Edu-Clima: A Didactic Tool for the Simulation and Modeling of Urban Climate

Edu-Clima: A Didactic Tool for the Simulation and Modeling of Urban Climate | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Edu-Clima: A Didactic Tool for the Simulation and Modeling of Urban Climate Henrique Nicolau Grillaud Maranholi, Flávia Maria de Moura Santos, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7743068/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Apr, 2026 Read the published version in Nativa → Version 1 posted You are reading this latest preprint version Abstract This article presents the development and validation of Edu-Clima, an open-source educational application for simulating urban surface temperature. The objective of this study was to develop and validate an urban surface temperature simulation application for educational purposes by comparing its results with those obtained via ENVI-met software and evaluating its reliability for use in teaching urban climatology. Edu-Clima was built in Python and uses libraries such as Streamlit, Pandas, Matplotlib, Seaborn, and Plotly to create an interactive and visually educational interface. The methodology involved comparing the results of Edu-Clima with those of ENVI-met, using 12 different scenarios, each with 23 simulations, totaling 276 simulations. The scenarios considered variables such as surface temperature, leaf area index (high, medium, low), ground cover (50% asphalt or concrete), and soil type (exposed or vegetated). The results revealed that although ENVI-met consistently presented higher temperatures, Edu-Clima followed a pattern closer to a "climate normal." Statistical analyses, including quadratic regression and PERMANOVA, revealed that, despite visual differences, there is no statistically significant distinction between the temperatures simulated by the two tools, validating Edu-Clima for educational purposes, in line with SDG 4 (Quality Education), which emphasizes the promotion of inclusive and accessible educational resources. However, Edu-Clima showed greater sensitivity in vegetated scenarios, indicating a limitation in evapotranspiration and shading modeling, which are more accurate in ENVI-met. In the future, we highlight the importance of improving the modeling of vegetation effects, including variables such as evapotranspiration, soil moisture, and shading. Environmental Sciences Environmental Education Urban Climate Simulation Climatology Teaching Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. INTRODUCTION The increasing degree of urbanization and the environmental challenges associated with city densification have intensified interest in tools that help understand and mitigate the impacts of the urban microclimate (Trane et al., 2023). Among the most relevant elements in this scenario are surface temperature, vegetation cover, and the types of materials used in urban soil, which are closely linked to the quality of life in urban areas and the thermal comfort of their inhabitants (Makvandi et al., 2023). In contemporary scenarios, which are characterized by extreme weather events, urban heat islands, and growing concern for environmental sustainability, understanding climatology has become an essential skill not only for specialists but also for public managers, educators, and citizens in general (Riuttanen et al., 2021). The ability to interpret and predict microclimatic patterns is fundamental for resilient urban planning and more informed decision-making (Maranholi & Santos, 2024). In this context, computational climate simulators have established themselves as essential tools for analyzing and predicting environmental variables (Camps-Valls et al., 2025). However, although robust platforms such as ENVI-met offer high precision and detail, their complexity and cost may limit their use in educational or preliminary planning contexts (Pacifici & Nieto-Tolosa, 2021). Given this, this research proposed the development of a proprietary application with an accessible interface and educational purpose for simulating surface temperature in different urban scenarios. The development of open-source applications that are easy to understand and use is essential to democratize access to these tools (Pacifici & Nieto-Tolosa, 2021; Kim & Kwon, 2025). When made available as open sources, such applications offer the opportunity for adaptation and customization, favoring their use not only by professionals but also by students and educators seeking accessible learning tools (Streb et al, 2021). In this sense, the development of the Edu-Clima application emerges as a strategic tool, enabling the simulation of different urban climate conditions in an accessible, didactic, and interactive way. By allowing users to explore variables such as surface temperature, vegetation cover, and soil types, the application not only facilitates the learning of complex concepts but also promotes awareness of the effects of urban choices on the local thermal environment, contributing to a more critical and engaged citizenry with respect to climate issues (Lozano-Díaz & Fernández-Prados, 2020; Kurokawa et al, 2023). This pedagogical character is directly linked to SDG 4 – Quality Education, as Edu-Clima presents itself as an innovative resource for expanding access to inclusive and equitable educational practices. By transforming technical data into easy-to-understand interactive experiences, the application strengthens the teaching of urban climatology, stimulates active learning, and promotes lifelong learning opportunities, both in school environments and in nonformal education spaces. Thus, the project contributes to the development of essential skills to address the challenges of sustainability and urban resilience. In this context, the present study investigated the feasibility of Edu-Clima as a reliable teaching tool for simulating urban surface temperature. Our hypothesis is that there is no statistically significant difference in the surface temperature results simulated by the Edu-Clima application and those obtained by the ENVI-met software. As a procedural premise, we sought to verify its reliability using statistical methods and multivariate analyses. The investigation thus aimed to contribute both to the improvement of teaching methodologies in urban climatology and to the democratization of digital tools applied to the sustainability of cities. Therefore, our goal was to develop and validate an open-source urban surface temperature simulation application by comparing its results with those obtained by the ENVI-met software and evaluating its reliability for application in the teaching of urban climatology. Thus, this paper is structured to detail the development and validation of the Edu-Clima application. The Methodology section describes the steps involved in building the application in Python, as well as the strategy for comparing the results with the ENVI-met software, including the variables and the 12 simulation scenarios. Next, the Results section presents statistical analyses, such as quadratic regression and PERMANOVA, which validate the reliability of Edu-Clima alongside ENVI-met. In the Discussion section, we further explore the positive points and limitations, especially in the scenarios analyzed. Finally, the conclusion summarizes the main findings, reinforces the application's contribution to the teaching of urban climatology, and points to avenues for future studies. 2. MATERIALS AND METHODS For the development of this research, an application was built to simulate different scenarios with different temperatures, types of ground cover, and leaf areas. To obtain more reliable results, scenarios were created via ENVI-met [1] , a three-dimensional microclimate simulation software used to model urban environments and assess the impacts of factors such as morphology, vegetation, and building materials on the local climate. This approach aimed to ensure the greatest possible similarity with the scenarios plotted in the developed application. 2.1. Application development The application was developed in Python [2] , with the support of several libraries and complementary technologies (Waskom, 2021). The analysis included an in-depth discussion of the technological resources employed, the methodology adopted during implementation, and the results obtained at the end of the process. To this end, a tool was created and implemented to compare temperatures provided by the OpenWeather API [3] and data recorded at specific geographic coordinates while also considering the associated climatic conditions. Program Operation The program uses the Streamlit library [4] to create an interactive web interface. The program workflow can be described in several steps: a) Data Collection: The program queries the OpenWeather API to obtain temperature data for a specific coordinate; b) Local Data Reading: Reading of locally recorded temperature data from CSV files using Pandas [5] ; c) Data Processing: The data are cleaned and processed to remove outliers and fill in missing values; d) Comparison and Analysis: API data and local data on temperature, leaf area, and type of ground cover (asphalt or concrete) and ground conditions (exposed or vegetated) are compared and analyzed; e) Visualization: The results are visualized in interactive graphs via Matplotlib [6] , Seaborn [7] , and Plotly. Streamlit [8] , which facilitates visualization in the web interface. Source Code The source code has been organized into modules to facilitate program maintenance and scalability. Thus, they will be presented below using "" (quotation marks). The main components include the following: a) inicio.py: Main file that initializes the Streamlit application and manages the user interface. b) The screen rendering uses the “columns” feature of the Streamlit library to divide the interface into three columns. c) The “center_column” renders “st.components.v1.html” – Streamlit uses the “components” function to transform the return of the “heatmap” function into an HTML file format understood by the browser; d) The "right" column rendered after the "center_column" call, shown above, contains the st.dataframe function (a method to inform Streamlit that a Dataframe will be rendered there). e) Next to the previous column is the "left" screen division, which complements the previous right one by rendering the bar chart, called in the chart by the function "st.plotly_chart(fig)", where: Fig: contains the "figure" of the rendered chart; St.plotly_chart is the streamlit function that renders the chart figure within the left division of the screen. f) sidebar.py: File containing the functions responsible for interacting with the OpenWeather API, generating the heatmap, and rendering the tool's side menu. g) Obter_temperatura_openweathermap: function that receives the latitude and longitude parameters to query the OpenWeather API, using the key obtained directly from their website, contained in the "api_key" object, for the climate information relevant to the scope of the project. h) The API returns a JSON file full of information, from which we extract the data we need. This functionality is located within the "try except" of the function. i) Folium: Library used to render the geographic map. In the code, it received the alias "fo." j) fo. Map – receives the parameter of the type of lines to be rendered on the map, and a zoom_start value of 13 – indicates that the map should render the coordinates with the zoom level appropriate for the search. k) fo. CircleMarker – renders the border of the circles that represent the temperature of the rendered region. The rendering of all circles that appear in the tool is carried out within the “for point-in-points” loop, where the search coordinates are queried in the API. l) Sidebar_dados_pesquisa – this function renders, in a side menu, all the fields in which the user enters temperature collection information. When invoked by the main.py file, it returns latitude_pesquisa, longitude_pesquisa, temperatura_final, and temperatura_sem_foliar. It consists of the following functions, among others: Latitude_search, a function that renders the latitude input field in the side menu. If the user does not enter a value, it will use the default value -15.612642354149061 to pass information to the API; Search_longitude: similar to the previous function, with a default value of -56.038829296302545 if no information is entered by the user; Search_temperature: renders the temperature input field collected by the user. By default, if the field is not filled in, it will contain the value obtained by the geographical coordinates above; Selected_vegetation: renders the options as a menu with 'high', 'medium', 'low', and 'manual' for user selection (in this case, an information entry field will be provided to the user); Percentage_ground_cover: renders the options 'asphalt', 'concrete', and 'manual' for user selection. The behavior of the 'manual' selection is the same as that of the previous selection. m) dataframe.py: This file contains the functions "dataframe" and "plot_graph," which return "df" and "fig,", respectively. n) dataframe: receives the arguments "search_latitude," "search_longitude," "search_temperature," and "regions" to render the dataframe displayed in the tool. The function uses Pandas to perform the operations. o) plot_graph: receives "df" as an argument and uses the Plotly library through the Express module to create the bar chart displayed in the tool. p) page_especs.py: file containing the page_config function. q) page_config: Streamlit is used, through the alias "st," to use the "set_page_config" method to inform the browser of interface details, which are page_title; page_icon; layout. 