Content area
Modeling urban microclimates is essential for assessing thermal comfort and the urban heat island (UHI) effect, particularly in the context of climate change. The UHI intensifies thermal discomfort, increases energy demand, and exacerbates health risks during extreme heat events. Accurate urban modeling is crucial for evaluating microclimatic conditions and developing effective mitigation strategies. However, traditional 3D modeling approaches often lack the efficiency and precision required to capture complex urban morphologies and integrating key environmental elements such as vegetation. This study presents an optimized workflow for large-scale 3D urban modeling that combines open-source geospatial data with programming and parametrisation tools to enhance the accuracy and scalability of urban studies. The methodology applied in Seville comprises data acquisition, processing, and modeling to produce a high-resolution urban environment model. Using Grasshopper and the ShrimpGIS plugin, spatial datasets of buildings and urban vegetation are processed to create a high-fidelity model. The resulting model is structured for integration into environmental analysis tools such as Ladybug Tools. This integration enables the direct assessment of design choices and morphological relationships for climate resilience, facilitating a detailed evaluation of urban microclimates and climate adaptation strategies. This approach provides urban planners and researchers with a replicable, efficient methodology to support evidence-based decisions for climate-responsive urban development.
Full text
1. Introduction
Global urbanization presents significant challenges for city planning and management, particularly in mitigating the urban heat island (UHI) effect. This phenomenon exacerbates thermal discomfort, increases energy demand, and intensifies the health risks associated with extreme heat events [1,2]. The urgency of addressing this issue is amplified by climate change projections indicating more frequent and intense heatwaves, especially in Mediterranean climates, where cities face compounded thermal stress [3]. To support decision-making for urban areas in the context of the global climate emergency, it is necessary to develop accurate models of urban configurations that enable the assessment of microclimatic conditions and the development of effective mitigation strategies [4]. Moreover, the complexity of urban systems demands integrated approaches that can simulate and capture both the physical and environmental dimensions of the UHI effect [5,6].
While advancements in three-dimensional (3D) urban modeling have improved the representation of built environments, significant limitations persist in terms of speed, accuracy, and efficiency, especially when capturing the complexity of urban morphology [7]. The level of detail in city models significantly influences their applicability for environmental simulations; there is a higher level of detail necessary for accurate microclimate analysis, but this is often difficult to achieve at urban scales [8].
Traditional 3D modeling approaches often rely on manual digitisation from 2D maps or the generation of models from terrestrial surveys. Both are labour-intensive and time-consuming, particularly in large urban areas with high geometric complexity [9]. More recent methods that use automated tools such as OpenStreetMap allow for the faster acquisition of volumetric urban data. However, these models often lack the necessary precision to represent real-world urban morphology, as they typically provide simplified building geometries and frequently exclude crucial information such as building heights [10]. The semantic richness of city models is equally important since the lack of attribute information beyond basic geometry limits their utility for sophisticated environmental analysis [11]. This is especially problematic in smaller urban areas and suburban contexts, where complete datasets are rarely available [12].
To improve efficiency and accuracy, recent studies have explored the integration of GIS-based spatial analysis with parametric modeling tools. These hybrid approaches allow for the automated generation of high-resolution urban models, enabling more detailed assessments of UHI dynamics and climate resilience strategies [13,14]. Unlike other methods, parametric workflows offer a distinct advantage by enabling rapid iteration and comparison of multiple design scenarios—such as variations in building geometry, material properties, and vegetation density—through adjustable parameters. This capability facilitates an iterative design process that allows planners to quickly test and optimize environmental performance criteria, leading to more efficient development of climate adaptation strategies [15]. Furthermore, recent research has emphasized that such parametric frameworks can bridge the gap between morphological modeling and environmental simulation by linking geometry parameters to microclimatic outcomes [16]. This connection has enabled more systematic exploration of design alternatives, where vegetation, materials, and form are dynamically parameterised to assess thermal comfort under changing climatic scenarios. However, most of these workflows still face scalability challenges, particularly when applied to heterogeneous Mediterranean urban fabrics, where fine-scale spatial variability significantly affects local heat stress. The need for robust, open-source methodologies that enhance urban microclimate research is particularly critical in cities with intricate spatial configurations, where the relationship between urban form and environmental performance is most complex [17].
Furthermore, a major limitation of existing urban models is their insufficient representation of vegetation, which plays a crucial role in mitigating UHI effects by reducing surface temperatures and improving thermal comfort [18,19]. Most current approaches either omit vegetation entirely or oversimplify its structural characteristics, neglecting variations in canopy density, tree species, and spatial distribution [20,21]. This simplification is problematic given that vegetation’s cooling efficacy depends significantly on its three-dimensional structure and spatial arrangement relative to built elements [22]. While high-resolution remote sensing techniques, such as LiDAR and hyperspectral imaging, offer more detailed vegetation data, their widespread application is constrained by high costs and computational demands [23,24]. Recent evaluations have shown that the level of detail in vegetation models can significantly impact the accuracy of microclimate simulations [25,26,27]. High-level-of-detail (LoD) tree models that incorporate volumetric canopies and species-dependent parameters have shown improved performance in representing radiative and shading processes, though at higher computational costs [26]. This trade-off highlights a persistent challenge: balancing geometric realism with simulation efficiency. Consequently, although vegetation geometry is often simplified at an urban scale, it is crucial to acknowledge how such abstractions—typically using cylindrical trunks and simplified canopy volumes—are acceptable for solar and shading analyses yet limit the precision of aerodynamic or evapotranspiration studies [26,27].
The integration of morphological analysis with environmental simulation represents another critical frontier in urban climate studies. The spatial composition and configuration of urban elements significantly influence local microclimates, necessitating methods that can quantify these relationships and translate them into design strategies [28]. This is particularly relevant for assessing the climate resilience of urban developments, where the interplay between building morphology, green infrastructure, and thermal performance must be evaluated holistically [29]. In this context, procedural modeling approaches offer promising avenues for generating and evaluating urban design alternatives based on environmental performance criteria [20,30].
To address these limitations, this research proposes an optimized workflow that integrates open-source geospatial data with parametrised 3D modeling techniques. Unlike traditional methods, which often lack precision or require expensive data sources, this approach integrates publicly available datasets to construct detailed urban models that include both built structures and vegetation. By building on recent advances in parametric modeling and spatial data integration, the methodology enables the generation of semantically rich 3D models suitable for microclimate analysis. The workflow specifically addresses the need for scalable solutions that can adapt to varying urban contexts and data availability, while incorporating vegetation representation that captures essential characteristics for environmental simulation. By streamlining data acquisition and processing, the proposed workflow enhances the accuracy, scalability, and applicability of the models across various urban environments, providing a robust foundation for climate-informed urban analysis and planning.
2. Materials and Methods
The proposed methodology, summarized in Figure 1, outlines an optimized workflow for large-scale 3D urban modeling to support climate-resilience planning. The process consists of three main phases: data acquisition, data processing, and 3D modeling. In the first phase, open-source geospatial data (e.g., building footprints and vegetation cover) are acquired from public portals. The subsequent data processing phase is performed using R Studio, where datasets are treated, spatially analyzed, and visualized to ensure quality, derive key 3D parameters, and prepare them for modeling. Finally, the processed data are integrated into the Rhinoceros 3D environment using the visual programming platform Grasshopper. Within this platform, the ShrimpGIS plugin is employed to import and process the files from the previous phase, converting them into 3D elements with attributes for integration into the parametric workflow. This process enables the creation of a detailed urban model ready for environmental analysis using tools such as Ladybug Tools [31] for assessing energy and thermal urban performance. The following sub-sections describe these three stages in detail.
