Temperature-Based Renewable Energy Forecasting: A Big Data Analysis for Sustainable Energy Planning in Bangladesh

Temperature-Based Renewable Energy Forecasting: A Big Data Analysis for Sustainable Energy Planning in Bangladesh | 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 Temperature-Based Renewable Energy Forecasting: A Big Data Analysis for Sustainable Energy Planning in Bangladesh Hasan Ahamed Alif This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7031666/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 27 Sep, 2025 Read the published version in Theoretical and Applied Climatology → Version 1 posted 12 You are reading this latest preprint version Abstract Sustainability in renewable energy involves utilizing energy sources intelligently to serve the present and yet be accessible for future generations. With Bangladesh shifting to renewable energy due to rising climate threats, it needs expert and data-based input to create sustainable infrastructure. The investigation uses a blend of statistical and deep learning methodologies to analyze the implications of temperature-based climatic patterns on the potential for solar energy in Rajshahi and Ishwardi in Bangladesh. We centered our study on past temperatures from 1980 to 2020, bringing in both linear regression and one of the top deep learning models, LSTM, as ideal ways to estimate climate in the future. It was found that both locations are seeing a significant increase in temperature, whereas the LSTM technique was superior at spotting irregular seasonal fluctuations and trends across years. We utilized the all-sky shortwave irradiance, clearness index of solar radiation, quantity of cloud cover, albedo, and near-surface temperature from the NASA-POWER datasets to estimate the solar energy potential. The research showed that Rajshahi had superior conditions for solar energy owing to the cleaner weather, fewer overcast days, and higher irradiance intensity. This research provides significant insights for regional policy interests and renewable energy planning via extensive visual data and the assessment of several factors. The report advocates for using AI and data to facilitate solar expansion. It highlights the crucial role of deep learning in promoting sustainability and ecologically safe energy in Bangladesh and other developing nations. Artificial Intelligence (AI) Renewable Energy Big Data Analysis LSTM Climate-Resilient Infrastructure NASA Power Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Climate change denotes a substantial and enduring alteration of global weather patterns, impacting areas as varied as the tropics and the polar regions(Abbass et al., 2022 ). Climate change has emerged as a significant global problem affecting corporations, individuals, and the environment(Song et al., 2023 ). Temperature significantly influences climate change by affecting weather patterns, ecosystems, food security, air quality, and mental health(Dellaripa et al., 2023 ). The requirement of lowering environmental externality, the energy security issue, the economic benefit, affordability, accessibility, and the sustainability agenda are forcing the globe to resort to the fast adoption of renewable energy sources.(Genc & Kosempel, 2023 ). Solar energy is categorized as a clean energy source that reduces greenhouse gas emissions compared to fossil fuels(Novas et al., 2021 ). Bangladesh must devise innovative solutions to its energy challenges, which are intricately linked to its climate vulnerability, to attain sustainable development and climate resilience(Al-Maruf et al., 2021 ). Bangladesh's rapid economic development and population expansion have resulted in a significant increase in energy consumption, posing a critical challenge in supplying energy to its industry and citizens(Miskat et al., 2023 ). Bangladesh, with its geographical limits mixed with socioeconomic dependency and inadequate access to resources, is among the most susceptible nations to climate-related challenges.(Al Mamun et al., 2023 ). Bangladesh's government has suggested high objectives for renewable energy to strengthen its energy sector and fulfill sustainable development goals (SDGs)(Munjer et al., 2023 ). Regional energy planning is crucial for formulating context-specific and pragmatic energy policies that promote sustainable development and the use of renewable energy(De Laurentis & Pearson, 2021 ). As Bangladesh wants to grow solar adoption, it confronts constraints relating to land scarcity, economic sustainability, and grid stability(Talut et al., 2022 ). The absence of region-specific evaluations of climatic adaptability in Bangladesh restricts the capacity to design effective adaptation solutions(Choi et al., 2021 ). Traditional forecasting methods frequently depend on historical data and statistical approaches to anticipate future climatic conditions(Zhou et al., 2023 ). Inadequate long-term climate data at the subnational level, mainly as a consequence of the use of indirect proxies, the limitations of the models that are now in use, and the challenges in validating them against data that has been observed(Hébert et al., 2022 ). Data restrictions, inconsistencies, and a comprehensive strategy to enhance data gathering and analysis are the key reasons for the paucity of climate data-driven research at the sub-national level(Hultman et al., 2020 ). There is a shortage of concentrated research that notably targets the distinct possibilities and problems that occur at sub-national levels, which might generate a gap in our knowledge of localized energy dynamics, where the bulk of research using AI in renewable energy forecasting mainly concentrates on national or large regional dimensions(Haupt et al., 2020 ). The proposed study is an effort to explore in detail the previous climate behavior and its repercussions on the feasibility of solar energy in two climatically diverse sites of Bangladesh- Rajshahi and Ishwardi. Using more than 40 years of temperature and solar-related environmental variables, the research employs a standard linear regression and state-of-the-art deep learning networks (particularly Long Short-Term Memory networks) to extract and forecast climatic fluctuations. Through this, the research can evaluate long-term warming trends and combine proxy variables like shortwave irradiance, clearness index, albedo, and cloud cover to estimate the solar feasibility of the areas involved. The basic purpose is to generate data-driven, spatially sensitive information that would ease the planning of sustainable energy and offer evidence-based policies about the deployment of renewable energy in the delicate climate of Bangladesh. This examination integrates long-term climate prediction and important solar viability indicators, thus reducing the gap between environmental information examination and practical Bangladesh energy planning. This study highlights the benefit of employing state-of-the-art deep learning approaches, particularly Long Short-Term Memory (LSTM) models, to depict nonlinear climatic patterns than more conventional statistical methods. More crucially, it is transforming the complex climatic dynamics into intelligence that might be acted upon in deploying solar energy, especially in locations like Rajshahi and Ishwardi, where optimization of resources is a big problem. The findings have been presented in the context of the national discussion on sustainable infrastructure, and they give a framework that may be copied in other places to plan renewable energy production at the local scale in the shifting climatic circumstances. In order to fulfill the research aims, we designed a hybrid analytical system that includes old statistics and recent AI methodologies. The linear regression was performed to calculate linear trends of baseline warming using historical temperature data between 1980 and 2020. Predictive modeling used Long Short-Term Memory (LSTM) neural networks to capture more complicated and nonlinear temporal dynamics. Also, climatic characteristics that include shortwave irradiance, clearness index, cloud cover, surface albedo, and temperature collected by NASA-POWER datasets were utilized to assess the feasibility of solar energy. Such a combination of methodologies enabled a consistent evaluation of the designated locations' climatic patterns and solar potential. 2. Methodology We use these regions to summarize and examine the outcome of this investigation. 2.1 Data Source and Description The research uses forty-year temperature data to evaluate the climate's influence on solar energy planning. The dataset's information was sourced from BMD, which documents meteorological occurrences in various regions of Bangladesh. Rajshahi and Ishwardi were selected for investigation due to their abundant sunshine, significance in agriculture, and pivotal involvement in renewable energy initiatives. The dataset contains daily maximum and minimum temperatures for each year from 1980 to 2020. The data were streamlined and organized by transforming them into monthly and annual average temperatures. Long-term data facilitates the identification of alterations in meteorological patterns, seasonal variations, and related factors that may impact solar energy initiatives. This case study evaluates the efficacy of solar photovoltaic technology in western Bangladesh using meteorological data. In addition, analyzing these two areas helps us compare inland climatic changes, which counts a lot for planning regional climate-related measures. It endorses the United Nations objectives for sustainable development (SDG 7 and SDG 13), emphasizing the accessibility and environmental sustainability of energy. The data spanning many decades (1980–2020) enables the identification of progressive warming trends, heightened temperature variability, and increased heatwave frequency, all of which are critical for renewable energy systems. 2.2 Computation of precipitation variability Several processes were required to prepare the temperature data so that analysis could be done with certainty. The data was first acquired from the Bangladesh Meteorological Department (BMD), covering daily maximum and minimum temperatures for Rajshahi and Ishwardi from 1980 to 2020. At first, gaps in the data were discovered and controlled by applying linear interpolation for those shorter holes. This step protected the time series from having unexpected or inexplicable changes. A basic statistical approach called Z-score analysis was employed to locate outliers. Extreme and well-documented measurements, for example, those signifying a heatwave, were left in the data to ensure the continuity of climatic patterns. All the data were then organized into monthly and yearly records so the analysis could be done properly. By combining the records, the researchers could observe how temperatures have changed over lengthy periods, which is essential for energy planning. Using Min-Max scaling, the raw temperature readings were modified for forecasting. This phase was crucial in ensuring that any machine learning models or statistical forecasts would be compatible. Thus, preprocessing the data made the information usable for credible energy estimates and regional analysis. 2.3 Tools and Technologies Used Data visualization and analysis were executed utilizing Python 3.11, and all the work was done within the Jupyter Notebook environment. Several typical scientific libraries were employed in the study. In my work, Pandas assisted with organizing, generating, and manipulating data, Matplotlib and Seaborn were used to illustrate general temperature variations, seasonal differences, and any odd data points. At the same time, NumPy was selected for introductory statistics. Analysis was further done using Google Colab, which made computing easier by offering quick access to its resources for code creation and execution. Matplotlib and Seaborn were used to produce all visual plots, ensuring they have a comparable appearance for simple interpretation. Because these approaches are versatile and practical, they were necessary for identifying temperature patterns that help forecast renewable energy in our research. 2.4 Analytical Approach We examined past temperature patterns to forecast the climate's future behavior using conventional statistical approaches and artificial intelligence techniques. Linear regression was employed as a reference model; however, we relied on LSTM neural networks to discover complicated, irregular changes throughout time. Applying this strategy lets us verify the efficacy of deep learning for climate-dependent activities such as renewable energy planning. 