Evaluating Urban Heat Island Mitigation Policies in Heritage Settings: An Integrated Analysis of Matera

Abstract

This study investigates the environmental parameters that contribute to the Urban Heat Island (UHI) effect in historic environments, with a particular focus on the UNESCO World Heritage City of Matera. The complex urban morphology of Matera, with its narrow streets and underground buildings, generates distinctive microclimates that intensify the UHI phenomenon, posing challenges for urban planning and heritage conservation. The main objective of the research is to identify which environmental parameters interact with Matera’s architectural and urban characteristics to intensify the UHI, and to propose mitigation strategies that balance heritage conservation with environmental sustainability. The research follows a mixed methodological approach in two phases. The first phase consisted of a comprehensive literature review, identifying gaps in previous studies and developing a methodological framework combining quantitative and qualitative techniques. The second phase involved empirical analysis using advanced techniques such as 3D laser scanning to model urban morphology, satellite image analysis to map the spatial distribution of the UHI, and the integration of historical and real-time meteorological data. The results show significant correlations between urban morphology and UHI intensity, suggesting strategic interventions such as green roofs and reflective materials to mitigate the effects. These findings provide valuable information for urban planners and policy makers, and highlight the importance of integrating sustainable approaches into heritage conservation.

1. Introduction

Urban heat islands (UHI) are a major environmental problem characterized by higher temperatures in urban areas compared to rural areas. This phenomenon results from several factors, including the extensive use of impervious surfaces, high building densities, and anthropogenic heat emissions. IHUs exacerbate energy consumption, increase levels of air pollutants and greenhouse gases, and negatively affect human health and comfort. In historical contexts such as Matera, a UNESCO World Heritage Site, the intensity of the UHI is related to the dense structure and lack of vegetation [1].

Matera has a unique urban morphology with narrow streets, densely built structures of calcarenite, and hypogean architecture with underground dwellings and water cisterns [2]. These characteristics generate specific microclimatic conditions which, in a scenario of urban overheating, pose particular challenges for urban planning and heritage conservation [3]. The impact of the UHI phenomenon in Matera has implications not only for environmental sustainability but also for the future development of the city’s cultural and architectural heritage. High urban temperatures can accelerate the degradation of historical materials, thus endangering the cultural heritage [4].

The main objective of this study is to identify the environmental parameters that contribute to the UHI effect in historic environments, with a particular focus on the street canyon of ‘Via delle Beccherie’ in Matera, Italy. The study aims to understand how environmental phenomena interact with the architectural and urban characteristics of the historic city, and how these interactions can intensify the UHI. Based on these findings, mitigation strategies will be developed that take into account both environmental sustainability and heritage conservation, without compromising the integrity of the historic buildings. The research recognizes the need to balance the preservation of Matera’s urban heritage with the demands of modern sustainability. The concept of sustainability is defined here in terms of the ability to integrate urban solutions that mitigate the effects of the UHI, such as the use of green roofs and reflective building materials, while preserving the historical and cultural integrity of the city. This responds to the need to adopt approaches that protect both the natural environment and the built heritage, ensuring urban development that is resilient to climate change while respecting Matera’s cultural identity.

1.1. Historical Background of Matera

Matera, located in southern Italy, is known for its complex historical architecture dating back to the Middle Ages. The ‘Sassi’ of Matera area, consisting of ancient neighborhoods built of calcarenite stone, is particularly notable for its rock-hewn dwellings, uniquely adapted to the local climatic conditions. The city has undergone a significant urban transformation since its foundation, from prehistoric settlements to an outstanding example of architectural and urban heritage adapted over the centuries. The interaction between these historic elements and contemporary sustainability requirements presents a unique conservation challenge. Urban interventions have been carried out to reconcile modern needs with the preservation of historic structures, including managing the urban microclimates created by the narrow streets and underground spaces [5].

Several historic cities face UHI challenges due to urban density and heritage preservation. Beijing has adopted green roofs and reflective materials to counter the rise in temperature [1]. Rome has expanded green spaces and used lighter materials to reduce UHI [6]. Cairo, proposed strategies include green roofs and improved urban ventilation [7]. Istanbul has restored historic buildings with thermal materials and introduced parks [6]. These strategies can inform UHI mitigation in Matera, where narrow streets and hypogean architecture pose similar challenges.

