that tropical regions exhibit a stronger radiative forcing response to changes in surface albedo. However, the precise impact of these changes on thermal comfort in tropical climates remains uncertain [31]. Xu et al., from the Massachusetts Institute of Technology (MIT), have also pointed out that experimental results are influenced by factors such as the location, experimental settings (e.g., model resolution), and specific metrics used to evaluate thermal comfort [26]. While much of the existing research focuses on radiative effects or temperature-based metrics, fewer studies have explored how material choices—such as wall and ground albedo—directly influence human thermal comfort in outdoor urban environments, particularly in tropical savanna climates, where intense solar radiation and rapid urbanization amplify heat stress [32]. Meanwhile, The Universal Thermal Climate Index (UTCI) and the Physiological Equivalent Temperature (PET) are increasingly recommended as comprehensive physiological metrics for assessing the impact of urban design choices on the outdoor thermal environment, rather than the unidimensional metrics of air temperature and radiation, wind speed, etc. [26].

Given the limited research on tropical savanna climates and the scarcity of studies differentiating ground and wall albedo effects measuring thermal comfort, this research aims to fill a significant gap. This study poses the following question: how do ground and wall albedo independently and interactively affect pedestrian thermal comfort (UTCI) in tropical savanna urban canyons, and what material combinations optimize sustainable urban design for heat mitigation and livability? The Universal Thermal Climate Index (UTCI), which integrates air temperature, humidity, wind, and radiation to assess human heat stress, offers a comprehensive measure for pedestrian thermal comfort [26]. Through a parametric simulation method, this study intends to simulate possible combinations of ground (albedo 0.2–0.8) and wall (0.05–0.9) albedo rates with emissivity fixed at 0.9 to control longwave effects and seek eventual patterns in relation to pedestrian thermal comfort. In contrast to temperature- and radiation-focused studies, this study takes a human-centered approach, using UTCI to evaluate how material selection impacts pedestrian comfort in tropical savanna climates. The study findings, tuned to tropical savanna climate conditions, are expected to contribute to the broader understanding of urban thermal comfort in tropical regions and provide valuable insights for urban designers and policymakers working in similar environments.

2. Materials and Methods

2.1. Overview

This study employed parametric simulations to evaluate ground and wall albedo effects on pedestrian thermal comfort in tropical savanna climates, aiming to inform sustainable urban design by optimizing material selections for heat mitigation and livability. The study used parametric simulations to evaluate ground and wall albedo effects on pedestrian thermal comfort in tropical savanna climates, measured via the Universal Thermal Climate Index (UTCI). The simulation technique builds multiple scenarios by combining pre-selected albedo values for ground and wall materials. The microclimate simulation software ENVI-met [33] was used to model the urban environment, varying ground and wall albedo (ground: 0.2–0.8; wall: 0.05–0.90), with emissivity fixed at 0.9 to control longwave radiation effects. A validated parameterization for tropical climates was applied, maintaining consistent grid resolution and weather boundary conditions across scenarios. A total of 117 receptors were placed at pedestrian height (1.5 m) across the model, with locations conserved for all scenarios. UTCI values were exported to Excel for post-processing, grouped by material combinations to identify patterns in thermal comfort.

2.2. Surface Material Selection

The ground albedo range (0.2–0.8) and wall albedo range (0.05–0.90) were selected to represent typical urban materials, reflecting their optical properties as documented in the literature and industry standards [10,16,17]. Ground albedo spans common pavement materials, from dark asphalt (albedo ~0.2) to light concrete (albedo ~0.8), covering the practical range for urban surfaces where lower albedos (<0.2) are rare and higher albedos (>0.8) are limited by maintenance and glare concerns [10,17]. Wall albedo encompasses a broader range due to the greater diversity of facade materials, from highly transmissive clear float glass (albedo 0.05, due to 90% transmission) to highly reflective aluminum (albedo 0.90), including concrete and insulated walls [19,34]. The wider wall albedo range (0.85 span vs. 0.6 for ground) reflects the varied optical behaviors of transparent, opaque, and metallic facades compared to pavements’ functional constraints. These ranges align with ASTM E1918 standards for albedo measurements and are consistent with urban design applications [10,19].