2.4.1 Baseline Trend Analysis using Linear Regression Linear regression follows a pattern as time passes. It is described as when one statement is added to another statement. The first equation is y = β₀ + β₁x + ε. …………. (1) Where: y = The average temperature of a place for the whole year x represents the year (for example, 1980, 1981, ...) Constantly or Intercept = β₀ The slope of the line y = β₁x is denoted β₁. ε = the unexpected component of the model Linear regression was applied to the data from Rajshahi and Ishwardi to determine the baseline warming trends. 2.4.2 Deep Learning-Based Forecasting using LSTM Since linear models could not handle all the disadvantages, we created an LSTM network as an alternative. LSTM may reflect patterns that continue over a long time, alter with the seasons, and are not linear. Preprocessing Steps : Min-Max Normalization: X_norm = (X − Xmin) / (Xmax − Xmin )……. (2) Sliding window transformation is used in supervised learning: Input X = [T(t − 5), ..., T(t − 1)] - the output is T(t) Dataset Split: A part of 80% was utilized for training, and the balance, 20%, was set aside for testing. LSTM Architecture : The history of this part comprises five timesteps. There are 50 LSTM units in the LSTM Layer. Dropout: 0.2 The last layer of the network has several nodes. Loss: Mean Squared Error Optimizer: Adam Early halting was utilized after 100 epochs. Evaluation Metric : We rely on Root Mean Square Error (RMSE) to analyze the forecasting models' performance, since it is a well-known tool in the domain. $$\:RMSE=\sqrt{MSE}=\sqrt{\frac{1}{N}\sum\:_{i=1}^{N}{({y}_{i}-{\widehat{y}}_{i})}^{2}}$$ The root square of MSE is the primary source of RMSE, which is relatively easy to interpret and offers information about average prediction errors. 2.5 Renewable Energy Mapping This paper adds solar energy potential estimates from proxies to the standard techniques of analyzing climates in Rajshahi and Ishwardi. Since we lacked ground-based solar information and extremely detailed solar energy performance from photovoltaic systems, we leaned on NASA POWER for satellite-collected environmental indicators. For instance, surface shortwave radiation, atmospheric clearness indices, cloud cover, albedo, and near-surface air temperature are some of them. It is a mechanism to examine whether solar can function in scarce-data locations using key climatic data like temperature and humidity. 2.6 Parameters Considered Five climatic characteristics for Rajshahi and Ishwardi were selected as they are known to impact solar energy potential: all-sky surface shortwave downward irradiance, insolation clearness index, cloud quantity, all-sky surface albedo, and temperature at 2 meters. The results were gathered from monthly data spanning many years. All of this information is accessible by merging the information on sunlight, sky clarity, and how much radiation the Earth’s surfaces reflect. 2.7 Rationale and Assumptions The logic suggests that more shortwave irradiance and decreased severity of cloud cover result in increased solar energy harvest. The clearness index offers transparency information, while albedo gives the fraction of sunlight reflected by the Earth back to space. By reviewing earlier portions of this text, we can claim that Rajshahi and Ishwardi areas are experiencing greater temperatures each year, suggesting better skies during the dry seasons, which makes solar PV technology more practicable. 2.8 Observational Insights from the Dataset To learn about the environment’s fundamental appropriateness, the relevant criteria were summarized for both research sites. In contrast, Rajshahi had greater average daily solar radiation (5.17 ± 0.21 kWh/m²/day), while being cloudier for less time (45.2 ± 6.1%) as compared to Ishwardi (5.09 ± 0.18 kWh/m²/day; 46.7 ± 6.5%). Inflowing water at both sites exhibited extremely comparable clearness indices (around 0.60–0.61) and albedo values (around 0.19–0.20), demonstrating that they both filtered radiation and reflected surface light in similar proportions. With these measures, judgments were taken to concentrate on various locations for renewable energy. 2.9 Integration with Climate-Aware Planning It also discusses how these elements might be coupled to LSTM-based weather forecasts to build adaptable solar energy plans. You should utilize tools that highlight the ideal times to gather solar energy, put solar panels in places with better weather conditions, and add weather-based activities into the algorithm. 2.10 Limitations This measuring approach is always constrained by its proxy character. Using forecasting models instead of real-time data frequently leads to inaccuracies. Besides, the impacts of concentrated aerosols in the air, dust on the surface, and unique effects on individual panels are not considered, limiting the model’s usefulness for practical application. Still, it offers a realistic beginning point for looking at the potential of solar energy in such places. 3. Results and discussion 3.1 Overview of Climate Parameters This section examines the disparities in key climatic characteristics influencing the solar energy potential of Rajshahi and Ishwardi. The factors include the magnitude of shortwave descending radiation reaching the Earth's surface, the insolation clearness index, cloud cover, surface albedo, and air temperature. We use statistics and graphics to elucidate the circumstances and analyze the climate from various perspectives. 3.2 Correlation Heatmap The following correlation matrices in Fig. 1 present an overview of the pair-wise relationship of five important climatic variables, namely clearness index, surface shortwave downward irradiance, surface albedo, cloud quantity, and 2 m air temperature at both Rajshahi and Ishwardi. The clearness index and irradiance exhibit a significant positive association (r > 0.9) in both areas, demonstrating that cleaner skies immediately predict more incident solar energy. The number of clouds is closely connected with both irradiance (r approx. -0.85) and clearness (r approx. -0.80), further highlighting the existence of cloud cover as a satellite of the solar resource availability. Albedo of the surfaces has a weak positive correlation with irradiance (r ≈ 0.3) and a weak negative correlation with cloud amount (r approximately equal to -0.2), which means that higher surface albedo can slightly increase reflected radiation. However, it is also partly covered by clouds. The air temperature is proportionally (but not substantially) connected to irradiance (r ≈ 0.5) and clearness (r = 0.2), such that warmer days are sunnier, but temperature changes also react to other meteorological factors. The comparison of the two places reveals relatively comparable patterns. However, the cloud-irradiance relation in Ishwardi is significantly stronger, perhaps due to the local peculiarities in the monsoonal cloud processes. These inter-parameter correlations combined reveal that the clearness index and temperature are the major predictors in our renewable-energy forecasting model, and they also illustrate how albedo and cloud cover may be utilized together to complement each other when site-specific energy output is to be forecasted. 3.3 Histogram Rajshahi Figure 2 depicts the monthly frequency distributions of five important climatic indicators in Rajshahi that suggest substantial seasonal dynamics that constitute the foundation of renewable-energy potential in the area. The clearness index likewise reaches its maximum in the range of 0.50 to 0.60 during the dry season (December to February) and correlates to irradiance more than 5.5 kW h m-2 day-1 1 suggesting a clear sky window when solar harvesting would be beneficial. On the other hand, the clearness and irradiance distributions shift to the left as cloud cover rises during the early monsoon season (June to August). This is because cloud-amount frequencies rise into the 30–80% frequencies with modal values in July and August, which mutes surface solar input. Surface albedo is unusually stable throughout the year at 0.13–0.15, demonstrating no seasonal fluctuation in reflectance despite slightly higher values during pre-monsoon dry months. Histograms of air temperatures range from modal values of about 15°C in January to values of approximately 32°C in May, and depict the change of Rajshahi between searing pre‐monsoon heat and cold winter climatic conditions. These linked distributions describe a high-resource dry season, a cloud-dominated monsoon season, and a steady albedo environment, which requires empirical context, temperature-based forecasting, and optimal system design. 3.4 Histogram Ishwardi Figure 3 illustrates the stacked monthly histograms of five climatic indicators in Ishwardi, which display prominent seasonal trends. The clearness index is concentrated in the range of 0.45 to 0.60, with the higher values happening in the dry winter season (December–February), and during monsoon (June–August), the cloud cover frequencies approach 50%. Surface irradiance follows the same pattern with the most significant levels (5.0–6.0 kW-h m-2 day-1) in spring (March-April) and considerable left-shift during monsoon clouds. Albedo is closely centered between 0.11 and 0.14 throughout the year, suggesting that the reflectance of the surface does not fluctuate considerably with the seasonal moisture fluctuations. Cloud‐amount histograms encompass the whole 0–100% range, but are concentrated between 20% and 70% in the noncore monsoon, rather than over 80% in July and August. Air temperaature patterns fluctuate from frigid January patterns of around 17°C to hot April-May patterns of about 32°C, with the monsoon months having temperatures that are tempered (24°C to 28°C). Combined, the distributions indicate the ability of Ishwardi to receive clear skies throughout the winter and spring seasons, the inhibitory monsoon season on irradiance, and the relatively uniform albedo environment, which are essential elements to incorporate in regional renewable‐energy Prediction. Table 1. Summary statistics of key climate indicators in Bangladesh's Rajshahi and Ishwardi regions. Values are presented as mean ± standard deviation, based on annual averages over the study period. Rajshahi consistently shows slightly higher shortwave irradiance and lower cloud cover, indicating comparatively more substantial solar potential. Indicator Rajshahi (Mean ± SD) Ishwardi (Mean ± SD) Temperature at 2m (°C) 26.42 ± 0.69 26.22 ± 0.72 Shortwave Irradiance (kWh/m²/day) 4.56 ± 0.18 4.56 ± 0.18 Insolation Clearness Index 0.49 ± 0.01 0.49 ± 0.01 Cloud Amount (%) 55.1 ± 3.2 55.1 ± 3.3 Surface Albedo 0.13 ± 0.005 0.13 ± 0.005 3.5 Temporal Temperature Trends (1980–2020) The average temperature variations per year were investigated by projecting them throughout 5 decades from the 1980s to the 2020s (Fig. 4 ). Every subplot depicts how temperatures vary from year to year, and what transpired in different places throughout the 10 years. In Fig. 4 a, there are not many temperature variations from one year to the next over the 1980s timeframe. Even if the trends were identical, Rajshahi experienced somewhat higher temperatures. In the years following the 1990s (Fig. 2 b–c), a steady temperature rise is obvious to discern. It can be observed that there are tiny oscillations, perhaps owing to both local weather patterns and broader atmospheric shifts. Figure 4 d shows that both the temperature and range of temperature values soared in the 2000s, suggesting that the swing from cold to hot temperatures got more severe. The situation in Rajshahi is increasingly urgent due to changes in land usage and the city's expansion. Figure 4 e demonstrates the steepest warming between the 2010s and 2050s, showing a warming rate of around + 0.024°C/year in Rajshahi and + 0.020°C/year in Ishwardi, as discovered. The information by decade highlights the necessity of observing how climate change proceeds, particularly for the future of solar and heat-ready constructions. Increasing mean temperatures show that LSTM is the best option for predicting, implying that the climate in both areas will get warmer. 3.6 Temperature Forecast Accuracy To determine how trustworthy statistical predicting techniques are for long-term fluctuations in Rajshahi and Ishwardi’s climate, linear regression was done using their historic temperature data from 1980 to 2020. The research was mainly targeted at assessing how successfully a simplistic linear model could simulate weather patterns over multiple years. In summary, linear regression may provide a rudimentary concept of the warming trend, even if it is not particularly good at forecasting it. Because the RMSE values are comparable, the exact consistent causes play a role in both climates. Table 2 This table presents the Root Mean Squared Error (RMSE) values for the linear regression models forecasting annual average temperatures in Rajshahi and Ishwardi. RMSE serves as a standard metric for evaluating model accuracy, with lower values indicating better performance Region RMSE Rajshahi 0.836 Ishwardi 0.847 Another look at the temperature data indicates that a linear model is not excellent at managing bigger changes between one year’s temperature and the next. That is sensible, since linear regression is confined to circumstances where change happens continuously and cannot handle rapid climate changes, which calls for employing stronger models like LSTM neural networks. Looking forward, it is expected that LSTM will fare better than linear regression in expressing the flexible and always-changing elements in climate data. Even so, the linear model Helps detect general changes in temperature and ensure that they are trending in a given direction. 