This map (see Figure 1) illustrates the delimitation of the historic city centre, highlighting two key zones. The red perimeter delimits an area of approximately 320,458.19 m², which corresponds to the total extension of the historic centre. On the other hand, the area in yellow, with a surface area of 20,060.15 m², marks the heritage core of the study.

The image (see Figure 2) shows some of the main attractions of Matera, Italy, a UNESCO World Heritage Site. Highlights include Civita, the historic centre of the city, with its particular urban landscape; the districts of Sasso Barisano and Sasso Caveoso, both characterized by troglodyte constructions dug into the limestone; and finally, the Murgia, a geological formation that contrasts with the urban environment, with a natural landscape of ravines and caves that underline the historical interaction between nature and human settlements in the region,

1.2. Objectives

The objectives of this study focus on identifying and analyzing the environmental parameters that contribute to the phenomenon of UHI in historic built environments, with a particular focus on the city of Matera. The aim is not only to understand these parameters but also to develop mitigation strategies that can be implemented without compromising the heritage and architectural value of the city [8]. To address this challenge, the following specific objectives have been defined:

• Identification of temporal patterns: a temporal clustering analysis is used to identify periods with similar temperature and humidity characteristics. This objective seeks to understand the seasonal and diurnal variations of the UHI in Matera and their impact on both architectural heritage and habitability [9].
• Scenario simulation: use the random forest regression model to simulate different scenarios, such as changes in vegetation NDVI (Normalized Difference Vegetation Index) or building materials (solar absorption coefficients), and predict their impact on the UHI. This simulation will assess how different interventions, such as green roofs or reflective materials, could mitigate the effects of the UHI in a sensitive urban and heritage context [8,10].
• Regulatory impact analysis: assess how regulatory changes, such as building codes and green roof mandates, could affect environmental indices and urban temperatures. This analysis will make it possible to anticipate the impact of these policies on the conservation of Matera’s historical heritage [11].
• Identification of dominant factors: spectral indices derived from remote sensing data are essential for analyzing and predicting UHI intensity, as they provide critical insights into land surface characteristics that influence temperature variations. The Normalized Difference Built-up Index (NDBI) identifies impervious surfaces, which contribute to heat accumulation due to their low albedo and high thermal storage capacity. Similarly, NDVI identifies green cover, which counteracts UHI through evapotranspiration and shading. Furthermore, the Normalized Difference Soil Index (NDSI) quantifies soil exposure and moisture content, both of which significantly affect surface temperature dynamics [12,13].

These objectives are designed to comprehensively understand the mechanisms contributing to UHI in historical contexts and develop practical and effective mitigation strategies consistent with preserving Matera’s cultural and architectural heritage.

2. Materials and Methods

The research adopts a mixed methodological approach. Initially, a review of 61 papers led to the selection of 20 studies focusing on the UHI phenomenon in street canyons. [13]. Secondly, the collection and analysis of empirical data, integrated with Machine Learning techniques using the Random Forest Model.

2.1. Literature Review

The main objective was to establish a solid theoretical basis and identify gaps in research related to UHI in historical contexts by reviewing the existing literature.

Matera’s hypogean structures and urban morphology influence internal and external temperatures [14]. In addition, to investigate how the thermal properties of the predominant building material, calcarenite, affect the urban climate and the internal comfort of dwellings [15]. As already discussed in the literature, data acquisition methods such as satellite data analysis and 3D scanning were reviewed to inform the design of the methodological framework [9]. The scatter diagram (see Figure 3) maps IHU research, highlighting urban settings, methodologies and parameters. Street canyons were most studied, with RANS/STKE (Reynolds-Averaged Navier-Stokes/Standard k-epsilon) dominant. LES/SLGSG (Large Eddy Simulation/Scale-Localized Generalized Subgrid-Scale) and RANS/ED (Reynolds-Averaged Navier-Stokes/Eddy Dissipation) were less frequent, showcasing methodological diversity.