(i)

Ground materials

The urban environment was modeled using loamy soil (albedo 0.15, emissivity 0.9) as the natural ground material (on unpaved surfaces) and maintained identically across all scenarios. Four concrete pavements with albedo values of 0.2 (dark), 0.3 (dirty), 0.5 (gray), and 0.8 (light) were used for roads and open spaces, all with emissivity of 0.9, to assess albedo’s impact on the UTCI. See Figure 1 for the detailed characteristics of the pavement materials used and Figure 2 for their 3D illustration in the ENVI-met models.

(ii)

Wall materials

Four wall materials were used on the buildings, corresponding to albedo values of 0.05 (clear float glass, W1), 0.3 (hollowed concrete block, W2), 0.45 (moderately insulated wall, W3), and 0.90 (aluminum, W4), all with emissivity of 0.9. For glass, albedo (0.05) serves as a parametric proxy for reflectivity, adjusted by its dominant transmittance effect, ensuring consistency across the range [16,34]. In Figure 3 below, the 3D views illustrate the wall surface variation based on identical urban geometries and their respective physical properties as captured from the ENVI-met material database (where wall albedo values are characterized as reflection). In all scenarios, a concrete slab (with hollow blocks) was used for the building roofs.

2.3. Surface Material Combinations

As shown in the table below (

Table 1. Parametric combination of wall and ground albedo.

Urban Geometry	Ground Albedo (Gi) G1 = 0.2, G2 = 0.3, G3 = 0.5, G4 = 0.8	Gi/Wall Albedo (Wi) W1 = 0.2, W2 = 0.3, W3 = 0.5, W4 = 0.8	Scenario No
UGref	UGref/G1	UGref/G1/W1	UGref/Mat1
		UGref/G1/W2	UGref/Mat 2
		UGref/G1/W3	UGref/Mat 3
		UGref/G1/W4	UGref/Mat 4
	UGref/G2	UGref/G2/W1	UGref/Mat 5
		UGref/G2/W2	UGref/Mat 6
		UGref/G2/W3	UGref/Mat 7
		UGref/G2/W4	UGref/Mat 8
	UGref/G3	UGref/G3/W1	UGref/Mat 9
		UGref/G3/W2	UGref/Mat 10
		UGref/G3/W3	UGref/Mat 11
		UGref/G3/W4	UGref/Mat 12
	UGref/G4	UGref/G4/W1	UGref/Mat 13
		UGref/G4/W2	UGref/Mat 14
		UGref/G4/W3	UGref/Mat 15
		UGref/G4/W4	UGref/Mat 16

), based on one identical Urban Geometry (UGref) with a street aspect ratio (height/width) of about 2.0, representing typical deep canyons in tropical savanna cities [35], the simulation scenarios are built by parametrically combining the four types of ground materials (named G1, G2, G3, and G4) and the four types of wall materials (named W1, W2, W3, and W4), forming 16 scenarios in total.

2.4. Target Climatic Area

The experiment targets tropical savanna climates (Aw per Köppen classification), a subset of tropical areas characterized by high solar radiation and distinct wet/dry seasons [36]. This focus is justified by their demographic importance (17.5% of the global population) and rapid urbanization, yet limited microclimate research [3,4,32]. Tropical climates (Figure 4) are of three types according to the Köppen climate classification: tropical rainforest (Af), tropical monsoon (Am), and tropical savanna (Aw/As), mainly distributed over Latin America, Sub-Saharan Africa, and Southeast Asia [36]. As studies have pointed out, 17.5% of the world’s population has lived in a tropical savanna climate (Aw) since the 90s [3], making it the second most populated. These regions have the fastest urban population growth, where the fastest urbanization is expected by 2050 [4]. A. Mellinger argued that tropical climates could be “highly detrimental” to human settlement and challenging for long-term economic development [3,4,32]. Given their urban demographic significance and research gaps, tropical savanna climates warrant focused study to guide urban planning [32].