3.7 Solar Energy Feasibility Comparison For the climatic parameters of solar energy, the data on the bar chart indicate the mean values for shortwave irradiance, insolation clearness index, cloud quantity, surface albedo, and temperature at 2 meters for both Rajshahi and Ishwardi. It briefly reviews whether solar energy can operate in both places. The daily irradiation from the sun is greater in Rajshahi than in Ishwardi. Higher exposure contributes to the potential of producing more PV energy. Similarly, the clearness index values for sun irradiation are comparable in Rajshahi and Ishwardi. Because the atmospheric index is clearer in Rajshahi, the sun’s energy reaches the land more quickly, suggesting that Rajshahi is ideally suited for solar power. The presence of clouds is usually greater in Ishwardi, lowering the consistency and power of sunshine. Rajshahi enjoys more steady sun radiation as it has fewer clouds than Dhaka. Ishwardi reflects more sunlight because its surface albedo is larger than that of other regions. However, as there is less solar energy reflection in Rajshahi, this could marginally aid in getting a better outcome with PV panels. The temperature in both places remained practically the same, Rajshahi simply being slightly hotter on average. High temperatures do not affect how much PV generates, yet they could modify the system’s efficiency and heating, and these aspects must be looked at in the design. Table 3 Summary of Solar Energy Indicators Indicator Mean Std Dev Region Shortwave Irradiance (kWh/m²/day) 3.53 0.47 Rajshahi Insolation Clearness Index 0.47 0.04 Rajshahi Cloud Amount (%) 32.35 9.97 Rajshahi Surface Albedo 0.14 0.01 Rajshahi Temperature at 2m (°C) 17.35 1.33 Rajshahi Shortwave Irradiance (kWh/m²/day) 3.63 0.46 Ishwardi Insolation Clearness Index 0.49 0.04 Ishwardi Cloud Amount (%) 31.3 10.45 Ishwardi Surface Albedo 0.13 0.01 Ishwardi Temperature at 2m (°C) 18.05 1.36 Ishwardi 3.8 Regional Comparison: Rajshahi vs. Ishwardi It pulls together a study of how Rajshahi and Ishwardi vary in climatic patterns and prospects for solar energy. Its base is a study into temperature trends, critical climatic elements for solar power, and the accuracy of the projections from each model. Both locations are growing warmer; however, Rajshahi heats up faster at + 0.024°C/year, while Ishwardi follows at + 0.020°C/year. This tendency corresponds with climate change in general and calls for increased efforts to embrace renewable energy in places vulnerable to heat. Performance projections built using linear regression and LSTM networks make it simpler to discover regional disparities. Both LSTM models functioned substantially better in both sites by following challenging patterns and providing trustworthy forecasts. Rajshahi and Ishwardi had greater RMSE in linear predictions, while both had lower RMSE in LSTM forecasts, indicating the success of AI in both domains. Rajshahi outperforms Ishwardi in everything evaluated when evaluating their solar energy potential. Slightly greater levels of shortwave irradiance, clearness index, and reduced average cloud cover may be seen, even though these slight variations imply that Rajshahi would be superior for producing solar energy routinely. In contrast, Ishwardi’s greater number of particles in the sky and somewhat brighter albedo might limit solar energy collection from the sun, mainly in the monsoon season. All in all, Rajshahi is somewhat better than Dhaka in terms of climate. Stability, predicting certainty, and energy capability give it a tiny edge in commencing the first solar projects. 3.9 Interpretation and Policy Implications Based on the study’s conclusions, Bangladesh may enhance its energy strategy at the regional and national levels. Because both Rajshahi and Ishwardi show evidence of constant warming, and solar energy has been verified using satellite climate data, climate intelligence should be employed in formulating energy plans. It is evident from LSTM findings that applying sophisticated AI technologies would assist in enhancing weather and energy monitoring systems. The data from these models may impact grid design, additional capacity construction, and how electricity will be delivered in the various seasons, notably for solar-dominant portfolios. With its excellent sun conditions, Rajshahi might be utilized as a trial site for targeted solar usage, notably in zones where people do not have access to the same electricity as big cities. The region might grow stronger and less dependent on fossil fuels by deploying solar panels on top of buildings here and in harmony with adjacent smart grids. Because of the greater cloud cover in Ishwardi, selecting solar energy systems in combination with wind or storage ones might be a wise choice for the whole year. Policymakers must be aware of micro-regional climatic differences when generating support for renewable energy and defining objectives for solar installations. The absence of high-resolution irradiance or panel efficiency statistics creates a significant gap in the data sets. Having additional stations for localized weather observation and extending remote-sensing abilities will make policies more reliable and ensure renewable energy is deployed where it is required in Bangladesh. 3.10 Limitations of the Study This study includes valuable insights regarding regional temperature, solar energy, and AI predictions for energy planning; however, several issues still hinder the results from being efficiently utilized or expanded. To begin with, proxy measures such as the clearness index, shortwave irradiance, and albedo are employed in the research to estimate solar energy capacity. While experts rely on these measures without PV output, they may not always accurately represent how solar panels perform in practice. Factors such as the angle of the solar panels, the amount of shading in the area, dust accumulation, and real-time panel effectiveness were not considered due to data limitations. Furthermore, the data needed for site selection was obtained from remote sensing archives, such as NASA-POWER, which, while providing accurate and reliable results, often lacks crucial information on fine local characteristics, particularly in highly developed or hilly areas. Although LSTM models performed better than standard linear regression in predicting temperatures, the research limited its approach to using only temperature time series without additional parameters. Incorporating multiple inputs, such as humidity, irradiance, and cloud cover, could strengthen the forecasting process and better represent the critical relationships in renewable energy systems. Changes in climate, such as sudden monsoons or cloud cover caused by cyclones, were not considered in the model. While these events do not occur consistently, they impact solar availability and require more complex computations using multiple models for optimal results. The conclusions of this research cannot be generalized to a global scale due to the region-specific nature of the study. Many other areas in Bangladesh face distinct challenges related to infrastructure and weather, and studying this concept in those regions will provide a more comprehensive national picture. Effectively harnessing real-time solar energy, developing models for panel performance, and addressing various climatic factors are viable next steps for researchers. Moreover, experts should review field findings and engage stakeholders to ensure the ideas translate into effective policy and practical solar planning. 4 Conclusion and Discussion The research studied the climate in Rajshahi and Ishwardi and estimated how efficiently solar energy can be utilized in these regions. Based on recorded temperatures and satellite sunlight readings, the author employed statistical approaches and AI to analyze the trends, variability, and potential for renewable energy. Both sections of Bangladesh have been found to warm at the same pace, with a temperature increase of 0.024°C/year in Rajshahi and 0.020°C/year in Ishwardi. These results are consistent with what is known about climate change and suggest that the area requires additional energy plans that can manage climate change. Forecasting using LSTM neural networks was considerably better than what could be obtained with linear regression. The RMSE for Rajshahi was 0.25°C and 0.22°C for Ishwardi, indicating that LSTM models effectively caught the varied changes in climate over time and across seasons. This indicates that applying deep learning methods aids in foreseeing climate change and that AI is necessary in energy planning systems. Rajshahi got a larger quantity of sunshine and had less cloud cover than Ishwardi in terms of solar power, suggesting that Rajshahi might be a better location for PV. Rajshahi’s insolation clearness index suggests the atmosphere is more translucent than elsewhere. Through boxplots, bar charts, and radar plots, these insights were offered, and the Stability and amplitude of the climatic parameters were visible in diverse places. Both places offer strong opportunities for solar expansion, although Rajshahi is significantly superior thanks to its brighter sky and consistent irradiance. During the months when there is less rain, the qualities become more beneficial since solar panels provide more power. This research shows why intelligent climate forecasting and regional climate diagnostics should be incorporated in developing national energy strategies. In Bangladesh, which wants to meet its renewable energy targets as specified in the SDGs, LSTM modeling and solar feasibility mapping are vital for prioritizing infrastructure development and sustaining constant energy security. However, there are constraints to this research investigation. Instead of utilizing actual information from PV plants, it based the model on satellite-recorded data, hence it omitted any conceivable impacts of dust, particles, and humidity. Consequently, the conveniences of economics and infrastructure did not impact where the locations were picked. In the future, we should incorporate real radiation sensors, reliable information on solar power plants’ production, and input from stakeholders to transform analysis into practical choices. Making more projections based on various policy situations and genuine climate data would help even more people discover solutions. Declarations ORCID Hasan Ahamed Alif - https://orcid.org/0009-0003-2456-0379 Funding This study did not obtain financing from any governmental, private, or not-for-profit source, and there was no outside support for financing the research. Author Contribution H.A.A. planned the study, obtained and examined the data, constructed the models, and analysed the data as well as generated all figures and visualisations in addition to drafting the text. H.A.A. glanced over and gave his permission on the final publishing of the work. Acknowledgement I want to thank Dhaanish Ahmed College of Engineering for allowing me to undertake this independent study as an undergraduate Student. I was in charge of the complete process, from designing, reviewing the data, developing models, visualizing, to ultimately documenting my work. I am also thankful to NASA-POWER and the Meteorological Department of Bangladesh (BMD) for offering various environmental data that allowed me to conduct this research. I studied Python-based libraries and AI approaches by myself and consistently applied them precisely to preserve precision and repeatability in the job. In addition, I appreciate other academic specialists and available resources outside my field, which played a role in increasing mutual knowledge and collaboration in this area of study. References Abbass K, Qasim MZ, Song H, Murshed M, Mahmood H, Younis I (2022) A review of the global climate change impacts, adaptation, and sustainable mitigation measures. Environ Sci Pollut Res 29(28):42539–42559. https://doi.org/10.1007/s11356-022-19718-6 Al Mamun A, Islam AR, Md. T, Alam GMM, Sarker MNI, Erdiaw-Kwasie