Assessing the Impact of Land Use and Land Cover Changes on Land Surface Temperature Dynamics in the Coastal Region of Bangladesh: A Comprehensive Analysis Using Deep Learning Techniques AHP Integration

Assessing the Impact of Land Use and Land Cover Changes on Land Surface Temperature Dynamics in the Coastal Region of Bangladesh: A Comprehensive Analysis Using Deep Learning Techniques AHP Integration | 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 Assessing the Impact of Land Use and Land Cover Changes on Land Surface Temperature Dynamics in the Coastal Region of Bangladesh: A Comprehensive Analysis Using Deep Learning Techniques AHP Integration Niloy Biswas, Kazi Saiful Islam This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6800634/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study examines the relationship between land use and land cover (LULC) changes and land surface temperature (LST) dynamics in the climate-vulnerable coastal regions of Bangladesh from 1990 to 2020. Utilizing multi-temporal Landsat imagery and advanced deep learning techniques, particularly Temporal Convolutional Networks (TCN), the research achieved a classification accuracy of 73% in 2020, outperforming other models such as XGBoost (71%). The findings reveal that rapid urbanization and deforestation are the principal drivers of increasing LST, with urban centers such as Khulna and Chittagong experiencing a temperature rise of up to 2.5°C over the study period. An Analytic Hierarchy Process (AHP)-based prioritization of LULC transitions identified agricultural-to-urban (weighted impact: 82%) and vegetation-to-urban conversions as the most significant contributors to LST escalation, whereas forested and water-covered areas were associated with relatively lower temperature increases. Seasonal analysis indicates a more pronounced warming during summer, with rural areas showing a mitigated rise due to residual vegetation cover. The study further underscores the compounding effects of climate change, suggesting that continued LULC transformations without adaptive measures could intensify future heat stress. To mitigate the urban heat island effect, the study recommends the implementation of green infrastructure, enforcement of forest conservation, and the promotion of climate-sensitive urban planning. By integrating deep learning with multi-criteria decision analysis, this research contributes a robust methodological framework and empirical insights to support sustainable land management and climate adaptation in coastal regions. Land Surface Temperature (LST) Temporal Convolutional Network (TCN) Analytic Hierarchy Process (AHP) Coastal Bangladesh Deep Learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Introduction The transformation of land use and land cover (LULC) in coastal regions, particularly in Bangladesh, has emerged as a critical environmental concern due to its significant impact on land surface temperature (LST) (Alam, 2013 ; Islam et al., 2021 ). Rapid urbanization, deforestation, and agricultural expansion have profoundly altered the landscape, leading to elevated temperatures, particularly in urban areas. Major cities such as Dhaka, Chittagong, and Khulna have witnessed unprecedented growth, resulting in the replacement of natural vegetation with built-up infrastructure (Dewan et al., 2012 ; Imran et al., 2021a ). This urban sprawl has intensified the urban heat island (UHI) effect, wherein densely built environments trap and retain heat, causing higher LST compared to surrounding rural areas (Imran et al., 2021b ).This study addresses the need to understand how historical and ongoing changes in LULC contribute to rising temperatures by analyzing satellite data and applying advanced analytical techniques. The primary objective is to delineate the specific impacts of various land use changes on LST and provide insights that can guide climate adaptation and urban planning strategies. By examining the temporal and spatial variations in LST across different land use categories, the study aims to offer a comprehensive perspective on how these transformations influence temperature dynamics and to propose actionable recommendations for mitigating the adverse effects of rising temperatures. Land Use and Land Cover (LULC) refers to the classification and characterization of land based on its physical cover and human utilization. LULC categories include urban, agricultural, forested, and barren areas (Beg, 2018 ; Chaves et al., 2020 ; Garouani et al., 2021 ). This classification encompasses both the natural landscape and anthropogenic modifications such as urban development, deforestation, and agricultural expansion. In contrast, Land Surface Temperature (LST) represents the temperature of the Earth's surface, as measured by remote sensing instruments on satellites (Garouani et al., 2021 ; Liu & Huang, 2018 ; Sobrino et al., 2020 ). LST is influenced by factors such as solar radiation absorption, surface emissivity, and land cover characteristics (Wang et al., 2018 ; Watson et al., 1997 ). The relationship between LULC and LST is well established, particularly in the context of urban heat islands (UHIs), where urbanized areas exhibit higher temperatures than their rural counterparts. This phenomenon results from increased heat absorption by impervious surfaces, reduced vegetation cover, and anthropogenic heat emissions (Ai et al., 2020a ; Filho et al., 2017 ). While the general correlation between urbanization and LST increase is well documented (Ai et al., 2020b ; Lu et al., 2021 ; Rehman et al., 2022 ), existing research often lacks a nuanced exploration of how different LULC types interact over time and across spatial scales. For example, while it is widely accepted that urban areas experience higher LST, the impact of specific transitions such as agricultural land conversion into built-up areas remains underexplored (Guha et al., 2020 ; H. Li et al., 2017 ; S. Li et al., 2022 ; Qiu et al., 2019 ). Additionally, the cumulative effects of gradual LULC changes on temperature dynamics require further investigation. Recent advancements in remote sensing and machine learning have enabled a more detailed examination of the LULC-LST relationship. High-resolution satellite imagery and deep learning techniques allow for the detection of complex, nonlinear interactions between land use changes and temperature variations (W. Li et al., 2017 ; X. Y. Lu et al., 2021 ). Traditional methods, such as statistical correlation and historical trend analysis, have provided foundational insights into these dynamics (W. Li et al., 2017 ). However, these approaches often oversimplify the interactions between land use changes and LST, failing to capture the intricate patterns and temporal variations (Zhou et al., 2021; Li et al., 2022 ). Additionally, conventional methods frequently rely on historical data without fully leveraging real-time or high-frequency data analysis (Ai et al., 2020a ; Liu & Huang, 2018 ; Talukdar et al., 2020 ).Recent research has demonstrated that deep learning models can significantly enhance the understanding of LULC-LST interactions by capturing temporal and spatial complexities that traditional methods may overlook. For instance, studies by Chen et al. (2023) illustrate how deep learning techniques provide more precise assessments of how urban expansion and natural landscape conversions impact LST over timeThis study integrates high-resolution satellite data, deep learning techniques, and