2.5. Software Reliability and Parameterization

ENVI-met, a high-resolution 3D microclimate simulation software, was selected for this study due to its robust capability to model complex urban environments, including radiative, convective, and conductive heat exchanges at the pedestrian level (1.5 m). Its advantages include detailed parameterization of surface albedo, shortwave (SW) and longwave (LW) radiant fluxes, and mean radiant temperature (MRT), which are critical for assessing albedo effects on thermal comfort [33,37]. Unlike other user-friendly simulation tools, such as RayMan or SOLWEIG, etc., ENVImet has a detailed material interaction modeling capability and multi-layer urban canopy representation, making it ideal for our hypothetical experiment [37]. In this study, ENVI-met v4.4.4 was used to conduct the 3D modeling and the outdoor thermal comfort calculation for each of the 16 scenarios. ENVI-met was chosen for its widespread use among urban microclimate researchers and confirmed reliability in tropical climates, with 8.5% of studies validating its performance [37]. A previous study screened ENVI-met’s validation metrics in tropical summer conditions, comparing simulations with field measurements in Bangkok, Thailand; Cuiaba, Brazil; Akure, Nigeria; and Pathanamthitta, India [35]. Overall, ENVI-met showed high reliability for air temperature, relative humidity, and the mean radiant temperature, with correlations (R²) exceeding 0.90 and acceptable errors (

Table 2. ENVI-met model validation metrics as reported in Bangkok, Cuiaba, Akure, and Pathanamthitta.

Location	Parameters	Model Validation Criteria
		Correlation (R2)	Error		Bias
Bangkok, Thailand [38]	MRT	0.91
Cuiaba, Brazil [39]	Air temperature	0.95–0.98	2.39 (RMSE)	2.00 (MAE)	1.29
	Relative humidity	0.91	14.32 (RMSE)	14.31 (MAE)	−14.31 (MBE)
		0.90	2.72 (RMSE)	4.21 (MAE)	−2.25(MBE)
Akure, Nigeria [40]	Air temperature	0.96–0.99	0.00–0.01 (NMSE)		0.01–0.06 (FB)
	Relative humidity	0.82–0.90	0.00–0.01 (NMSE)		0.01–0.07 (FB)
Pathanamthitta, Kerala (India) [41]	Air temperature	0.80–0.92	0.58–0.72 (RMSE)	0.48–0.77 (MAE)

) [38,39,40,41].

Studies have shown that the accuracy of the ENVI-met simulation is determined by the grid resolution, domain size, and boundary conditions. Reported ENVI-met calibrations in tropical areas showed that the simulation settings are considerably similar. Apart from reasonable differences in the domain size, the simulation period, and the start time, the simulation duration is typically more than 48 h, and the grid resolution is around 2 m or 2.5 m in x, y, and z directions. For meteorological boundary conditions, the wind speed is not more than 3 m/s, the wind direction is often south or southeast, the air temperatures vary between 24 °C and 38 °C, with relative humidities between 50% and 90%, and there is a clear sky condition. Based on the reported simulation settings and their accuracy (

Table 3. Model configuration and initialization parameters’ values.

Parameters	Values/Configuration
Climate type	Warm-humid (Aw)
Simulated summer period	1 September–3 September (represent peak summer conditions in tropical savanna climates [36])
Simulation duration	72 h (ensure model stabilization [37])
Start time	6:00 A.M.
Spatial resolution (grid size)	2 m × 2 m × 2 m
Domain Size	360 m × 360 m × 100 m
Wind speed (m/s)	3 m/s
Wind direction (°)	265° (Southwest)
Air temperature (°C)	25.1–29
Relative Humidity (%)	85–90
Sky condition	Clear

), calibrating ENVI-met simulations around the reported values should give acceptable results. This study used ENVI-met V4, calibrated based on Morakinyo et al. [40] in Akure, Nigeria, for a high correlation and low errors. This parametric study employs hypothetical urban canyon scenarios (aspect ratio 2.0) to isolate albedo effects, leveraging Morakinyo et al.’s validation in a similar tropical savanna climate (R² > 0.96) and ENVI-met’s robustness across urban morphologies [35,37]. Site-specific validation was deemed unnecessary due to the hypothetical design and prior validations, though this introduces potential uncertainty discussed in Section 4.3. The domain size used here is 360 m × 360 m × 100 m, corresponding to 180 × 180 grids in the x and y directions. In the z direction, the first grid subdivision function is applied with a 10% telescoping grid above 2 m. The standard TKE model is applied. The 72 h simulation period (1–3 September) represents peak summer conditions in tropical savanna climates, characterized by high solar radiation and dry-season weather [36], with 48 h for model stabilization and 24 h for analysis, following standard ENVI-met protocols [37].