MO, Bhandari H, Mallick J (2023) Livelihood vulnerability of char land communities to climate change and natural hazards in Bangladesh: An application of livelihood vulnerability index. Nat Hazards 115(2):1411–1437. https://doi.org/10.1007/s11069-022-05599-y Al-Maruf A, Mira SA, Rida TN, Rahman MS, Sarker PK, Jenkins JC (2021) Piloting a Weather-Index-Based Crop Insurance System in Bangladesh: Understanding the Challenges of Financial Instruments for Tackling Climate Risks. Sustainability 13(15):8616. https://doi.org/10.3390/su13158616 Choi Y-W, Campbell DJ, Aldridge JC, Eltahir EAB (2021) Near-term regional climate change over Bangladesh. Clim Dyn 57(11–12):3055–3073. https://doi.org/10.1007/s00382-021-05856-z De Laurentis C, Pearson PJG (2021) Policy-relevant insights for regional renewable energy deployment. Energy Sustain Soc 11(1):19. https://doi.org/10.1186/s13705-021-00295-4 Dellaripa PF, Bush T, Miller FW, Feldman CH (2023) The Climate Emergency and the Health of Our Patients: The Role of the Rheumatologist. Arthritis Rheumatol 75(1):1–3. https://doi.org/10.1002/art.42279 Genc TS, Kosempel S (2023) Energy Transition and the Economy: A Review Article. Energies 16(7):2965. https://doi.org/10.3390/en16072965 Haupt SE, McCandless TC, Dettling S, Alessandrini S, Lee JA, Linden S, Petzke W, Brummet T, Nguyen N, Kosović B, Wiener G, Hussain T, Al-Rasheedi M (2020) Combining Artificial Intelligence with Physics-Based Methods for Probabilistic Renewable Energy Forecasting. Energies , 13 (8), 1979. https://doi.org/10.3390/en13081979 Hébert R, Herzschuh U, Laepple T (2022) Ocean temperature fluctuations overprint millennial-scale climate variability over land. Nat Geosci 15(11):899–905. https://doi.org/10.1038/s41561-022-01056-4 Hultman NE, Clarke L, Frisch C, Kennedy K, McJeon H, Cyrs T, Hansel P, Bodnar P, Manion M, Edwards MR, Cui R, Bowman C, Lund J, Westphal MI, Clapper A, Jaeger J, Sen A, Lou J, Saha D, O’Neill J (2020) Fusing subnational with national climate action is central to decarbonization: The case of the United States. Nat Commun 11(1):5255. https://doi.org/10.1038/s41467-020-18903-w Miskat MI, Sarker P, Chowdhury H, Chowdhury T, Rahman MS, Hossain N, Chowdhury P, Sait SM (2023) Current Scenario of Solar Energy Applications in Bangladesh: Techno-Economic Perspective, Policy Implementation, and Possibility of the Integration of Artificial Intelligence. Energies 16(3):1494. https://doi.org/10.3390/en16031494 Munjer MA, Hasan MZ, Hossain MK, Rahman MF (2023) The Obstruction and Advancement in the Sustainable Energy Sector to Achieve SDG in Bangladesh. Sustainability 15(5):3913. https://doi.org/10.3390/su15053913 Novas N, Garcia RM, Camacho JM, Alcayde A (2021) Advances in Solar Energy towards Efficient and Sustainable Energy. Sustainability 13(11):6295. https://doi.org/10.3390/su13116295 Song J, Tong G, Chao J, Chung J, Zhang M, Lin W, Zhang T, Bentler PM, Zhu W (2023) Data-driven pathway analysis and forecast of global warming and sea level rise. Sci Rep 13(1):5536. https://doi.org/10.1038/s41598-023-30789-4 Talut M, Bahaj AS, James P (2022) Solar Power Potential from Industrial Buildings and Impact on Electricity Supply in Bangladesh. Energies 15(11):4037. https://doi.org/10.3390/en15114037 Zhou Z, Tang W, Li M, Cao W, Yuan Z (2023) A Novel Hybrid Intelligent SOPDEL Model with Comprehensive Data Preprocessing for Long-Time-Series Climate Prediction. Remote Sensing , 15 (7), 1951. https://doi.org/10.3390/rs15071951 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 27 Sep, 2025 Read the published version in Theoretical and Applied Climatology → Version 1 posted Editorial decision: Revision requested 23 Jul, 2025 Reviews received at journal 23 Jul, 2025 Reviews received at journal 22 Jul, 2025 Reviews received at journal 17 Jul, 2025 Reviewers agreed at journal 14 Jul, 2025 Reviewers agreed at journal 13 Jul, 2025 Reviewers agreed at journal 12 Jul, 2025 Reviewers agreed at journal 12 Jul, 2025 Reviewers invited by journal 11 Jul, 2025 Editor assigned by journal 03 Jul, 2025 Submission checks completed at journal 03 Jul, 2025 First submitted to journal 02 Jul, 2025 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-7031666","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":484849852,"identity":"80a8799b-d318-42d0-b863-ffcff0636948","order_by":0,"name":"Hasan Ahamed Alif","email":"data:image/png;base64,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","orcid":"","institution":"Dhaanish Ahmed College of Engineering","correspondingAuthor":true,"prefix":"","firstName":"Hasan","middleName":"Ahamed","lastName":"Alif","suffix":""}],"badges":[],"createdAt":"2025-07-02 17:23:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7031666/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7031666/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00704-025-05746-y","type":"published","date":"2025-09-27T15:57:47+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":86780325,"identity":"8a4baf0a-43b1-4784-858f-c33bb06fed72","added_by":"auto","created_at":"2025-07-15 13:19:31","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":137829,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation heatmaps of key climate indicators comparing Rajshahi and Ishwardi regions of Bangladesh..\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7031666/v1/6845d3edd8b52e36125ccd3f.png"},{"id":86780327,"identity":"4a790965-33f3-444e-b1f4-5954de5f1185","added_by":"auto","created_at":"2025-07-15 13:19:31","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":837936,"visible":true,"origin":"","legend":"\u003cp\u003eHistograms of monthly distributions for key climate parameters in the Rajshahi region of Bangladesh.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7031666/v1/70840447c501264bf2108a68.png"},{"id":86781399,"identity":"7f3f4c8e-48b8-402f-94ad-6f5fcfe11266","added_by":"auto","created_at":"2025-07-15 13:27:31","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":971941,"visible":true,"origin":"","legend":"\u003cp\u003eHistograms of monthly distributions for key climate parameters in the Ishwardi region of Bangladesh.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7031666/v1/a68e87938aed653a34526241.png"},{"id":86780326,"identity":"f823d418-2159-4775-88ff-1daf92a358dd","added_by":"auto","created_at":"2025-07-15 13:19:31","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":180271,"visible":true,"origin":"","legend":"\u003cp\u003eDecadal \u0026nbsp;\u0026nbsp;Trends of Annual Average Temperature in Rajshahi and Ishwardi (1980–2020)\u003cbr\u003e\n \u0026nbsp;(a–e) Line graphs demonstrate the yearly average temperature spanning five \u0026nbsp;\u0026nbsp;decades, contrasting Rajshahi (green) to Ishwardi (blue). Each subplot \u0026nbsp;\u0026nbsp;represents one decade. Rajshahi routinely displays somewhat higher \u0026nbsp;\u0026nbsp;temperatures, and both locations show a progressive warming trend over time, \u0026nbsp;\u0026nbsp;with sharper rises reported in the 2000s and 2010s. These patterns are \u0026nbsp;\u0026nbsp;symptomatic of increasing climatic circumstances and underscore the necessity \u0026nbsp;\u0026nbsp;for adaptive planning in climate-sensitive industries.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7031666/v1/5d0ad7e1b69b7b338ac9db66.png"},{"id":86781774,"identity":"9620aa16-52ab-42cd-a8ae-156b7fcd0a0e","added_by":"auto","created_at":"2025-07-15 13:35:31","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":91365,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3. Coefficient regression values for different \u0026nbsp;\u0026nbsp;climate predictors within each climate region. (a) Boxplots indicating the \u0026nbsp;\u0026nbsp;range of coefficient regressions (variable scale of y-axis). (b). Maps showing \u0026nbsp;\u0026nbsp;the spatial distribution of the coefficient of regression values for \u0026nbsp;\u0026nbsp;different climate variables. (c) Bar charts of the significance ratio (ratio \u0026nbsp;\u0026nbsp;of stations with significant response at 5% to the total number of stations) \u0026nbsp;\u0026nbsp;of the climatic variables within each climate region.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7031666/v1/f9f55d3de1ec47ec22df39b9.png"},{"id":92431773,"identity":"1ef1318a-d87b-48f3-b05d-00f7b26b9f44","added_by":"auto","created_at":"2025-09-29 16:10:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2953200,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7031666/v1/0bc5d310-1aab-4818-9468-97aad43824ac.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Temperature-Based Renewable Energy Forecasting: A Big Data Analysis for Sustainable Energy Planning in Bangladesh","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eClimate change denotes a substantial and enduring alteration of global weather patterns, impacting areas as varied as the tropics and the polar regions(Abbass et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Climate change has emerged as a significant global problem affecting corporations, individuals, and the environment(Song et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Temperature significantly influences climate change by affecting weather patterns, ecosystems, food security, air quality, and mental health(Dellaripa et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The requirement of lowering environmental externality, the energy security issue, the economic benefit, affordability, accessibility, and the sustainability agenda are forcing the globe to resort to the fast adoption of renewable energy sources.(Genc \u0026amp; Kosempel, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Solar energy is categorized as a clean energy source that reduces greenhouse gas emissions compared to fossil fuels(Novas et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Bangladesh must devise innovative solutions to its energy challenges, which are intricately linked to its climate vulnerability, to attain sustainable development and climate resilience(Al-Maruf et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Bangladesh's rapid economic development and population expansion have resulted in a significant increase in energy consumption, posing a critical challenge in supplying energy to its industry and citizens(Miskat et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Bangladesh, with its geographical limits mixed with socioeconomic dependency and inadequate access to resources, is among the most susceptible nations to climate-related challenges.(Al Mamun et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Bangladesh's government has suggested high objectives for renewable energy to strengthen its energy sector and fulfill sustainable development goals (SDGs)(Munjer et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Regional energy planning is crucial for formulating context-specific and pragmatic energy policies that promote sustainable development and the use of renewable energy(De Laurentis \u0026amp; Pearson, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). As Bangladesh wants to grow solar adoption, it confronts constraints relating to land scarcity, economic sustainability, and grid stability(Talut et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The absence of region-specific evaluations of climatic adaptability in Bangladesh restricts the capacity to design effective adaptation solutions(Choi et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Traditional forecasting methods frequently depend on historical data and statistical approaches to anticipate future climatic conditions(Zhou et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Inadequate long-term climate data at the subnational level, mainly as a consequence of the use of indirect proxies, the limitations of the models that are now in use, and the challenges in validating them against data that has been observed(H\u0026eacute;bert et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Data restrictions, inconsistencies, and a comprehensive strategy to enhance data gathering and analysis are the key reasons for the paucity of climate data-driven research at the sub-national level(Hultman et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). There is a shortage of concentrated research that notably targets the distinct possibilities and problems that occur at sub-national levels, which might generate a gap in our knowledge of localized energy dynamics, where the bulk of research using AI in renewable energy forecasting mainly concentrates on national or large regional dimensions(Haupt et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The proposed study is an effort to explore in detail the previous climate behavior and its repercussions on the feasibility of solar energy in two climatically diverse sites of Bangladesh- Rajshahi and Ishwardi. Using more than 40 years of temperature and solar-related environmental variables, the research employs a standard linear regression and state-of-the-art deep learning networks (particularly Long Short-Term Memory networks) to extract and forecast climatic fluctuations. Through this, the research can evaluate long-term warming trends and combine proxy variables like shortwave irradiance, clearness index, albedo, and cloud cover to estimate the solar feasibility of the areas involved. The basic purpose is to generate data-driven, spatially sensitive information that would ease the planning of sustainable energy and offer evidence-based policies about the deployment of renewable energy in the delicate climate of Bangladesh.