2.2. Empirical Data Gathering and Analysis

The second phase of the study focused on data collection and analysis, using advanced techniques in line with the research objectives and designed to provide a comprehensive analysis.

• Data were obtained from four sources (see Figure 4):
• Topographic surveys: detailed 3D laser scanning models of the urban canyon were obtained [9].
• Satellite Image Analysis: the spatial distribution of the UHI effect in Matera was analyzed using satellite imagery, focusing on Sentinel-2 data for its high spatial and spectral resolution [9]. The spectral analysis examined key environmental factors, including vegetation (NDVI), built-up areas (NDBI), and soil exposure (NDSI), highlighting their roles in heat retention and urban microclimate regulation [16].
• Meteorological data integration: real-time indoor and outdoor temperature and relative humidity data were collected to understand local climatic conditions [17].
• Analysis of the thermal properties of calcarenite: this study assesses the solar absorption rate of calcarenite and its impact on UHI dynamics in Matera [18]. Experimental analyses at ESPOL measured its thermal behavior, integrating data into a random forest regression model to identify key UHI factors and explore mitigation strategies that preserve architectural heritage.

Finally, the collected data were integrated into a random forest regression model to identify the dominant factors affecting UHI intensity and to predict outcomes under different scenarios.

2.3. Conceptual Breakdown of Random Forest Regression

The Random Forest Regression model has proven to be an effective tool in environmental and urban studies due to its ability to capture nonlinear relationships and process multiple heterogeneous variables. Recent research has applied this approach to evaluate urban heat island (UHI) mitigation strategies. For instance, Ref. [19] used Random Forest to predict the impact of green roofs on urban temperature in Belgian cities, combining satellite data with three-dimensional parameters. Following this approach, the present study implemented a Random Forest Regression model to analyze UHI in Matera, integrating environmental and urban factors. This methodology enabled the identification of key determinants influencing outdoor temperature and the proposal of mitigation strategies compatible with heritage conservation.

Multiple Decision Trees: The model consists of T decision trees, each making an independent prediction ŷ for a given input x. Averaging Predictions: The final prediction ŷ is obtained by averaging the predictions from all TTT trees, ensuring robustness and reducing variance. Mathematical Representation: The prediction for an input xxx in a Random Forest model with T trees is given by:

$$\mathrm{ŷ}={\displaystyle \frac{1}{T}}{\sum }_{t=1}^T\mathrm{ŷ}\mathrm{t}\left(\mathrm{x}\right)$$

where:

• ŷ is the final predicted value.
• T is the total number of trees in the Random Forest.
• ŷt(x) is the prediction of the t-th tree for the input x.

For data integration, a Random Forest Regression model is proposed that integrates heterogeneous data: dynamic heterogeneous (external meteorology; internal temperature/humidity/pressure) and static heterogeneous (calcarenite building material characteristics). The proposed analysis seeks to understand whether the solar absorption coefficient of calcarenite and the spectral indices (NDBI/NDSI/NDVI) extracted by photogrammetric analysis condition the outdoor/indoor temperature in the city of Matera.

The Random Forest Regression model has been trained, and we have extracted the feature importances and visualized the partial dependence plots (see Table S1). The following perspectives emerge from the analysis:

a. 

Indoor Temperature Analysis: Feature Importances:

• External Temperature: 63.8%; External Humidity: 35.9%; External Pressure: 0.24%

The environmental indices (NDBI, NDSI, NDVI) and solar absorption coefficients (Sample699, Sample700) had negligible importance in predicting indoor temperature (Figure 5).

b. 

External Temperature Analysis: Feature Importance:

• NDSI: 38.9%; NDBI: 32.7%; NDVI: 25.4%; Sample699: 1.5%; Sample700: 1.4%.

c. 

Partial Dependence Plots:

The partial dependence plots for the most important features in the external temperature analysis confirm that NDSI, NDBI, and NDVI significantly impact external temperature.