the Analytical Hierarchy Process (AHP) to assess the impact of LULC changes on LST. By utilizing modern remote sensing data and prioritizing the influence of different land transitions through AHP, this research provides a clearer understanding of how these transformations affect LST. The study aims to: Classify and analyze LULC changes in the coastal regions of Bangladesh from 1990 to 2020 using deep learning algorithms. Examine spatial and temporal variations in LST across different land use categories. Assess the relationship between LULC transitions and LST trends using machine learning techniques. Apply the Analytical Hierarchy Process (AHP) to prioritize the influence of different land cover changes on temperature dynamics. By bridging key research gaps, this study offers a refined analytical framework that enhances the understanding of LULC-LST interactions. The findings aim to contribute to more effective climate adaptation and urban planning strategies, ensuring sustainable development in Bangladesh’s coastal regions. Materials & Methods 2.1. Study Area This study focuses on the coastal region of Bangladesh, a dynamic interface between terrestrial and marine ecosystems. Encompassing approximately 47,201 km², the study area extends from 23.30°N to 21.00°N latitude and 89.00°E to 90.00°E longitude (Fig. 1 ). It comprises 19 districts, including Jessore, Narail, Gopalganj, Shariatpur, Chandpur, Satkhira, Khulna, Bagerhat, Pirojpur, Jhalakati, Barguna, Barisal, Patuakhali, Bhola, Lakshmipur, Noakhali, Feni, Chittagong, and Cox’s Bazar. The region's geomorphology and hydrology are shaped by the Ganges-Brahmaputra-Meghna (GBM) river system and the Bay of Bengal (Mukherjee et al., 2009 ).The 710 km-long coastline hosts diverse ecosystems, including the Sundarbans (6,017 km²), tidal flats, estuaries, seagrass beds, and numerous islands (Rogers & Goodbred, 2014 ; Umitsu, 1997 ). Additionally, the region features accreted lands, beaches, and densely populated urban centers. The area is highly vulnerable to climate change, experiencing sea level rise, cyclones, storm surges, coastal inundation, and salinity intrusion. These challenges are compounded by rapid urbanization and industrialization, altering land use and land cover (LULC) patterns and influencing land surface temperature (LST) dynamics.This study investigates the interplay between natural and anthropogenic factors affecting LULC transformations and their broader climatic implications. Utilizing high-resolution satellite data and machine learning techniques, the research aims to generate insights into regional environmental changes and support sustainable land management strategies. 2.2 Data Collection This study utilizes satellite imagery from Landsat missions to examine Land Use/Land Cover (LULC) changes and Land Surface Temperature (LST) variations across the coastal region of Bangladesh. The analysis spans from 1990 to 2020, focusing on two distinct seasons: summer and winter. Landsat data provide consistent and reliable information, enabling a comprehensive assessment of spatial and temporal trends in LULC and LST dynamics. 2.2.1 Satellite Data Collection for LST and LULC The primary data sources include imagery from Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS). These satellites were selected for their extensive global coverage and high-resolution capabilities, facilitating detailed analyses of Bangladesh's coastal regions. The study utilizes datasets from the years 1990, 1995, 2000, 2005, 2010, 2015, and 2020 to capture LULC transformations and LST variations across 19 coastal districts. LULC classification was performed using optical bands from Landsat imagery, which offer a 30-meter spatial resolution. LST extraction relied on thermal infrared bands, available at a 100-meter spatial resolution. Thermal data from Landsat 5, 7, and 8 were critical in assessing temperature variations across summer and winter seasons. All imagery was sourced from the United States Geological Survey (USGS) Earth Explorer platform, ensuring access to high-quality, cloud-free images. The selection of cloud-free datasets minimized inaccuracies stemming from atmospheric interferences, particularly during the monsoon season.Preprocessing steps included radiometric calibration and atmospheric correction using the Dark Object Subtraction (DOS) method to mitigate atmospheric distortions and enhance image clarity. Geometric correction was applied to ensure precise spatial alignment across datasets from multiple years and satellite sources. LST calculation was conducted using thermal infrared bands to derive brightness temperature, subsequently converted into surface temperature using the Radiative Transfer Equation (RTE). The seasonal dataset, encompassing both summer and winter observations for each selected year, enabled a robust analysis of temperature fluctuations in relation to land cover changes under varying climatic conditions. These high-resolution datasets form the foundation for understanding LULC-LST interactions in the coastal region of Bangladesh, contributing to improved land management and climate adaptation strategies. 2.2.2 Data Sources for LULC and LST Analysis Data for this study were collected for the following years: 1990, 1995, 2000, 2005, 2010, 2015, and 2020. Each year’s data included summer and winter seasons to ensure a comprehensive understanding of seasonal LST dynamics. The data sources are as follows: Table 1 Data sources for Study Year Satellite Sensor Data Type Spatial Resolution Temporal Coverage Season 1990 Landsat 5 TM LULC, LST 30m (LULC), 120m (LST) Dry and Wet Seasons Summer, Winter 1995 Landsat 5 TM LULC, LST 30m (LULC), 120m (LST) Dry and Wet Seasons Summer, Winter 2000 Landsat 7 ETM+ LULC, LST 30m (LULC), 60m (LST) Dry and Wet Seasons Summer, Winter 2005 Landsat 7 ETM+ LULC, LST 30m (LULC), 60m (LST) Dry and Wet Seasons Summer, Winter 2010 Landsat 7 ETM+ LULC, LST 30m (LULC), 60m (LST) Dry and Wet Seasons Summer, Winter 2015 Landsat 8 OLI/TIRS LULC, LST 30m (LULC), 100m (LST) Dry and Wet Seasons Summer, Winter 2020 Landsat 8 OLI/TIRS LULC, LST 30m (LULC), 100m (LST) Dry and Wet Seasons Summer, Winter 2.3 LULC Classification LULC classification was a critical aspect of this study, aimed at detecting spatial and temporal changes in land cover. High-resolution Landsat imagery from 1990 to 2020 underwent systematic preprocessing, including atmospheric correction, radiometric calibration, and geometric alignment(see Fig. 2 ). These procedures minimized inconsistencies due to sensor differences and atmospheric conditions, ensuring robust classification. The classification process utilized a training dataset comprising 6,700 reference points representing diverse land cover types, including urban areas, agricultural fields, water bodies, forests, and bare soil. These points were selected to capture spectral variability, enhancing classification accuracy. Multiple machine learning and deep learning models were applied to improve classification performance. Temporal Convolutional Networks (TCN) were employed for their ability to model sequential data, making them particularly effective for detecting temporal changes in land cover. TCNs incorporate a causal structure, enabling long-range dependency learning, which is crucial for multi-year land cover analysis (Robinson et al., 2021 ). Extreme Gradient Boosting (XGBoost), a decision-tree-based ensemble learning method, was integrated due to its high efficiency