2.1. Data Acquisition
The first stage involves acquiring geospatial datasets that contain essential information for large-scale urban modeling. This process involves identifying reliable data sources that provide spatial representations of urban elements such as buildings, vegetation, and land surfaces. Datasets are selected based on their level of detail, format compatibility, and availability, prioritizing open-access sources to ensure replicability across various urban contexts. This approach minimizes the reliance on proprietary or costly data, which can limit the method’s generalisability.
Geospatial data can be obtained from various sources, including national mapping agencies, municipal open data portals, and global repositories. To illustrate the global availability and diversity of these open-access resources, Table 1 provides a representative compilation of key data portals relevant to urban modeling. This compilation serves as a practical reference for identifying sources based on geographic coverage and data type, thereby facilitating the method’s replication in different urban contexts. Common formats include Geography Markup Language (GML), GeoJSON, or Shapefile (SHP), ensuring smooth integration into spatial analysis and modeling workflows.
It is important to note that the specific data requirements—such as Level of Detail (LoD), mandatory attributes, and spatial resolution—are not rigidly defined in this methodology, as they are inherently dependent on the specific research objectives, the scale of analysis, the chosen environmental simulation tools, and the data accessible for the case study location. The workflow presented is therefore a flexible framework designed to adapt to these variable conditions. However, to ensure the model’s basic functionality and the reliability of subsequent environmental simulations, a foundational set of data requirements is recommended. While the workflow can operate with minimal inputs, the absence of key attributes may limit the analytical depth or accuracy of the results. For instance, building footprints are fundamental for generating the built mass, but without height information, a 3D volumetric extrusion is impossible. Similarly, a geolocated point dataset of individual trees becomes significantly more valuable for microclimatic analysis when including essential attributes like height, crown dimensions, and species to accurately parametrize shading and evapotranspiration effects. The key principle is to build the model upon verified and validated data from reliable sources, ensuring the robustness of the results. To provide a practical guide for implementation, Table 2 outlines the core datasets and highly recommended attributes, highlighting the potential implications if these data are missing or of low quality.
The inclusion of attributes such as the construction year and building use enables more advanced analyses, such as estimating the thermal properties of the building stock and identifying priority areas for energy retrofitting. Similarly, a detailed point-based vegetation inventory with species information allows for accurate modeling of shading patterns and transpiration cooling effects. The acquired datasets can include additional geometric and descriptive attributes beyond those listed here. Resolution and reliability are key selection factors, as higher resolution and reliability will ensure greater accuracy in the subsequent modeling and simulation processes.
To further enhance the conceptual clarity and reproducibility of the proposed workflow, Figure 2 presents a lightweight data model that captures the core entities, attributes, and relationships underlying our methodology. This UML-style diagram illustrates the fundamental data structure while maintaining the framework’s inherent adaptability to different urban contexts and data availability scenarios.
This flexible approach is particularly valuable regarding formal data standards. In terms of data standardization and interoperability, the workflow is conceptually compatible with semantic frameworks such as CityGML. Specifically, its JSON-based encoding, CityJSON, provides a standardized schema for representing urban objects that aligns well with our methodology. This compatibility ensures that the workflow can leverage officially curated 3D city models when available, using CityJSON as an interoperable bridge between institutional data sources and the parametric modeling environment. However, the implementation maintains flexibility to accommodate the diverse data landscapes found across different municipalities, where such comprehensive standardized datasets may not yet be available, without compromising the core analytical objectives.
Once all required datasets are obtained, their consistency and completeness are evaluated. Since geospatial data are often sourced from different institutions and platforms, discrepancies in coordinate systems, missing attributes, or redundant information may arise. Addressing these inconsistencies is a key aspect of the subsequent data processing phase.
2.2. Data Processing
After acquisition, the raw spatial datasets undergo a processing phase using R Studio to ensure their reliability and compatibility with the 3D modeling workflow. This open-source programming environment was selected for its robust capabilities in computing, spatial data manipulation, and visualization, allowing for efficient handling of large geospatial datasets through scripting. The processing begins with harmonizing coordinate reference systems, as datasets from different sources may use varying projections. Standardizing these systems ensures spatial accuracy and prevents misalignments during modeling. Next, the data are cleaned and filtered to remove duplicates, correct inconsistencies, and address missing values. Additionally, a spatial filtering step is performed to select only those elements located within the study area boundaries, eliminating any data outside the zone of interest. This process is critical for reducing data volume and computational load, thereby streamlining the subsequent 3D modeling stages by focusing solely on relevant urban elements. The objective is to maintain high data integrity while ensuring that all relevant urban elements are accurately represented and optimized for further processing.
Following data cleaning, additional processing is performed in R Studio to ensure that the spatial attributes required for 3D modeling are properly defined. Sometimes, the available datasets do not provide attributes in the exact format needed for urban modeling. For example, instead of direct building height values, the data may contain the number of floors, requiring an estimation of height per floor to derive a new building height attribute. The scripting capabilities of R Studio facilitate these attribute processing and spatial operations efficiently. Once these derived attributes are established, the processed datasets are structured and exported in SHP file format to ensure compatibility with geospatial processing tools and seamless integration.
2.3. Three-Dimensional Urban Modeling
The final phase involves generating a three-dimensional urban model using Grasshopper in combination with the ShrimpGIS plugin. This open-source tool enables the direct import of geospatial datasets and their conversion into urban geometry, ensuring a seamless workflow from data acquisition to 3D representation.
The advantage of Grasshopper lies in its ability to incorporate various analytical plugins that extend its functionality. The ShrimpGIS plugin enables the direct import and manipulation of SHP files. Building footprints are extruded based on height attributes derived in earlier phases, ensuring volumetric accuracy. Urban vegetation is modelled parametrically, with attributes such as canopy radius and trunk height determining its geometric representation. Trees are typically generated as cylindrical volumes to balance computational efficiency with geometric realism, allowing their shading and microclimatic influence to be incorporated into subsequent analyses. Furthermore, through the use of plugins such as Ladybug Tools [31], Grasshopper facilitates not only microclimatic and thermal analysis but also energy, lighting, and thermal simulations at an urban scale. The proposed workflow is also designed to be compatible with other Grasshopper plugins that could complement these environmental assessments or address more specific analytical or procedural needs. Tools such as Eddy3D [32] for wind flow analysis or Colibri [33] for optimization algorithms could be readily integrated into the framework, enhancing its scope and application potential. This multi-faceted analysis is essential for evaluating urban performance, as it allows for the exploration of how the built environment interacts with its microclimate and energy flows. The 3D model, therefore, becomes a key tool for addressing urban challenges, such as UHI, energy efficiency, and sustainable design.
The workflow is also designed to incorporate additional geospatial layers—such as roads, pavements, and blue infrastructure—when data is available. Furthermore, the definition of surface material properties (e.g., albedo, emissivity, thermal mass) for all urban elements is a critical step for accurate microclimatic simulation. In contexts where this information is not available as geospatial data, these properties are assigned within the parametric environment of Grasshopper based on on-site measurements and validated datasets from technical specifications. This ensures the model empirically accounts for the radiative and thermal characteristics of different surfaces.
By integrating these parametric tools and environmental analysis plugins, the proposed methodology provides an efficient and scalable approach to 3D urban modeling, specifically tailored for climate-related assessments. The following section presents its application in a real case study, demonstrating how this workflow can be optimally integrated into an urban climate analysis framework through parametric tools.