These findings underscore the importance of considering external environmental conditions such as temperature and humidity in urban planning. While the solar absorption coefficients influence external temperature, their impact is relatively minor compared to the environmental indices (see Figure 5 and Figure 6). This insight can guide policymakers and urban planners in effectively formulating strategies to manage urban temperatures [12].

3. Results

The results of this research provide detailed insight into how UHI affects Matera’s architectural and urban heritage and how urban planning strategies can help mitigate these effects without compromising the city’s historic value.

3.1. Review of Methodologies Used in Previous UHI Studies to Inform the Design of Our Research Framework

Previous studies have analyzed the climatic and architectural characteristics of Matera’s hypogean structures, emphasizing their role in mitigating indoor temperature variations and reducing artificial cooling needs. The high thermal mass of calcarenite stabilizes indoor microclimates, promoting passive cooling and limiting temperature fluctuations [20,21]. However, while beneficial indoors, calcarenite’s thermal inertia can exacerbate UHI in dense urban areas by trapping heat, particularly in confined spaces with limited ventilation [22,23]. Matera’s intricate road network and compact built environment intensify this effect [4,5]. A review of methodologies highlights remote sensing as a key tool in UHI research (see Table S2), often combined with GIS analysis to map temperature distribution and urban characteristics [24]. Ground-based measurements validate these data, offering insights into local microclimates [25,26]. Our study integrates dynamic and static data into a random forest regression model, incorporating environmental indices (NDBI, NDSI, NDVI), solar absorption coefficients, and meteorological variables [27].

3.2. Analysis of Data Collected at Matera

3.2.1. Temporal Clustering

The Temporal Clustering in the Random Forest Regression (RFR) model was applied to identify seasonal patterns in temperature, humidity, and pressure in Matera. By grouping data into specific periods, significant fluctuations in relative humidity and atmospheric pressure were observed, along with trends in temperature variation over the years. This approach allowed the detection of critical periods where Urban Heat Island (UHI) intensity peaks, particularly during summer [28].

Additionally, dimensionality reduction using PCA (Principal Component Analysis) highlighted the relationship between these parameters, facilitating the identification of recurring trends and improving the model’s accuracy in predicting climatic variations in the urban environment (see

Table 1. Meteorological Data. Source: Weather in 07-04-2023 in Matera, Italy (TIMEANDDATE.COM), (see Figures S1–S3).

External_Temperature	External_Humidity	Cluster
20	0.7	0
23.5	0.575	1
22	0.68	2
26	0.42	3

).

In the scatter plot (see Figure 7), the blue dots represent the actual observed temperature and humidity data, while the red x’s indicate model predicted values. When a blue dot coincides with a red X, it means that the observed value and the predicted value are very similar or equal (see Table S3). The monitored data from the home on Via Castelvecchio No. 6 reveal microclimatic conditions influenced by layout and ventilation (see Figure 8). Higher humidity is observed in the bathroom and entrance hall, while the bedroom benefits from thermal contributions from the rear wall [13,29].

3.2.2. Scenario Simulation

Scenarios were simulated in the random forest regression (RFR) model to predict the impact of changes in vegetation cover (NDVI) and building materials (solar absorption coefficients). This analysis integrates satellite images of four points in the analysis area(see Figure 9); thermal properties of materials and urban geometry to assess their role in local temperature variations [30,31].

• Satellite Imagery Analysis.

The trends reveal that NDBI has slightly declined, suggesting a reduction in construction activity or a shift toward sustainable practices. In contrast, NDSI has increased, indicating decreasing soil cover, which may exacerbate UHI effects. Meanwhile, NDVI shows an upward trend, highlighting an increase in vegetation that could mitigate urban overheating. Seasonal variations indicate that NDBI peaks in warmer months, reflecting construction activity, while NDSI fluctuates seasonally, peaking mid-year and decreasing in winter due to variations in solar exposure and surface conditions (see Table S4).

The time series analysis (2016–2023) examines annual and monthly variations of NDBI, NDSI, and NDVI using Sentinel-2 imagery (10 m, 5-day revisit) and Google Earth Engine (GEE) (see Figure 10 and Figure 11). Illustrates automated land cover trend extraction, enabling a detailed assessment of urban environmental dynamics.