and accuracy in classification tasks. By constructing sequential decision trees that correct prior errors, XGBoost effectively handled complex feature interactions and missing data, enhancing classification performance (Bui et al., 2021 ).CatBoost, optimized for categorical data processing, was also employed. Its ordered boosting mechanism mitigates overfitting, improving generalization in LULC classification tasks (Prokhorenkova et al., 2017 ). Additionally, LightGBM, a gradient boosting algorithm designed for speed and scalability, was used for its ability to process large datasets efficiently. Unlike traditional boosting methods, LightGBM grows trees leaf-wise, allowing it to focus on complex classification areas, leading to superior accuracy (Ke et al., 2017 ).H2O.ai, a comprehensive machine learning platform, streamlined the model development process by supporting multiple algorithms and facilitating seamless comparisons (Madni et al., 2023 ). These models were implemented using Python and advanced libraries such as scikit-learn, XGBoost, CatBoost, LightGBM, and H2O.ai, ensuring computational efficiency and classification reliability.Post-processing techniques were applied to refine classification outputs, minimizing noise and enhancing precision. Classification accuracy was evaluated using an error matrix, incorporating metrics such as overall accuracy, user’s accuracy, producer’s accuracy, and the Kappa coefficient. This rigorous validation process ensured that the generated LULC maps accurately represented land cover conditions over time. Table 2 Description of LULC Types LULC Type Description Built-Up Areas covered with impervious surfaces such as buildings, roads, concrete, and asphalt structures. Waterbodies Water features including rivers, lakes, reservoirs, and other inundated areas. Forest Areas with dense tree cover, including both natural and reforested forests. This category includes: Hilly Forest : Forested areas in hilly or mountainous regions with varying tree densities. Mangrove Forest : Coastal forests characterized by mangrove trees, typically found in tidal areas and estuaries. Vegetation Green areas such as grasslands, parks, green belts, and other non-forest vegetative covers. Agricultural Land Cropland and open land used for farming, including fields, plantations, and other agricultural activities. 2.3.1 LST Analysis The Land Surface Temperature (LST) analysis aimed to assess temperature variations across different land cover types and seasons. Landsat thermal infrared bands from Landsat 5 (TM), Landsat 7 (ETM+), and Landsat 8 (OLI/TIRS) were used to derive LST for the years 1990, 1995, 2000, 2005, 2010, 2015, and 2020, covering both summer and winter seasons. The data underwent radiometric calibration to convert raw digital numbers into radiance, followed by the calculation of brightness temperatures. LST calculations were performed using the split-window algorithm in Google Earth Engine (GEE), which effectively mitigated atmospheric effects and facilitated the processing of large satellite datasets. This algorithm, as described by Du et al. ( 2015 ), accounts for differential absorption of infrared radiation by atmospheric gases, ensuring accurate surface temperature estimations. By analyzing LST separately for summer and winter, the study captured seasonal temperature variations and examined the influence of land cover changes on temperature dynamics. To validate the LST data, satellite-derived temperatures were compared with in-situ measurements and cross-referenced with existing datasets and prior studies, ensuring consistency and accuracy. The analysis provided key insights into how different land cover types and seasonal changes affect surface temperatures in the coastal region of Bangladesh. 2.3.4. Accuracy assessment The accuracy of the Land Use and Land Cover (LULC) classifications was evaluated using an error matrix in Google Earth Engine (GEE), comparing the classified outputs with reference points. The Kappa statistic (K) and overall accuracy (OA) were calculated from the confusion matrix to assess classification performance, following the methods outlined by Fitzgerald & Lees ( 1994 ) and Rwanga et al. ( 2017 ) (Eq. 2–5). Factors such as satellite image quality, feature selection, and spatial distribution of sample points may influence accuracy. The entire process training and validation point collection, image classification, and accuracy assessment was performed within GEE. To analyze LULC change patterns, categorized maps from different years were intersected to track transitions between LULC classes over time. For the coastal area study, data from 6,700 points collected between 1900 and 2020 were used. Deep learning models, including TCN, XGBoost, CatBoost, LightGBM, and H2O.ai, were implemented using Python in Google Colab to analyze and model the data. $$\:\text{O}\text{v}\text{e}\text{r}\text{a}\text{l}\text{l}\:\text{a}\text{c}\text{c}\text{u}\text{r}\text{a}\text{c}\text{y}=\:\frac{\text{T}\text{o}\text{t}\text{a}\text{l}\:\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{c}\text{o}\text{r}\:\text{r}\text{e}\text{c}\text{t}\text{e}\text{d}\:\text{c}\text{l}\text{a}\text{s}\text{s}\text{i}\text{f}\text{i}\text{e}\text{d}\:\text{p}\text{i}\text{x}\text{e}\text{l}\text{x}\:\left(\text{d}\text{i}\text{a}\text{g}\text{o}\text{n}\text{a}\text{l}\right)}{\text{t}\text{o}\text{t}\text{a}\text{l}\:\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{r}\text{e}\text{f}\text{e}\text{r}\text{e}\text{n}\text{c}\text{e}\:\text{p}\text{i}\text{x}\text{e}\text{l}\text{s}}\:*100\:\:\:\:\:\:\:\:\left(2\right)$$ $$\:\text{U}\text{s}\text{e}\text{r}\:\text{A}\text{c}\text{c}\text{u}\text{r}\text{a}\text{c}\text{y}=\:\frac{\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{c}\text{o}\text{r}\:\text{r}\text{e}\text{c}\text{t}\text{l}\text{y}\:\text{c}\text{l}\text{a}\text{s}\text{s}\text{i}\text{f}\text{i}\text{e}\text{d}\:\text{p}\text{i}\text{x}\text{e}\text{l}\text{x}\text{s}\:\text{i}\text{n}\:\text{e}\text{a}\text{c}\text{h}\:\text{c}\text{a}\text{t}\text{a}\text{g}\text{o}\text{r}\text{y}\:(\text{d}\text{i}\text{a}\text{g}\text{o}\text{n}\text{a}\text{l}}{\text{t}\text{o}\text{t}\text{a}\text{l}\:\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{r}\text{e}\text{f}\text{e}\text{r}\text{e}\text{n}\text{c}\text{e}\:\text{p}\text{i}\text{x}\text{e}\text{l}\text{s}\:\text{i}\text{n}\:\text{e}\text{a}\text{c}\text{h}\:\text{c}\text{a}\text{t}\text{e}\text{g}\text{o}\text{r}\text{y}\:\left(\text{r}\text{o}\text{w}\:\text{t}\text{o}\text{t}\text{a}\text{l}\right)}\:*100\:\:\:\:\left(3\right)$$ $$\:\text{P}.\text{A}=\:\frac{\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{c}\text{o}\text{r}\:\text{r}\text{e}\text{c}\text{t}\text{l}\text{y}\:\text{c}\text{l}\text{a}\text{s}\text{s}\text{i}\text{f}\text{i}\text{e}\text{d}\:\text{p}\text{i}\text{x}\text{e}\text{l}\text{x}\text{s}\:\text{i}\text{n}\:\text{e}\text{a}\text{c}\text{h}\:\text{c}\text{a}\text{t}\text{e}\text{g}\text{o}\text{r}\text{y}\:\left(\text{d}\text{i}\text{a}\text{g}\text{o}\text{n}\text{a}\text{l}\right)}{\text{t}\text{o}\text{t}\text{a}\text{l}\:\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{r}\text{e}\text{f}\text{e}\text{r}\text{e}\text{n}\text{c}\text{e}\:\text{p}\text{i}\text{x}\text{e}\text{l}\text{s}\:\text{i}\text{n}\:\text{e}\text{a}\text{c}\text{h}\:\text{c}\text{a}\text{t}\text{e}\text{g}\text{o}\text{r}\text{y}\:\left(\text{c}\text{o}\text{l}\text{u}\text{m}\text{n}\:\text{t}\text{o}\text{t}\text{a}\text{l}\right)}\:*100\:\:\:\:\left(4\right)$$ $$\:\text{K}.