3. Test Case Application and Results
3.1. Case Study Description
The proposed methodology was applied to the neighbourhood of Los Diez Mandamientos (37°22′ N, 5°58′ W), located in the southern district of Seville, Spain. This area is situated in the Guadalquivir River valley at an approximate altitude of 7 m above sea level and represents a critical case study within the Mediterranean climate zone. This climate is characterized by hot, dry summers, during which extreme heat events pose significant health risks. With summer temperatures frequently exceeding 40 °C, these severe climatic conditions create an urgent need for effective mitigation strategies. The neighbourhood’s selection is particularly relevant due to its high socioeconomic vulnerability, which compounds these environmental risks, thereby amplifying residents’ exposure to heat stress. Factors such as an older housing stock with poor thermal insulation, lower income levels, and a higher proportion of elderly residents have been identified as key contributors to its heightened sensitivity to heat impacts.
This combination of socioeconomic factors and unfavourable urban morphological characteristics makes this neighbourhood a priority area for climate adaptation interventions. The case study focuses on a defined area of approximately 1 km2 within Los Diez Mandamientos, which was delineated for detailed analysis (Figure 3). This context presents an exemplary scenario for demonstrating the potential of the proposed workflow, as it addresses the critical need to optimize thermal comfort during the summer months and mitigate the effects of extreme heat through improved urban design. To implement such strategies, an accurate and detailed 3D model is first required; this model must be capable of simulating microclimatic behaviour at the neighbourhood scale. The methodology developed in this research directly addresses this need, providing not only the final 3D model but also enabling its continuous analysis, refinement, and adaptation throughout the process. Furthermore, the methodology enables the model to be parametrised and integrated into a multifaceted environmental workflow within Grasshopper. This creates a comprehensive framework for evaluating and designing targeted mitigation measures.
3.2. Optimized Modeling Implementation
3.2.1. Data Acquisition
A robust and detailed dataset was fundamental for creating an accurate representation of the urban environment in Los Diez Mandamientos. The selection of specific data sources was guided by the principles of accuracy, official status, open access, and granularity of attributes, ensuring the model’s suitability for microclimatic analysis.
The building data came exclusively from the national Dirección General del Catastro (Figure 4a), the official Spanish Cadastre. This source was selected over alternatives, such as OpenStreetMap, for several critical reasons. Firstly, as a legal-administrative register, it provides an authoritative and highly accurate representation of building footprints and their subdivision into individual cadastral parcels (which define the geometry of volumes). This level of detail is essential for representing actual urban density. Secondly, the Cadastre contains invaluable descriptive attributes that are not commonly found in other open sources, such as the year of construction and renovation (useful for assessing building materials and thermal performance), the building use, and the number of floors per volume. While the building height was not explicitly provided, the number of floors allowed for a reliable estimation, as detailed in the processing phase. Crucially, the Spanish Cadastre is a key component of Spain’s Infrastructure for Spatial Information in Europe (INSPIRE) implementation. This status means the dataset follows pan-European standards for data formats, metadata, and web services. This compliance ensures not only consistency and interoperability but also long-term stability and accessibility. The dataset was directly downloaded in the INSPIRE-compliant XML/GML format from the official Cadastre INSPIRE service [34]. This dataset, updated in the first quarter of 2024, represents the most reliable and feature-rich open data source for this purpose.
Similarly, the vegetation data came from the Seville City Council’s Open Data Portal [35]. This municipal source (Figure 4b) was prioritized because it offers a curated and updated tree inventory, a level of detail that regional or global land cover datasets (e.g., CORINE Land Cover) cannot provide at the neighbourhood scale. This inventory includes the precise geolocation of individual trees and shrubs, along with key attributes for modeling: the tree species, height, and crown perimeter. The species data allows for classification into deciduous/evergreen and enables the parametrisation of species-specific transpiration rates. The data were provided in SHP format. The crown perimeter was a crucial attribute, as it enabled the calculation of the canopy radius, which directly influences shading simulations. The dataset, last updated in 2024, is part of the city’s forestry management plan, which guarantees its operational accuracy and makes it uniquely suited for a high-resolution study focused on thermal comfort.
These two primary datasets, both from official and open-access sources, provided a complementary foundation: the Cadastre for the precise geometry of the built environment and the City Council’s inventory for the explicit location and characteristics of the green infrastructure. Their selection minimizes costs and licencing restrictions while maximizing the methodological reproducibility in other Spanish and European contexts with similar data infrastructures. The subsequent phase involved processing these raw datasets in RStudio to transform them into an analysis-ready format for 3D modeling.
3.2.2. Data Processing
The acquired geospatial datasets underwent a systematic processing phase in R Studio (v4.3.1) to ensure geometric integrity, attribute completeness, and interoperability for the 3D modeling workflow. This phase was critical for transforming the raw, multi-source data into a unified and analysis-ready format. Figure 5 details the processing pipeline, which we executed using a suite of specialized R packages to ensure reproducibility and precision. The sf package was employed for all spatial data operations and transformations, while data manipulation and attribute computation were managed using dplyr, and ggplot2 was utilized for data visualization and quality control throughout the processing stages.
The cadastral building data, initially in INSPIRE-compliant GML format, was first read into RStudio. A critical initial step involved verifying and standardizing the Coordinate Reference System (CRS) to ensure spatial consistency across all datasets. The st_transform function from the sf package was employed to project all geospatial data into the WGS84 geographic coordinate system (EPSG:4326). This standardization guaranteed that all geospatial elements—buildings, vegetation, and other features—were perfectly aligned in a common coordinate framework, forming a coherent and accurate base for the subsequent integration and 3D geometric generation.
Following CRS verification, the building dataset underwent comprehensive data cleaning and attribute filtering. Non-essential administrative attributes that were irrelevant for 3D modeling and environmental analysis (e.g., parcel identifiers, the height below ground, the number of floors below ground). This step optimized dataset size and processing efficiency. The integrity of critical attributes was rigorously validated, with particular attention to the numberOfFloorsAboveGround field. Records with missing or null values in this essential attribute were identified and, following validation against complementary data sources, were assigned a default value of 1 floor to ensure completeness of the 3D model.
A key processing task was the derivation of the building height attribute. The raw dataset provided the number of floors (numberOfFloorsAboveGround) for each cadastral volume but not the absolute height. To address this, a height estimation algorithm was implemented. The floor height value of 3.2 m was determined through empirical analysis of the predominant building typologies within the study area, representing the average height derived from characteristic residential structures in this context. A new building_height attribute was programmatically calculated using the formula:
(1)
This parameter, while a generalization, provides a volumetrically consistent and reliable estimate for neighbourhood-scale environmental simulations, forming the basis for the subsequent geometric extrusion.
For its part, the municipal tree inventory, provided as a SHP file, was processed to convert descriptive attributes into actionable geometric parameters (Figure 6). The dataset underwent a rigorous quality control and cleaning procedure before geometric parameterization. Non-essential administrative fields were excluded, and the completeness of critical attributes—specifically tree_species, tree_height, and canopy_perimeter—was verified.
Each tree was represented as a georeferenced point. The dataset’s canopy_perimeter attribute was transformed into a canopy_radius to define the horizontal extent of the tree canopy in the 3D model. This conversion was performed using the geometric relationship:
(2)
This derived radius is crucial for accurately modeling the vegetation’s shading effects.