• Thermal properties of calcarenite.

The high thermal mass of calcarenite, analyzed for its solar absorption properties, significantly contributes to heat retention, particularly in narrow urban canyons, where heat dissipation is limited. This material stores solar radiation during the day and releases it gradually at night, intensifying UHI effects [18,32]. Laboratory analyses confirmed that calcarenite exhibits high solar absorption coefficients, influencing local microclimates and shaping temperature variations within the built environment (see

Table 2. Solar Absorption Coefficient of Calcarenite. Source: Laboratory Results, ESPOL Ecuador.

Sample699	Sample700
ASTM E 891: 67.70	ASTM E 891: 67.89
ASTM E 892: 65.35	ASTM E 892: 65.41
ASTM G (173)	ASTM g 173 (Direct Circumsolar): 66.22
ASTM G (173) (Hemispherical Tilt @ 37 degrees): 64.90	ASTM g 173 Hemispherical Tilt@ 37 degrees): 64.99

).

The above images (Figure 12 and Figure 13) explain the climatic characteristics that can be affected by the morphology of the urban canyon and their consequences on thermal comfort. Morphological aspects generate changes in environmental parameters impacting the energy heat balance of urban areas.

• Topographic Survey Analysis.

High-precision 3D models of Via delle Beccherie were generated using advanced laser scanning technology (see Figure 12), enabling a detailed examination of the relationship between urban geometry and UHI intensity. The findings indicate a strong correlation between the aspect ratio of street canyons (height-to-width ratio) and localized temperature variations [33,34]; (see Figure 13). Narrower streets with higher aspect ratios experience intensified UHI effects, primarily due to restricted airflow and increased heat retention by surrounding surfaces with high thermal mass.

The scenario analysis evaluated the effects of increased NDVI (through green roofs and urban parks) and modified solar absorption coefficients (reflective materials) on external temperature (see Table S5). The predicted temperature variations under different conditions revealed nuanced impacts on UHI intensity. In the NDVI scenarios, an increase in vegetation resulted in a predicted external temperature of 23.51 °C, while reducing vegetation led to a slight temperature decrease to 23.30 °C. This suggests that complex interactions with urban materials and microclimate conditions influence the cooling effect of vegetation.

The concentration of minimum intensity values in the LiDAR point cloud indicates areas with lower reflectivity, which could be associated with an increase in relative humidity, especially in areas influenced by the urban canyon morphology (see Figure 14). For solar absorption scenarios, results varied based on the sample and reflectivity properties. Increasing solar absorption for Sample699 led to a predicted temperature of 23.12 °C, while decreasing it resulted in 23.13 °C. Similarly, for Sample700, an increase in solar absorption led to a temperature of 23.11 °C, while a decrease maintained the temperature at 23.13 °C. Further material property analysis revealed variations in solar absorption coefficients based on standard testing protocols. The ASTM E891 test yielded values of 67.70 for Sample 699 and 67.89 for Sample 700, while ASTM E892 produced 65.35 and 65.41, respectively. Under direct circumsolar conditions (ASTM G173), Sample 699 exhibited a coefficient of 66.07, while Sample 700 measured 66.22. The hemispherical tilt test at 37 degrees revealed absorption values of 64.90 for Sample699 and 64.99 for Sample700. These findings indicate that material reflectivity plays a role in mitigating urban heat retention, although its effects on temperature variations remain marginal.

The scenario analysis underscores the multifaceted interactions between vegetation, urban materials, and thermal properties. While increasing NDVI contributes to urban greenery, its influence on cooling is not linear and depends on additional environmental variables. Meanwhile, reflective building materials show a consistent, though minor, reduction in external temperature, reinforcing their potential as an effective UHI mitigation strategy when combined with green infrastructure.