\text{C}\left(\text{T}\right)=\frac{\text{T}\text{o}\text{t}\text{a}\text{l}\:\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{S}\text{a}\text{m}\text{p}\text{l}\text{e}\text{*}\text{T}\text{o}\text{t}\text{a}\text{l}\:\text{N}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{C}\text{o}\text{r}\:\text{r}\text{e}\text{c}\text{t}\text{e}\text{d}\:\text{S}\text{a}\text{m}\text{p}\text{l}\text{e}\:-\:\sum\:\left(\text{c}\text{o}\text{l}.\text{t}\text{o}\text{t}\:\text{*}\:\text{r}\text{o}\text{w}\:\text{t}\text{o}\text{t}\right)}{\left(\text{T}\text{o}\text{t}\text{a}\text{l}\:\text{n}\text{u}\text{m}\text{b}\text{e}\text{r}\:\text{o}\text{f}\:\text{S}\text{a}\text{m}\text{p}\text{l}\text{e}\right)2\:-\:\sum\:\left(\text{c}\text{o}\text{l}.\text{t}\text{o}\text{t}\:\text{*}\:\text{r}\text{o}\text{w}\:\text{t}\text{o}\text{t}\right)}\:*100\:\:\:\:\:\left(5\right)\:\:\:$$ Here P.A = Producer Accuracy K.C (T) = Kappa Coefficient For the Land Surface Temperature (LST) analysis, which covered both summer and winter seasons from 1990 to 2020, Landsat satellite data were used to extract temperature values. The accuracy of the LST estimates was assessed by comparing the derived values with in-situ temperature data from local meteorological stations. Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) were calculated to quantify estimation errors. This accuracy assessment highlighted the strengths and limitations of the models, confirming that the LST analysis and classification models are reliable and effectively capture the spatiotemporal dynamics of land use and temperature changes. These results provide a robust basis for the study’s conclusions regarding climate adaptation and urban planning strategies in the coastal region. 2.4 Integration of AHP for LULC Transition Analysis To assess the impacts of LULC transitions on LST, this study incorporated the Analytical Hierarchy Process (AHP) to evaluate the relative importance of different LULC transitions in influencing temperature changes. The LULC categories considered include Forest, Waterbodies, Vegetation, Agricultural Land, and Urban (Built-up). A pairwise comparison matrix was developed to assign relative importance to transitions between these categories, using a scale of 1 to 9, where 1 indicates equal importance and 9 represents the highest level of importance. Table 3 Importance of Land use Categories From-To Forest Waterbodies Vegetation Agricultural Land Urban (Built-up) Forest 1 3 5 7 9 Waterbodies 1/3 1 5 7 9 Vegetation 1/5 1/5 1 5 9 Agricultural Land 1/7 1/7 1/5 1 9 Urban (Built-up) 1/9 1/9 1/9 1/9 1 2.4.1 Normalization and Priority Weight Calculation Each element of the matrix was normalized by dividing it by the sum of its respective column. The normalized values were used to calculate priority weights, which represent the relative influence of each LULC transition on LST. Table 4 Priority of Land use Categories LULC Transition Priority Weight Forest 0.396 Waterbodies 0.250 Vegetation 0.131 Agricultural Land 0.082 Urban (Built-up) 0.036 2.4.2 Consistency Check To ensure the reliability of the pairwise comparison matrix, a consistency ratio (CR) was calculated using the Equation. The CR value was found to be less than 0.1, confirming that the matrix is consistent and the derived weights are valid. $$\:CI=\frac{{\lambda\:}_{m}-n}{n-1}$$ 2.4.3 Spatial Representation The calculated priority weights were applied to LULC change maps to create spatial representations of the impacts of these transitions on LST. These maps identified hotspots of temperature rise, highlighting areas where specific LULC transitions, such as deforestation and urban expansion, had the most significant effects. Results 3.1 Land Use and Land Cover Classification The Land Use and Land Cover (LULC) classification of the coastal region of Bangladesh, shown in 3, provides a comprehensive spatial distribution of land cover types in 1990. The classification includes five primary categories: agriculture, forest, vegetation, water bodies, and built-up areas. The temporal analysis, presented in 5, outlines the changes in these categories from 1990 to 2020, facilitating a deeper understanding of land cover dynamics over three decades.In 1990, agricultural land was the dominant category, covering 35% of the coastal zone. This included cropland and other open areas used for agricultural purposes. Vegetation, comprising forests, grasslands, and green belts, accounted for 30% of the land cover. Water bodies, including rivers, lakes, and wetlands, constituted 10% of the region, emphasizing the importance of water features within the coastal ecosystem. Built-up areas, encompassing urban settlements, infrastructure, and ports, occupied 5% of the land.This classification serves as a baseline for understanding land cover changes and provides essential context for analyzing the implications of these shifts on environmental and urban dynamics Table 5 Year-wise Distribution of LULC Year Vegetation (km²) Vegetation (%) Forest (km²) Forest (%) Waterbodies (km²) Waterbodies (%) Built-Up (km²) Built-Up (%) Agricultural Land (km²) Agricultural Land (%) 1990 14,160 30.00 9,440 20.00 4,720 10.00 2,360 5.00 16,520 35.00 1995 13,620 28.80 9,440 20.00 5,000 10.60 2,720 5.75 16,880 35.73 2000 13,260 28.13 9,440 20.00 4,720 10.00 3,000 6.35 17,580 37.27 2005 12,860 27.20 9,440 20.00 5,000 10.60 3,080 6.52 18,080 38.30 2010 12,600 26.74 9,600 20.30 4,800 10.20 3,400 7.20 18,600 39.49 2015 12,340 26.12 9,700 20.50 4,000 8.50 3,800 8.05 19,000 40.25 2020 9,440 20.00 4,720 10.00 3,776 8.00 9,440 20.00 19,824 42.00 Over the decades, the distribution of land cover types in the coastal region of Bangladesh has undergone significant changes, as shown in Table 5 and Figs. 3 and 4 . By 1995, agricultural land had increased slightly to 35.73%, reflecting a gradual expansion of cropland and agricultural activities. In contrast, vegetation decreased to 28.80%, indicating a reduction in natural green spaces. Water bodies rose to 10.60%, and built-up areas increased to 5.75%, marking the early stages of urban expansion. These trends continued into the early 2000s, with agricultural land growing further to 37.27%, while vegetation declined to 28.13%. Built-up areas increased to 6.35%, reflecting the expansion of urban and industrial development, while the extent of water bodies remained stable, indicating consistent water features in the region. From 2005 to 2010, agricultural land further increased to 38.30%, and built-up areas expanded to 6.52%. Vegetation continued to decrease, reaching 27.20%, while water bodies remained stable at 10.60%, showing minor fluctuations in the region’s water distribution. These changes reflect the ongoing influence of urbanization and environmental pressures on land cover.The period from 2010 to 2020 saw even more pronounced shifts. By 2010, agricultural land had reached 39.49%, and built-up areas expanded to 7.20%. Vegetation continued to decline, dropping to 26.74%, while water bodies slightly reduced to 10.20%. By 2015, agricultural land rose to 40.25%, with built-up areas increasing to 8.05%. Vegetation declined further to 26.12%, and water bodies decreased to 8.50%. By 2020, agricultural land reached 42.00%, and built-up areas surged dramatically to 20.00%. Vegetation fell significantly to 20.00%, while water bodies decreased to 8.00%. Throughout the study period, forest