2.2. Parameters for scenario production: ENVI-met × Edu-Clima For the simulations to be as similar as possible, the input parameters were defined for both ENVI-met and the developed application. The following elements were used to construct the scenarios: surface temperature, leaf area index, ground cover (asphalt or concrete), and soil type (vegetated or exposed). Thus, for temperature, the times and values were described (Table 1). Table 1 Times and values arbitrarily assigned for the simulations 00:00 hours - 20.98°C 1:00 p.m. - 26.03°C 02:00 hours - 20.03°C 2:00 p.m. - 25.98°C 3:00 a.m. - 19.28°C 3:00 p.m. - 26.48°C 4:00 a.m. - 18.71°C 4:00 p.m. - 26.35°C 5:00 a.m. - 18.09°C 5:00 p.m. - 25.19°C 6:00 a.m. - 18.33°C 6:00 p.m. - 24.84°C 7:00 a.m. - 19.01°C 7:00 p.m. - 24.25°C 8:00 a.m. - 20.47°C 8:00 p.m. - 23.73°C 9:00 a.m. - 22.20°C 9:00 p.m. - 23.05°C 10:00 a.m. - 23.60°C 10:00 p.m. - 22.44°C 11:00 a.m. - 24.66°C 11:00 p.m. - 21.82°C 12:00 p.m. - 24.96°C For the leaf area index, three different levels were established: high LAI (leaf area index), where an estimated canopy cover of greater than 90% was assigned; medium LAI, where an estimated canopy cover of approximately 50% was assigned; and low LAI, where an estimated canopy cover value of less than 15% was assigned. For the soil cover layer, two types of materials were considered: concrete and asphalt. For these materials, although both ENVI-met and Edu-clima allow for different percentages, a constant soil cover percentage of 50% was established. Regarding the type of soil used to produce the scenarios, two types were defined: exposed soil, where there is no type of construction material, or even natural cover. The other type of soil defined was vegetated soil, where natural grass cover was considered. The parameters can be seen in Table 2. Table 2 Parameters for constructing simulation scenarios LAI 1 – High Soil cover – asphalt (50%) Soil type – exposed LAI – Medium Soil cover – asphalt (50%) Soil type – exposed LAI – Low Ground cover – asphalt (50%) Soil type – exposed LAI – High Ground cover – concrete (50%) Soil type – exposed LAI – Medium Ground cover – Concrete (50%) Soil type – exposed LAI – Low Ground cover – Concrete (50%) Soil type – exposed LAI – High Ground cover – asphalt (50%) Soil type – Vegetated LAI – Medium Ground cover – asphalt (50%) Soil type – Vegetated LAI – Low Ground cover – asphalt (50%) Soil type – Vegetated LAI – High Ground cover – concrete (50%) Soil type – Vegetated LAI – Medium Ground cover – Concrete (50%) Soil type – Vegetated LAI – Low Ground cover – Concrete (50%) Soil type – Vegetated 1 LAI – Leaf Area Index Finally, after the parameters and constituent elements of the scenarios were defined, a total of 12 different configurations were established. For each scenario, 23 simulations were performed, totaling 276 simulations. This set was initially run in ENVI-met software, considering the real scenario, and later reproduced in Edu-Clima, resulting in another 276 simulations. For nomenclature purposes, each scenario was named according to its arbitrary characteristics: a) AEB – Asphalt with exposed surface and low leaf area index; b) AEM – Asphalt ground cover with exposed surface and medium leaf area index; c) AEA – Asphalt ground cover with exposed surface and high leaf area index; d) AVB – Asphalt ground cover with a vegetated surface and low leaf area index; e) AVM – Vegetated asphalt surface ground cover and medium leaf area index; f) AVA – Vegetated asphalt surface with a high leaf area index; g) CEB – Concrete ground cover with exposed surface and low leaf area index; h) CEM – Concrete ground cover with exposed surface and medium leaf area index; i) CEA – Concrete ground cover with exposed surface and high leaf area index; j) CVB – Concrete ground cover with a vegetated surface and low leaf area index; k) CVM – Concrete ground cover with vegetated surface and medium leaf area index; l) CVA – Concrete soil cover with a vegetated surface and high leaf area index. 2.3. Data exploration and statistical analysis To explore the results, the deviation (°C) was calculated from the difference between the simulated temperatures in ENVI-met and Edu-Clima to verify the existence of climatological anomalies. The temperature variation (%) between the simulators was calculated by dividing the deviation value (°C) by the average temperature of the simulators (Almazroui et al., 2021; Twardosz et al., 2021). The coefficient of variation was calculated by dividing the standard deviation of temperatures throughout the day by their average. We applied the paired t test to compare the average temperatures between the two simulators to determine whether there was a significant difference between them in the different scenarios evaluated (Aslam, 2021). To verify whether time (X-axis) influenced temperature (Y-axis) in both simulators for the different scenarios, cubic regression analysis was applied, estimated by the least squares method using Minitab 19 statistical software, with an analysis of variance at a 5% probability of rejecting the null hypothesis (Aslam, 2021). Multivariate statistics were also explored, with principal component analysis (PCoA) based on Euclidean distances, followed by PERMANOVA to verify whether the observations made between the simulators behaved differently. [1] You can find more information and download the free version of this software on its official website: https://envi-met.com/. [2] Python is an interpreted, high-level, dynamically typed programming language. It is widely used due to its simplicity and robustness, facilitating rapid application development. Its vast collection of libraries makes it ideal for scientific and data analysis projects. Python Software Foundation (PSF) license, which is permissive and allows commercial use. [3] The OpenWeather API provides access to global weather data, including forecasts for temperature, humidity, precipitation, and more. The API is often used in applications that require accurate and up-to-date weather data. Available under different subscription plans, including free and paid options. [4] Streamlit is a Python library that allows you to quickly and easily create interactive web applications. It is particularly useful for data visualization and dashboards, making it a popular choice among data scientists. Apache 2.0 license, a permissive license that allows use, modification, and distribution. [5] Pandas is a software library written for the Python language for data manipulation and analysis. It provides data structures and operations for manipulating numerical tables and time series. [6] A 2D plotting library in Python that produces high-quality figures in a variety of formats and environments. [7] Based on Matplotlib, Seaborn provides a high-level interface for creating attractive statistical graphics. [8] An interactive graphics library that allows the creation of complex visualizations with zoom, pan , and hover capabilities . 3. RESULTS 3.1 Edu-clima The result of the application development process, named Edu-clima (accessible at https://simuladorclimaticodoutoradomaranholi.streamlit.app/), consisted of the creation of a functional, practical, interactive, and educational application aimed at simulating surface temperature in different urban scenarios (Fig. 1). The design was intended to be simple and understandable for the user. The central goal was to transform technical and abstract content, such as thermal behavior in urban areas, into an interactive and accessible experience capable of engaging students and educators at different levels of education. Unlike robust simulators such as ENVI-met, whose use requires advanced technical knowledge and specific computational infrastructure, Edu-Clima was designed on three main premises: simplified usability, didactic visualization, and integration with open data. Edu-Clima sensitivity to small variations in input parameters such as type of cover, vegetation, and time of day—allows users to experiment with scenarios and observe their thermal implications, which enhances engagement in pedagogical strategies such as investigative teaching, problem-based learning (PBL), and gamification. Another carefully designed feature was the presentation of a table with temperatures in neighborhoods in the city of Cuiabá, Mato Grosso (Fig. 2), where the person using the app can immediately make comparisons with different locations in the city. Although we used the coordinates of Cuiabá, Brazil, as the basis for the app's simulations and validations, it is important to note that the system is not limited to this location. Edu-Clima was developed with a flexible architecture, allowing users to enter any geographic coordinates to simulate the microclimate of any urban area in the world if the user adheres to the characteristic metric of the local climate. The interface based on the Streamlit library, for example, allows for fluid and intuitive navigation, whereas the use of libraries such as Folium, Plotly , and Pandas enables the creation of interactive maps, dynamic graphs, and real-time tables—elements that reinforce visual understanding and interactivity with the phenomenon being studied. 3.2 Data Verification and Validation In the real scenario, as expected, the temperature (°C) varied according to time (hours) (Fig. 3), with a nonlinear relationship between the variables, represented by a cubic polynomial trend curve and coefficient of determination ( R² ) of 0.8788. °The analysis of variance (ANOVA) of the regression was significant (p<0.05), suggesting that the temperature variation in the real scenario was not influenced by the time of day. The maximum temperature recorded during the period was 26.48°C at 3:00 p.m., whereas the minimum temperature reached 18.9°C at 5:00 a.m. The prevailing average temperature was 22.63°C (∘ C), indicating a moderate thermal regime for the city of Cuiabá, Brazil (Table 3). The variability of temperatures in the dataset was characterized by a standard deviation of 2.73°C and a coefficient of variation of 12.06%, suggesting moderate temperature dispersion (Table 3), which implies stable daily thermal conditions at the time and place of study A clear visual difference between the two climate simulators, ENVI-met and Edu-Clima, was evident for all the scenarios studied. The temperatures simulated by ENVI-met were consistently higher than those simulated by Edu-Clima. When comparing the simulations performed with ENVI-met and Edu-Clima for the different scenarios, the observed behavior line of Edu-Clima follows a pattern closer to a climatic normal, presenting a lower temperature variation at dusk. In contrast, in ENVI-met, for all scenarios, the temperature continued to rise throughout the day or remained stable from dusk to nightfall (between 5 p.m. and 11 p.m.) (Fig. 4 and Fig. 5). The climate behavior simulated by Edu-Clima for the city of Cuiabá, MT, reflects the trend observed in the real scenario. As in the city, the application results in the highest temperatures between 10 a.m. and 2 p.m., with a gradual reduction throughout the evening. The regression analysis presented in Fig. 4 and Fig. 5 reveals that the temperature varies with time for the different soil cover scenarios, whether they are asphalt with exposed surfaces, different leaf area indices, or even asphalt with vegetated surfaces. This pattern is also repeated for concrete ground covers, with or without exposed surfaces, at different leaf area indices. This can be seen in Fig. 4, where the temperature provided by ENVI-met on the asphalt ground cover with an exposed surface and a low leaf area index (LAI) at 8 a.m. was significantly lower than that at 2 a.m. Analysis of the temperature data using a third-order regression curve revealed that the thermal behavior in the study area does not follow a normal distribution, indicating complex and nonlinear variability. Interestingly, the observations generated in both applications within the different scenarios showed a similar behavior pattern in the polynomial line. The coefficient of determination (R²) is a statistical indicator used to assess the quality of a regression model. It measures the proportion of variation in the data that is explained by the model. The R² values found in our equations indicate that the model is a reliable tool for describing the temperature trend of the data. An analysis of variance of the regressions for the different scenarios revealed that time had a significant influence on temperature in both the simulators and all the scenarios (Fig. 6; p < 0.05). The average variation in the scenarios with exposed surfaces ranged from 13.27% (AEB) to 37.49% (CEA) (Fig. 6). In the scenarios with vegetated surfaces, the variation observed ranged from 27.27% (AVB) to 47.11% (CVA). These values suggest that the variation in