2.6. Output Processing

The simulation outputs were processed in the Biomet tool incorporated in ENVI-met v4.4.4 for thermal comfort calculation. The thermal comfort index UTCI was calculated in each case based on the ISO 7730 standard values [42] for human biological parameters. The output UTCI values were then exported to Excel for post-processing. A total of 117 receptors were inserted at a 1.5 m height, distributed and classified on N-S streets, E-W streets, and public spaces, to capture varied solar exposure, as E-W streets and open spaces receive more radiation than N-S streets [35]; also, the distribution along buildings (Figure 5) ensures that some receptors are shaded by nearby buildings, which later allows differences between shaded and unshaded receptors to be observed. This stratified distribution of receptors is intended to minimize spatial uncertainty; with the relatively high receptor density across the model, Standard Errors in the collected UTCI (0.07–0.2, Appendix A) were sufficiently low to ensure acceptable precision.

UTCI data tables were generated from 6:00 to 18:00 on 2 September, representing peak daytime conditions after model stabilization. Hourly UTCI calculations align with standard ENVI-met practices, balancing computational efficiency and accuracy for diurnal trends [37], with low Standard Errors (0.07–0.2, Appendix A) indicating reliable results. A finer 30 min resolution could capture transient solar peaks, a potential refinement noted in Section 4.3. Since there are 16 models, in total, 208 tables have been generated containing the outputs of the 117 receptors, from which the results are plotted and analyzed. The UTCI data collected at all receptors showed an overall Standard Error of about 0.2 on average; when separately computed for shaded and unshaded receptors, even lower Standard Errors were obtained (0.07 in average).

3. Results

3.1. UTCI Variation According to Ground Materials

Ground albedo (0.2–0.8) had a minimal impact on the UTCI across all wall albedo scenarios (0.05–0.90) and street orientations, with a mean amplitude of 0.44 °C (Figure 6). Linear regression showed a low correlation (R = 3–10%, R² = 0.001–0.009) and flat slopes (0.35–1.16), with p-values mostly above 0.05, indicating statistical insignificance (see Appendix B,

Table A2. Linear regression statistics for X variable = Gi, Y variable = UTCI.

	Gi W1			Gi W2			Gi W3			Gi W4
	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded
Multiple R	0.03	0.31	0.06	0.07	0.42	0.18	0.06	0.39	0.14	0.10	0.55	0.23
R Square	0.00	0.10	0.00	0.00	0.17	0.03	0.00	0.15	0.02	0.01	0.31	0.05
Adjusted R Square	0.00	0.09	0.00	0.00	0.17	0.03	0.00	0.15	0.02	0.01	0.30	0.05
Standard Error	2.56	0.45	0.60	2.25	0.55	0.45	2.35	0.53	0.49	2.78	0.57	0.83
Observations	468	184	284	468	184	284	468	184	284	468	184	284
Slope	0.35	0.64	0.16	0.64	1.08	0.35	0.57	0.99	0.31	1.16	1.64	0.85
Coef. Standard Error	0.52	0.15	0.16	0.45	0.18	0.12	0.47	0.17	0.13	0.56	0.18	0.21
t Stat	0.67	4.39	1.00	1.41	6.16	3.06	1.21	5.78	2.41	2.06	8.96	3.96
p-value	0.50	0.00	0.32	0.16	0.00	0.00	0.23	0.00	0.02	0.04	0.00	0.00

). The effect was slightly stronger in highly reflective wall environments (e.g., W4, p = 0.04), with amplitudes of 0.22 °C (W1), 0.41 °C (W2), 0.37 °C (W3), and 0.75 °C (W4).

Figure 6 illustrates UTCI variations with ground albedo across wall albedo scenarios, showing a minimal impact and distinct clusters: unshaded receptors (higher UTCI) and shaded receptors (lower UTCI, ~5 °C difference, ranging 4.2–5.7 °C). Regression analyses for shaded and unshaded areas revealed ground albedo’s greater relevance in shaded areas (p ≈ 0, slopes 0.64–1.08), where higher albedo slightly increased thermal stress, particularly with reflective walls (Figure 6d). This suggests that high ground albedo may reduce the cooling benefits of reflective walls, even in shaded areas.