land, including both hilly and mangrove forests, remained relatively stable, indicating some resilience in these ecosystems. These trends highlight the complex interplay between natural processes and human activities in Bangladesh’s coastal zone. The expansion of agricultural and built-up areas, coupled with the decline in vegetation and water bodies, underscores the need for effective land management strategies. Sustainable practices are essential to balance development with environmental preservation, especially in a region as vulnerable and dynamic as Bangladesh’s coastal zone. 4.2 Accuracy Assessment Accuracy assessment is essential in land use and land cover (LULC) classification, as it evaluates the reliability of classified images by comparing them with an accurate reference dataset or ground truth. This study used five deep learning algorithms Temporal Convolutional Network (TCN), XGBoost, CatBoost, LightGBM, and H2O.ai to assess LULC classifications for the years 1990, 1995, 2000, 2005, 2010, 2015, and 2020. The performance of each algorithm was evaluated based on accuracy and Kappa values, as summarized in Table 6 .TCN showed gradual improvement over time, with accuracy increasing from 0.62 and Kappa from 0.61 in 1995 to 0.71 and 0.70, respectively, by 2020. TCN consistently outperformed XGBoost, achieving a peak accuracy of 0.73 and a Kappa value of 0.69 in 2020. In comparison, CatBoost and LightGBM performed slightly lower, with CatBoost reaching a maximum accuracy of 0.67 and LightGBM 0.65 in 2020. H2O.ai demonstrated competitive results, with accuracy and Kappa values of 0.69 and 0.67, respectively, in 2020. Overall, TCN and H2O.ai were identified as the top-performing algorithms, with TCN showing substantial improvement over the study period. Table 6 Comparative Assessment of Supervised Models Algorithm 2020 Ac 2020 ka 2015 Ac 2015 Ka 2010 Ac 2010 Ka 2005 Ac 2005 Ka 2000 Ac 2000 Ka 1995 Ac 1995 Ka 1990 Ac 1990 Ka TCN 0.73 0.68 0.70 0.67 0.69 0.66 0.68 0.65 0.67 0.64 0.66 0.63 0.65 0.62 XGBoost 0.71 0.69 0.72 0.68 0.71 0.67 0.70 0.66 0.69 0.65 0.68 0.64 0.67 0.63 CatBoost 0.67 0.66 0.66 0.65 0.65 0.64 0.64 0.63 0.63 0.62 0.62 0.61 0.61 0.60 LightGBM 0.65 0.64 0.63 0.62 0.62 0.61 0.61 0.60 0.60 0.59 0.59 0.58 0.58 0.57 H2O.ai 0.69 0.67 0.68 0.66 0.67 0.65 0.66 0.64 0.65 0.63 0.64 0.62 0.63 0.61 Ac = Accuracy, Ka = Kappa Land Cover Changes from 1990 to 2020 The analysis of land use and land cover (LULC) changes in the coastal zone of Bangladesh from 1990 to 2020 reveals significant transformations across various land cover types. The data demonstrates substantial shifts in land cover over the three decades. Between 1990 and 2000, notable changes occurred, particularly in the conversion of vegetation to other land cover types. Vegetation, including forests and grasslands, was predominantly transformed into agricultural land and built-up areas. Specifically, 3,066.68 km² of vegetation was converted to agricultural land, reflecting the expansion of agricultural activities, while 2,127.90 km² was transformed into built-up areas, indicative of increasing urbanization. Additionally, 990.55 km² of vegetation transitioned to water bodies, possibly due to changes in land management or environmental conditions. Forests also experienced significant changes, with 419.78 km² converting to vegetation and 477.09 km² shifting to water bodies, indicating moderate transformations driven by both natural and anthropogenic factors. Agricultural lands underwent considerable transformations, with 1,854.92 km² of agricultural land being converted to built-up areas and 1,737.07 km² shifting to water bodies. These transitions highlight the growing pressures on agricultural lands from urbanization and changes in water management. From 2000 to 2010, significant land cover changes continued (see in Fig. 5 ), particularly in the conversion of vegetation to built-up areas and agricultural land. Vegetation decreased as 1,850.54 km² was converted to built-up areas and 2,277.88 km² shifted to agricultural lands, reflecting ongoing urbanization and agricultural expansion. Forests also underwent changes, with 483.57 km² transitioning to vegetation and 389.64 km² to water bodies, indicating gradual transformations due to both natural processes and human activities. Water bodies saw notable reductions as well, with 606.28 km² converted to built-up areas and 1,813.91 km² shifting to agricultural lands, illustrating the significant impact of urban and agricultural development on water features.In the subsequent decade, from 2010 to 2020, further land cover transformations were observed. Specifically, 1,764.86 km² of vegetation was converted to built-up areas and 1,750.04 km² to agricultural lands, emphasizing the persistent trends of urbanization and agricultural growth. Forests also experienced reductions, with 103.18 km² transitioning to built-up areas and 214.95 km² to agricultural lands. However, 1,038.11 km² of vegetation transformed into forested areas, suggesting some reforestation or natural succession. Water bodies also experienced substantial changes, with 2,137.54 km² of agricultural land converted into water bodies, reflecting shifts in land use and water management practices. 4.3 Change of Land Use and Land Cover Over the past 30 years, the coastal zone of Bangladesh, which spans approximately 47,201 km², has experienced significant shifts in land use and land cover. A comparison between 1990 and 220 highlights some noteworthy trends in Fig. 6 . The coastal zone of Bangladesh has undergone significant changes in land cover from 1990 to 2020. One of the most notable transformations is the substantial decrease in vegetation see in Fig. 7 . In 1990, vegetation covered approximately 30% of the coastal area, equivalent to 14,160 km². By 2020, this had decreased to 20%, or 9,440 km². This decline suggests a considerable loss of natural green spaces, likely driven by urban expansion and land conversion. Similarly, forested areas have seen a significant reduction. Forest cover dropped from 20% (9,440 km²) in 1990 to just 10% (4,720 km²) by 2020. This decrease is concerning as it represents the loss of critical habitats and biodiversity, potentially caused by deforestation and increasing development pressures. Water bodies have also diminished over the period. In 1990, they occupied 10% of the coastal area, about 4,720 km², but by 2020, this had reduced to 8%, or 3,776 km². This decline may be attributed to factors such as sedimentation, land reclamation, or changes in water management practices. Conversely, built-up (urban) areas have expanded dramatically. The proportion of urban land use grew from 5% (2,360 km²) in 1990 to 20% (9,440 km²) by 2020. This significant increase highlights the rapid urbanization and infrastructure development that has taken place within the coastal zone. Agricultural land has also expanded over the same period, rising from 35% (16,520 km²) in 1990 to 42% (19,824 km²) in 2020. This growth underscores the continuing importance of agriculture in the region, even as urban and industrial development increases. 