temperatures between simulators may present an acceptable range between average, maximum, and minimum temperatures. The cumulative coefficient of variation for the scenarios with vegetated and exposed surfaces remained below 11%, indicating that the data collected from both simulators tended toward normality when compared. This is because the cumulative coefficients of variation were close to the individual coefficients of variation of the temperatures extracted from each simulator over time. The appropriate measures for expressing variability are generally the standard deviation (absolute variability) and the coefficient of variation (relative variability), which vary between 2.3°C and 2.8°C between the urban scenarios and the simulations. The coefficients of variation ranged from 8.63% to 12.29% (Fig. 6 and Fig. 7). These results reinforce that the simulations performed in ENVI-met and Edu-Clima are within an acceptable range, which validates the simulator developed in relation to the temperatures obtained throughout the simulations and scenarios. The values of the coefficients of variation, which are less than 30%, corroborate the absence of a significant difference between the temperatures projected by the Edu-Clima and ENVI-met simulators. The average temperature deviation between ENVI-met and Edu-Clima varied considerably depending on the scenario. In the scenario of asphalt soil cover with an exposed surface and a low leaf area index (LAI), the average temperature deviation was 3.84°C (Fig. 7a). In contrast, in the scenario of a concrete ground cover with a vegetated surface and a high leaf area index (LAI), the temperature deviation between the simulators reached 11.5°C (Fig. 7f). This occurred because the temperatures simulated by Edu-Clima in the LAI scenario were lower than those projected by ENVI-met. In the AEB scenario, although the Edu-Clima temperatures were lower than those in ENVI-met, the percentage variations were smaller than those in CVA (average variation of 13.27% versus 47.11%). These results reveal a greater temperature variation between the simulators for scenarios with vegetated surfaces (Fig. 7), as well as for exposed surfaces with a high leaf area index, suggesting that the Edu-Clima simulator demonstrates greater sensitivity in simulations of vegetated scenarios. However, when comparing the temperatures provided by ENVI-met and Edu-Clima within each scenario, a significant difference was observed by the paired t test in all the simulated scenarios (p < 0.05), leading to the rejection of the null hypothesis, i.e., the temperatures given by the simulators are different for the times covered. Multivariate analysis, using Principal Coordinate Analysis (PCoA), was able to summarize the variability in temperatures across all the different scenarios and times covered. All the data could be explained well by the first two axes of the PCoA: axis 1 represented 98.02% of the total variation, whereas axis 2 explained only 1.85%. The graphical visualization (Fig. 8) shows a gap between the temperature values simulated by ENVI-met and those simulated by Edu-Clima. Additionally, PCoA revealed that Edu-Clima's climate indicators were positively correlated, whereas ENVI-met's indicators tended to occupy the negative quadrants of the graph. The p value obtained from the PERMANOVA, which was applied to the PCoA results (Fig. 8), confirmed the difference visualized graphically. The analysis revealed that the tools followed a similar pattern of response to environmental variables, such as soil type and vegetation cover. The differences observed are largely a consequence of the intentional simplification of the Edu-Clima model, which focuses on ease of use and accessibility at the expense of the computational accuracy of ENVI-met. 4. DISCUSSION From a methodological point of view, Edu-Clima plays a strategic role by aligning itself with trends in digital education and climate citizenship. Recent studies, such as that by Zourmpakis et al. (2023), noted that the use of narratives and visual elements in science teaching, such as graphs, maps, and simulations, not only increases student engagement but also improves their conceptual understanding and motivation to learn. The sensitivity of the Edu-Clima app to small variations in input parameters, such as type of cover, vegetation, and time of day, allows users to experiment with different scenarios and observe their thermal implications in real time. This functionality makes it a relevant tool for active teaching strategies, such as inquiry-based teaching, problem-based learning (PBL), and gamification. This orientation is linked to SDG 4, considering the importance of methodologies and tools that favor the expansion of access to knowledge and critical training in the field of environmental education (Vivar & Peñalvo, 2023). Interactive tables and graphs play a key role in the learning process by transforming complex data into accessible and dynamic visual representations. These tools allow students to develop more sophisticated interpretive skills, going beyond simple literal reading to identify trends, patterns, and relationships in the data. Studies show that both high school students and teachers in training face challenges in adequately interpreting these representations, highlighting the need for pedagogical practices that promote the critical and effective use of tables and graphs in the classroom (Perin & Campo, 2022; Castro et al., 2024). The development of these skills contributes to more active, analytical learning that is connected to the reality of data in the contemporary world. A relevant aspect is the use of good software development practices, such as code modularization and organization by separate functions and scripts, which favor maintenance, reusability, and future improvements to the tool. This open architecture, in addition to promoting transparency, allows the educational and scientific community to contribute improvements and adaptations, transforming Edu-clima into a collaborative platform, an important aspect highlighted by Tozzi et al. (2024). The climate behavior observed in the city of Cuiabá-MT, used as the setting for the simulations, follows the same trend that was observed in Edu-clima, with higher temperatures occurring between 10:00 a.m. and 2:00 p.m. and decreasing at dusk (Callejas & Krüger, 2022). In the analysis of the coefficient of variation, the index of 30% is considered a limit; values above this value suggest high irregularity, whereas values below this value indicate greater regularity of temperature in relation to the average (Taye & Njuho, 2008). The third-order regression curve revealed that this temperature does not follow a normal pattern, unlike the simulation verified in Edu-Clima. The coefficient of determination (R²) indicates the proportion of data variation explained by the model, which, through the regression equation produced, was able to represent the behavior of the temperature observations during the analyzed period (Chicco et al., 2021). The equations extracted from the models can make temperature predictions at times not explored in the different simulations, which would help fill gaps in the data series. By analyzing the regressions for different scenarios, we confirmed that time (time of day) strongly influenced the temperature under both the Edu-Clima application and ENVI-met. This significant result corroborates what the scientific literature already points out: the time of day is a determining factor for thermal variations (Kou et al., 2022). This influence becomes even more evident in our results, given the wide temperature range recorded in the simulations of both tools, reinforcing the direct impact of the passage of time on the temperature behavior in all the scenarios. The ranges verified in the simulations are also within the normal range of climatological records for the region of Cuiabá, Mato Grosso (Machado & Gonzalez, 2021; Carmo, Reboita et al., 2023). Importantly, although the PERMANOVA p value (0.00010) points to statistically significant differences between the temperatures simulated by Edu-Clima and ENVI-met, this distinction does not invalidate the quality of the application for educational purposes. The analysis showed that the tools follow a similar standard in response to environmental variables, such as soil type and vegetation cover. The differences observed are largely a consequence of the intentional simplification of the Edu-Clima model, which focuses on ease of use and accessibility at the expense of the computational accuracy of ENVI-met. Therefore, the PERMANOVA results reinforce that, for teaching purposes, the Edu-Clima model is reliable, as it reproduces the trends and fundamental concepts of urban climatology in a clear and interactive way (Belouafa, 2017). 5. CONCLUSION Although ENVI-met has a robust database and is one of the main climate simulation tools used, the simulator created for teaching purposes has a simple and intuitive interface for designing basic climate parameters, such as average temperature, minimum temperature, or their progression over time. The analyses also highlighted greater deviations in scenarios with dense vegetation, especially when combined with vegetated soils, as in the case of the CVA (Concrete, Vegetated, High LAI) scenario. These deviations of more than 11°C and percentage variations close to 47% indicate that Edu-Clima does not yet accurately incorporate the effects of evapotranspiration, natural shading, and soil moisture, aspects that are better modeled by ENVI-met. This limitation, however, is understandable and acceptable within the scope of a tool built with open-source code, fed thus far with simplified data for educational use. Finally, the results suggest that Edu-clima fulfills its purpose of being a reliable, accessible, and functional educational simulator. The analysis also shows that the tool can be understood as a contribution aligned with SDG 4, as it favors educational practices that integrate technological innovation and training focused on sustainability. Although its limitations in relation to vegetation modeling are evident, these limitations do not compromise its validity as a teaching tool. In contrast, they reveal points for future development, allowing the tool to advance in robustness without losing its didactic and intuitive essence. STATEMENTS AND DECLARATIONS Ethical Approval : Not applicable. Informed Consent : Not applicable. Statement Regarding Research Involving Human Participants and/or Animals : Not applicable. Consent to Participate : Informed consent was obtained from all individual participants included in the study. Consent to Publish : All authors read and approved the final manuscript. Funding: This work was funded by the authors themselves. Competing Interests : The authors declare no competing interests. References Almazroui, M., Abid, M. A., Ashfaq, M., Islam, M. N., Kamil, S., Rashid, I. U., ... & Sylla, M. B. (2021). Assessment of CMIP6 performance and projected temperature and precipitation changes over South America. Earth Systems and Environment , 5(2), 155-183. Aslam, M. (2021). Retracted Article: Neutrosophic Statistical Test For Counts in Climatology. Scientific Reports , 11(1), 17806. Belouafa, S., Belafkih, B., Benhar, S., Habti, F., Hamdouch, S., Tayane, S., ... & Abourriche, A. (2017). Ferramentas e abordagens estatísticas para validar métodos analíticos: metodologia e exemplos práticos. Revista Internacional de Metrologia e Engenharia da Qualidade , 8, 9. Callejas, I. J. A., & Krüger, E. (2022). Microclimate and thermal perception in courtyards located in a tropical savannah climate. International journal of biometeorology , 66(9), 1877–1890. Camps-Valls, G., Castelletti, A., Cohrs, K. H., Fernández-Torres, M. Á., Höhl, A., Pacal, A., ... & Williams, T. (2025). Artificial intelligence for modeling and understanding extreme weather and climate events. Nature Communications , 16(1), 1919. Carmo, E. L. I., Reboita, M. S., & Marques, R. (2023). Evolução Temporal das Variáveis Atmosféricas associadas a Casos de Frentes Frias Fortes em Cuiabá, MT, entre 1996 e 2015 . Revista Brasileira de Geografia Física , 16(1), 145-154. Castro, E. R., Barreto, M. C., Nascimento, F. J., & Sousa, G. L. (2024). Construction and Interpretation of Statistical Tables and Graphs: A Look at the Attitude of Teachers in Training. Teaching in Review , 31, 1–24. Chicco, D., Warrens, M. J., & Jurman, G. (2021). The Coefficient Of Determination R-Squared Is More Informative than SMAPE, MAE, MAPE, MSE and RMSE in Regression analysis evaluation. Peerj computer science , 7, e623. Kim, D. Y., & Kwon, O. (2025). How Open Source Artificial Intelligence Democratizes and Improves Science: Evidence from Structural Biology. Available at SSRN 5219337 . Kou, Y., Chen, J., Cheng, B., Liu, Y., Wang, C., Xian, D., … & Tang, L. (2022). Factors affecting urban climate at different times of the day in China: a case study in Yibin, a riverside mountain city. Nature-Based Solutions , 2, 100043. Kurokawa, H., Igei, K., Kitsuki, A., Kurita, K., Managi, S., Nakamuro, M., … & Sakano, A. (2023). Improvement Impact Of Nudges Incorporated In Environmental Education On Students Environmental Knowledge, Attitudes, And Behaviors. Journal