3.2.3. Regulatory Impact Analysis

The Regulatory Impact Analysis in the Random Forest Regression (RFR) model was applied to evaluate how regulatory changes, such as building codes and green roof mandates, influence environmental indices and Urban Heat Island (UHI) intensity. The model integrated data adjustments to simulate the impact of regulatory measures, including increased NDVI values for green roofs and modified solar absorption coefficients for new building materials. Additionally, 3D survey data was utilized to visualize the potential impact of these changes on the urban landscape while preserving Matera’s historical integrity. The analysis examined various regulatory scenarios, predicting their effects on external temperature. Simulating the implementation of green roofs by increasing NDVI by 0.2 from its mean value resulted in a predicted temperature of 23.51 °C (see Table S6). In contrast, using reflective building materials (Sample699 and Sample700) lowered the predicted temperature to 23.13 °C, indicating their potential for mitigating UHI effects.

The findings suggest that green roofs, despite adding vegetation, have a more complex impact on temperature regulation, influenced by factors such as humidity and building density. Meanwhile, reflective building materials, which reduce solar absorption, effectively lower external temperatures, highlighting their potential as a UHI mitigation strategy. These results underscore the importance of regulatory measures in urban planning, demonstrating that implementing green infrastructure and advanced building materials can significantly enhance environmental sustainability and urban resilience [35,36].

3.2.4. Dominant Factors Analysis

The Dominant Factors Analysis in the Random Forest Regression (RFR) model was applied to integrate dynamic and static data, including NDVI, NDBI, NDSI, and meteorological variables, to determine the key predictors influencing Urban Heat Island (UHI) intensity [37,38]. The feature importance analysis identified NDSI (38.97%) and NDBI (32.74%) as the most significant predictors of UHI intensity, followed by NDVI (25.43%).

The Partial Dependence Plots (PDPs) further illustrate the influence of these variables, highlighting that higher NDSI values correlate with increased temperatures, indicating that greater soil exposure and lower moisture retention contribute to surface heat accumulation (Figure 15). Similarly, NDBI strongly impacts temperature variations, with higher values associated with elevated temperatures due to the heat absorption of impervious surfaces. In contrast, NDVI plays a lesser role, yet its higher values contribute to lower temperatures, reinforcing the cooling effect of vegetation through shading and evapotranspiration (see Table S7).

The interaction analysis between NDVI and NDBI suggests that the built environment moderates the impact of vegetation on temperature, while NDVI and NDSI interactions show complex patterns, indicating that soil cover levels influence NDVI’s cooling effect. These findings confirm that NDSI and NDBI are the dominant factors driving UHI intensity, and their interactions with NDVI reveal that vegetation alone does not consistently result in temperature reductions. This underscores the need for a holistic approach to urban planning, where increasing vegetation must be complemented by managing impervious surfaces (NDBI) and soil cover (NDSI) to optimize cooling effects and enhance environmental sustainability.

4. Discussion

4.1. Temporal Clustering and Seasonal Patterns

The temporal clustering analysis identified distinct periods with similar temperature and humidity characteristics, revealing significant seasonal trends. These findings are consistent with previous studies that highlight the seasonal variability of UHI effects, particularly the intensification during summer months due to higher temperatures and reduced humidity. For instance, research by [16,39] demonstrated that UHI effects are more pronounced during warmer seasons, corroborating our findings in Matera. This seasonal insight is critical for developing targeted mitigation strategies that address the periods of highest UHI intensity.

4.2. Impact of Vegetation and Building Materials

The impact of vegetation and building materials on UHI was analyzed using NDVI, NDBI, and NDSI. Surprisingly, increasing NDVI slightly raised outdoor temperatures; while decreasing it led to cooling. Partial dependence plots (PDP) (Figure 16) revealed that NDVI’s effect depends on NDBI and NDSI levels. A correlation matrix (Figure 17) confirmed a positive correlation between NDVI and external temperature, suggesting hidden interactions. The model captured complex local effects, explaining these counterintuitive findings. Research indicates that tree cooling efficiency is overestimated by 60% during extreme heat waves [40].