3.2. UTCI Variation According to Wall Materials

Wall albedo (0.05–0.90) showed a non-linear effect on the UTCI across all ground albedo scenarios (0.2–0.8), with emissivity fixed at 0.9 (Figure 7). For instance, with ground albedo G1 = 0.2, the UTCI increased by 1.4 °C as wall albedo rose from 0.05 to 0.3 and then decreased at an average rate of −0.0475 °C per 0.01 albedo increase to 0.90. Similar patterns occurred for G2 (0.3, +1.48 °C, −0.0455 °C), G3 (0.5, +1.5 °C, −0.0445 °C), and G4 (0.8, +1.6 °C, −0.0413 °C), with amplitudes of −2.80 °C (G1), −2.69 °C (G2), −2.64 °C (G3), and −2.47 °C (G4). Figure 7 shows these trends, with the UTCI peaking at intermediate wall albedo (0.3–0.45) and declining at higher values.

To explore the non-linear relationship between wall albedo and the UTCI, a third-order polynomial regression model was applied to the data. With all ground scenarios considered, the polynomial regressions between wall albedo and UTCI showed a modest fit, with R² values around 0.14 and F(3, 464) between 24.57 and 25.97, with p values < 0.001. The regression equations are shown in Figure 7 where x represents wall albedo. The polynomial regression statistics are summarized in Appendix B,

Table A3. Polynomial regression statistics for X variable = Wi (wall albedo), Y variable = UTCI.

	G1 Wi			G2 Wi			G3 Wi			G4 Wi
	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded
Multiple R	0.38	0.94	0.81	0.38	0.93	0.82	0.37	0.92	0.82	0.37	0.90	0.82
R Square	0.14	0.88	0.66	0.14	0.86	0.67	0.14	0.85	0.67	0.14	0.81	0.67
Adjusted R Square	0.14	0.88	0.66	0.14	0.86	0.66	0.13	0.85	0.67	0.13	0.80	0.67
Standard Error	2.60	0.48	0.66	2.53	0.51	0.62	2.49	0.53	0.60	2.39	0.59	0.56
Observations	468	184	284	468	184	284	468	184	284	468	184	284

, where one can notice that despite the relatively low R² in the overall model, the relationship becomes much stronger when data are separated by shading conditions. For shaded areas, the polynomial regression yielded higher R² values between 0.80 and 0.88, indicating a very strong correlation between wall albedo and the UTCI. For unshaded areas, the R² was between 0.66 and 0.67, also reflecting a significant fit. Notably, this non-linear pattern challenges the assumption that higher albedo linearly reduces heat stress. Indeed, the parallel attempt of a linear regression showed relatively moderate correlations (R = 18–24% in Appendix B,

Table A4. Linear regression statistics for X variable = Wi (wall albedo), Y variable = UTCI.

	G1 Wi			G2 Wi			G3 Wi			G4 Wi
	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded	Overall	Shaded	Unshaded
Multiple R	0.24	0.58	0.51	0.22	0.53	0.49	0.21	0.51	0.47	0.18	0.43	0.42
R Square	0.06	0.33	0.26	0.05	0.29	0.24	0.04	0.26	0.22	0.03	0.19	0.18
Adjusted R Square	0.05	0.33	0.26	0.05	0.28	0.23	0.04	0.26	0.22	0.03	0.18	0.17
Standard Error	2.72	1.13	0.97	2.65	1.15	0.93	2.62	1.16	0.92	2.52	1.19	0.89
Observations	468	184	284	468	184	284	468	184	284	468	184	284
Slope	−2.13	−2.56	−1.85	−1.93	−2.33	−1.67	−1.84	−2.22	−1.59	−1.53	−1.85	−1.32
Coef. Standard Error	0.41	0.27	0.19	0.40	0.27	0.18	0.39	0.28	0.18	0.38	0.28	0.17
t Stat	−5.23	−9.52	−9.95	−4.87	−8.53	−9.31	−4.69	−8.03	−8.98	−4.06	−6.51	−7.79
p-value	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00

), compared to the higher correlation obtained based on the 3rd-order polynomial regression statistics. The authors would argue that material properties like transparency and density influenced the outcomes. For example, clear float glass (W1, albedo 0.05, 90% transmission) produced less reflective and longwave heat than opaque materials, reducing pedestrian-level heat stress. Specifically, W1 exhibited a mean radiant temperature (MRT) of 52.82 °C, which is approximately 11.3% lower than concrete (W2, albedo 0.3, MRT 59.52 °C) and 7.8% lower than moderately insulated walls (W3, albedo 0.45, MRT 57.31 °C). This reduction is driven by differences in radiant fluxes: W1’s shortwave (SW) flux was 151.33 W/m², a decrease of ~16.1% compared to W2 (180.39 W/m²) and ~14.3% compared to W3 (176.58 W/m²), while its longwave (LW) flux was 471.32 W/m², ~1.4% lower than W2 (477.75 W/m²) and ~0.3% lower than W3 (472.64 W/m²), as illustrated in Figure 8. These reductions in radiant fluxes directly contribute to a lower UTCI for W1 compared to W2 and W3, as high transmission minimizes the reflection and absorption of solar radiation, reducing both the reflected shortwave and emitted longwave radiation at the pedestrian level (~1.5 m). This will be discussed further in the next section (Section 4.2).