of Environmental Management , 325, 116612. Lozano-Díaz, A., & Fernández-Prados, J., S. (2020). Educating digital citizens: An opportunity to critical and activist perspective of sustainable development goals. Sustainability , 12(18), 7260. Machado, C. S. D., & Gonzalez, A. Z. D. (2021). Climate Variability in the Cerrado Biome of Mato Grosso During the Last Decades. Revista Equador , 10(2), 207–227. Makvandi, M., Chai, H., Fu, J., Khodabakhshi, Z., Li, W., Ou, X., ... & Horimbere, E. D. L. J. (2023). Mitigation of urban heat for climate change adaptation: an eco-sustainable design strategy to improve environmental performance amid rapid urbanization. Atmosphere , 14(4), 638. Maranholi, H. N. G., & Santos, F. M. M. (2024). O Método Científico e a Compreensão das Variáveis Ambientais no Contexto Educacional: Uma Revisão. Cadernos Cajuína , 9(5), e249525. Pacifici, M., & Nieto-Tolosa, M. (2021). Comparing ENVI-Met and grasshopper modeling strategies to assess local thermal stress and urban heat island effects. In Urban microclimate modeling for comfort and energy studies (pp. 293-316). Cham: Springer International Publishing. Perin, A. P., & Campo, C. R. (2022). Leitura e interpretação de gráficos estatísticos por alunos do 2º ano do ensino médio. Revista Baiana de Educação Matemática , 3(01), e202227. Riuttanen, L., Ruuskanen, T., Äijälä, M., & Lauri, A. (2021). Society needs experts with climate change competencies–what is the role of higher education in atmospheric and Earth system sciences?. Tellus B: Chemical and Physical Meteorology , 73(1), 1–14. Streb, V., Ferreira, M. D., Gomes, A. F., Reginatto, A. A., Siqueira Cecchin, A. de, & da Rocha, KM (2021). Teoria x Prática: Panorama inicial sobre a inserção das Tecnologias Digitais no Ensino Superior presencial e a distância na UFSM. Revista Brasileira de Desenvolvimento , 7 (4), 41318-41331. Taye, G., & Njuho, P. (2008). Monitoring field variability using confidence interval for coefficient of variation. Communications in Statistics—Theory and Methods , 37(6), 831–846. Tozzi, C. C., de Oliveira, I. D. S. B., Bonicenha, L. C., Campanin, M. A. A., Dona, R. A. M., Onofre, V., & Andreza, W. G. G. (2024). Mídias digitais na educação online: o impacto da linguagem audiovisual e ferramentas colaborativas. Revista Ibero-Americana de Humanidades, Ciências e Educação , 10(10), 3723-3729. Trane, M., Giovanardi, M., Pejovic, A., & Pollo, R. (2023). Visão geral sobre ferramentas de modelagem de clima e microclima urbano e seu papel para alcançar os Objetivos de Desenvolvimento Sustentável. In: Arquitetura e Design para a Indústria 4.0: Teoria e Prática (pp. 247-267). Cham: Springer International Publishing. Twardosz, R., Walanus, A., & Guzik, I. (2021). Warming in Europe: recent trends in annual and seasonal temperatures. Pure and Applied Geophysics , 178(10), 4021-4032. Vivar, J. M. F., & Peñalvo, F. J. G. (2023). Reflections on the ethics, potential, and challenges of Artificial Intelligence in the context of Quality Education (SDG 4). Comunicar: Revista científica de comunicación y educación , (74), 37-47. Waskom, M. L. (2021). Seaborn: statistical data visualization. Journal of Open Source Software , 6(60), 3021. Zourmpakis, A. I., Kalogiannakis, M., & Papadakis, S. (2023). Adaptive Gamification In Science Education: An Analysis Of The Impact Of Implementation And Adapted Game Elements On Students’ Motivation. Computers , 12(7), 143. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 14 Apr, 2026 Read the published version in Nativa → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7743068","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":523652461,"identity":"b95254ea-b7b2-47e2-abac-01db460e6888","order_by":0,"name":"Henrique Nicolau Grillaud Maranholi","email":"","orcid":"","institution":"Universidade Federal de Mato Grosso","correspondingAuthor":false,"prefix":"","firstName":"Henrique","middleName":"Nicolau Grillaud","lastName":"Maranholi","suffix":""},{"id":523652462,"identity":"e6040b86-f152-4623-bd82-5e547a0e3beb","order_by":1,"name":"Flávia Maria de Moura Santos","email":"","orcid":"","institution":"Universidade Federal de Mato Grosso","correspondingAuthor":false,"prefix":"","firstName":"Flávia","middleName":"Maria de Moura","lastName":"Santos","suffix":""},{"id":523652463,"identity":"888dd7db-585e-4b3f-b457-b0e99ff01af0","order_by":2,"name":"Victor Hugo Maranholi","email":"","orcid":"","institution":"Universidade Federal de Mato Grosso","correspondingAuthor":false,"prefix":"","firstName":"Victor","middleName":"Hugo","lastName":"Maranholi","suffix":""},{"id":523652464,"identity":"982ab6a9-2dcd-471a-b7de-ef793327521a","order_by":3,"name":"Wennder Tharso Oliveira da Silva Martins","email":"","orcid":"","institution":"Universidade Federal de Mato Grosso","correspondingAuthor":false,"prefix":"","firstName":"Wennder","middleName":"Tharso Oliveira da Silva","lastName":"Martins","suffix":""},{"id":523652465,"identity":"78501e43-7a3d-4f4a-9c62-07c59159ce1e","order_by":4,"name":"Wellington Fava Roque-Maranholi","email":"data:image/png;base64,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","orcid":"","institution":"Universidade Federal de Mato Grosso","correspondingAuthor":true,"prefix":"","firstName":"Wellington","middleName":"Fava","lastName":"Roque-Maranholi","suffix":""}],"badges":[],"createdAt":"2025-09-29 14:23:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7743068/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7743068/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.31413/nat.v14i2.20963","type":"published","date":"2026-04-15T00:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":97421129,"identity":"713f8adc-9ce7-4150-9a06-8bbf1f80e05a","added_by":"auto","created_at":"2025-12-04 08:33:53","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":449345,"visible":true,"origin":"","legend":"\u003cp\u003eFront-end of the Edu-clima application.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/6294a8d91f204799da0729ee.png"},{"id":97666825,"identity":"192c38c5-b5b5-436d-83ed-52d889d264b2","added_by":"auto","created_at":"2025-12-08 09:22:11","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":149975,"visible":true,"origin":"","legend":"\u003cp\u003eView of the table of neighborhoods in the city of Cuiabá/MT\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/ddc187d353885a173e70addc.png"},{"id":97421128,"identity":"aff3433d-29f6-4f9f-9d02-eeed12baf74e","added_by":"auto","created_at":"2025-12-04 08:33:53","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":21545,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal distribution of temperature on paved urban surfaces with a low leaf area index, represented by the real scenario\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/d131c30090bcd026854cff4a.png"},{"id":97421133,"identity":"cecbcda1-ea12-480b-9992-b33aa3176c20","added_by":"auto","created_at":"2025-12-04 08:33:53","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":146094,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal distribution of temperature on paved urban surfaces with different surfaces and leaf area indices\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/3ed48b6a9774d13dca647a3b.png"},{"id":97666341,"identity":"a8cf198e-bdb7-4586-b171-01a8aa9d3ae2","added_by":"auto","created_at":"2025-12-08 09:21:02","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":148654,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal distribution of temperature on concrete urban surfaces with different surfaces and leaf area indices\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/2c4025b72f1f90da24427005.png"},{"id":97421131,"identity":"fde78077-45f3-49b6-83cc-7ccef963606f","added_by":"auto","created_at":"2025-12-04 08:33:53","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":341460,"visible":true,"origin":"","legend":"\u003cp\u003eComparative tables of climatological metrics and statistical analysis of nonvegetated surfaces with different types of cover and leaf area indices. A p value (p) less than 0.05 (p=0.05) indicates a statistically significant difference. \"N\" is equal to the sample size.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/b62478304dc1f6e021bf4be2.png"},{"id":97421134,"identity":"235a2bba-8bf5-4374-ad70-6cd5a73159d8","added_by":"auto","created_at":"2025-12-04 08:33:53","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":335181,"visible":true,"origin":"","legend":"\u003cp\u003eComparative tables of climatological metrics and statistical analysis of vegetated surfaces under different soil cover types and leaf area indices. A p value (p) less than 0.05 (p=0.05) indicates a statistically significant difference. \"N\" is equal to the sample size.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/57c95b9e9bbabba6688c9b02.png"},{"id":97667267,"identity":"9b96c39b-de1e-4a43-9428-7fde204c9d12","added_by":"auto","created_at":"2025-12-08 09:23:10","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":56746,"visible":true,"origin":"","legend":"\u003cp\u003ePrincipal Coordinate Analysis (PCoA) of the distributions of simulation scenarios between Envi-Met and Edu-Clima according to the extracted climatological variables\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/a30e90862dd786472bb9282c.png"},{"id":107437501,"identity":"d16539ff-f057-4d28-8fc1-3ab62238bbbf","added_by":"auto","created_at":"2026-04-21 13:32:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1837660,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7743068/v1/93f005d2-b3c2-4233-aa81-96f94556702a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Edu-Clima: A Didactic Tool for the Simulation and Modeling of Urban Climate","fulltext":[{"header":"1.\tINTRODUCTION","content":"\u003cp\u003eThe increasing degree of urbanization and the environmental challenges associated with city densification have intensified interest in tools that help understand and mitigate the impacts of the urban microclimate (Trane et al., 2023). Among the most relevant elements in this scenario are surface temperature, vegetation cover, and the types of materials used in urban soil, which are closely linked to the quality of life in urban areas and the thermal comfort of their inhabitants (Makvandi et al., 2023).\u003c/p\u003e\n\u003cp\u003eIn contemporary scenarios, which are characterized by extreme weather events, urban heat islands, and growing concern for environmental sustainability, understanding climatology has become an essential skill not only for specialists but also for public managers, educators, and citizens in general (Riuttanen et al., 2021). The ability to interpret and predict microclimatic patterns is fundamental for resilient urban planning and more informed decision-making (Maranholi \u0026amp; Santos, 2024).\u003c/p\u003e\n\u003cp\u003eIn this context, computational climate simulators have established themselves as essential tools for analyzing and predicting environmental variables (Camps-Valls et al., 2025). However, although robust platforms such as ENVI-met offer high precision and detail, their complexity and cost may limit their use in educational or preliminary planning contexts (Pacifici \u0026amp;amp; Nieto-Tolosa, 2021). Given this, this research proposed the development of a proprietary application with an accessible interface and educational purpose for simulating surface temperature in different urban scenarios.\u003c/p\u003e\n\u003cp\u003eThe development of open-source applications that are easy to understand and use is essential to democratize access to these tools (Pacifici \u0026amp; Nieto-Tolosa, 2021; Kim \u0026amp; Kwon, 2025). When made available as open sources, such applications offer the opportunity for adaptation and customization, favoring their use not only by professionals but also by students and educators seeking accessible learning tools (Streb et al, 2021).\u003c/p\u003e\n\u003cp\u003eIn this sense, the development of the Edu-Clima application emerges as a strategic tool, enabling the simulation of different urban climate conditions in an accessible, didactic, and interactive way. By allowing users to explore variables such as surface temperature, vegetation cover, and soil types, the application not only facilitates the learning of complex concepts but also promotes awareness of the effects of urban choices on the local thermal environment, contributing to a more critical and engaged citizenry with respect to climate issues (Lozano-Díaz \u0026amp; Fernández-Prados, 2020; Kurokawa et al, 2023).\u003c/p\u003e\n\u003cp\u003eThis pedagogical character is directly linked to SDG 4 – Quality Education, as Edu-Clima presents itself as an innovative resource for expanding access to inclusive and equitable educational practices. By transforming technical data into easy-to-understand interactive experiences, the application strengthens the teaching of urban climatology, stimulates active learning, and promotes lifelong learning opportunities, both in school environments and in nonformal education spaces. Thus, the project contributes to the development of essential skills to address the challenges of sustainability and urban resilience.