Furthermore, the tree_species attribute was leveraged to classify each specimen into broader physiological categories: deciduous or evergreen. This classification, executed via a lookup table cross-referencing species name with their phenological traits. This is a fundamental input for dynamic annual simulations, as it allows the model to account for seasonal variations in solar exposure and transpiration rates, particularly during the critical leaf-off period in winter. This significantly enhances the realism of microclimatic analyses.
To precisely define the study area and optimize computational efficiency, a spatial filter was applied to both datasets. A circular buffer with a 1 km radius was generated around a central georeferenced point within the Los Diez Mandamientos neighbourhood. Using spatial join operations, both the building footprints and tree points were clipped to this boundary. This step ensured that the subsequent 3D model and simulations focused exclusively on the area of interest, eliminating extraneous data and reducing processing loads.
Finally, the fully processed and harmonized datasets—which comprised building footprints with derived height attributes and tree points with calculated radius and phenological classifications—were exported as ESRI Shapefiles (.shp). This format was selected for its robust compatibility with the downstream 3D modeling environment in Grasshopper, ensuring a seamless transition to the geometric generation phase.
3.2.3. Three-Dimensional Urban Modeling
The final phase of the methodology involved translating the processed geospatial data into a semantically rich 3D urban model, which serves as the foundational geometry for subsequent environmental simulations. This process was executed within the visual programming environment of Grasshopper for Rhino 3D, utilizing a parametric workflow that ensures both scalability and adaptability for future studies (Figure 7a).
The processed Shapefiles, containing the validated building footprints with derived height attributes and the corrected tree points with calculated canopy radius, were imported into Grasshopper using the ShrimpGIS plugin. This plugin acts as a critical bridge between geospatial data and the parametric environment, enabling the direct reading and geometric interpretation of vector data within the 3D modeling workspace.
For building generation, the workflow utilized fundamental Grasshopper components to extrude each building footprint. The geometry attribute, containing the individual footprint polygons for each building volume, was directly coupled with the corresponding building_height attribute derived during the data processing phase. This straightforward component-based extrusion process transformed the 2D cadastral parcels into precise volumetric masses, accurately capturing the heterogeneous urban fabric where different volumes of the same building often have varying heights.
For vegetation representation, a component-based workflow was implemented in Grasshopper to generate simplified 3D tree geometries at each georeferenced point. This workflow used the tree_height and derived canopy_radius attributes. For each tree, the trunk was modelled as a vertical cylinder based on the trunk height. The canopy was geometrically represented not as a volume, but as a single, flat circular surface positioned at the top of the trunk, with its radius defined by the calculated canopy_radius attribute. While this representation is a significant simplification of complex biological forms, it was strategically chosen to maximize computational efficiency for subsequent urban-scale simulations. This simplified geometry is often sufficient for accurately capturing the essential shading effects in solar radiation and microclimatic analyses, as the primary interaction is through the projection of shadows. The prior classification of species as deciduous or evergreen remains crucial, as it allows for the dynamic adjustment of this surface’s properties (e.g., transparency, reflectivity) in seasonal studies to simulate leaf-on and leaf-off conditions.
The resulting integrated 3D model (Figure 7b) provides a high-fidelity digital twin of the Los Diez Mandamientos neighbourhood, accurately capturing the spatial distribution, volumetric relationships, and key morphological characteristics of both built and natural elements. A visual comparison with an aerial orthoimage (Figure 8) confirms the model’s geometric accuracy in representing the actual urban morphology of Seville. This geometric precision and the preserved semantic attributes are essential for reliable microclimate and thermal comfort analyses. They ensure that the physical interactions between buildings, vegetation, and open spaces are correctly represented in simulation engines.
To enable the subsequent microclimatic analysis, material properties were assigned to all surfaces in the model. As detailed geospatial data on road pavements, sidewalk materials, and building surface finishes was not available for the case study area, these were defined parametrically within Grasshopper based on on-site measurements and technical specifications from municipal construction standards. Table 3 summarizes the assigned properties, which are crucial for simulating albedo, thermal mass, and long-wave radiation exchange.
The simulation workflow was validated against empirical measurements collected in the same case study area during June 2024. The validation demonstrated strong performance with urban air temperature predictions, achieving R2 = 0.99 and a mean bias error of 0.21 °C when compared to in situ monitoring data, following ASHRAE Guideline 14:2002 standards.
3.3. Urban Microclimate Analysis
Following the optimized modeling implementation, the generated 3D urban model was employed in a two-stage environmental simulation workflow to quantify the Urban Heat Island (UHI) effect and its implications for human thermal comfort at the neighbourhood scale.
3.3.1. Urban Microclimate Generation and UHI Intensity
To accurately simulate urban microclimates, a rural meteorological baseline was established using the TMYx Seville weather file. This file was sourced from the TMYx (Typical Meteorological Year extended) dataset hosted on the
This rural weather file was subsequently transformed into an urbanized microclimate file using the Urban Weather Generator (UWG) [39] within the Dragonfly plugin for Ladybug Tools. The UWG refines air temperature data by incorporating key urban parameters derived directly from the 3D urban model, including building density and morphology, material properties (albedo and emissivity), vegetation cover, and anthropogenic heat loads. The use of a large-scale urban model is critical for this process, as the UHI is a mesoscale phenomenon. The 1 km2 model captures the aggregate thermal inertia of the built environment and the integrated effect of distributed heat sources, which a smaller model would fail to represent adequately.
The UHI intensity was quantified by calculating the difference between urban and rural near-surface air temperatures at 2 a.m. (UHI = ∑ Turban,2am − ∑ Trural,2am), following the methodology established by Martí Ezpeleta and Royé [40] and grounded in the theoretical framework of Oke et al. [41]. This nocturnal hour was selected as it typically marks the peak intensity of the UHI effect, when urban materials release heat absorbed during the day into a stable nocturnal boundary layer.
The analysis revealed a significant and persistent UHI effect, with the simulated data showing an annual average UHI intensity of 3.49 °C at 2 a.m. Figure 9 displays the comparative annual air temperature profiles, which confirm a consistent nocturnal temperature offset. This pattern is further detailed in Figure 10, which illustrates the average hourly UHI intensity throughout the year and clearly captures the characteristic diurnal pattern: a minimum during the day and a peak at night.
3.3.2. UHI’s Impact on Radiative Comfort
Although air temperature is a fundamental metric, human thermal perception is heavily influenced by radiant heat exchange. The Mean Radiant Temperature (MRT) is defined as the uniform temperature of an imaginary enclosure where the radiant heat transfer from the human body equals that in the actual non-uniform environment [42]. It is a critical driver of thermal comfort indices like the Universal Thermal Climate Index (UTCI), since radiative fluxes affect thermal sensation more than air temperature alone.
To assess the UHI’s impact on radiative comfort, the annual average difference in MRT was calculated for each point in a high-resolution grid (one point every 2 m) across the neighbourhood, using the formula: ΔMRT = MRTurban, 2am − MRTrural, 2am. At 2:00 a.m., in the absence of solar radiation, differences in MRT primarily reflect disparities in long-wave radiation emitted by urban surfaces, as well as sky temperature and urban geometry. Thus, the ΔMRT can be interpreted as a measure of the UHI’s radiative component that directly impacts a person’s energy balance.
The spatial variation of the ΔMRT across the study area is mapped in Figure 11. The results show that the average ΔMRT across all analysis points is 2.3 °C. This indicates that, on average, a person in the urban environment analyzed experiences a radiant field that is 2.3 °C warmer than in the rural reference.