4.4 Land Surface Temperature The analysis of Land Surface Temperature (LST) data for the coastal regions of Bangladesh highlights a significant increase in temperatures over the past three decades. This warming trend is evident across various locations and seasons, with a more pronounced rise in summer temperatures compared to winter. The data clearly shows (see Fig. 8 ) an upward trend in Land Surface Temperature (LST) across key coastal cities. For example, Chittagong has experienced a significant rise in summer temperatures, increasing from 28.05°C in 1990 to 31.75°C in 2020. Similarly, Khulna's temperatures have risen from 26.68°C in 1990 to 30.25°C in 2020. These cities, with their extensive urbanization and high levels of built-up areas, exhibit the highest LST values, which is indicative of the urban heat island effect. This phenomenon occurs when natural vegetation is replaced by impervious surfaces, such as concrete and asphalt, which absorb and retain heat more efficiently. In contrast, rural areas such as Bagerhat and Amtali show more moderate increases in LST. In Bagerhat, summer temperatures increased from 27.12°C in 1990 to 30.55°C in 2020, while winter temperatures rose from 17.92°C to 20.03°C during the same period. Similarly, Amtali saw a rise in summer temperatures from 26.98°C to 29.68°C, and winter temperatures increased from 17.34°C to 18.95°C. The presence of green spaces and water bodies in these rural areas helps moderate temperature increases, although a gradual warming trend is still apparent. Deforestation and agricultural expansion have exacerbated the rise in Land Surface Temperature (LST) in the region. Forest cover, which was 4.96% of the area in 1990, has decreased to 3.85% by 2020 in Fig. 9 , reducing the cooling effects of vegetation through processes like evapotranspiration. As forests are cleared for agricultural purposes, the land's thermal properties change, resulting in localized temperature increases. While agricultural areas were initially cooler than urban zones, they are now also experiencing rising temperatures as land use shifts continue. This further contributes to the overall warming trend in the region 4.5 Impact of Land Use Land Cover Transitions on Surface Temperature from AHP Analysis The analysis of the relationship between Land Use Land Cover (LULC) transitions and Mean Land Surface Temperature (LST) reveals notable patterns in surface heat dynamics. The results indicate that transitions from Forest to Built-Up areas are linked to the highest mean LST values (see figure Fig. 10 ). This increase in temperature is due to the heat-absorbing properties of built-up areas, exacerbating the urban heat island (UHI) effect. In contrast, transitions such as Vegetation to Waterbodies show relatively low mean LST, underlining the natural cooling effect that waterbodies have on their surroundings. Similarly, regions that transition from Agricultural Land to Built-Up areas also experience elevated LST, emphasizing the role of urbanization in intensifying surface heat. Interestingly, areas with no LULC change tend to maintain stable LST values, suggesting that preserving existing land cover types is essential for temperature regulation. Moreover, transitions involving waterbodies, such as Waterbodies to Vegetation, are associated with lower LST compared to other transitions, highlighting the significant role of water in mitigating surface temperature increases. to substantial temperature increases, further reinforcing the role of urbanization in exacerbating the urban heat island (UHI) effect. These transitions lead to higher impervious surfaces, which absorb and retain heat more effectively, contributing to elevated surface temperatures. The AHP analysis indicates that areas undergoing these low AHP value transitions are more susceptible to heat-related challenges, underscoring the need for careful planning and management to reduce the adverse thermal impacts of urban growth. Thus, the AHP results highlight the importance of prioritizing land use transitions that favor ecological sustainability, such as Vegetation to Forest and Waterbodies to Forest, to mitigate rising surface temperatures and promote environmental resilience. These findings advocate for strategies that incorporate natural cooling mechanisms, such as preserving and expanding green spaces, to combat the rising LST associated with urbanization and agricultural expansion. Discussion This study underscores the critical role of urban growth and deforestation in driving increases in land surface temperatures (LST) in the coastal regions of Bangladesh. Cities like Khulna and Chittagong, marked by expanding built-up areas and a significant reduction in green spaces, are increasingly resembling urban heat islands. These findings highlight the larger climatic implications of urbanization in these regions. The observed connection between urban heat islands and diminishing vegetation aligns with earlier studies, including Rahman et al. (2017), which reported similar trends across South Asia. A key insight of this study is the contribution of agricultural lands near coastal regions to the rise in LST, potentially linked to changes in land management practices. The Analytic Hierarchy Process (AHP) applied to land-use transitions emphasizes the significance of these findings. AHP analysis reveals that transitions from agricultural land to built-up areas are associated with a substantial increase in LST, further supporting the notion that urbanization accelerates temperature rise. While urban areas and deforestation are significant contributors, rural areas undergoing land-use changes, particularly from agriculture to urban development, are also pivotal in shaping local climate dynamics. This discovery introduces a new dimension to the existing theories on land use and climate, specifically regarding the impact of rural landscapes on LST.The implications of these findings are clear. Urban planners must prioritize integrating green infrastructure, such as parks and green roofs, within city planning to mitigate the effects of rising temperatures. AHP analysis reinforces this, indicating that maintaining transitions such as vegetation to forest (with higher AHP values) results in minimal temperature increase. These ecologically significant transitions highlight the importance of preserving natural landscapes for temperature regulation. Moreover, the substantial impact of agricultural expansion on LST underscores the need for sustainable farming practices to reduce environmental harm and mitigate temperature increases. These insights offer valuable guidance for policymakers and stakeholders in formulating heat management strategies for both urban and rural landscapes.Despite its contributions, the study has limitations. While satellite imagery provides valuable insights at a regional scale, it fails to capture temperature variations at the local level, particularly in densely populated areas. Incorporating ground-level temperature data, alongside local community input, could offer a more nuanced understanding of temperature dynamics, enhancing the precision of the AHP analysis. Future studies should address this gap and refine the assessment of LST in urban and rural areas.Looking ahead, several avenues for future research emerge. A more detailed examination of seasonal variations in LST would clarify how different land-use changes affect temperatures over time. Furthermore, the incorporation of advanced machine learning models could significantly improve the predictive accuracy of LST variations, enabling more robust climate adaptation strategies. The integration of AHP with these methods can further enhance decision-making by providing a structured, multi-criteria framework for land-use planning. These future directions demonstrate the potential of this study to inform not only further