\u003c/p\u003e\n\u003cp\u003eIn this context, the present study investigated the feasibility of Edu-Clima as a reliable teaching tool for simulating urban surface temperature. Our hypothesis is that there is no statistically significant difference in the surface temperature results simulated by the Edu-Clima application and those obtained by the ENVI-met software.\u003c/p\u003e\n\u003cp\u003eAs a procedural premise, we sought to verify its reliability using statistical methods and multivariate analyses. The investigation thus aimed to contribute both to the improvement of teaching methodologies in urban climatology and to the democratization of digital tools applied to the sustainability of cities.\u003c/p\u003e\n\u003cp\u003eTherefore, our goal was to develop and validate an open-source urban surface temperature simulation application by comparing its results with those obtained by the ENVI-met software and evaluating its reliability for application in the teaching of urban climatology.\u003c/p\u003e\n\u003cp\u003eThus, this paper is structured to detail the development and validation of the Edu-Clima application. The Methodology section describes the steps involved in building the application in Python, as well as the strategy for comparing the results with the ENVI-met software, including the variables and the 12 simulation scenarios. Next, the Results section presents statistical analyses, such as quadratic regression and PERMANOVA, which validate the reliability of Edu-Clima alongside ENVI-met. In the Discussion section, we further explore the positive points and limitations, especially in the scenarios analyzed. Finally, the conclusion summarizes the main findings, reinforces the application's contribution to the teaching of urban climatology, and points to avenues for future studies.\u003c/p\u003e"},{"header":"2.\tMATERIALS AND METHODS","content":"\u003cp\u003eFor the development of this research, an application was built to simulate different scenarios with different temperatures, types of ground cover, and leaf areas. To obtain more reliable results, scenarios were created via \u003cem\u003eENVI-met\u003c/em\u003e\u003csup\u003e\u003csup\u003e[1]\u003c/sup\u003e\u003c/sup\u003e, a three-dimensional microclimate simulation software used to model urban environments and assess the impacts of factors such as morphology, vegetation, and building materials on the local climate. This approach aimed to ensure the greatest possible similarity with the scenarios plotted in the developed application.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1. \u003c/strong\u003eApplication development\u003c/p\u003e\n\u003cp\u003eThe application was developed in \u003cem\u003ePython\u003csup\u003e\u003cstrong\u003e\u003csup\u003e[2]\u003c/sup\u003e\u003c/strong\u003e\u003c/sup\u003e\u003c/em\u003e, with the support of several libraries and complementary technologies (Waskom, 2021). The analysis included an in-depth discussion of the technological resources employed, the methodology adopted during implementation, and the results obtained at the end of the process. To this end, a tool was created and implemented to compare temperatures provided by the \u003cem\u003eOpenWeather\u003c/em\u003e \u003cem\u003eAPI\u003csup\u003e\u003cstrong\u003e\u003csup\u003e[3]\u003c/sup\u003e\u003c/strong\u003e\u003c/sup\u003e\u003c/em\u003e and data recorded at specific geographic coordinates while also considering the associated climatic conditions.\u003c/p\u003e\n\u003cp\u003eProgram Operation\u003c/p\u003e\n\u003cp\u003eThe program uses the \u003cem\u003eStreamlit\u003c/em\u003e \u003cem\u003elibrary\u003csup\u003e\u003cstrong\u003e\u003csup\u003e[4]\u003c/sup\u003e\u003c/strong\u003e\u003c/sup\u003e\u003c/em\u003e to create an interactive web interface. The program workflow can be described in several steps:\u003c/p\u003e\n\u003cp\u003ea) Data Collection: The program queries the \u003cem\u003eOpenWeather\u003c/em\u003e \u003cem\u003eAPI\u003c/em\u003e to obtain temperature data for a specific coordinate;\u003c/p\u003e\n\u003cp\u003eb) Local Data Reading: Reading of locally recorded temperature data from CSV files using Pandas\u003csup\u003e\u003csup\u003e[5]\u003c/sup\u003e\u003c/sup\u003e;\u003c/p\u003e\n\u003cp\u003ec) Data Processing: The data are cleaned and processed to remove outliers and fill in missing values;\u003c/p\u003e\n\u003cp\u003ed) Comparison and Analysis: API data and local data on temperature, leaf area, and type of ground cover (asphalt or concrete) and ground conditions (exposed or vegetated) are compared and analyzed;\u003c/p\u003e\n\u003cp\u003ee) Visualization: The results are visualized in interactive graphs via \u003cem\u003eMatplotlib\u003csup\u003e\u003cstrong\u003e\u003csup\u003e[6]\u003c/sup\u003e\u003c/strong\u003e\u003c/sup\u003e, Seaborn\u003csup\u003e\u003cstrong\u003e\u003csup\u003e[7]\u003c/sup\u003e\u003c/strong\u003e\u003c/sup\u003e\u003c/em\u003e, and \u003cem\u003ePlotly. Streamlit\u003c/em\u003e\u003csup\u003e\u003csup\u003e[8]\u003c/sup\u003e\u003c/sup\u003e, which facilitates visualization in the web interface.\u003c/p\u003e\n\u003cp\u003eSource Code\u003c/p\u003e\n\u003cp\u003eThe source code has been organized into modules to facilitate program maintenance and scalability. Thus, they will be presented below using \u0026quot;\u0026quot; (quotation marks). The main components include the following:\u003c/p\u003e\n\u003cp\u003ea) inicio.py: Main file that initializes the \u003cem\u003eStreamlit\u003c/em\u003e application and manages the user interface.\u003c/p\u003e\n\u003cp\u003eb) The screen rendering uses the \u0026ldquo;columns\u0026rdquo; feature of the \u003cem\u003eStreamlit\u003c/em\u003e library to divide the interface into three columns.\u003c/p\u003e\n\u003cp\u003ec) The \u0026ldquo;center_column\u0026rdquo; renders \u0026ldquo;st.components.v1.html\u0026rdquo; \u0026ndash; \u003cem\u003eStreamlit \u003c/em\u003euses the \u0026ldquo;components\u0026rdquo; function to transform the return of the \u0026ldquo;heatmap\u0026rdquo; function into an HTML file format understood by the browser;\u003c/p\u003e\n\u003cp\u003ed) The \u0026quot;right\u0026quot; column rendered after the \u0026quot;center_column\u0026quot; call, shown above, contains the st.dataframe function (a method to inform \u003cem\u003eStreamlit \u003c/em\u003ethat a Dataframe will be rendered there).\u003c/p\u003e\n\u003cp\u003ee) Next to the previous column is the \u0026quot;left\u0026quot; screen division, which complements the previous right one by rendering the bar chart, called in the chart by the function \u0026quot;st.plotly_chart(fig)\u0026quot;, where: Fig: contains the \u0026quot;figure\u0026quot; of the rendered chart; St.plotly_chart is the \u003cem\u003estreamlit\u003c/em\u003e function that renders the chart figure within the left division of the screen.\u003c/p\u003e\n\u003cp\u003ef) sidebar.py: File containing the functions responsible for interacting with the OpenWeather API, generating the heatmap, and rendering the tool\u0026apos;s side menu.\u003c/p\u003e\n\u003cp\u003eg) Obter_temperatura_openweathermap: function that receives the latitude and longitude parameters to query the \u003cem\u003eOpenWeather\u003c/em\u003e API, using the key obtained directly from their website, contained in the \u0026quot;api_key\u0026quot; object, for the climate information relevant to the scope of the project.\u003c/p\u003e\n\u003cp\u003eh) The API returns a \u003cem\u003eJSON\u003c/em\u003e file full of information, from which we extract the data we need. This functionality is located within the \u0026quot;try except\u0026quot; of the function.\u003c/p\u003e\n\u003cp\u003ei) Folium: Library used to render the geographic map. In the code, it received the alias \u0026quot;fo.\u0026quot;\u003c/p\u003e\n\u003cp\u003ej) fo. Map \u0026ndash; receives the parameter of the type of lines to be rendered on the map, and a zoom_start value of 13 \u0026ndash; indicates that the map should render the coordinates with the zoom level appropriate for the search.\u003c/p\u003e\n\u003cp\u003ek) fo. CircleMarker \u0026ndash; renders the border of the circles that represent the temperature of the rendered region. The rendering of all circles that appear in the tool is carried out within the \u0026ldquo;for point-in-points\u0026rdquo; loop, where the search coordinates are queried in the API.\u003c/p\u003e\n\u003cp\u003el) Sidebar_dados_pesquisa \u0026ndash; this function renders, in a side menu, all the fields in which the user enters temperature collection information. When invoked by the main.py file, it returns latitude_pesquisa, longitude_pesquisa, temperatura_final, and temperatura_sem_foliar. It consists of the following functions, among others: Latitude_search, a function that renders the latitude input field in the side menu. If the user does not enter a value, it will use the default value -15.612642354149061 to pass information to the API; Search_longitude: similar to the previous function, with a default value of -56.038829296302545 if no information is entered by the user; Search_temperature: renders the temperature input field collected by the user. By default, if the field is not filled in, it will contain the value obtained by the geographical coordinates above; Selected_vegetation: renders the options as a menu with \u0026apos;high\u0026apos;, \u0026apos;medium\u0026apos;, \u0026apos;low\u0026apos;, and \u0026apos;manual\u0026apos; for user selection (in this case, an information entry field will be provided to the user); Percentage_ground_cover: renders the options \u0026apos;asphalt\u0026apos;, \u0026apos;concrete\u0026apos;, and \u0026apos;manual\u0026apos; for user selection. The behavior of the \u0026apos;manual\u0026apos; selection is the same as that of the previous selection.\u003c/p\u003e\n\u003cp\u003em) dataframe.py: This file contains the functions \u0026quot;dataframe\u0026quot; and \u0026quot;plot_graph,\u0026quot; which return \u0026quot;df\u0026quot; and \u0026quot;fig,\u0026quot;, respectively.\u003c/p\u003e\n\u003cp\u003en) dataframe: receives the arguments \u0026quot;search_latitude,\u0026quot; \u0026quot;search_longitude,\u0026quot; \u0026quot;search_temperature,\u0026quot; and \u0026quot;regions\u0026quot; to render the dataframe displayed in the tool. The function uses Pandas to perform the operations.\u003c/p\u003e\n\u003cp\u003eo) plot_graph: receives \u0026quot;df\u0026quot; as an argument and uses the Plotly library through the Express module to create the bar chart displayed in the tool.\u003c/p\u003e\n\u003cp\u003ep) page_especs.py: file containing the page_config function.\u003c/p\u003e\n\u003cp\u003eq) page_config: Streamlit is used, through the alias \u0026quot;st,\u0026quot; to use the \u0026quot;set_page_config\u0026quot; method to inform the browser of interface details, which are page_title; page_icon; layout.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2. \u003c/strong\u003eParameters for scenario production: ENVI-met \u0026times; Edu-Clima\u003c/p\u003e\n\u003cp\u003eFor the simulations to be as similar as possible, the input parameters were defined for both ENVI-met and the developed application. The following elements were used to construct the scenarios: surface temperature, leaf area index, ground cover (asphalt or concrete), and soil type (vegetated or exposed). Thus, for temperature, the times and values were described (Table 1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1 \u003c/strong\u003eTimes and values arbitrarily assigned for the simulations\u003c/p\u003e\n\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"311\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e00:00 hours - 20.98\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e1:00 p.m. - 26.03\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e02:00 hours - 20.03\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e2:00 p.m. - 25.98\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e3:00 a.m. - 19.28\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e3:00 p.m. - 26.48\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e4:00 a.m. - 18.71\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e4:00 p.m. - 26.35\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e5:00 a.m. - 18.09\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e5:00 p.m. - 25.19\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e6:00 a.m. - 18.33\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e6:00 p.m. - 24.84\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e7:00 a.m. - 19.01\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e7:00 p.m. - 24.25\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e8:00 a.m. - 20.47\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e8:00 p.m. - 23.73\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e9:00 a.m. - 22.20\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e9:00 p.m. - 23.05\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e10:00 a.m. - 23.60\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e10:00 p.m. - 22.44\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e11:00 a.m. - 24.66\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e11:00 p.m. - 21.82\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \u003cp\u003e12:00 p.m. - 24.96\u0026deg;C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 155px;\"\u003e\n \n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\n\u003cp\u003eFor the leaf area index, three different levels were established: high LAI (leaf area index), where an estimated canopy cover of greater than 90% was assigned; medium LAI, where an estimated canopy cover of approximately 50% was assigned; and low LAI, where an estimated canopy cover value of less than 15% was assigned.