\u003c/p\u003e\u003cp\u003eThis examination integrates long-term climate prediction and important solar viability indicators, thus reducing the gap between environmental information examination and practical Bangladesh energy planning. This study highlights the benefit of employing state-of-the-art deep learning approaches, particularly Long Short-Term Memory (LSTM) models, to depict nonlinear climatic patterns than more conventional statistical methods. More crucially, it is transforming the complex climatic dynamics into intelligence that might be acted upon in deploying solar energy, especially in locations like Rajshahi and Ishwardi, where optimization of resources is a big problem. The findings have been presented in the context of the national discussion on sustainable infrastructure, and they give a framework that may be copied in other places to plan renewable energy production at the local scale in the shifting climatic circumstances.\u003c/p\u003e\u003cp\u003eIn order to fulfill the research aims, we designed a hybrid analytical system that includes old statistics and recent AI methodologies. The linear regression was performed to calculate linear trends of baseline warming using historical temperature data between 1980 and 2020. Predictive modeling used Long Short-Term Memory (LSTM) neural networks to capture more complicated and nonlinear temporal dynamics. Also, climatic characteristics that include shortwave irradiance, clearness index, cloud cover, surface albedo, and temperature collected by NASA-POWER datasets were utilized to assess the feasibility of solar energy. Such a combination of methodologies enabled a consistent evaluation of the designated locations' climatic patterns and solar potential.\u003c/p\u003e"},{"header":"2. Methodology","content":"\u003cp\u003eWe use these regions to summarize and examine the outcome of this investigation.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Data Source and Description\u003c/h2\u003e\u003cp\u003eThe research uses forty-year temperature data to evaluate the climate's influence on solar energy planning. The dataset's information was sourced from BMD, which documents meteorological occurrences in various regions of Bangladesh. Rajshahi and Ishwardi were selected for investigation due to their abundant sunshine, significance in agriculture, and pivotal involvement in renewable energy initiatives. The dataset contains daily maximum and minimum temperatures for each year from 1980 to 2020. The data were streamlined and organized by transforming them into monthly and annual average temperatures. Long-term data facilitates the identification of alterations in meteorological patterns, seasonal variations, and related factors that may impact solar energy initiatives. This case study evaluates the efficacy of solar photovoltaic technology in western Bangladesh using meteorological data. In addition, analyzing these two areas helps us compare inland climatic changes, which counts a lot for planning regional climate-related measures. It endorses the United Nations objectives for sustainable development (SDG 7 and SDG 13), emphasizing the accessibility and environmental sustainability of energy. The data spanning many decades (1980\u0026ndash;2020) enables the identification of progressive warming trends, heightened temperature variability, and increased heatwave frequency, all of which are critical for renewable energy systems.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Computation of precipitation variability\u003c/h2\u003e\u003cp\u003eSeveral processes were required to prepare the temperature data so that analysis could be done with certainty. The data was first acquired from the Bangladesh Meteorological Department (BMD), covering daily maximum and minimum temperatures for Rajshahi and Ishwardi from 1980 to 2020.\u003c/p\u003e\u003cp\u003eAt first, gaps in the data were discovered and controlled by applying linear interpolation for those shorter holes. This step protected the time series from having unexpected or inexplicable changes. A basic statistical approach called Z-score analysis was employed to locate outliers. Extreme and well-documented measurements, for example, those signifying a heatwave, were left in the data to ensure the continuity of climatic patterns. All the data were then organized into monthly and yearly records so the analysis could be done properly. By combining the records, the researchers could observe how temperatures have changed over lengthy periods, which is essential for energy planning. Using Min-Max scaling, the raw temperature readings were modified for forecasting. This phase was crucial in ensuring that any machine learning models or statistical forecasts would be compatible. Thus, preprocessing the data made the information usable for credible energy estimates and regional analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Tools and Technologies Used\u003c/h2\u003e\u003cp\u003eData visualization and analysis were executed utilizing Python 3.11, and all the work was done within the Jupyter Notebook environment. Several typical scientific libraries were employed in the study. In my work, Pandas assisted with organizing, generating, and manipulating data, Matplotlib and Seaborn were used to illustrate general temperature variations, seasonal differences, and any odd data points. At the same time, NumPy was selected for introductory statistics. Analysis was further done using Google Colab, which made computing easier by offering quick access to its resources for code creation and execution. Matplotlib and Seaborn were used to produce all visual plots, ensuring they have a comparable appearance for simple interpretation. Because these approaches are versatile and practical, they were necessary for identifying temperature patterns that help forecast renewable energy in our research.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Analytical Approach\u003c/h2\u003e\u003cp\u003eWe examined past temperature patterns to forecast the climate's future behavior using conventional statistical approaches and artificial intelligence techniques. Linear regression was employed as a reference model; however, we relied on LSTM neural networks to discover complicated, irregular changes throughout time. Applying this strategy lets us verify the efficacy of deep learning for climate-dependent activities such as renewable energy planning.\u003c/p\u003e\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\u003ch2\u003e2.4.1 Baseline Trend Analysis using Linear Regression\u003c/h2\u003e\u003cp\u003eLinear regression follows a pattern as time passes. It is described as when one statement is added to another statement.\u003c/p\u003e\u003cp\u003eThe first equation is y\u0026thinsp;=\u0026thinsp;β₀ + β₁x\u0026thinsp;+\u0026thinsp;ε. \u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;. (1)\u003c/p\u003e\u003cp\u003eWhere:\u003c/p\u003e\u003cp\u003ey\u0026thinsp;=\u0026thinsp;The average temperature of a place for the whole year\u003c/p\u003e\u003cp\u003ex represents the year (for example, 1980, 1981, ...)\u003c/p\u003e\u003cp\u003eConstantly or Intercept\u0026thinsp;=\u0026thinsp;β₀\u003c/p\u003e\u003cp\u003eThe slope of the line y\u0026thinsp;=\u0026thinsp;β₁x is denoted β₁.\u003c/p\u003e\u003cp\u003eε\u0026thinsp;=\u0026thinsp;the unexpected component of the model\u003c/p\u003e\u003cp\u003eLinear regression was applied to the data from Rajshahi and Ishwardi to determine the baseline warming trends.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\u003ch2\u003e2.4.2 Deep Learning-Based Forecasting using LSTM\u003c/h2\u003e\u003cp\u003eSince linear models could not handle all the disadvantages, we created an LSTM network as an alternative. LSTM may reflect patterns that continue over a long time, alter with the seasons, and are not linear.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePreprocessing Steps\u003c/b\u003e:\u003c/p\u003e\u003cp\u003eMin-Max Normalization:\u003c/p\u003e\u003cp\u003eX_norm = (X\u0026thinsp;\u0026minus;\u0026thinsp;Xmin) / (Xmax\u0026thinsp;\u0026minus;\u0026thinsp;Xmin )\u0026hellip;\u0026hellip;. (2)\u003c/p\u003e\u003cp\u003eSliding window transformation is used in supervised learning:\u003c/p\u003e\u003cp\u003eInput X = [T(t\u0026thinsp;\u0026minus;\u0026thinsp;5), ..., T(t\u0026thinsp;\u0026minus;\u0026thinsp;1)] - the output is T(t)\u003c/p\u003e\u003cp\u003eDataset Split: A part of 80% was utilized for training, and the balance, 20%, was set aside for testing.\u003c/p\u003e\u003cp\u003e\u003cb\u003eLSTM Architecture\u003c/b\u003e:\u003c/p\u003e\u003cp\u003eThe history of this part comprises five timesteps.\u003c/p\u003e\u003cp\u003eThere are 50 LSTM units in the LSTM Layer.\u003c/p\u003e\u003cp\u003eDropout: 0.2\u003c/p\u003e\u003cp\u003eThe last layer of the network has several nodes.\u003c/p\u003e\u003cp\u003eLoss: Mean Squared Error\u003c/p\u003e\u003cp\u003eOptimizer: Adam\u003c/p\u003e\u003cp\u003eEarly halting was utilized after 100 epochs.\u003c/p\u003e\u003cp\u003e\u003cb\u003eEvaluation Metric\u003c/b\u003e:\u003c/p\u003e\u003cp\u003eWe rely on Root Mean Square Error (RMSE) to analyze the forecasting models' performance, since it is a well-known tool in the domain.\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:RMSE=\\sqrt{MSE}=\\sqrt{\\frac{1}{N}\\sum\\:_{i=1}^{N}{({y}_{i}-{\\widehat{y}}_{i})}^{2}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe root square of MSE is the primary source of RMSE, which is relatively easy to interpret and offers information about average prediction errors.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Renewable Energy Mapping\u003c/h2\u003e\u003cp\u003eThis paper adds solar energy potential estimates from proxies to the standard techniques of analyzing climates in Rajshahi and Ishwardi. Since we lacked ground-based solar information and extremely detailed solar energy performance from photovoltaic systems, we leaned on NASA POWER for satellite-collected environmental indicators. For instance, surface shortwave radiation, atmospheric clearness indices, cloud cover, albedo, and near-surface air temperature are some of them. It is a mechanism to examine whether solar can function in scarce-data locations using key climatic data like temperature and humidity.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Parameters Considered\u003c/h2\u003e\u003cp\u003eFive climatic characteristics for Rajshahi and Ishwardi were selected as they are known to impact solar energy potential: all-sky surface shortwave downward irradiance, insolation clearness index, cloud quantity, all-sky surface albedo, and temperature at 2 meters. The results were gathered from monthly data spanning many years. All of this information is accessible by merging the information on sunlight, sky clarity, and how much radiation the Earth\u0026rsquo;s surfaces reflect.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Rationale and Assumptions\u003c/h2\u003e\u003cp\u003eThe logic suggests that more shortwave irradiance and decreased severity of cloud cover result in increased solar energy harvest. The clearness index offers transparency information, while albedo gives the fraction of sunlight reflected by the Earth back to space. By reviewing earlier portions of this text, we can claim that Rajshahi and Ishwardi areas are experiencing greater temperatures each year, suggesting better skies during the dry seasons, which makes solar PV technology more practicable.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e2.8 Observational Insights from the Dataset\u003c/h2\u003e\u003cp\u003eTo learn about the environment\u0026rsquo;s fundamental appropriateness, the relevant criteria were summarized for both research sites. In contrast, Rajshahi had greater average daily solar radiation (5.17\u0026thinsp;\u0026plusmn;\u0026thinsp;0.21 kWh/m\u0026sup2;/day), while being cloudier for less time (45.2\u0026thinsp;\u0026plusmn;\u0026thinsp;6.1%) as compared to Ishwardi (5.09\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18 kWh/m\u0026sup2;/day; 46.7\u0026thinsp;\u0026plusmn;\u0026thinsp;6.5%). Inflowing water at both sites exhibited extremely comparable clearness indices (around 0.60\u0026ndash;0.61) and albedo values (around 0.19\u0026ndash;0.20), demonstrating that they both filtered radiation and reflected surface light in similar proportions. With these measures, judgments were taken to concentrate on various locations for renewable energy.