The study reveals a combined exposure for residents, with an average air temperature increase of 3.49 °C (UHI effect) and an average radiant temperature increase (ΔMRT) of 2.3 °C. As air temperature and MRT influence human energy balance through distinct physiological mechanisms (convective vs. radiative heat exchange), their concurrent elevation presents a multifaceted thermal stress. This finding highlights the critical need for urban heat mitigation strategies that address not only the convective environment but also the complex radiative exchanges that are crucial for thermal comfort.
4. Discussion, Limitations, and Research Directions
The case study application demonstrates the methodology’s practical value in urban climate assessment. The generation of a detailed digital twin facilitated the identification of significant microclimatic variations within the Los Diez Mandamientos neighbourhood. The quantified urban heat island intensity of 3.49 °C at 2 a.m. aligns with findings from similar Mediterranean urban fabrics [3,43], while the spatial analysis of ΔMRT provides novel insights into the neighbourhood’s radiant environment. These results, supported by validation against empirical measurements in the same case study area, confirm the workflow’s capacity to support detailed environmental assessments that inform climate-responsive urban planning.
The relationship between urban morphology and microclimatic conditions suggests several mitigation approaches. Areas exhibiting higher ΔMRT values, typically associated with reduced sky view factors and materials with high thermal capacity, would benefit from interventions that enhance radiative heat loss and reduce heat storage. Crucially, the descriptive attributes from the Cadastre and municipal vegetation inventory enable targeted implementation of these strategies. The year of construction attribute identifies buildings constructed before thermal regulations (pre-1980 in the Spanish context) as priority candidates for envelope retrofits, especially within intense UHI zones. Building use classification further allows for customized approaches: residential structures may benefit most from reflective roofs and improved insulation, while commercial buildings might prioritize shading systems and ventilation enhancements. Meanwhile, the vegetation dataset provides essential intelligence for strategic greening: tree species classification distinguishes deciduous from evergreen specimens, informing placement decisions to optimize seasonal shading and solar access, while canopy dimensions help identify under-vegetated areas where new planting would yield maximum cooling benefits The integration of reflective materials, vegetative shading, and strategic urban geometry modifications could thus be strategically deployed based on this multi-layered semantic analysis to effectively address both the convective and radiative components of urban overheating. These findings align with previous research emphasizing the importance of multi-faceted heat mitigation strategies [28,44] while demonstrating how data-rich urban models can optimize their implementation.
Despite the workflow’s advantages, several limitations warrant consideration alongside practical compromise solutions for real-world applications. The accuracy of the generated models remains dependent on source data quality. The accuracy of microclimate simulations is further influenced by urban data completeness, particularly regarding surface material properties. For instance, while material assignments were based on measured values (Table 2), the absence of detailed spatial data on road pavements and building finishes introduces uncertainties. Sensitivity analysis indicates that albedo variations (±0.1) can affect simulated temperatures by 0.3–0.7 °C in Mediterranean climates [45], suggesting our reported UHI values may have a corresponding margin of error. For practical implementation, we recommend developing typological material libraries based on local construction characteristics when comprehensive data is unavailable. Similarly, the simplification of building height estimation can be addressed through typological height libraries derived from local construction patterns and building typologies, providing sufficiently accurate estimates for neighbourhood-scale analysis while respecting regional architectural characteristics. In any case, the application of both approaches should be coupled with sensitivity analysis protocols to quantify how these attribute uncertainties might affect simulation outcomes.
Computational demands present another practical constraint. The modeling and microclimate simulation of the study area with a high-resolution measurement grid (2 m spacing) required approximately 2 h and 15 min of processing time on a workstation with an AMD Ryzen 7 4800H processor, 32 GB RAM, and integrated AMD Radeon Graphics. Memory usage peaked at approximately 22 GB during the simulation phase. These requirements scale non-linearly with area size and grid resolution, suggesting that modeling larger districts or implementing finer grids would necessitate optimized data structures or high-performance computing resources. For rapid iterative design studies, strategic district segmentation and simplified vegetation representations can reduce processing time while preserving essential microclimatic interactions.
The intentional flexibility in data modeling, while accommodating diverse urban contexts, may affect reproducibility. Researchers must adapt attribute definitions to specific case studies, though the provided processing codes (Figure 4 and Figure 5) serve as adaptable templates. Crucially, we emphasize that any practical implementation should include case-specific validation through targeted field measurements to calibrate model parameters and verify simulation accuracy for reliable planning decisions.
Future research should address several promising directions with specific technical considerations. Building upon the validated workflow presented here, immediate research priorities include the detailed morphological segmentation of thermal results to correlate geometric parameters with microclimatic performance. The analysis of ΔMRT distributions by Sky View Factor, building height variability, and street canyon proportions will provide actionable insights for climate-responsive urban design. This deeper analytical layer, essential for translating environmental data into targeted mitigation strategies, exceeds the scope of the current methodological paper but represents a critical direction for ongoing research.
Another direction is the integration with computational fluid dynamics (CFD)to enhance comprehensive environmental assessment, particularly for evaluating ventilation and wind patterns. A practical pathway for this integration involves leveraging Eddy3D, a CFD plugin already compatible with Grasshopper, which would enable direct wind simulation within the existing parametric workflow. This approach avoids interoperability problems between software platforms. However, it still has high computational demands, which we propose addressing through multi-scale modeling strategies that focus detailed CFD analysis on critical areas identified by the current radiative model.
Furthermore, coupling the geometric model with socio-economic data layers represents a critical direction for enabling equitable urban planning. This integration builds directly upon our parallel existing vulnerability research in Seville, still in the process of being published, which has identified Los Diez Mandamientos as a high-priority area based on multifactorial analysis combining thermal exposure, demographic vulnerability (elderly population, unemployment rates), and energy poverty indicators related to building age and income levels. The technical challenge of data resolution mismatches between building-level geometric data and census tract socio-economic data can be addressed through spatial disaggregation techniques that align census tract data with building-level attributes from the Cadastre. This approach will transform the current model into a decision-support tool that prioritizes interventions for the most vulnerable populations, ensuring that climate adaptation strategies address both environmental and social equity dimensions. Cross-disciplinary collaboration frameworks between urban designers, social scientists, and local communities will be essential to operationalize this equitable approach.
5. Conclusions
The Urban Heat Island (UHI) effect exacerbates thermal stress, energy demand, and health risks in Mediterranean cities like Seville. This study responds by developing an accurate urban microclimate model to better understand and mitigate this phenomenon. An optimized workflow was developed for large-scale 3D urban modeling by integrating open-source geospatial data with parametric tools in Grasshopper and ShrimpGIS, enabling detailed representation of both buildings and vegetation. The methodology was applied to the Los Diez Mandamientos neighbourhood in southern Seville—an area of high climatic and social vulnerability—to analyze the local nocturnal microclimate and the combined influence of air temperature and Mean Radiant Temperature (MRT) on thermal comfort.
The proposed methodology demonstrates significant advantages in urban modeling efficiency through its integrated open-source approach. The workflow combines geospatial data processing in RStudio with parametric modeling in Grasshopper. This integration significantly reduces manual processing time and addresses two key shortcomings of previous studies: the labour-intensive nature of traditional modeling and the insufficient representation of urban vegetation. This streamlined process enables researchers to focus computational resources on simulation and analysis rather than model creation, addressing a key limitation identified in previous urban modeling studies. The automated data handling and processing pipeline ensures consistent geometric accuracy while maintaining semantic richness in the resulting 3D models, providing a reliable foundation for environmental analysis.