\u003c/p\u003e\n\u003cp\u003eFor the soil cover layer, two types of materials were considered: concrete and asphalt. For these materials, although both ENVI-met and Edu-clima allow for different percentages, a constant soil cover percentage of 50% was established.\u003c/p\u003e\n\u003cp\u003eRegarding the type of soil used to produce the scenarios, two types were defined: exposed soil, where there is no type of construction material, or even natural cover. The other type of soil defined was vegetated soil, where natural grass cover was considered. The parameters can be seen in Table 2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2 \u003c/strong\u003eParameters for constructing simulation scenarios\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"595\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003eLAI \u003csup\u003e1\u003c/sup\u003e \u0026ndash; High\u003c/p\u003e\n \u003cp\u003eSoil cover \u0026ndash; asphalt (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; exposed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003eLAI \u0026ndash; Medium\u003c/p\u003e\n \u003cp\u003eSoil cover \u0026ndash; asphalt (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; exposed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 199px;\"\u003e\n \u003cp\u003eLAI \u0026ndash; Low\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; asphalt (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; exposed\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003eLAI \u0026ndash; High\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; concrete (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; exposed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLAI \u0026ndash; Medium\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; Concrete (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; exposed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 199px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLAI \u0026ndash; Low\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; Concrete (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; exposed\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003eLAI \u0026ndash; High\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; asphalt (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; Vegetated\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLAI \u0026ndash; Medium\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; asphalt (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; Vegetated\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 199px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLAI \u0026ndash; Low\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; asphalt (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; Vegetated\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003eLAI \u0026ndash; High\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; concrete (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; Vegetated\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLAI \u0026ndash; Medium\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; Concrete (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; Vegetated\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 199px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLAI \u0026ndash; Low\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGround cover \u0026ndash; Concrete (50%)\u003c/p\u003e\n \u003cp\u003eSoil type \u0026ndash; Vegetated\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003csup\u003e1\u003c/sup\u003e LAI \u0026ndash; Leaf Area Index\u003c/p\u003e\n\u003cp\u003eFinally, after the parameters and constituent elements of the scenarios were defined, a total of 12 different configurations were established. For each scenario, 23 simulations were performed, totaling 276 simulations. This set was initially run in ENVI-met software, considering the real scenario, and later reproduced in Edu-Clima, resulting in another 276 simulations.\u003c/p\u003e\n\u003cp\u003eFor nomenclature purposes, each scenario was named according to its arbitrary characteristics:\u003c/p\u003e\n\u003cp\u003ea) AEB \u0026ndash; Asphalt with exposed surface and low leaf area index;\u003c/p\u003e\n\u003cp\u003eb) AEM \u0026ndash; Asphalt ground cover with exposed surface and medium leaf area index;\u003c/p\u003e\n\u003cp\u003ec) AEA \u0026ndash; Asphalt ground cover with exposed surface and high leaf area index;\u003c/p\u003e\n\u003cp\u003ed) AVB \u0026ndash; Asphalt ground cover with a vegetated surface and low leaf area index;\u003c/p\u003e\n\u003cp\u003ee) AVM \u0026ndash; Vegetated asphalt surface ground cover and medium leaf area index;\u003c/p\u003e\n\u003cp\u003ef) AVA \u0026ndash; Vegetated asphalt surface with a high leaf area index;\u003c/p\u003e\n\u003cp\u003eg) CEB \u0026ndash; Concrete ground cover with exposed surface and low leaf area index;\u003c/p\u003e\n\u003cp\u003eh) CEM \u0026ndash; Concrete ground cover with exposed surface and medium leaf area index;\u003c/p\u003e\n\u003cp\u003ei) CEA \u0026ndash; Concrete ground cover with exposed surface and high leaf area index;\u003c/p\u003e\n\u003cp\u003ej) CVB \u0026ndash; Concrete ground cover with a vegetated surface and low leaf area index;\u003c/p\u003e\n\u003cp\u003ek) CVM \u0026ndash; Concrete ground cover with vegetated surface and medium leaf area index;\u003c/p\u003e\n\u003cp\u003el) CVA \u0026ndash; Concrete soil cover with a vegetated surface and high leaf area index.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3. \u003c/strong\u003eData exploration and statistical analysis\u003c/p\u003e\n\u003cp\u003eTo explore the results, the deviation (\u0026deg;C) was calculated from the difference between the simulated temperatures in ENVI-met and Edu-Clima to verify the existence of climatological anomalies. The temperature variation (%) between the simulators was calculated by dividing the deviation value (\u0026deg;C) by the average temperature of the simulators (Almazroui et al., 2021; Twardosz et al., 2021). The coefficient of variation was calculated by dividing the standard deviation of temperatures throughout the day by their average. We applied the paired t test to compare the average temperatures between the two simulators to determine whether there was a significant difference between them in the different scenarios evaluated (Aslam, 2021).\u003c/p\u003e\n\u003cp\u003eTo verify whether time (X-axis) influenced temperature (Y-axis) in both simulators for the different scenarios, cubic regression analysis was applied, estimated by the least squares method using Minitab 19 statistical software, with an analysis of variance at a 5% probability of rejecting the null hypothesis (Aslam, 2021). Multivariate statistics were also explored, with principal component analysis (PCoA) based on Euclidean distances, followed by PERMANOVA to verify whether the observations made between the simulators behaved differently.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[1]\u003c/sup\u003e\u003c/sup\u003e You can find more information and download the free version of this software on its official website: https://envi-met.com/.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[2]\u003c/sup\u003e\u003c/sup\u003e Python is an interpreted, high-level, dynamically typed programming language. It is widely used due to its simplicity and robustness, facilitating rapid application development. Its vast collection of libraries makes it ideal for scientific and data analysis projects. Python Software Foundation (PSF) license, which is permissive and allows commercial use.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[3]\u003c/sup\u003e\u003c/sup\u003e The OpenWeather API provides access to global weather data, including forecasts for temperature, humidity, precipitation, and more. The API is often used in applications that require accurate and up-to-date weather data. Available under different subscription plans, including free and paid options.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[4]\u003c/sup\u003e\u003c/sup\u003e Streamlit is a Python library that allows you to quickly and easily create interactive web applications. It is particularly useful for data visualization and dashboards, making it a popular choice among data scientists. Apache 2.0 license, a permissive license that allows use, modification, and distribution.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[5]\u003c/sup\u003e\u003c/sup\u003e Pandas is a software library written for the Python language for data manipulation and analysis. It provides data structures and operations for manipulating numerical tables and time series.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[6]\u003c/sup\u003e\u003c/sup\u003e A 2D plotting library in \u003cem\u003ePython \u003c/em\u003ethat produces high-quality figures in a variety of formats and environments.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[7]\u003c/sup\u003e\u003c/sup\u003e Based on \u003cem\u003eMatplotlib, Seaborn \u003c/em\u003eprovides a high-level interface for creating attractive statistical graphics.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u003csup\u003e[8]\u003c/sup\u003e\u003c/sup\u003e An interactive graphics library that allows the creation of complex visualizations with zoom, \u003cem\u003epan\u003c/em\u003e, and \u003cem\u003ehover \u003c/em\u003ecapabilities\u003cem\u003e.\u003c/em\u003e\u003c/p\u003e"},{"header":"3.\tRESULTS","content":"\u003cp\u003e\u003cstrong\u003e3.1\u0026nbsp;\u0026nbsp;\u003c/strong\u003eEdu-clima\u003c/p\u003e\n\u003cp\u003eThe result of the application development process, named Edu-clima (accessible at https://simuladorclimaticodoutoradomaranholi.streamlit.app/), consisted of the creation of a functional, practical, interactive, and educational application aimed at simulating surface temperature in different urban scenarios (Fig. 1). The design was intended to be simple and understandable for the user.\u003c/p\u003e\n\u003cp\u003eThe central goal was to transform technical and abstract content, such as thermal behavior in urban areas, into an interactive and accessible experience capable of engaging students and educators at different levels of education.\u003c/p\u003e\n\u003cp\u003eUnlike robust simulators such as ENVI-met, whose use requires advanced technical knowledge and specific computational infrastructure, Edu-Clima was designed on three main premises: simplified usability, didactic visualization, and integration with open data.\u003c/p\u003e\n\u003cp\u003eEdu-Clima sensitivity to small variations in input parameters such as type of cover, vegetation, and time of day\u0026mdash;allows users to experiment with scenarios and observe their thermal implications, which enhances engagement in pedagogical strategies such as investigative teaching, problem-based learning (PBL), and gamification.\u003c/p\u003e\n\u003cp\u003eAnother carefully designed feature was the presentation of a table with temperatures in neighborhoods in the city of Cuiab\u0026aacute;, Mato Grosso (Fig. 2), where the person using the app can immediately make comparisons with different locations in the city.\u003c/p\u003e\n\u003cp\u003eAlthough we used the coordinates of Cuiab\u0026aacute;, Brazil, as the basis for the app\u0026apos;s simulations and validations, it is important to note that the system is not limited to this location. Edu-Clima was developed with a flexible architecture, allowing users to enter any geographic coordinates to simulate the microclimate of any urban area in the world if the user adheres to the characteristic metric of the local climate.\u003c/p\u003e\n\u003cp\u003eThe interface based on the \u003cem\u003eStreamlit\u003c/em\u003e library, for example, allows for fluid and intuitive navigation, whereas the use of libraries such as \u003cem\u003eFolium, Plotly\u003c/em\u003e, and Pandas enables the creation of interactive maps, dynamic graphs, and real-time tables\u0026mdash;elements that reinforce visual understanding and interactivity with the phenomenon being studied.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 \u0026nbsp;\u003c/strong\u003eData Verification and Validation\u003c/p\u003e\n\u003cp\u003eIn the real scenario, as expected, the temperature (\u0026deg;C) varied according to time (hours) (Fig. 3), with a nonlinear relationship between the variables, represented by a cubic polynomial trend curve and coefficient of determination (\u003csup\u003eR\u0026sup2;\u003c/sup\u003e) of 0.8788. \u0026deg;The analysis of variance (ANOVA) of the regression was significant (p\u0026amp;lt;0.05), suggesting that the temperature variation in the real scenario was not influenced by the time of day. The maximum temperature recorded during the period was 26.48\u0026deg;C at 3:00 p.m., whereas the minimum temperature reached 18.9\u0026deg;C at 5:00 a.m. The prevailing average temperature was 22.63\u0026deg;C (∘\u0026nbsp;C), indicating a moderate thermal regime for the city of Cuiab\u0026aacute;, Brazil (Table 3).