The economic feasibility of Urban Heat Island (UHI) mitigation strategies could be assessed through a cost-benefit analysis, using cost estimation methods found in existing literature, as it was not possible to derive case-specific costs for Matera. This analysis would evaluate the financial implications of implementing green roofs and reflective materials, estimating potential benefits such as energy savings (20% for green roofs, 15% for reflective materials), improved thermal comfort ($100 per person per year), and health benefits ($50 per person per year) [41]. A net present value (NPV) analysis over a 20-year period with a 5% discount rate could yield estimated costs of $15,512,210.34 and total benefits of $81,004,367.23, resulting in a net benefit of $65,492,156.88. These projections suggest that UHI mitigation measures may be economically viable, offering potential long-term financial and environmental benefits [42]. Scenario simulations align with previous studies [43,44], demonstrating that increased vegetation and reflective materials significantly reduce UHI. The upward NDVI trend in Matera suggests a shift towards more greenery, reinforcing its role in urban cooling [9]. Policymakers should consider these strategies for sustainable urban planning [45,46]. These results allow us to move on to a more detailed discussion of the effectiveness of the proposed strategies, which will be analyzed in the next section.

These findings align with previous research indicating that the effectiveness of vegetation in UHI mitigation depends on urban morphology and material properties. [19] found that, in Belgian cities, green roofs reduced urban temperatures, but their impact varied significantly with built density and surface materials. Similarly, our results show that while NDVI has a moderating effect, NDBI and NDSI remain the dominant drivers of temperature variations in Matera. This reinforces the need for integrated strategies combining green infrastructure with material-based solutions, especially in heritage cities with architectural constraints.

4.3. Regulatory Impact and Urban Planning

The regulatory impact analysis highlights the influence of building codes and green roof mandates on improving environmental indices and mitigating urban temperatures. Sensitivity analysis revealed that within the tested range, increasing NDVI (+0.1, +0.2, +0.3, and +0.4) did not significantly alter external temperature, which remained constant at 23.51 °C. This outcome suggests either limitations in the predictive model or the dominance of other environmental variables, such as NDBI and NDSI, in shaping urban microclimate dynamics. These findings align with previous studies emphasizing the importance of regulatory interventions in UHI mitigation [47].

Further analysis integrating 3D survey data underscores the role of urban morphology in shaping regulatory frameworks. The topographic survey identified a strong correlation between high aspect ratios in urban canyons and increased UHI intensity, reinforcing the necessity of improved ventilation and the use of reflective materials in urban planning [48,49]. These results further demonstrate that while vegetation plays a role in moderating surface temperatures, its effectiveness may be context-dependent, influenced by factors such as surface material properties, urban geometry, and meteorological conditions. The lack of a significant cooling effect within the tested NDVI range suggests a potential threshold effect, where the benefits of increased vegetation become apparent only beyond a certain level of coverage. Additionally, the model may not fully capture key spectral and physical parameters—such as albedo, soil moisture, and evapotranspiration rates—which are crucial for a comprehensive assessment of vegetation’s cooling potential.

These findings highlight the complexity of urban heat mitigation strategies and the need for a multifaceted approach that integrates regulatory measures, urban morphology considerations, and vegetation-based interventions. Future research should explore higher NDVI increments beyond +0.4 and incorporate additional urban climate variables to refine the understanding of vegetation’s contribution to reducing UHI effects. Urban and architectural solutions focus on high-reflective materials, phase change materials (PCM), and green roofs. Simulations show green roofs reduce cooling demand by 30% in hot climates [50,51]. Combined with passive ventilation, these strategies optimize airflow, crucial for cities like Rome and Cairo.

4.4. Dominant Factors Influencing UHI

Analysis 4 identifies NDSI and NDBI as the dominant factors influencing outdoor temperature predictions. Feature importance analysis shows that NDSI (38.97%) and NDBI (32.74%) significantly impact external temperature, while NDVI (25.43%) has a lesser effect. Partial dependence plots indicate that higher NDSI and NDBI values increase temperatures, with NDVI’s effect moderated by these indices. Combined scenario analysis shows:

• Increasing NDVI and decreasing NDBI: Temperature = 23.55 °C
• Increasing both NDVI and NDBI: Temperature = 23.65 °C
• Decreasing both NDVI and NDBI: Temperature = 23.42 °C
• Decreasing NDVI and increasing NDBI: Temperature = 23.26 °C