4. Discussion

4.1. Ground Versus Wall Albedo: Differences and Interdependencies

Multiple studies have analyzed the effects of urban canopy materials on the outdoor temperature, radiation, and thermal comfort, often concluding that high-albedo surfaces can produce an outdoor heat-mitigation effect [16,17,18]. This study, measuring pedestrian thermal comfort via the Universal Thermal Climate Index (UTCI) in tropical savanna climates, found that ground albedo (0.2–0.8) had a minimal impact (mean UTCI amplitude 0.44 °C), while wall albedo (0.05–0.90) significantly reduced the UTCI (up to −2.80 °C), with emissivity fixed at 0.9 for all materials. The common knowledge on the subject is that high-albedo materials reflect a larger portion of solar radiation, reducing heat absorption and longwave emission, thus improving outdoor thermal conditions [17,18]. Our results, however, align with Erell et al. [15], who clarified that while high-albedo grounds reduce surface temperatures, increased reflected radiation may elevate pedestrian heat stress, particularly in unshaded areas. Huang et al. [14] reported 2.6 °C air temperature reductions with retro-reflective pavements, contrasting our minimal UTCI impact, likely due to their focus on meteorological metrics rather than physiological comfort. Yang et al. [24] similarly cautioned that high-albedo pavements can increase radiant heat exposure, supporting our observed 5 °C UTCI difference between shaded and unshaded receptors.

The variation in wall albedo, in contrast, produced a notable influence on thermal comfort, with UTCI reductions ranging from −2.47 to −2.80 °C across scenarios. Drawing from Erell et al. [15], the difference in the surface orientation relative to pedestrians’ positions and the larger total area of wall surfaces compared to ground surfaces in deep urban canyons (aspect ratio 2.0) may account for these findings. Pedestrians are directly exposed to most solar radiation reflected from the ground, increasing heat stress with high ground albedo, whereas much of the reflected radiation from walls is emitted above pedestrian height, reducing exposure [12,13,17]. This is consistent with Tsekeri et al. [19], who found that reflective facades reduce cooling energy by 30%, and Lee and Mayer [20], who reported improved pedestrian comfort with high-albedo walls. Salvati et al. [21] demonstrated that reflective walls enhance urban canyon albedo, lowering street-level temperatures, aligning with our results. We hypothesize that the greater wall area in deep canyons amplifies these effects, as suggested by Ali-Toudert and Mayer (2006) [43], though further validation is needed.

A critical interdependency emerged: high ground albedo was moderately detrimental in highly reflective wall environments (e.g., W4, albedo 0.90, p = 0.04), particularly in shaded areas, offsetting wall albedo’s cooling potential. This effect was less pronounced in unshaded areas, suggesting that reflected radiation from high-albedo grounds interacts adversely with reflective walls. High-albedo grounds (e.g., light concrete, albedo 0.8) reflect more shortwave radiation directly to pedestrians, increasing the radiant load on vertical body surfaces and raising the UTCI by up to 0.75 °C compared to low-albedo grounds (e.g., dark asphalt, albedo 0.2) [15]. In deep canyons with reflective walls (e.g., aluminum, albedo 0.9), this ground-reflected radiation strikes the walls, which secondarily reflect a portion back to pedestrian height (1.5 m), amplifying thermal stress, particularly in shaded areas where reflected fluxes dominate [12,17]. This interaction reduces the cooling effectiveness of high-albedo walls, as the net UTCI reduction (e.g., 2.47 °C with high ground albedo) is slightly less than it would be with low ground albedo (closer to 2.80 °C). For example, combining high-albedo asphalt (albedo 0.8) with reflective aluminum walls (albedo 0.9) increased the UTCI by 0.75 °C compared to low-albedo asphalt, diminishing the wall cooling benefits. This finding, not widely discussed in prior studies [15], underscores a complex interplay between the ground and wall albedo. As far as urban design and surface material choice are concerned, the results presented in this paper imply the following: (1) the emphasis should be less on ground materials and more attention should be paid to wall material selection for effective heat mitigation during tropical savanna summers; (2) particularly in a highly reflective wall environment, it would be preferable to maintain low ground albedo to avoid offsetting the cooling impact of the highly reflective walls; for instance, urban designers could prioritize reflective wall coatings (e.g., white paints, albedo ~0.8) paired with low-albedo asphalt (albedo ~0.2) to maximize cooling in deep canyons, optimizing cost-effectiveness. These recommendations challenge the conventional focus on high-albedo pavements, advocating, instead, for a strategic approach to material selection in tropical savanna urban canyons.