Most importantly, the parametric nature of the resulting digital twin transforms urban climate analysis from a diagnostic exercise into an active design tool. Urban designers can rapidly generate and evaluate multiple design alternatives within the Grasshopper environment, receiving immediate microclimatic feedback. This includes testing shading implementations, adjusting vegetation density and distribution, or assessing material changes. This research demonstrates that the synergy between open-source data processing and parametric modeling provides a powerful, replicable framework not only for urban climate analysis but for iterative, evidence-based design. The workflow enables designers to move beyond understanding existing conditions to actively shaping more resilient urban environments through rapid prototyping and simulation of design alternatives. The methodology provides immediate environmental performance feedback, thereby bridging the gap between analysis and action in climate-responsive planning. This empowers designers to optimize interventions before implementation and create more sustainable, comfortable urban spaces.
Conceptualization, J.D.-B., R.E. and A.A.; methodology, J.D.-B.; software, J.D.-B.; validation, J.D.-B.; formal analysis, J.D.-B.; investigation, J.D.-B.; data curation, J.D.-B.; writing—original draft preparation, J.D.-B.; writing—review and editing, J.D.-B., R.E. and A.A.; visualization, J.D.-B.; supervision, R.E. and A.A. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
The data presented in this study are available on request from the corresponding author due to privacy or ethical restrictions.
The authors gratefully acknowledge the Dirección General del Catastro of Spain for making their data publicly available through INSPIRE-compliant web services, and the Centro de Datos del Ayuntamiento de Sevilla for providing open spatial data through the Sevilla Open Data portal. The availability of this high-quality, standardized geospatial data was fundamental to this study.
The authors declare no conflicts of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Figure 1 Workflow of the proposed urban modeling optimization method.
Figure 2 Conceptual diagram of the data model for urban microclimate modeling. The diagram shows the main entities, their basic attributes, and the containment relationships within the study area. The cardinality notation “1” indicates exactly one instance, while “0..*” indicates from zero to many instances (i.e., the relationship is optional and can include multiple elements).
Figure 3 Geographical context of the case study, showing the location of Seville within Spain and the detailed boundaries of the Los Diez Mandamientos neighbourhood study area.
Figure 4 Open-data portals utilized in the case study: (a) Official Cadastre INSPIRE service providing building geometries and attributes; (b) Local government open data platform supplying urban vegetation datasets.
Figure 5 Buildings data processing general code in RStudio.
Figure 6 Trees data processing general code in RStudio.
Figure 7 (a) 3D Urban Modeling Workflow in Grasshopper; (b) Perspective view of the model.
Figure 8 Accuracy of the 3D Urban Model in Representing the Built Environment of Seville: (a) Aerial photograph (Source: Google Earth Pro, 2025); (b) Aerial view of the 3D urban model.
Figure 9 Comparative annual air temperature profiles: rural vs. urban conditions.
Figure 10 Comparative analysis of average hourly air temperatures: rural weather file vs. urban weather file with derived UHI intensity profile, showing characteristic nocturnal peak.
Figure 11 Nocturnal radiant environment mapping: spatial distribution of ΔMRT (MRTurban − MRTrural) at 2:00 a.m. across the study area.
Compilation of representative open-access geospatial data portals for urban modeling.
| Portal | Geographic Coverage | Primary Data Types | Common Formats | Key Characteristics/Licence | URL |
|---|---|---|---|---|---|
| Bhuvan | India | Orthoimagery, boundaries, infrastructure, hydrology | GeoTIFF, SHP, WMS | National geospatial portal of India. | |
| Copernicus (CORINE) | Europe | CORINE Land Cover, vegetation covers, protected sites | SHP, GeoTIFF, WMS/WFS | Pan-European environmental monitoring data. | |
| Data.gov (USA) | United States | Infrastructure, transportation, environment, public health | SHP, GeoJSON, APIs | Central portal for US open government data. | |
| EarthWorks (Stanford) | Global | Multidisciplinary collection of spatial and cartographic data | SHP, GeoJSON, KML | Academic institution repository. | |
| ECLAC Geoportal | Latin America & Caribbean | Geographic indicators, spatial statistics, socioeconomic development | SHP, GeoJSON | Official UN organization. | |
| ESA/NASA Earth Observation | Global | Satellite observations, land cover, elevation, NDVI, climate | GeoTIFF, SHP, GeoJSON | Data from official space agencies. | |
| EU Open Data Portal | Europe & Global | European statistics, environment, transport, climate | SHP, GeoJSON, GML | Official open data portal of the EU. | |
| GADM | Global | Administrative boundaries | SHP, GeoJSON, GPKG | Widely used in research; note version. | |
| Geoscience Australia | Australia | Cartography, Digital Elevation Models (DEM), boundaries, land use | SHP, GeoTIFF, WMS/WFS | National geoscience data portal of Australia. | |
| GISCO/Eurostat (EU) | Europe | Administrative boundaries, territorial statistics, networks | GML, SHP, GeoJSON | Official statistical data from the European Union. | |
| GSI (Japan) | Japan | Topographic maps, transport, hydrography, DEM | SHP, GeoTIFF, WMS | National mapping portal of Japan. | |
| IBGE (Brazil) | Brazil | Maps, networks, boundaries, orthophotos | SHP, GeoTIFF | Official Brazilian Institute of Geography and Statistics. | |
| IDB Open Data | Latin America & Caribbean | Economic, social, and environmental indicators | SHP, GeoJSON | Inter-American Development Bank. | |
| INSPIRE Geoportal (EU) | Europe | INSPIRE Themes: Administrative units, land use, transport, buildings, hydrography | GML, SHP, GeoJSON | European standard; GML is prevalent; detailed metadata available. | |
| ISRIC World Soil Data | Global | Soil properties, maps, soil profiles | SHP, GeoJSON, WFS | Data from an international scientific institution. | |
| Microsoft Building Footprints | United States & other selected countries | Building footprints | GeoJSON, SHP | Official Microsoft building footprints collection. | |
| Natural Earth | Global | Countries, cities, hydrography, relief | SHP (primary) | Public domain. | |
| Open Government Portal (Canada) | Canada | Boundaries, land parcels, transportation, natural resources | SHP, GeoJSON, GPKG | Federal open data portal of Canada. | |
| OpenStreetMap/Geofabrik | Global | Buildings, roads, land use, Points of Interest, parks | SHP, GeoJSON, PBF | Collaborative source; Open Database License (ODbL). | |
| Ordnance Survey OpenData (UK) | United Kingdom | Base maps, networks, buildings, transport | SHP, GeoJSON | Official open cartography of Great Britain. | |
| US Census TIGER/Line | United States | Boundaries, roads, hydrography, census tracts | SHP, GeoJSON | Official US census geographic data resource. | |
| USGS | United States & Global | DEM, topography, geology, land cover | GeoTIFF, SHP, WMS | Authoritative source for elevation and geological data. | |
| WorldMap (Harvard) | Global | Transportation, demography, economy, infrastructure | SHP, GeoJSON | Harvard University’s mapping platform. |
Core data requirements and implications of data gaps for urban microclimate modeling.
| Urban | Core Dataset | Minimum/Highly Recommended Attributes | Implications of Missing or Low-Quality Data |
|---|---|---|---|
| Buildings | Building Footprints (Polygons) | - Building Height (absolute value) or Number of Floors (for derivation). | Without height information, 3D extrusion is impossible. Lack of construction year prevents assessment of thermal performance. Missing building use data limits the accuracy of occupancy and internal load patterns in energy simulations. Missing identifiers complicate data management and attribute joining. |
| Vegetation | Tree Inventory (Points) | - Tree Height. | Without a point-based inventory with geometric attributes (height, crown), vegetation cannot be modelled parametrically, and its shading and cooling effects are poorly represented. Lack of species data prevents parametrisation of phenology and transpiration rates, reducing microclimate simulation accuracy. |
| Context | Study Area Boundary (Polygon) | - Clearly defined boundary for spatial filtering. | Lack of a defined boundary can lead to processing unnecessary data, increasing computational load and complexity. |
Material properties assigned to urban surfaces for microclimate simulation.