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e2.9 Integration with Climate-Aware Planning\u003c/h2\u003e\u003cp\u003eIt also discusses how these elements might be coupled to LSTM-based weather forecasts to build adaptable solar energy plans. You should utilize tools that highlight the ideal times to gather solar energy, put solar panels in places with better weather conditions, and add weather-based activities into the algorithm.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e2.10 Limitations\u003c/h2\u003e\u003cp\u003eThis measuring approach is always constrained by its proxy character. Using forecasting models instead of real-time data frequently leads to inaccuracies. Besides, the impacts of concentrated aerosols in the air, dust on the surface, and unique effects on individual panels are not considered, limiting the model\u0026rsquo;s usefulness for practical application. Still, it offers a realistic beginning point for looking at the potential of solar energy in such places.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results and discussion","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Overview of Climate Parameters\u003c/h2\u003e\n \u003cp\u003eThis section examines the disparities in key climatic characteristics influencing the solar energy potential of Rajshahi and Ishwardi. The factors include the magnitude of shortwave descending radiation reaching the Earth\u0026apos;s surface, the insolation clearness index, cloud cover, surface albedo, and air temperature. We use statistics and graphics to elucidate the circumstances and analyze the climate from various perspectives.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Correlation Heatmap\u003c/h2\u003e\n \u003cp\u003eThe following correlation matrices in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e present an overview of the pair-wise relationship of five important climatic variables, namely clearness index, surface shortwave downward irradiance, surface albedo, cloud quantity, and 2 m air temperature at both Rajshahi and Ishwardi. The clearness index and irradiance exhibit a significant positive association (r\u0026thinsp;\u0026gt;\u0026thinsp;0.9) in both areas, demonstrating that cleaner skies immediately predict more incident solar energy. The number of clouds is closely connected with both irradiance (r approx. -0.85) and clearness (r approx. -0.80), further highlighting the existence of cloud cover as a satellite of the solar resource availability. Albedo of the surfaces has a weak positive correlation with irradiance (r\u0026thinsp;\u0026asymp;\u0026thinsp;0.3) and a weak negative correlation with cloud amount (r approximately equal to -0.2), which means that higher surface albedo can slightly increase reflected radiation. However, it is also partly covered by clouds. The air temperature is proportionally (but not substantially) connected to irradiance (r\u0026thinsp;\u0026asymp;\u0026thinsp;0.5) and clearness (r\u0026thinsp;=\u0026thinsp;0.2), such that warmer days are sunnier, but temperature changes also react to other meteorological factors. The comparison of the two places reveals relatively comparable patterns. However, the cloud-irradiance relation in Ishwardi is significantly stronger, perhaps due to the local peculiarities in the monsoonal cloud processes. These inter-parameter correlations combined reveal that the clearness index and temperature are the major predictors in our renewable-energy forecasting model, and they also illustrate how albedo and cloud cover may be utilized together to complement each other when site-specific energy output is to be forecasted.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Histogram Rajshahi\u003c/h2\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e depicts the monthly frequency distributions of five important climatic indicators in Rajshahi that suggest substantial seasonal dynamics that constitute the foundation of renewable-energy potential in the area. The clearness index likewise reaches its maximum in the range of 0.50 to 0.60 during the dry season (December to February) and correlates to irradiance more than 5.5 kW h m-2 day-1 1 suggesting a clear sky window when solar harvesting would be beneficial. On the other hand, the clearness and irradiance distributions shift to the left as cloud cover rises during the early monsoon season (June to August). This is because cloud-amount frequencies rise into the 30\u0026ndash;80% frequencies with modal values in July and August, which mutes surface solar input. Surface albedo is unusually stable throughout the year at 0.13\u0026ndash;0.15, demonstrating no seasonal fluctuation in reflectance despite slightly higher values during pre-monsoon dry months. Histograms of air temperatures range from modal values of about 15\u0026deg;C in January to values of approximately 32\u0026deg;C in May, and depict the change of Rajshahi between searing pre‐monsoon heat and cold winter climatic conditions. These linked distributions describe a high-resource dry season, a cloud-dominated monsoon season, and a steady albedo environment, which requires empirical context, temperature-based forecasting, and optimal system design.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4 Histogram Ishwardi\u003c/h2\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e illustrates the stacked monthly histograms of five climatic indicators in Ishwardi, which display prominent seasonal trends. The clearness index is concentrated in the range of 0.45 to 0.60, with the higher values happening in the dry winter season (December\u0026ndash;February), and during monsoon (June\u0026ndash;August), the cloud cover frequencies approach 50%. Surface irradiance follows the same pattern with the most significant levels (5.0\u0026ndash;6.0 kW-h m-2 day-1) in spring (March-April) and considerable left-shift during monsoon clouds. Albedo is closely centered between 0.11 and 0.14 throughout the year, suggesting that the reflectance of the surface does not fluctuate considerably with the seasonal moisture fluctuations. Cloud‐amount histograms encompass the whole 0\u0026ndash;100% range, but are concentrated between 20% and 70% in the noncore monsoon, rather than over 80% in July and August. Air temperaature patterns fluctuate from frigid January patterns of around 17\u0026deg;C to hot April-May patterns of about 32\u0026deg;C, with the monsoon months having temperatures that are tempered (24\u0026deg;C to 28\u0026deg;C). Combined, the distributions indicate the ability of Ishwardi to receive clear skies throughout the winter and spring seasons, the inhibitory monsoon season on irradiance, and the relatively uniform albedo environment, which are essential elements to incorporate in regional renewable‐energy Prediction.\u003c/p\u003eTable 1. Summary statistics of key climate indicators in Bangladesh\u0026apos;s Rajshahi and Ishwardi regions. Values are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation, based on annual averages over the study period. Rajshahi consistently shows slightly higher shortwave irradiance and lower cloud cover, indicating comparatively more substantial solar potential.\u003cbr\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tabd\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eIndicator\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRajshahi (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eIshwardi (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTemperature at 2m (\u0026deg;C)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26.42\u0026thinsp;\u0026plusmn;\u0026thinsp;0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26.22\u0026thinsp;\u0026plusmn;\u0026thinsp;0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eShortwave Irradiance (kWh/m\u0026sup2;/day)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.56\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.56\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsolation Clearness Index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCloud Amount (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e55.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e55.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurface Albedo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.13\u0026thinsp;\u0026plusmn;\u0026thinsp;0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.13\u0026thinsp;\u0026plusmn;\u0026thinsp;0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n \u003ch2\u003e3.5 Temporal Temperature Trends (1980\u0026ndash;2020)\u003c/h2\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eThe average temperature variations per year were investigated by projecting them throughout 5 decades from the 1980s to the 2020s (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e). Every subplot depicts how temperatures vary\u003c/p\u003e\n \u003cp\u003efrom year to year, and what transpired in different places throughout the 10 years.\u003c/p\u003e\n \u003cp\u003eIn Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea, there are not many temperature variations from one year to the next over the 1980s timeframe. Even if the trends were identical, Rajshahi experienced somewhat higher temperatures. In the years following the 1990s (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eb\u0026ndash;c), a steady temperature rise is obvious to discern. It can be observed that there are tiny oscillations, perhaps owing to both local weather patterns and broader atmospheric shifts.\u003c/p\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ed shows that both the temperature and range of temperature values soared in the 2000s, suggesting that the swing from cold to hot temperatures got more severe. The situation in Rajshahi is increasingly urgent due to changes in land usage and the city\u0026apos;s expansion. Figure\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ee demonstrates the steepest warming between the 2010s and 2050s, showing a warming rate of around +\u0026thinsp;0.024\u0026deg;C/year in Rajshahi and +\u0026thinsp;0.020\u0026deg;C/year in Ishwardi, as discovered.\u003c/p\u003e\n \u003cp\u003eThe information by decade highlights the necessity of observing how climate change proceeds, particularly for the future of solar and heat-ready constructions. Increasing mean temperatures show that LSTM is the best option for predicting, implying that the climate in both areas will get warmer.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n \u003ch2\u003e3.6 Temperature Forecast Accuracy\u003c/h2\u003e\n \u003cp\u003eTo determine how trustworthy statistical predicting techniques are for long-term fluctuations in Rajshahi and Ishwardi\u0026rsquo;s climate, linear regression was done using their historic temperature data from 1980 to 2020. The research was mainly targeted at assessing how successfully a simplistic linear model could simulate weather patterns over multiple years.\u003c/p\u003e\n \u003cp\u003eIn summary, linear regression may provide a rudimentary concept of the warming trend, even if it is not particularly good\u003c/p\u003e\n \u003cp\u003eat forecasting it. Because the RMSE values are comparable, the exact consistent causes play a role in both climates.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"char\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eThis table presents the Root Mean Squared Error (RMSE) values for the linear regression models forecasting annual average temperatures in Rajshahi and Ishwardi. RMSE serves as a standard metric for evaluating model accuracy, with lower values indicating better performance\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRegion\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRMSE\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRajshahi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.836\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIshwardi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.847\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eAnother look at the temperature data indicates that a linear model is not excellent at managing bigger changes between one year\u0026rsquo;s temperature and the next. That is sensible, since linear regression is confined to circumstances where change happens continuously and cannot handle rapid climate changes, which calls for employing stronger models like LSTM neural networks.