\u003c/p\u003e\n\u003cp\u003eThe variability of temperatures in the dataset was characterized by a standard deviation of 2.73\u0026deg;C and a coefficient of variation of 12.06%, suggesting moderate temperature dispersion (Table 3), which implies stable daily thermal conditions at the time and place of study\u003c/p\u003e\n\u003cp\u003eA clear visual difference between the two climate simulators, ENVI-met and Edu-Clima, was evident for all the scenarios studied. The temperatures simulated by ENVI-met were consistently higher than those simulated by Edu-Clima.\u003c/p\u003e\n\u003cp\u003eWhen comparing the simulations performed with ENVI-met and Edu-Clima for the different scenarios, the observed behavior line of Edu-Clima follows a pattern closer to a climatic normal, presenting a lower temperature variation at dusk. In contrast, in ENVI-met, for all scenarios, the temperature continued to rise throughout the day or remained stable from dusk to nightfall (between 5 p.m. and 11 p.m.) (Fig. 4 and Fig. 5).\u003c/p\u003e\n\u003cp\u003eThe climate behavior simulated by Edu-Clima for the city of Cuiab\u0026aacute;, MT, reflects the trend observed in the real scenario. As in the city, the application results in the highest temperatures between 10 a.m. and 2 p.m., with a gradual reduction throughout the evening.\u003c/p\u003e\n\u003cp\u003eThe regression analysis presented in Fig. 4 and Fig. 5 reveals that the temperature varies with time for the different soil cover scenarios, whether they are asphalt with exposed surfaces, different leaf area indices, or even asphalt with vegetated surfaces. This pattern is also repeated for concrete ground covers, with or without exposed surfaces, at different leaf area indices. This can be seen in Fig. 4, where the temperature provided by ENVI-met on the asphalt ground cover with an exposed surface and a low leaf area index (LAI) at 8 a.m. was significantly lower than that at 2 a.m.\u003c/p\u003e\n\u003cp\u003eAnalysis of the temperature data using a third-order regression curve revealed that the thermal behavior in the study area does not follow a normal distribution, indicating complex and nonlinear variability. Interestingly, the observations generated in both applications within the different scenarios showed a similar behavior pattern in the polynomial line.\u003c/p\u003e\n\u003cp\u003eThe coefficient of determination (R\u0026sup2;) is a statistical indicator used to assess the quality of a regression model. It measures the proportion of variation in the data that is explained by the model. The R\u0026sup2; values found in our equations indicate that the model is a reliable tool for describing the temperature trend of the data.\u003c/p\u003e\n\u003cp\u003eAn analysis of variance of the regressions for the different scenarios revealed that time had a significant influence on temperature in both the simulators and all the scenarios (Fig. 6; p \u0026lt; 0.05). The average variation in the scenarios with exposed surfaces ranged from 13.27% (AEB) to 37.49% (CEA) (Fig. 6). In the scenarios with vegetated surfaces, the variation observed ranged from 27.27% (AVB) to 47.11% (CVA). These values suggest that the variation in temperatures between simulators may present an acceptable range between average, maximum, and minimum temperatures.\u003c/p\u003e\n\u003cp\u003eThe cumulative coefficient of variation for the scenarios with vegetated and exposed surfaces remained below 11%, indicating that the data collected from both simulators tended toward normality when compared. This is because the cumulative coefficients of variation were close to the individual coefficients of variation of the temperatures extracted from each simulator over time.\u003c/p\u003e\n\u003cp\u003eThe appropriate measures for expressing variability are generally the standard deviation (absolute variability) and the coefficient of variation (relative variability), which vary between 2.3\u0026deg;C and 2.8\u0026deg;C between the urban scenarios and the simulations. The coefficients of variation ranged from 8.63% to 12.29% (Fig. 6 and Fig. 7).\u003c/p\u003e\n\u003cp\u003eThese results reinforce that the simulations performed in ENVI-met and Edu-Clima are within an acceptable range, which validates the simulator developed in relation to the temperatures obtained throughout the simulations and scenarios. The values of the coefficients of variation, which are less than 30%, corroborate the absence of a significant difference between the temperatures projected by the Edu-Clima and ENVI-met simulators.\u003c/p\u003e\n\u003cp\u003eThe average temperature deviation between ENVI-met and Edu-Clima varied considerably depending on the scenario. In the scenario of asphalt soil cover with an exposed surface and a low leaf area index (LAI), the average temperature deviation was 3.84\u0026deg;C (Fig. 7a). In contrast, in the scenario of a concrete ground cover with a vegetated surface and a high leaf area index (LAI), the temperature deviation between the simulators reached 11.5\u0026deg;C (Fig. 7f). This occurred because the temperatures simulated by Edu-Clima in the LAI scenario were lower than those projected by ENVI-met. In the AEB scenario, although the Edu-Clima temperatures were lower than those in ENVI-met, the percentage variations were smaller than those in CVA (average variation of 13.27% versus 47.11%).\u003c/p\u003e\n\u003cp\u003eThese results reveal a greater temperature variation between the simulators for scenarios with vegetated surfaces (Fig. 7), as well as for exposed surfaces with a high leaf area index, suggesting that the Edu-Clima simulator demonstrates greater sensitivity in simulations of vegetated scenarios. However, when comparing the temperatures provided by ENVI-met and Edu-Clima within each scenario, a significant difference was observed by the paired t test in all the simulated scenarios (p \u0026lt; 0.05), leading to the rejection of the null hypothesis, i.e., the temperatures given by the simulators are different for the times covered.\u003c/p\u003e\n\u003cp\u003eMultivariate analysis, using Principal Coordinate Analysis (PCoA), was able to summarize the variability in temperatures across all the different scenarios and times covered. All the data could be explained well by the first two axes of the PCoA: axis 1 represented 98.02% of the total variation, whereas axis 2 explained only 1.85%.\u003c/p\u003e\n\u003cp\u003eThe graphical visualization (Fig. 8) shows a gap between the temperature values simulated by ENVI-met and those simulated by Edu-Clima. Additionally, PCoA revealed that Edu-Clima\u0026apos;s climate indicators were positively correlated, whereas ENVI-met\u0026apos;s indicators tended to occupy the negative quadrants of the graph. The p value obtained from the PERMANOVA, which was applied to the PCoA results (Fig. 8), confirmed the difference visualized graphically.\u003c/p\u003e\n\u003cp\u003eThe analysis revealed that the tools followed a similar pattern of response to environmental variables, such as soil type and vegetation cover. The differences observed are largely a consequence of the intentional simplification of the Edu-Clima model, which focuses on ease of use and accessibility at the expense of the computational accuracy of ENVI-met.\u003c/p\u003e"},{"header":"4.\tDISCUSSION","content":"\u003cp\u003eFrom a methodological point of view, Edu-Clima plays a strategic role by aligning itself with trends in digital education and climate citizenship. Recent studies, such as that by Zourmpakis et al. (2023), noted that the use of narratives and visual elements in science teaching, such as graphs, maps, and simulations, not only increases student engagement but also improves their conceptual understanding and motivation to learn.\u003c/p\u003e\n\u003cp\u003eThe sensitivity of the Edu-Clima app to small variations in input parameters, such as type of cover, vegetation, and time of day, allows users to experiment with different scenarios and observe their thermal implications in real time. This functionality makes it a relevant tool for active teaching strategies, such as inquiry-based teaching, problem-based learning (PBL), and gamification. This orientation is linked to SDG 4, considering the importance of methodologies and tools that favor the expansion of access to knowledge and critical training in the field of environmental education (Vivar \u0026amp; Peñalvo, 2023).\u003c/p\u003e\n\u003cp\u003eInteractive tables and graphs play a key role in the learning process by transforming complex data into accessible and dynamic visual representations. These tools allow students to develop more sophisticated interpretive skills, going beyond simple literal reading to identify trends, patterns, and relationships in the data. Studies show that both high school students and teachers in training face challenges in adequately interpreting these representations, highlighting the need for pedagogical practices that promote the critical and effective use of tables and graphs in the classroom (Perin \u0026amp;amp; Campo, 2022; Castro et al., 2024). The development of these skills contributes to more active, analytical learning that is connected to the reality of data in the contemporary world.\u003c/p\u003e\n\u003cp\u003eA relevant aspect is the use of good software development practices, such as code modularization and organization by separate functions and scripts, which favor maintenance, reusability, and future improvements to the tool. This open architecture, in addition to promoting transparency, allows the educational and scientific community to contribute improvements and adaptations, transforming Edu-clima into a collaborative platform, an important aspect highlighted by Tozzi et al. (2024).\u003c/p\u003e\n\u003cp\u003eThe climate behavior observed in the city of Cuiabá-MT, used as the setting for the simulations, follows the same trend that was observed in Edu-clima, with higher temperatures occurring between 10:00 a.m. and 2:00 p.m. and decreasing at dusk (Callejas \u0026amp; Krüger, 2022). In the analysis of the coefficient of variation, the index of 30% is considered a limit; values above this value suggest high irregularity, whereas values below this value indicate greater regularity of temperature in relation to the average (Taye \u0026amp; Njuho, 2008).\u003c/p\u003e\n\u003cp\u003eThe third-order regression curve revealed that this temperature does not follow a normal pattern, unlike the simulation verified in Edu-Clima. The coefficient of determination (R²) indicates the proportion of data variation explained by the model, which, through the regression equation produced, was able to represent the behavior of the temperature observations during the analyzed period (Chicco et al., 2021). The equations extracted from the models can make temperature predictions at times not explored in the different simulations, which would help fill gaps in the data series.\u003c/p\u003e\n\u003cp\u003eBy analyzing the regressions for different scenarios, we confirmed that time (time of day) strongly influenced the temperature under both the Edu-Clima application and ENVI-met. This significant result corroborates what the scientific literature already points out: the time of day is a determining factor for thermal variations (Kou et al., 2022).\u003c/p\u003e\n\u003cp\u003eThis influence becomes even more evident in our results, given the wide temperature range recorded in the simulations of both tools, reinforcing the direct impact of the passage of time on the temperature behavior in all the scenarios. The ranges verified in the simulations are also within the normal range of climatological records for the region of Cuiabá, Mato Grosso (Machado \u0026amp; Gonzalez, 2021; Carmo, Reboita et al., 2023).\u003c/p\u003e\n\u003cp\u003eImportantly, although the PERMANOVA p value (0.00010) points to statistically significant differences between the temperatures simulated by Edu-Clima and ENVI-met, this distinction does not invalidate the quality of the application for educational purposes. The analysis showed that the tools follow a similar standard in response to environmental variables, such as soil type and vegetation cover. The differences observed are largely a consequence of the intentional simplification of the Edu-Clima model, which focuses on ease of use and accessibility at the expense of the computational accuracy of ENVI-met. Therefore, the PERMANOVA results reinforce that, for teaching purposes, the Edu-Clima model is reliable, as it reproduces the trends and fundamental concepts of urban climatology in a clear and interactive way (Belouafa, 2017).\u003c/p\u003e"},{"header":"5.\tCONCLUSION","content":"\u003cp\u003eAlthough ENVI-met has a robust database and is one of the main climate simulation tools used, the simulator created for teaching purposes has a simple and intuitive interface for designing basic climate parameters, such as average temperature, minimum temperature, or their progression over time.