| Building Component | Construction Description | U (W/m2K) |
|---|---|---|
| Façade | 24 cm solid brick, unventilated air cam, septum of double hollow brick and gypsum plaster. | 1.44 |
| Roof | Andalusian style roof with slope formation with lime and charcoal concrete and loose screed. | 1.29 |
| Slab | Reinforced concrete slab with ceramic vaults and hydraulic tile flooring. | 1.57 |
| Frames | Aluminium frames without thermal break and blinds integrated in the façade enclosure | 5.88 |
| Glazing | Single glazing | 5.59 |
| Infiltration (ACH) | 0.35 [ | |
| Window/Façade ratio (%) | 15 | |
| Pavement material | Construction Description | U (W/m2K) |
| Concrete | Medium rough, thermal conductivity 1.73 W/mK, density 2243 Kg/m3, specific heat 837 J/KgK, thermal absorptance 0.9, solar and visible absorptance 0.65 | 3.58 |
| Asphalt | Medium rough, thermal conductivity 0.75 W/mK, density 2360 Kg/m3, specific heat 920 J/KgK, thermal absorptance 0.93, solar and visible absorptance 0.87 | 2.33 |
| Dry Sand | Rough, thermal conductivity 0.33 W/mK, density 1555 Kg/m3, specific heat 800 J/KgK, thermal absorptance 0.85, solar and visible absorptance 0.65 | 1.29 |
1. Oke, T.R.; Mills, G.; Christen, A.; Voogt, J.A. Urban Climates; Cambridge University Press: Cambridge, UK, 2017; ISBN 9781139016476
2. Oke, T.R. The Energetic Basis of the Urban Heat Island. Q. J. R. Meteorol. Soc.; 1982; 108, pp. 1-24. [DOI: https://dx.doi.org/10.1002/qj.49710845502]
3. Cos, J.; Doblas-Reyes, F.; Jury, M.; Marcos, R.; Bretonnière, P.-A.; Samsó, M. The Mediterranean Climate Change Hotspot in the CMIP5 and CMIP6 Projections. Earth Syst. Dyn.; 2022; 13, pp. 321-340. [DOI: https://dx.doi.org/10.5194/esd-13-321-2022]
4. Santamouris, M. Recent Progress on Urban Overheating and Heat Island Research. Integrated Assessment of the Energy, Environmental, Vulnerability and Health Impact. Synergies with the Global Climate Change. Energy Build.; 2020; 207, 109482. [DOI: https://dx.doi.org/10.1016/j.enbuild.2019.109482]
5. Quan, S.J.; Bansal, P. A Systematic Review of GIS-Based Local Climate Zone Mapping Studies. Build. Environ.; 2021; 196, 107791. [DOI: https://dx.doi.org/10.1016/j.buildenv.2021.107791]
6. He, L.; Geng, X.-W.; Huo, H.-Y.; Lian, Y.; Xi, Q.; Feng, W.; Tu, M.-C.; Leng, P. Simulation of Urban Thermal Environment Based on Urban Weather Generator: Narrative Review. Urban Sci.; 2025; 9, 275. [DOI: https://dx.doi.org/10.3390/urbansci9070275]
7. Biljecki, F.; Ledoux, H.; Stoter, J. An Improved LOD Specification for 3D Building Models. Comput. Environ. Urban Syst.; 2016; 59, pp. 25-37. [DOI: https://dx.doi.org/10.1016/j.compenvurbsys.2016.04.005]
8. Biljecki, F.; Ledoux, H.; Stoter, J.; Zhao, J. Formalisation of the Level of Detail in 3D City Modeling. Comput. Environ. Urban Syst.; 2014; 48, pp. 1-15. [DOI: https://dx.doi.org/10.1016/j.compenvurbsys.2014.05.004]
9. Rottensteiner, F. Semi-Automatic Extraction of Buildings Based on Hybrid Adjustment Using 3D Surface Models and Management of Building Data in a TIS; Technische Universität Wien: Vienna, Austria, 2001.
10. Pađen, I.; García-Sánchez, C.; Ledoux, H. Towards Automatic Reconstruction of 3D City Models Tailored for Urban Flow Simulations. Front. Built Environ.; 2022; 8, 899332. [DOI: https://dx.doi.org/10.3389/fbuil.2022.899332]
11. Rook, M.; Biljecki, F.; Diakité, A.A. Towards automatic semantic labelling of 3D city models. Proceedings of the 11th 3D Geoinfo Conference; Athens, Greece, 20–21 October 2016; pp. 23-30.
12. Fan, H.; Zipf, A.; Fu, Q.; Neis, P. Quality Assessment for Building Footprints Data on OpenStreetMap. Int. J. Geogr. Inf. Sci.; 2014; 28, pp. 700-719. [DOI: https://dx.doi.org/10.1080/13658816.2013.867495]
13. Nakata-Osaki, C.M.; Souza, L.C.L.; Rodrigues, D.S. THIS—Tool for Heat Island Simulation: A GIS Extension Model to Calculate Urban Heat Island Intensity Based on Urban Geometry. Comput. Environ. Urban Syst.; 2018; 67, pp. 157-168. [DOI: https://dx.doi.org/10.1016/j.compenvurbsys.2017.09.007]
14. Back, Y.; Kumar, P.; Bach, P.M.; Rauch, W.; Kleidorfer, M. Integrating CFD-GIS Modeling to Refine Urban Heat and Thermal Comfort. Sci. Total Environ.; 2023; 858, 159729. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.159729]
15. Peeters, A.; Etzion, Y. Automated Recognition of Urban Objects for Morphological Urban Analysis. Comput. Environ. Urban Syst.; 2012; 36, pp. 573-582. [DOI: https://dx.doi.org/10.1016/j.compenvurbsys.2012.05.002]
16. HosseiniHaghighi, S.; Izadi, F.; Padsala, R.; Eicker, U. Using Climate-Sensitive 3D City Modeling to Analyze Outdoor Thermal Comfort in Urban Areas. ISPRS Int. J. Geoinf.; 2020; 9, 688. [DOI: https://dx.doi.org/10.3390/ijgi9110688]
17. Waly, N.M.; Hassan, H.; Murata, R.; Sailor, D.J.; Mahmoud, H. Correlating the Urban Microclimate and Energy Demands in Hot Climate Contexts: A Hybrid Review. Energy Build.; 2023; 295, 113303. [DOI: https://dx.doi.org/10.1016/j.enbuild.2023.113303]
18. Armson, D.; Stringer, P.; Ennos, A.R. The Effect of Tree Shade and Grass on Surface and Globe Temperatures in an Urban Area. Urban For. Urban Green.; 2012; 11, pp. 245-255. [DOI: https://dx.doi.org/10.1016/j.ufug.2012.05.002]
19. Gill, S.E.; Handley, J.F.; Ennos, A.R.; Pauleit, S. Adapting Cities for Climate Change: The Role of the Green Infrastructure. Built Environ.; 2007; 33, pp. 115-133. [DOI: https://dx.doi.org/10.2148/benv.33.1.115]
20. Evola, G.; Costanzo, V.; Magrì, C.; Margani, G.; Marletta, L.; Naboni, E. A Novel Comprehensive Workflow for Modeling Outdoor Thermal comfort and Energy Demand in Urban Canyons: Results and Critical Issues. Energy Build.; 2020; 216, 109946. [DOI: https://dx.doi.org/10.1016/j.enbuild.2020.109946]
21. Javanroodi, K.; Mahdavinejad, M.; Nik, V.M. Impacts of Urban Morphology on Reducing Cooling Load and Increasing Ventilation Potential in Hot-Arid Climate. Appl. Energy; 2018; 231, pp. 714-746. [DOI: https://dx.doi.org/10.1016/j.apenergy.2018.09.116]
22. Buccolieri, R.; Santiago, J.-L.; Rivas, E.; Sanchez, B. Review on Urban Tree Modeling in CFD Simulations: Aerodynamic, Deposition and Thermal Effects. Urban For. Urban Green.; 2018; 31, pp. 212-220. [DOI: https://dx.doi.org/10.1016/j.ufug.2018.03.003]