\u003c/p\u003e\n \u003cp\u003eLooking forward, it is expected that LSTM will fare better than linear regression in expressing the flexible and always-changing elements in climate data. Even so, the linear model\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eHelps detect general changes in temperature and ensure that they are trending in a given direction.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\n \u003ch2\u003e3.7 Solar Energy Feasibility Comparison\u003c/h2\u003e\n \u003cp\u003eFor the climatic parameters of solar energy, the data on the bar chart indicate the mean values for shortwave irradiance, insolation clearness index, cloud quantity, surface albedo, and temperature at 2 meters for both Rajshahi and Ishwardi. It briefly reviews whether solar energy can operate in both places. The daily irradiation from the sun is greater in Rajshahi than in Ishwardi. Higher exposure contributes to the potential of producing more PV energy. Similarly, the clearness index values for sun irradiation are comparable in Rajshahi and Ishwardi. Because the atmospheric index is clearer in Rajshahi, the sun\u0026rsquo;s energy reaches the land more quickly, suggesting that Rajshahi is ideally suited for solar power. The presence of clouds is usually greater in Ishwardi, lowering the consistency and power of sunshine. Rajshahi enjoys more steady sun radiation as it has fewer clouds than Dhaka. Ishwardi reflects more sunlight because its surface albedo is larger than that of other regions. However, as there is less solar energy reflection in Rajshahi, this could marginally aid\u003c/p\u003e\n \u003cp\u003ein getting a better outcome with PV panels. The temperature in both places remained practically the same, Rajshahi simply being\u003c/p\u003e\n \u003cp\u003eslightly hotter on average. High temperatures do not affect how much PV generates, yet they could modify the system\u0026rsquo;s efficiency and heating, and these aspects must be looked at in the design.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eSummary of Solar Energy Indicators\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eIndicator\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eStd Dev\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRegion\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eShortwave Irradiance (kWh/m\u0026Acirc;\u0026sup2;/day)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRajshahi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsolation Clearness Index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRajshahi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCloud Amount (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e32.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRajshahi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurface Albedo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRajshahi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTemperature at 2m (\u0026Acirc;\u0026deg;C)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e17.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRajshahi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eShortwave Irradiance (kWh/m\u0026Acirc;\u0026sup2;/day)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIshwardi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsolation Clearness Index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIshwardi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCloud Amount (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e31.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIshwardi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurface Albedo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIshwardi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTemperature at 2m (\u0026Acirc;\u0026deg;C)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIshwardi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\n \u003ch2\u003e3.8 Regional Comparison: Rajshahi vs. Ishwardi\u003c/h2\u003e\n \u003cp\u003eIt pulls together a study of how Rajshahi and Ishwardi vary in climatic patterns and prospects for solar energy. Its base is a study into temperature trends, critical climatic elements for solar power, and the accuracy of the projections from each model.\u003c/p\u003e\n \u003cp\u003eBoth locations are growing warmer; however, Rajshahi heats up faster at +\u0026thinsp;0.024\u0026deg;C/year, while Ishwardi follows at +\u0026thinsp;0.020\u0026deg;C/year. This tendency corresponds with climate change in general and calls for increased efforts to embrace renewable energy in places vulnerable to heat.\u003c/p\u003e\n \u003cp\u003ePerformance projections built using linear regression and LSTM networks make it simpler to discover regional disparities. Both LSTM models functioned substantially better in both sites by following challenging patterns and providing trustworthy forecasts. Rajshahi and Ishwardi had greater RMSE in linear predictions, while both had lower RMSE in LSTM forecasts, indicating the success of AI in both domains.\u003c/p\u003e\n \u003cp\u003eRajshahi outperforms Ishwardi in everything evaluated when evaluating their solar energy potential. Slightly greater levels of shortwave irradiance, clearness index, and reduced average cloud cover may be seen, even though these slight variations imply that Rajshahi would be superior for producing solar energy routinely.\u003c/p\u003e\n \u003cp\u003eIn contrast, Ishwardi\u0026rsquo;s greater number of particles in the sky and somewhat brighter albedo might limit solar energy collection from the sun, mainly in the monsoon season.\u003c/p\u003e\n \u003cp\u003eAll in all, Rajshahi is somewhat better than Dhaka in terms of climate.\u003c/p\u003e\n \u003cp\u003eStability, predicting certainty, and energy capability give it a tiny edge in commencing the first solar projects.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\n \u003ch2\u003e3.9 Interpretation and Policy Implications\u003c/h2\u003e\n \u003cp\u003eBased on the study\u0026rsquo;s conclusions, Bangladesh may enhance its energy strategy at the regional and national levels. Because both Rajshahi and Ishwardi show evidence of constant warming, and solar energy has been verified using satellite climate data, climate intelligence should be employed in formulating energy plans.\u003c/p\u003e\n \u003cp\u003eIt is evident from LSTM findings that applying sophisticated AI technologies would assist in enhancing weather and energy monitoring systems. The data from these models may impact grid design, additional capacity construction, and how electricity will be delivered in the various seasons, notably for solar-dominant portfolios. With its excellent sun conditions, Rajshahi might be utilized as a trial site for targeted solar usage, notably in zones where people do not have access to the same electricity as big cities. The region might grow stronger and less dependent on fossil fuels by deploying solar panels on top of buildings here and in harmony with adjacent smart grids. Because of the greater cloud cover in Ishwardi, selecting solar energy systems in combination with wind or storage ones might be a wise choice for the whole year. Policymakers must be aware of micro-regional climatic differences when generating support for renewable energy and defining objectives for solar installations. The absence of high-resolution irradiance or panel efficiency statistics creates a significant gap in the data sets. Having additional stations for localized weather observation and extending remote-sensing abilities will make policies more reliable and ensure renewable energy is deployed where it is required in Bangladesh.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e\n \u003ch2\u003e3.10 Limitations of the Study\u003c/h2\u003e\n \u003cp\u003eThis study includes valuable insights regarding regional temperature, solar energy, and AI predictions for energy planning; however, several issues still hinder the results from being efficiently utilized or expanded. To begin with, proxy measures such as the clearness index, shortwave irradiance, and albedo are employed in the research to estimate solar energy capacity. While experts rely on these measures without PV output, they may not always accurately represent how solar panels perform in practice. Factors such as the angle of the solar panels, the amount of shading in the area, dust accumulation, and real-time panel effectiveness were not considered due to data limitations. Furthermore, the data needed for site selection was obtained from remote sensing archives, such as NASA-POWER, which, while providing accurate and reliable results, often lacks crucial information on fine local characteristics, particularly in highly developed or hilly areas. Although LSTM models performed better than standard linear regression in predicting temperatures, the research limited its approach to using only temperature time series without additional parameters. Incorporating multiple inputs, such as humidity, irradiance, and cloud cover, could strengthen the forecasting process and better represent the critical relationships in renewable energy systems. Changes in climate, such as sudden monsoons or cloud cover caused by cyclones, were not considered in the model. While these events do not occur consistently, they impact solar availability and require more complex computations using multiple models for optimal results. The conclusions of this research cannot be generalized to a global scale due to the region-specific nature of the study. Many other areas in Bangladesh face distinct challenges related to infrastructure and weather, and studying this concept in those regions will provide a more comprehensive national picture. Effectively harnessing real-time solar energy, developing models for panel performance, and addressing various climatic factors are viable next steps for researchers. Moreover, experts should review field findings and engage stakeholders to ensure the ideas translate into effective policy and practical solar planning.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4 Conclusion and Discussion","content":"\u003cp\u003eThe research studied the climate in Rajshahi and Ishwardi and estimated how efficiently solar energy can be utilized in these regions. Based on recorded temperatures and satellite sunlight readings, the author employed statistical approaches and AI to analyze the trends, variability, and potential for renewable energy. Both sections of Bangladesh have been found to warm at the same pace, with a temperature increase of 0.024\u0026deg;C/year in Rajshahi and 0.020\u0026deg;C/year in Ishwardi. These results are consistent with what is known about climate change and suggest that the area requires additional energy plans that can manage climate change. Forecasting using LSTM neural networks was considerably better than what could be obtained with linear regression. The RMSE for Rajshahi was 0.25\u0026deg;C and 0.22\u0026deg;C for Ishwardi, indicating that LSTM models effectively caught the varied changes in climate over time and across seasons. This indicates that applying deep learning methods aids in foreseeing climate change and that AI is necessary in energy planning systems. Rajshahi got a larger quantity of sunshine and had less cloud cover than Ishwardi in terms of solar power, suggesting that Rajshahi might be a better location for PV. Rajshahi\u0026rsquo;s insolation clearness index suggests the atmosphere is more translucent than elsewhere. Through boxplots, bar charts, and radar plots, these insights were offered, and the\u003c/p\u003e\u003cp\u003eStability and amplitude of the climatic parameters were visible in diverse places. Both places offer strong opportunities for solar expansion, although Rajshahi is significantly superior thanks to its brighter sky and consistent irradiance. During the months when there is less rain, the qualities become more beneficial since solar panels provide more power. This research shows why intelligent climate forecasting and regional climate diagnostics should be incorporated in developing national energy strategies. In Bangladesh, which wants to meet its renewable energy targets as specified in the SDGs, LSTM modeling and solar feasibility mapping are vital for prioritizing infrastructure development and sustaining constant energy security. However, there are constraints to this research investigation. Instead of utilizing actual information from PV plants, it based the model on satellite-recorded data, hence it omitted any conceivable impacts of dust, particles, and humidity. Consequently, the conveniences of economics and infrastructure did not impact where the locations were picked. In the future, we should incorporate real radiation sensors, reliable information on solar power plants\u0026rsquo; production, and input from stakeholders to transform analysis into practical choices. Making more projections based on various policy situations and genuine climate data would help even more people discover solutions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eORCID\u003c/h2\u003e\u003cp\u003eHasan Ahamed Alif - \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://orcid.org/0009-0003-2456-0379\u003c/span\u003e\u003cspan address=\"https://orcid.org/0009-0003-2456-0379\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis study did not obtain financing from any governmental, private, or not-for-profit source, and there was no outside support for financing the research.