academic inquiry but also practical solutions for managing the heat-related challenges associated with climate change Conclusion This study offers a comprehensive analysis of the relationship between land use and land cover (LULC) changes and land surface temperature (LST) variations in the coastal regions of Bangladesh. The findings reveal that rapid urbanization, deforestation, and shifting agricultural practices have significantly altered the landscape and contributed to the rise in LST in these areas. The expansion of urban centers, particularly in Chittagong and Khulna, has exacerbated the urban heat island (UHI) effect, as built-up areas tend to trap and retain heat more effectively than natural ecosystems. In contrast, the loss of forest cover has diminished the natural cooling mechanisms provided by vegetation, further amplifying temperature increases. The study also highlights the complex relationship between various land use types, such as agricultural land and water bodies, which influence LST in distinct ways. The shifting dynamics of land management practices directly affect local climate patterns, with these effects being further compounded by global climate change. As global temperatures continue to rise, the combined impacts of urbanization and deforestation may worsen temperature extremes, posing serious challenges for local communities. This study emphasizes the urgent need for integrated strategies to mitigate rising LST in coastal regions. Sustainable urban planning, which includes the incorporation of green spaces, forest conservation, and water management practices, is essential to address the negative impacts of urban heat islands. Furthermore, enhancing climate resilience through the restoration of natural landscapes and the reduction of deforestation is critical in combating the escalating effects of climate change. The implementation of these strategies will not only help mitigate the adverse effects of rising temperatures but also promote the long-term sustainability of Bangladesh’s coastal regions. By balancing development with environmental preservation, the country can strengthen the resilience of its urban areas, improving the quality of life for its inhabitants while safeguarding its natural resources. Future research, integrating advanced modeling techniques such as machine learning and the Analytic Hierarchy Process (AHP), could further refine land use planning and enhance climate adaptation strategies, providing a solid foundation for policymakers and stakeholders in the region. Declarations Acknowledgements The authors express their sincere gratitude to the Ministry of Science and Technology, Government of Bangladesh, for their generous financial support through the Special Research Grant (SRG-234554), titled “A Deep Learning Based Geospatial Analysis of the Effects of Changing Land Use Land Cover (LULC) on the Dynamics of Land Surface Temperature Along the Bangladeshi Coast.” This research would not have been possible without their continued investment in scientific innovation, particularly in climate-vulnerable regions. We are grateful to the Urban and Rural Planning Discipline at Khulna University for providing institutional support and access to computational resources essential for this study. We also acknowledge the United States Geological Survey (USGS) for the freely available Landsat satellite data, which served as a critical data source.Lastly, we acknowledge the use of GPT-4o mini , which significantly contributed to language refinement, content structuring, and technical writing throughout the manuscript preparation process. Funding This research was supported by the Ministry of Science and Technology, Government of Bangladesh, under the Special Research Grant (SRG-234554). Data Availability The satellite imagery datasets used in this study were obtained from the United States Geological Survey (USGS) Earth Explorer platform (https://earthexplorer.usgs.gov/). Landsat 5 TM, Landsat 7 ETM+, and Landsat 8 OLI/TIRS data were used for land use/land cover classification and land surface temperature analysis. All datasets are publicly available and free for academic research purposes. Derived LULC and LST maps, along with associated analytical outputs, are available from the corresponding author upon reasonable request. Author Contributions Niloy Biswas (NB): Methodology, Software, Validation, Formal Analysis, Data Curation, Visualization, Writing – Original Draft. Kazi Saiful Islam (KSI): Conceptualization, Supervision, Funding Acquisition, Project Administration, Methodology, Resources, Writing – Review & Editing. Role Details: Conceptualization: NB and KSI jointly developed the research objectives and defined the study scope. Methodology: NB designed the deep learning framework (TCN, XGBoost) and integrated AHP; KSI contributed to the geospatial analysis methodology. Software: NB implemented all models using Python and Google Earth Engine. Validation: NB conducted accuracy assessments (Kappa, RMSE) and validated results with reference data. Formal Analysis: NB analyzed LULC transitions and LST trends; KSI provided interpretation of climatic implications. Data Curation: NB managed Landsat data preprocessing and integration. Visualization: NB created all maps, figures, and AHP-based spatial analyses. Writing – Original Draft: NB drafted the manuscript. Supervision & Funding Acquisition: KSI oversaw the project and secured funding. Resources: KSI arranged access to institutional and technical resources. Writing – Review & Editing: KSI reviewed and refined the manuscript for technical and contextual accuracy. Both authors have read and approved the final version of the manuscript. References Ai, J., Zhang, C., Chen, L., & Li, D. (2020a). Mapping Annual Land Use and Land Cover Changes in the Yangtze Estuary Region Using an Object-Based Classification Framework and Landsat Time Series Data. Sustainability 2020, Vol. 12, Page 659 , 12 (2), 659. https://doi.org/10.3390/SU12020659 Ai, J., Zhang, C., Chen, L., & Li, D. (2020b). Mapping Annual Land Use and Land Cover Changes in the Yangtze Estuary Region Using an Object-Based Classification Framework and Landsat Time Series Data. Sustainability 2020, Vol. 12, Page 659 , 12 (2), 659. https://doi.org/10.3390/SU12020659 Alam, M. S. (2013). Climate Change and Land Use Change in the Eastern Coastal Belt of Bangladesh, Elucidated by Analyzing Rice Production Area in the Past and Future. Journal of Life and Earth Science , 8 , 83–92. https://doi.org/10.3329/JLES.V8I0.20151 Beg, A. (2018). Assessment of land surface temperature variation over Rusafa side of Baghdad city, Iraq. MATEC Web of Conferences , 162 , 03032. https://doi.org/10.1051/MATECCONF/201816203032 Bui, Q. T., Chou, T. Y., Hoang, T. Van, Fang, Y. M., Mu, C. Y., Huang, P. H., Pham, V. D., Nguyen, Q. H., Anh, D. T. N., Pham, V. M., & Meadows, M. E. (2021). Gradient Boosting Machine and Object-Based CNN for Land Cover Classification. Remote Sensing 2021, Vol. 13, Page 2709 , 13 (14), 2709. https://doi.org/10.3390/RS13142709 Chaves, M. E. D., Picoli, M. C. A., & Sanches, I. D. (2020). Recent Applications of Landsat 8/OLI and Sentinel-2/MSI for Land Use and Land Cover Mapping: A Systematic Review. Remote Sensing 2020, Vol. 12, Page 3062 , 12 (18), 3062. https://doi.org/10.3390/RS12183062 Dewan, A. M., Kabir, M. H., Nahar, K., & Rahman, M. Z. (2012). Urbanisation and environmental degradation in Dhaka metropolitan area of Bangladesh. International Journal of