23. Jia, J.; Wang, L.; Yao, Y.; Jing, Z.; Zhai, Y.; Ren, Z.; He, X.; Li, R.; Zhang, X.; Chen, Y.
24. Yip, K.H.A.; Liu, R.; Wu, J.; Hau, B.C.H.; Lin, Y.; Zhang, H. Community-Based Plant Diversity Monitoring of a Dense-Canopy and Species-Rich Tropical Forest Using Airborne LiDAR Data. Ecol. Indic.; 2024; 158, 111346. [DOI: https://dx.doi.org/10.1016/j.ecolind.2023.111346]
25. Ortega-Córdova, L. Urban Vegetation Modeling 3D Levels of Detail, Delft University of Technology. Master’s Thesis; TU Delft: Delft, The Netherlands, August 2018.
26. Xu, H.; Wang, C.C.; Shen, X.; Zlatanova, S. Evaluating the Performance of High Level-of-Detail Tree Models in Microclimate Simulation. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci.; 2022; X-4/W3-2022, pp. 277-284. [DOI: https://dx.doi.org/10.5194/isprs-annals-X-4-W3-2022-277-2022]
27. Krayenhoff, E.S.; Broadbent, A.M.; Zhao, L.; Georgescu, M.; Middel, A.; Voogt, J.A.; Martilli, A.; Sailor, D.J.; Erell, E. Cooling Hot Cities: A Systematic and Critical Review of the Numerical Modeling Literature. Environ. Res. Lett.; 2021; 16, 053007. [DOI: https://dx.doi.org/10.1088/1748-9326/abdcf1]
28. Kouklis, G.-R.; Yiannakou, A. The Contribution of Urban Morphology to the Formation of the Microclimate in Compact Urban Cores: A Study in the City Center of Thessaloniki. Urban Sci.; 2021; 5, 37. [DOI: https://dx.doi.org/10.3390/urbansci5020037]
29. Salvati, A.; Palme, M.; Inostroza, L. Key Parameters for Urban Heat Island Assessment in A Mediterranean Context: A Sensitivity Analysis Using the Urban Weather Generator Model. IOP Conf. Ser. Mater. Sci. Eng.; 2017; 245, 082055. [DOI: https://dx.doi.org/10.1088/1757-899X/245/8/082055]
30. Wai, C.Y.; Tariq, M.A.U.R.; Chau, H.-W.; Muttil, N.; Jamei, E. A Simulation-Based Study on the Impact of Parametric Design on Outdoor Thermal Comfort and Urban Overheating. Land; 2024; 13, 829. [DOI: https://dx.doi.org/10.3390/land13060829]
31. Roudsari, M.S.; Pak, M. Ladybug: A Parametric Environmental Plugin for Grasshopper to Help Designers Create an Environmentally-Conscious Design. Proceedings of the BS 2013: 13th Conference of the International Building Performance Simulation Association; Chambery, France, 26–28 August 2013; pp. 3128-3135.
32. Kastner, P.; Dogan, T. Eddy3D: A Toolkit for Decoupled Outdoor Thermal Comfort Simulations in Urban Areas. Build. Environ.; 2022; 212, 108639. [DOI: https://dx.doi.org/10.1016/j.buildenv.2021.108639]
33. Tomasetti, T. Core Studio. Colibri, in TT Toolbox for Grasshopper. 2017; Available online: http://Core.Thorntontomasetti.Com/Tt-Toolbox-for-Grasshopper/ (accessed on 10 October 2025).
34. Dirección General del Catastro Servicios INSPIRE de Cartografía Catastral. Available online: https://www.catastro.hacienda.gob.es/ (accessed on 21 April 2025).
35. Centro de Datos del Ayuntamiento de Sevilla Datos Espaciales Abiertos de Sevilla. Available online: https://cda-idesevilla.opendata.arcgis.com/ (accessed on 21 April 2025).
36. Fernández-Agüera, J. Caracterización de La Estanqueidad al Aire de La Vivienda Social En Andalucía. Ph.D. Thesis; Universidad de Sevilla: Sevilla, Spain, 2018.
37. Lawrie, L.K.; Crawley, D.B. Development of Global Typical Meteorological Years (TMYx). Available online: https://climate.onebuilding.org/WMO_Region_6_Europe/ESP_Spain/ (accessed on 2 October 2025).
38.
39. Bueno, B.; Norford, L.; Hidalgo, J.; Pigeon, G. The Urban Weather Generator. J. Build. Perform. Simul.; 2013; 6, pp. 269-281. [DOI: https://dx.doi.org/10.1080/19401493.2012.718797]
40. Martí Ezpeleta, A.; Royé, D. Intensidad y Duración Del Estrés Térmico En Verano En El Área Urbana de Madrid. Geographicalia; 2021; pp. 95-113. [DOI: https://dx.doi.org/10.26754/ojs_geoph/geoph.2021735202]
41. Oke, T.R.; Johnson, G.T.; Steyn, D.G.; Watson, I.D. Simulation of Surface Urban Heat Islands under “ideal” Conditions at Night Part 2: Diagnosis of Causation. Bound. Layer Meteorol.; 1991; 56, pp. 339-358. [DOI: https://dx.doi.org/10.1007/BF00119211]
42. Fanger, P.O. Thermal Comfort. Analysis and Applications in Environmental Engineering; Danish Technical Press: Copenhagen, Denmark, 1970.
43. Núñez Peiró, M.; Sánchez-Guevara Sánchez, C.; Neila González, F.J. Update of the Urban Heat Island of Madrid and Its Influence on the Building’s Energy Simulation. Sustainable Development and Renovation in Architecture, Urbanism and Engineering; Springer International Publishing: Berlin/Heidelberg, Germany, 2017; pp. 339-350. ISBN 9783319514420
44. Sola-Caraballo, J.; Lopez-Cabeza, V.P.; Roa-Fernández, J.; Rivera-Gomez, C.; Galan-Marin, C. Assessing and Upgrading Urban Thermal Resilience of a Spanish MoMo Neighbourhood over the Span of 1960–2080. Build. Environ.; 2024; 256, 111485. [DOI: https://dx.doi.org/10.1016/j.buildenv.2024.111485]
45. Campra, P. Numerical Simulation of the Effects of Increasing Urban Albedo on Air Temperatures and Quality over Madrid City (Spain) by Coupled WRF/CMAQ Atmospheric Chemistry Model. Understanding of Atmospheric Systems with Efficient Numerical Methods for Observation and Prediction; IntechOpen: London, UK, 2019.
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.