These results highlight that NDBI changes impact temperature more than NDVI changes. The interactions suggest that merely increasing vegetation is insufficient; effective UHI mitigation requires reducing built-up areas while increasing vegetation to achieve cooling effects. Temporal clustering highlights the need to address seasonal variations, especially during peak summer. The analysis of vegetation and building materials underscores that increasing green areas alone is insufficient for Urban Heat Island (UHI) mitigation without simultaneously addressing built-up surfaces and the solar absorption properties of materials [52]. In Matera, the widespread use of calcarenite, a material with a high solar absorption coefficient, further complicates efforts to reduce urban temperatures

Regulatory impact analysis confirms that UHI mitigation policies are economically viable, with long-term benefits outweighing implementation costs. The dominant factors analysis identifies NDSI and NDBI as the primary drivers of UHI, emphasizing the need for a balanced approach that not only enhances NDVI but also reduces solar absorption in built environments [53,54]. However, historical satellite data reveals that a decrease in NDSI can disrupt environmental equilibrium, potentially exacerbating UHI effects rather than mitigating them. These findings highlight the necessity of an integrated strategy that harmonizes built-up area management, vegetation expansion, and solar absorption reduction. A comprehensive approach that considers urban morphology, material properties, and vegetation cover is essential to effectively regulate urban temperatures while maintaining environmental stability.

This study contributes empirical data on Matera, demonstrating green infrastructure and sustainable materials’ viability in heritage settings. The random forest regression model effectively identifies key UHI factors, informing future research and urban policies. Economic analysis supports green roofs and reflective materials, highlighting long-term savings and health benefits. Findings emphasize a holistic UHI mitigation approach, balancing historic preservation with sustainability, strengthening urban resilience, and guiding policymakers [55].

5. Conclusions and Recommendations

Advanced models such as random forest regression provide a detailed understanding of the factors contributing to UHI, facilitating the development of effective mitigation strategies. The study revealed several key findings on the effects of UHI on Matera. The temporal clustering analysis highlighted the importance of addressing seasonal variations, especially during the summer when the UHI effect intensifies. The results showed that increasing green areas alone cannot mitigate UHI in Matera due to the complex interaction between NDVI, NDBI, and NDSI and calcarenite’s high solar absorption coefficient. NDSI and NDBI were identified as the most influential factors in predicting external temperature, suggesting that management of built-up areas and reduction of solar absorption are essential. In addition, the economic analysis showed that UHI mitigation policies, such as the implementation of green roofs and reflective materials, are economically viable in the long term, with benefits significantly outweighing the costs. The decrease in NDSI, according to historical satellite imagery, contributes to the imbalance of other environmental parameters, exacerbating the effects of UHI.

Based on these findings, it is recommended that integrated strategies be implemented, which include reducing solar absorption of building materials, managing built-up areas, and increasing vegetation. Developing urban planning policies that promote adequate ventilation and reflective materials is crucial, balancing historic preservation with modern sustainability needs. In addition, it is essential to continue research through integrated data analysis to deepen the relationships and correlations of variables contributing to the UHI phenomenon using advanced methods such as random forest regression modeling. Maintaining continuous monitoring and updating models with new data will ensure the accuracy and relevance of mitigation strategies. Finally, UHI mitigation strategies in Matera need to integrate both urban planning and architectural design. While planning can guide the implementation of green spaces and space management, architectural design is essential to select appropriate materials that reduce heat absorption and improve natural ventilation. Both approaches need to be coordinated to ensure a sustainable and heritage-friendly solution.

6. Patents

The results of this preliminary research correspond to a research grant for the PhD internship in Cities and Landscapes: Architecture, Archeology, Cultural Heritage, History and Resources at DiCEM—Department of European and Mediterranean Culture: Architecture, Environment, and Culture. The research was conducted at DiCEM|CTM—House of Emerging Technology in Matera, Italy, under the project Adaptive Microcity: Digital Twin Experimentation = AFM & Ai+D, at the University of Basilicata.