4.2. Beyond Albedo: The Role of Material Properties

From previous studies on urban surface material selection as a heat-mitigation strategy, there have been suggestions that the albedo of surfaces determines their capacity to increase or reduce heat stress in urban areas [18,19]. Our results suggest that albedo alone may not be sufficient to predict the relative capacity of wall materials to improve outdoor thermal comfort in tropical savanna climates, particularly when transparency is involved. Indeed, a first attempt of linear fitting of the data (Figure 9a) showed a week linear correlation (R2 between 0.05 and 0.33) compared to a hypothetical case in Figure 9b where only opaque materials are considered (clear float glass data set removed), in which case the linear regression gave a higher correlation (R2 between 0.18 and 0.89). This demonstrates that assuming a linear correlation between wall albedo and thermal comfort cannot account for all materials, in this case, when transparent material is considered.

The clear float glass employed in the study (W1, albedo 0.05, 90% transmission) did not fit the expected pattern; despite having a low albedo, it did not worsen outdoor thermal comfort as one would predict based on albedo alone. Instead, the clear float glass produced less direct reflective heat and longwave radiation than opaque materials, such as concrete (W2, albedo 0.3), which absorbs and re-emits significant longwave radiation [44]. As specified earlier in the results section, the clear float glass resulted in an MRT ~11.3% lower than W2 and ~7.8% lower than W3. Indeed, the W1’s shortwave flux was ~16.1% lower than W2 and ~14.3% lower than W3, while its longwave flux was ~1.4% lower than W2 and ~0.3% lower than W3. This occurs because W1’s high transmission allows most incoming solar radiation to pass through rather than being reflected or absorbed, reducing both the reflected shortwave radiation and the heat available for longwave emission. Consequently, the lower radiant load at the pedestrian level decreases the MRT and, in turn, the UTCI, highlighting the role of material transmittance in mitigating heat stress beyond albedo effects alone. This observation would confirm the study of Cuce et al. [34], who found that transparent facades increase indoor heat gain in tropical climates, while the high transmission may reduce pedestrian-level heat stress compared to opaque materials [20].

The non-linear UTCI response to wall albedo (peaking at 0.3–0.45, Figure 7) reflects complex radiative and material interactions. The polynomial regressions (with R² reaching up to 0.88 in shaded areas, Appendix B) capture this pattern, driven by material-specific radiative behaviors. High-albedo aluminum (W4, 0.9) minimizes absorption, reflecting shortwave radiation above the pedestrian level, thus reducing the UTCI [19]. In contrast, intermediate-albedo concrete (W2, 0.3) absorbs more radiation, increasing longwave emission and the UTCI. The thermal mass also plays a role: aluminum (2700 kg/m³) has lower heat storage than concrete (2200 kg/m³), reducing diurnal heat release and enhancing cooling [45,46]. These findings highlight the interplay of shortwave reflection, longwave emission, transparency, and thermal mass, challenging linear albedo assumptions. Obviously, some wall materials perform better than others, but the absence of a single-property pattern upon which a steady reduction in the UTCI hypothetically depends suggests that the heat mitigation capacity of wall materials is determined by a combination of properties. Bulbaai and Halman [16] emphasized combining reflective coatings with an optimal building orientation in tropical climates, and our results extend this by advocating for multi-indicator evaluations, including albedo, emissivity, transmission, density, thermal mass, etc. For urban designers, numerical simulations can optimize wall material choices by accounting for these properties. For example, semi-reflective coatings or textured aluminum panels could balance thermal and aesthetic considerations. The exploratory glass analysis reinforces this complexity, as transparency altered the expected albedo–UTCI relationships, underscoring the need for case-specific numerical simulations. While radiant flux data was not directly measured, the UTCI reductions with high-albedo walls align with reduced longwave emission, as reported by Lee and Mayer [20]. For urban designers, instead of directly assuming that higher wall albedo would have better heat mitigation, numerical simulations, as employed in this study, are critical for evaluating the effect of wall material choices on outdoor thermal comfort, ensuring optimal outcomes in tropical savanna climates.