\u003c/p\u003e\n\u003cp\u003eThe analyses also highlighted greater deviations in scenarios with dense vegetation, especially when combined with vegetated soils, as in the case of the CVA (Concrete, Vegetated, High LAI) scenario. These deviations of more than 11°C and percentage variations close to 47% indicate that Edu-Clima does not yet accurately incorporate the effects of evapotranspiration, natural shading, and soil moisture, aspects that are better modeled by ENVI-met. This limitation, however, is understandable and acceptable within the scope of a tool built with open-source code, fed thus far with simplified data for educational use.\u003c/p\u003e\n\u003cp\u003eFinally, the results suggest that Edu-clima fulfills its purpose of being a reliable, accessible, and functional educational simulator. The analysis also shows that the tool can be understood as a contribution aligned with SDG 4, as it favors educational practices that integrate technological innovation and training focused on sustainability. Although its limitations in relation to vegetation modeling are evident, these limitations do not compromise its validity as a teaching tool. In contrast, they reveal points for future development, allowing the tool to advance in robustness without losing its didactic and intuitive essence.\u003c/p\u003e"},{"header":"STATEMENTS AND DECLARATIONS","content":"\u003cp\u003e\u003cstrong\u003eEthical Approval\u003c/strong\u003e: Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInformed Consent\u003c/strong\u003e: Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatement Regarding Research Involving Human Participants and/or Animals\u003c/strong\u003e: Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Participate\u003c/strong\u003e: Informed consent was obtained from all individual participants included in the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publish\u003c/strong\u003e: All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e This work was funded by the authors themselves.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e: The authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAlmazroui, M., Abid, M. A., Ashfaq, M., Islam, M. N., Kamil, S., Rashid, I. U., ... \u0026amp; Sylla, M. B. (2021). Assessment of CMIP6 performance and projected temperature and precipitation changes over South America. \u003cem\u003eEarth Systems and Environment\u003c/em\u003e, 5(2), 155-183.\u003c/li\u003e\n \u003cli\u003eAslam, M. (2021). Retracted Article: Neutrosophic Statistical Test For Counts in Climatology. \u003cem\u003eScientific Reports\u003c/em\u003e, 11(1), 17806.\u003c/li\u003e\n \u003cli\u003eBelouafa, S., Belafkih, B., Benhar, S., Habti, F., Hamdouch, S., Tayane, S., ... \u0026amp; Abourriche, A. (2017). Ferramentas e abordagens estat\u0026iacute;sticas para validar m\u0026eacute;todos anal\u0026iacute;ticos: metodologia e exemplos pr\u0026aacute;ticos. \u003cem\u003eRevista Internacional de Metrologia e Engenharia da Qualidade\u003c/em\u003e, 8, 9.\u003c/li\u003e\n \u003cli\u003eCallejas, I. J. A., \u0026amp; Kr\u0026uuml;ger, E. (2022). Microclimate and thermal perception in courtyards located in a tropical savannah climate. \u003cem\u003eInternational journal of biometeorology\u003c/em\u003e, 66(9), 1877\u0026ndash;1890.\u003c/li\u003e\n \u003cli\u003eCamps-Valls, G., Castelletti, A., Cohrs, K. H., Fern\u0026aacute;ndez-Torres, M. \u0026Aacute;., H\u0026ouml;hl, A., Pacal, A., ... \u0026amp; Williams, T. (2025). Artificial intelligence for modeling and understanding extreme weather and climate events.\u0026nbsp;\u003cem\u003eNature Communications\u003c/em\u003e, 16(1), 1919.\u003c/li\u003e\n \u003cli\u003eCarmo, E. L. I., Reboita, M. S., \u0026amp; Marques, R. (2023). Evolu\u0026ccedil;\u0026atilde;o Temporal das Vari\u0026aacute;veis Atmosf\u0026eacute;ricas associadas a Casos de Frentes Frias Fortes em Cuiab\u0026aacute;, MT, entre 1996 e 2015\u003cem\u003e. Revista Brasileira de Geografia F\u0026iacute;sica\u003c/em\u003e, 16(1), 145-154.\u003c/li\u003e\n \u003cli\u003eCastro, E. R., Barreto, M. C., Nascimento, F. J., \u0026amp; Sousa, G. L. (2024). Construction and Interpretation of Statistical Tables and Graphs: A Look at the Attitude of Teachers in Training. \u003cem\u003eTeaching in Review\u003c/em\u003e, 31, 1\u0026ndash;24.\u003c/li\u003e\n \u003cli\u003eChicco, D., Warrens, M. J., \u0026amp; Jurman, G. (2021). The Coefficient Of Determination R-Squared Is More Informative than SMAPE, MAE, MAPE, MSE and RMSE in Regression analysis evaluation. \u003cem\u003ePeerj computer science\u003c/em\u003e, 7, e623.\u003c/li\u003e\n \u003cli\u003eKim, D. Y., \u0026amp; Kwon, O. (2025). How Open Source Artificial Intelligence Democratizes and Improves Science: Evidence from Structural Biology. \u003cem\u003eAvailable at SSRN 5219337\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eKou, Y., Chen, J., Cheng, B., Liu, Y., Wang, C., Xian, D., \u0026hellip; \u0026amp; Tang, L. (2022). Factors affecting urban climate at different times of the day in China: a case study in Yibin, a riverside mountain city. \u003cem\u003eNature-Based Solutions\u003c/em\u003e, 2, 100043.\u003c/li\u003e\n \u003cli\u003eKurokawa, H., Igei, K., Kitsuki, A., Kurita, K., Managi, S., Nakamuro, M., \u0026hellip; \u0026amp; Sakano, A. (2023). Improvement Impact Of Nudges Incorporated In Environmental Education On Students Environmental Knowledge, Attitudes, And Behaviors. \u003cem\u003eJournal of Environmental Management\u003c/em\u003e, 325, 116612.\u003c/li\u003e\n \u003cli\u003eLozano-D\u0026iacute;az, A., \u0026amp; Fern\u0026aacute;ndez-Prados, J., S. (2020). Educating digital citizens: An opportunity to critical and activist perspective of sustainable development goals. \u003cem\u003eSustainability\u003c/em\u003e, 12(18), 7260.\u003c/li\u003e\n \u003cli\u003eMachado, C. S. D., \u0026amp; Gonzalez, A. Z. D. (2021). Climate Variability in the Cerrado Biome of Mato Grosso During the Last Decades.\u0026nbsp;\u003cem\u003eRevista Equador\u003c/em\u003e, 10(2), 207\u0026ndash;227.\u003c/li\u003e\n \u003cli\u003eMakvandi, M., Chai, H., Fu, J., Khodabakhshi, Z., Li, W., Ou, X., ... \u0026amp; Horimbere, E. D. L. J. (2023). Mitigation of urban heat for climate change adaptation: an eco-sustainable design strategy to improve environmental performance amid rapid urbanization.\u0026nbsp;\u003cem\u003eAtmosphere\u003c/em\u003e, 14(4), 638.\u003c/li\u003e\n \u003cli\u003eMaranholi, H. N. G., \u0026amp; Santos, F. M. M. (2024). O M\u0026eacute;todo Cient\u0026iacute;fico e a Compreens\u0026atilde;o das Vari\u0026aacute;veis Ambientais no Contexto Educacional: Uma Revis\u0026atilde;o.\u0026nbsp;\u003cem\u003eCadernos Caju\u0026iacute;na\u003c/em\u003e, 9(5), e249525.\u003c/li\u003e\n \u003cli\u003ePacifici, M., \u0026amp; Nieto-Tolosa, M. (2021). Comparing ENVI-Met and grasshopper modeling strategies to assess local thermal stress and urban heat island effects. In \u003cem\u003eUrban microclimate modeling for comfort and energy studies\u0026nbsp;\u003c/em\u003e(pp. 293-316). Cham: Springer International Publishing.\u003c/li\u003e\n \u003cli\u003ePerin, A. P., \u0026amp; Campo, C. R. (2022). Leitura e interpreta\u0026ccedil;\u0026atilde;o de gr\u0026aacute;ficos estat\u0026iacute;sticos por alunos do 2\u0026ordm; ano do ensino m\u0026eacute;dio. \u003cem\u003eRevista Baiana de Educa\u0026ccedil;\u0026atilde;o Matem\u0026aacute;tica\u003c/em\u003e, 3(01), e202227.\u003c/li\u003e\n \u003cli\u003eRiuttanen, L., Ruuskanen, T., \u0026Auml;ij\u0026auml;l\u0026auml;, M., \u0026amp; Lauri, A. (2021). Society needs experts with climate change competencies\u0026ndash;what is the role of higher education in atmospheric and Earth system sciences?.\u0026nbsp;\u003cem\u003eTellus B: Chemical and Physical Meteorology\u003c/em\u003e, 73(1), 1\u0026ndash;14.\u003c/li\u003e\n \u003cli\u003eStreb, V., Ferreira, M. D., Gomes, A. F., Reginatto, A. A., Siqueira Cecchin, A. de, \u0026amp; da Rocha, KM (2021). Teoria x Pr\u0026aacute;tica: Panorama inicial sobre a inser\u0026ccedil;\u0026atilde;o das Tecnologias Digitais no Ensino Superior presencial e a dist\u0026acirc;ncia na UFSM.\u0026nbsp;\u003cem\u003eRevista Brasileira de Desenvolvimento\u003c/em\u003e, 7 (4), 41318-41331.\u003c/li\u003e\n \u003cli\u003eTaye, G., \u0026amp; Njuho, P. (2008). Monitoring field variability using confidence interval for coefficient of variation. \u003cem\u003eCommunications in Statistics\u0026mdash;Theory and Methods\u003c/em\u003e, 37(6), 831\u0026ndash;846.\u003c/li\u003e\n \u003cli\u003eTozzi, C. C., de Oliveira, I. D. S. B., Bonicenha, L. C., Campanin, M. A. A., Dona, R. A. M., Onofre, V., \u0026amp; Andreza, W. G. G. (2024). M\u0026iacute;dias digitais na educa\u0026ccedil;\u0026atilde;o online: o impacto da linguagem audiovisual e ferramentas colaborativas. \u003cem\u003eRevista Ibero-Americana de Humanidades, Ci\u0026ecirc;ncias e Educa\u0026ccedil;\u0026atilde;o\u003c/em\u003e, 10(10), 3723-3729.\u003c/li\u003e\n \u003cli\u003eTrane, M., Giovanardi, M., Pejovic, A., \u0026amp; Pollo, R. (2023). Vis\u0026atilde;o geral sobre ferramentas de modelagem de clima e microclima urbano e seu papel para alcan\u0026ccedil;ar os Objetivos de Desenvolvimento Sustent\u0026aacute;vel. \u003cem\u003eIn: Arquitetura e Design para a Ind\u0026uacute;stria 4.0: Teoria e Pr\u0026aacute;tica\u003c/em\u003e (pp. 247-267). Cham: Springer International Publishing.\u003c/li\u003e\n \u003cli\u003eTwardosz, R., Walanus, A., \u0026amp; Guzik, I. (2021). Warming in Europe: recent trends in annual and seasonal temperatures. \u003cem\u003ePure and Applied Geophysics\u003c/em\u003e, 178(10), 4021-4032.\u003c/li\u003e\n \u003cli\u003eVivar, J. M. F., \u0026amp; Pe\u0026ntilde;alvo, F. J. G. (2023). Reflections on the ethics, potential, and challenges of Artificial Intelligence in the context of Quality Education (SDG 4).\u0026nbsp;\u003cem\u003eComunicar: Revista cient\u0026iacute;fica de comunicaci\u0026oacute;n y educaci\u0026oacute;n\u003c/em\u003e, (74), 37-47.\u003c/li\u003e\n \u003cli\u003eWaskom, M. L. (2021). Seaborn: statistical data visualization. \u003cem\u003eJournal of Open Source Software\u003c/em\u003e, 6(60), 3021.\u003c/li\u003e\n \u003cli\u003eZourmpakis, A. I., Kalogiannakis, M., \u0026amp; Papadakis, S. (2023). Adaptive Gamification In Science Education: An Analysis Of The Impact Of Implementation And Adapted Game Elements On Students\u0026rsquo; Motivation. \u003cem\u003eComputers\u003c/em\u003e, 12(7), 143.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Environmental Sciences, Environmental Education, Urban Climate Simulation, Climatology Teaching","lastPublishedDoi":"10.21203/rs.3.rs-7743068/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7743068/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This article presents the development and validation of Edu-Clima, an open-source educational application for simulating urban surface temperature. The objective of this study was to develop and validate an urban surface temperature simulation application for educational purposes by comparing its results with those obtained via ENVI-met software and evaluating its reliability for use in teaching urban climatology. Edu-Clima was built in Python and uses libraries such as Streamlit, Pandas, Matplotlib, Seaborn, and Plotly to create an interactive and visually educational interface. The methodology involved comparing the results of Edu-Clima with those of ENVI-met, using 12 different scenarios, each with 23 simulations, totaling 276 simulations. The scenarios considered variables such as surface temperature, leaf area index (high, medium, low), ground cover (50% asphalt or concrete), and soil type (exposed or vegetated). The results revealed that although ENVI-met consistently presented higher temperatures, Edu-Clima followed a pattern closer to a \"climate normal.\" Statistical analyses, including quadratic regression and PERMANOVA, revealed that, despite visual differences, there is no statistically significant distinction between the temperatures simulated by the two tools, validating Edu-Clima for educational purposes, in line with SDG 4 (Quality Education), which emphasizes the promotion of inclusive and accessible educational resources. However, Edu-Clima showed greater sensitivity in vegetated scenarios, indicating a limitation in evapotranspiration and shading modeling, which are more accurate in ENVI-met. In the future, we highlight the importance of improving the modeling of vegetation effects, including variables such as evapotranspiration, soil moisture, and shading.","manuscriptTitle":"Edu-Clima: A Didactic Tool for the Simulation and Modeling of Urban Climate","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-04 08:33:48","doi":"10.21203/rs.3.rs-7743068/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b1dd7b8e-abd2-4699-858f-a10ebe09e033","owner":[],"postedDate":"December 4th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-21T13:32:28+00:00","versionOfRecord":{"articleIdentity":"rs-7743068","link":"https://doi.org/10.31413/nat.v14i2.20963","journal":{"identity":"nativa","isVorOnly":true,"title":"Nativa"},"publishedOn":"2026-04-15 00:00:00","publishedOnDateReadable":"April 15th, 2026"},"versionCreatedAt":"2025-12-04 08:33:48","video":"","vorDoi":"10.31413/nat.v14i2.20963","vorDoiUrl":"https://doi.org/10.31413/nat.v14i2.20963","workflowStages":[]},"version":"v1","identity":"rs-7743068","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7743068","identity":"rs-7743068","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}