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eH.A.A. planned the study, obtained and examined the data, constructed the models, and analysed the data as well as generated all figures and visualisations in addition to drafting the text. H.A.A. glanced over and gave his permission on the final publishing of the work.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eI want to thank Dhaanish Ahmed College of Engineering for allowing me to undertake this independent study as an undergraduate Student. I was in charge of the complete process, from designing, reviewing the data, developing models, visualizing, to ultimately documenting my work. I am also thankful to NASA-POWER and the Meteorological Department of Bangladesh (BMD) for offering various environmental data that allowed me to conduct this research. I studied Python-based libraries and AI approaches by myself and consistently applied them precisely to preserve precision and repeatability in the job. In addition, I appreciate other academic specialists and available resources outside my field, which played a role in increasing mutual knowledge and collaboration in this area of study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbbass K, Qasim MZ, Song H, Murshed M, Mahmood H, Younis I (2022) A review of the global climate change impacts, adaptation, and sustainable mitigation measures. Environ Sci Pollut Res 29(28):42539\u0026ndash;42559. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11356-022-19718-6\u003c/span\u003e\u003cspan address=\"10.1007/s11356-022-19718-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl Mamun A, Islam AR, Md. T, Alam GMM, Sarker MNI, Erdiaw-Kwasie MO, Bhandari H, Mallick J (2023) Livelihood vulnerability of char land communities to climate change and natural hazards in Bangladesh: An application of livelihood vulnerability index. Nat Hazards 115(2):1411\u0026ndash;1437. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11069-022-05599-y\u003c/span\u003e\u003cspan address=\"10.1007/s11069-022-05599-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl-Maruf A, Mira SA, Rida TN, Rahman MS, Sarker PK, Jenkins JC (2021) Piloting a Weather-Index-Based Crop Insurance System in Bangladesh: Understanding the Challenges of Financial Instruments for Tackling Climate Risks. Sustainability 13(15):8616. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su13158616\u003c/span\u003e\u003cspan address=\"10.3390/su13158616\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoi Y-W, Campbell DJ, Aldridge JC, Eltahir EAB (2021) Near-term regional climate change over Bangladesh. Clim Dyn 57(11\u0026ndash;12):3055\u0026ndash;3073. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-021-05856-z\u003c/span\u003e\u003cspan address=\"10.1007/s00382-021-05856-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDe Laurentis C, Pearson PJG (2021) Policy-relevant insights for regional renewable energy deployment. Energy Sustain Soc 11(1):19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13705-021-00295-4\u003c/span\u003e\u003cspan address=\"10.1186/s13705-021-00295-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDellaripa PF, Bush T, Miller FW, Feldman CH (2023) The Climate Emergency and the Health of Our Patients: The Role of the Rheumatologist. Arthritis Rheumatol 75(1):1\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/art.42279\u003c/span\u003e\u003cspan address=\"10.1002/art.42279\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGenc TS, Kosempel S (2023) Energy Transition and the Economy: A Review Article. Energies 16(7):2965. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/en16072965\u003c/span\u003e\u003cspan address=\"10.3390/en16072965\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaupt SE, McCandless TC, Dettling S, Alessandrini S, Lee JA, Linden S, Petzke W, Brummet T, Nguyen N, Kosović B, Wiener G, Hussain T, Al-Rasheedi M (2020) Combining Artificial Intelligence with Physics-Based Methods for Probabilistic Renewable Energy Forecasting. \u003cem\u003eEnergies\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(8), 1979. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/en13081979\u003c/span\u003e\u003cspan address=\"10.3390/en13081979\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eH\u0026eacute;bert R, Herzschuh U, Laepple T (2022) Ocean temperature fluctuations overprint millennial-scale climate variability over land. Nat Geosci 15(11):899\u0026ndash;905. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41561-022-01056-4\u003c/span\u003e\u003cspan address=\"10.1038/s41561-022-01056-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHultman NE, Clarke L, Frisch C, Kennedy K, McJeon H, Cyrs T, Hansel P, Bodnar P, Manion M, Edwards MR, Cui R, Bowman C, Lund J, Westphal MI, Clapper A, Jaeger J, Sen A, Lou J, Saha D, O\u0026rsquo;Neill J (2020) Fusing subnational with national climate action is central to decarbonization: The case of the United States. Nat Commun 11(1):5255. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-020-18903-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-18903-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiskat MI, Sarker P, Chowdhury H, Chowdhury T, Rahman MS, Hossain N, Chowdhury P, Sait SM (2023) Current Scenario of Solar Energy Applications in Bangladesh: Techno-Economic Perspective, Policy Implementation, and Possibility of the Integration of Artificial Intelligence. Energies 16(3):1494. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/en16031494\u003c/span\u003e\u003cspan address=\"10.3390/en16031494\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMunjer MA, Hasan MZ, Hossain MK, Rahman MF (2023) The Obstruction and Advancement in the Sustainable Energy Sector to Achieve SDG in Bangladesh. Sustainability 15(5):3913. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su15053913\u003c/span\u003e\u003cspan address=\"10.3390/su15053913\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNovas N, Garcia RM, Camacho JM, Alcayde A (2021) Advances in Solar Energy towards Efficient and Sustainable Energy. Sustainability 13(11):6295. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su13116295\u003c/span\u003e\u003cspan address=\"10.3390/su13116295\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSong J, Tong G, Chao J, Chung J, Zhang M, Lin W, Zhang T, Bentler PM, Zhu W (2023) Data-driven pathway analysis and forecast of global warming and sea level rise. Sci Rep 13(1):5536. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-023-30789-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-30789-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTalut M, Bahaj AS, James P (2022) Solar Power Potential from Industrial Buildings and Impact on Electricity Supply in Bangladesh. Energies 15(11):4037. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/en15114037\u003c/span\u003e\u003cspan address=\"10.3390/en15114037\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou Z, Tang W, Li M, Cao W, Yuan Z (2023) A Novel Hybrid Intelligent SOPDEL Model with Comprehensive Data Preprocessing for Long-Time-Series Climate Prediction. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(7), 1951. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs15071951\u003c/span\u003e\u003cspan address=\"10.3390/rs15071951\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"theoretical-and-applied-climatology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taac","sideBox":"Learn more about [Theoretical and Applied Climatology](https://www.springer.com/journal/704)","snPcode":"704","submissionUrl":"https://submission.nature.com/new-submission/704/3","title":"Theoretical and Applied Climatology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Artificial Intelligence (AI), Renewable Energy, Big Data Analysis, LSTM, Climate-Resilient Infrastructure, NASA Power","lastPublishedDoi":"10.21203/rs.3.rs-7031666/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7031666/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSustainability in renewable energy involves utilizing energy sources intelligently to serve the present and yet be accessible for future generations. With Bangladesh shifting to renewable energy due to rising climate threats, it needs expert and data-based input to create sustainable infrastructure. The investigation uses a blend of statistical and deep learning methodologies to analyze the implications of temperature-based climatic patterns on the potential for solar energy in Rajshahi and Ishwardi in Bangladesh. We centered our study on past temperatures from 1980 to 2020, bringing in both linear regression and one of the top deep learning models, LSTM, as ideal ways to estimate climate in the future. It was found that both locations are seeing a significant increase in temperature, whereas the LSTM technique was superior at spotting irregular seasonal fluctuations and trends across years. We utilized the all-sky shortwave irradiance, clearness index of solar radiation, quantity of cloud cover, albedo, and near-surface temperature from the NASA-POWER datasets to estimate the solar energy potential. The research showed that Rajshahi had superior conditions for solar energy owing to the cleaner weather, fewer overcast days, and higher irradiance intensity. This research provides significant insights for regional policy interests and renewable energy planning via extensive visual data and the assessment of several factors. The report advocates for using AI and data to facilitate solar expansion. It highlights the crucial role of deep learning in promoting sustainability and ecologically safe energy in Bangladesh and other developing nations.\u003c/p\u003e","manuscriptTitle":"Temperature-Based Renewable Energy Forecasting: A Big Data Analysis for Sustainable Energy Planning in Bangladesh","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-15 13:19:26","doi":"10.21203/rs.3.rs-7031666/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-23T17:28:19+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-23T15:29:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-23T03:03:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-17T06:32:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"329711740791564820112972966591063048560","date":"2025-07-14T04:21:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"291702963376599836696839603409579951222","date":"2025-07-13T15:33:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"60203075722373865111434913395355780798","date":"2025-07-13T03:15:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"203214762495439787596076282567966520111","date":"2025-07-12T06:37:49+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-11T15:07:19+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-03T23:09:07+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-03T23:07:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"Theoretical and Applied Climatology","date":"2025-07-02T17:14:46+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"theoretical-and-applied-climatology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taac","sideBox":"Learn more about [Theoretical and Applied Climatology](https://www.springer.com/journal/704)","snPcode":"704","submissionUrl":"https://submission.nature.com/new-submission/704/3","title":"Theoretical and Applied Climatology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"5fe41ea5-c0cb-4ddd-adba-489e05501339","owner":[],"postedDate":"July 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-09-29T16:09:44+00:00","versionOfRecord":{"articleIdentity":"rs-7031666","link":"https://doi.org/10.1007/s00704-025-05746-y","journal":{"identity":"theoretical-and-applied-climatology","isVorOnly":false,"title":"Theoretical and Applied Climatology"},"publishedOn":"2025-09-27 15:57:47","publishedOnDateReadable":"September 27th, 2025"},"versionCreatedAt":"2025-07-15 13:19:26","video":"","vorDoi":"10.1007/s00704-025-05746-y","vorDoiUrl":"https://doi.org/10.1007/s00704-025-05746-y","workflowStages":[]},"version":"v1","identity":"rs-7031666","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7031666","identity":"rs-7031666","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}