4.3. Limitations and Future Research

This study’s findings are constrained by the hypothetical urban geometry employed, which may not generalize to shallower canyons, open plazas, or other urban forms in tropical savanna climates. The fixed emissivity assumption (0.9), used to isolate albedo effects, simplifies radiative transfer but overlooks material-specific variations, potentially affecting longwave radiation and UTCI in a real-world context [42]. The study’s hypothetical parametric design relies on ENVI-met validated performance in similar climates (e.g., Akure, Nigeria, R² > 0.96 [39]). While suitable for theoretically isolating albedo effects, this parametric approach may introduce uncertainty in site-specific applicability. The hourly UTCI resolution, while standard for ENVI-met simulations [36], may miss transient solar radiation peaks around noon, though low Standard Errors (0.07–0.2, Appendix B) indicate reliable steady-state results.

The hypothetical simulation of a relatively extreme albedo like aluminum (albedo 0.9) to observe the behavior trend of the UTCI showed significant thermal comfort improvements, but the practical implementation of such high wall albedo requires considerations of durability, maintenance, and visual comfort. For instance, aluminum’s high reflectivity may cause glare, potentially affecting pedestrian comfort, and soiling can reduce albedo over time [16,45]. In tropical savanna climates, frequent rainfall may mitigate dust accumulation, and maintenance costs also remain a factor. For practical purposes, urban designers should balance thermal benefits with practical constraints, potentially using semi-reflective coatings (e.g., white paints, albedo ~0.8) or textured surfaces to minimize glare while retaining cooling potential [15].

Future research should explore diverse urban geometries, such as shallow canyons or plazas, to enhance the generalizability of albedo effects for sustainable urban design. A site-specific field study in tropical savanna regions could complement this study and improve real-world applicability, while sensitivity analyses of emissivity variations could refine radiative transfer models, optimizing material selection for heat mitigation. These efforts will build on our findings, providing urban designers with robust tools to create resilient, livable urban environments in rapidly urbanizing tropical savanna climates [16,17,37,46].

5. Conclusions

This study investigated the impact of ground and wall albedo on pedestrian thermal comfort in tropical savanna climates using parametric simulations in ENVI-met [33]. The findings guide sustainable urban design by identifying optimal material strategies to mitigate heat stress and enhance livability in rapidly urbanizing tropical regions. The analysis combined four ground materials (albedo 0.2–0.8) and four wall materials (albedo 0.05–0.90), with emissivity fixed at 0.9. The UTCI, a physiological metric integrating air temperature, humidity, wind, and radiation, was recorded at pedestrian height (1.5 m) in streets and courtyards. We observed that ground albedo had a minimal impact on the UTCI (mean amplitude 0.44 °C, R = 3–9%), consistent with Erell et al. [15], while wall albedo significantly reduced the UTCI (up to −2.80 °C), with a non-linear effect influenced by transparency [19,20].

Transparent walls (e.g., clear glass) and opaque walls (e.g., concrete, aluminum) exhibited distinct thermal behaviors, indicating that albedo alone is insufficient to predict wall material performance, as supported by Cuce et al. [34]. High ground albedo offsets the cooling potential of reflective walls, particularly in shaded areas, highlighting critical interdependencies. For sustainable urban design, these results advocate for prioritizing reflective wall materials (e.g., white coatings, albedo ~0.8) paired with low-albedo grounds (e.g., asphalt, albedo ~0.2) to maximize cooling, reduce energy demands, and improve urban livability, challenging assumptions about universal high-albedo benefits [14,24]. Urban designers should employ case-specific simulations to optimize material choices. Limitations include the single geometry (aspect ratio 2.0) and exclusion of factors like vegetation or wind, suggesting future research on diverse configurations and material properties (e.g., texture, thermal mass) to enhance thermal comfort models [43]. This study underscores wall albedo’s primacy for sustainable heat mitigation in tropical savanna climates, offering evidence-based strategies for creating resilient, livable urban environments.