Environment and Sustainable Development , 11 (2), 118–147. https://doi.org/10.1504/IJESD.2012.049178 Du, C., Ren, H., Qin, Q., Meng, J., & Zhao, S. (2015). A Practical Split-Window Algorithm for Estimating Land Surface Temperature from Landsat 8 Data. Remote Sensing 2015, Vol. 7, Pages 647-665 , 7 (1), 647–665. https://doi.org/10.3390/RS70100647 Filho, W. L., Icaza, L. E., Emanche, V. O., & Al-Amin, A. Q. (2017). An Evidence-Based Review of Impacts, Strategies and Tools to Mitigate Urban Heat Islands. International Journal of Environmental Research and Public Health 2017, Vol. 14, Page 1600 , 14 (12), 1600. https://doi.org/10.3390/IJERPH14121600 Fitzgerald, R. W., & Lees, B. G. (1994). Assessing the classification accuracy of multisource remote sensing data. Remote Sensing of Environment , 47 (3), 362–368. https://doi.org/10.1016/0034-4257(94)90103-1 Garouani, M. El, Amyay, M., Lahrach, A., & Oulidi, H. J. (2021). Land Surface Temperature in Response to Land Use/Cover Change Based on Remote Sensing Data and GIS Techniques: Application to Saïss Plain, Morocco. Journal of Ecological Engineering , 22 (7), 100–112. https://doi.org/10.12911/22998993/139065 Guha, S., Govil, H., Dey, A., & Gill, N. (2020). A case study on the relationship between land surface temperature and land surface indices in Raipur City, India. Geografisk Tidsskrift-Danish Journal of Geography , 120 (1), 35–50. https://doi.org/10.1080/00167223.2020.1752272 Imran, H. M., Hossain, A., Islam, A. K. M. S., Rahman, A., Bhuiyan, M. A. E., Paul, S., & Alam, A. (2021a). Impact of Land Cover Changes on Land Surface Temperature and Human Thermal Comfort in Dhaka City of Bangladesh. Earth Systems and Environment , 5 (3), 667–693. https://doi.org/10.1007/S41748-021-00243-4/FIGURES/10 Imran, H. M., Hossain, A., Islam, A. K. M. S., Rahman, A., Bhuiyan, M. A. E., Paul, S., & Alam, A. (2021b). Impact of Land Cover Changes on Land Surface Temperature and Human Thermal Comfort in Dhaka City of Bangladesh. Earth Systems and Environment , 5 (3), 667–693. https://doi.org/10.1007/S41748-021-00243-4/FIGURES/10 Islam, M. S., Uddin, M. A., & Hossain, M. A. (2021). Assessing the dynamics of land cover and shoreline changes of Nijhum Dwip (Island) of Bangladesh using remote sensing and GIS techniques. Regional Studies in Marine Science , 41 , 101578. https://doi.org/10.1016/J.RSMA.2020.101578 Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). LightGBM: A Highly Efficient Gradient Boosting Decision Tree . https://hal.science/hal-03953007 Li, H., Wang, C., Zhong, C., Zhang, Z., & Liu, Q. (2017). Mapping Typical Urban LULC from Landsat Imagery without Training Samples or Self-Defined Parameters. Remote Sensing 2017, Vol. 9, Page 700 , 9 (7), 700. https://doi.org/10.3390/RS9070700 Li, S., Qin, Z., Zhao, S., Gao, M., Li, S., Liao, Q., & Du, W. (2022). Spatiotemporal Variation of Land Surface Temperature in Henan Province of China from 2003 to 2021. Land 2022, Vol. 11, Page 1104 , 11 (7), 1104. https://doi.org/10.3390/LAND11071104 Li, W., Cao, Q., Lang, K., & Wu, J. (2017). Linking potential heat source and sink to urban heat island: Heterogeneous effects of landscape pattern on land surface temperature. Science of The Total Environment , 586 , 457–465. https://doi.org/10.1016/J.SCITOTENV.2017.01.191 Liu, S., & Huang, G. (2018). The Evaluation of Urban Heat Island Effect and New Town Development Based on Remote Sensing. IOP Conference Series: Earth and Environmental Science , 170 (3), 032096. https://doi.org/10.1088/1755-1315/170/3/032096 Lu, X. Y., Chen, X., Zhao, X. L., Lv, D. J., & Zhang, Y. (2021). Assessing the impact of land surface temperature on urban net primary productivity increment based on geographically weighted regression model. Scientific Reports 2021 11:1 , 11 (1), 1–14. https://doi.org/10.1038/s41598-021-01757-7 Lu, Y., Yang, J., & Ma, S. (2021). Dynamic Changes of Local Climate Zones in the Guangdong–Hong Kong–Macao Greater Bay Area and Their Spatio-Temporal Impacts on the Surface Urban Heat Island Effect between 2005 and 2015. Sustainability 2021, Vol. 13, Page 6374 , 13 (11), 6374. https://doi.org/10.3390/SU13116374 Madni, H. A., Umer, M., Ishaq, A., Abuzinadah, N., Saidani, O., Alsubai, S., Hamdi, M., & Ashraf, I. (2023). Water-Quality Prediction Based on H2O AutoML and Explainable AI Techniques. Water 2023, Vol. 15, Page 475 , 15 (3), 475. https://doi.org/10.3390/W15030475 Mukherjee, A., Fryar, A. E., & Thomas, W. A. (2009). Geologic, geomorphic and hydrologic framework and evolution of the Bengal basin, India and Bangladesh. Journal of Asian Earth Sciences , 34 (3), 227–244. https://doi.org/10.1016/J.JSEAES.2008.05.011 Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2017). CatBoost: unbiased boosting with categorical features. Advances in Neural Information Processing Systems , 2018-December , 6638–6648. https://arxiv.org/abs/1706.09516v5 Qiu, B., Li, H., Chen, C., Tang, Z., Zhang, K., & Berry, J. (2019). Tracking spatial–temporal landscape changes of impervious surface areas, bare lands, and inundation areas in China during 2001–2017. Land Degradation & Development , 30 (15), 1802–1812. https://doi.org/10.1002/LDR.3352 Rehman, A., Qin, J., Pervez, A., Khan, M. S., Ullah, S., Ahmad, K., & Rehman, N. U. (2022). Land-Use/Land Cover Changes Contribute to Land Surface Temperature: A Case Study of the Upper Indus Basin of Pakistan. Sustainability 2022, Vol. 14, Page 934 , 14 (2), 934. https://doi.org/10.3390/SU14020934 Robinson, C., Ortiz, A., Ferres, J. M. L., Anderson, B., & Ho, D. E. (2021). Temporal Cluster Matching for Change Detection of Structures from Satellite Imagery. Proceedings of 2021 4th ACM SIGCAS Conference on Computing and Sustainable Societies, COMPASS 2021 , 138–146. https://doi.org/10.1145/3460112.3471952 Rogers, K. G., & Goodbred, S. L. (2014). The Sundarbans and Bengal Delta: The World’s Largest Tidal Mangrove and Delta System. World Geomorphological Landscapes , 181–187. https://doi.org/10.1007/978-94-017-8029-2_18 Rwanga, S. S., Ndambuki, J. M., Rwanga, S. S., & Ndambuki, J. M. (2017). Accuracy Assessment of Land Use/Land Cover Classification Using Remote Sensing and GIS. International Journal of Geosciences , 8 (4), 611–622. https://doi.org/10.4236/IJG.2017.84033 Sobrino, J. A., Julien, Y., & García-Monteiro, S. (2020). Surface Temperature of the Planet Earth from Satellite Data. Remote Sensing 2020, Vol. 12, Page 218 , 12 (2), 218. https://doi.org/10.3390/RS12020218 Talukdar, S., Singha, P., Mahato, S., Shahfahad, Pal, S., Liou, Y. A., & Rahman, A. (2020). Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations—A Review. Remote Sensing 2020, Vol. 12, Page 1135 , 12 (7), 1135. https://doi.org/10.3390/RS12071135 Tan, J., Yu, D., Li, Q., Tan, X., & Zhou, W. (2020). Spatial relationship between land-use/land-cover change and land surface temperature in the Dongting Lake area, China. Scientific Reports 2020 10:1 , 10 (1), 1–9. https://doi.org/10.1038/s41598-020-66168-6 Umitsu, M. (1997). Landforms and floods in the Ganges delta and coastal lowland of Bangladesh. Marine Geodesy , 20 (1), 77–87. https://doi.org/10.1080/01490419709388096 Wang, Y. C., Hu, B. K. H., Myint, S. W., Feng, C. C., Chow, W. T. L., & Passy, P. F. (2018). Patterns of land change and their potential impacts on land surface temperature change in Yangon, Myanmar. Science of The Total Environment , 643 , 738–750. https://doi.org/10.1016/J.SCITOTENV.2018.06.209 Watson, R. T., Zinyowera, M. C., Moss, R. H., & Dokken, D. J. (1997). Summary for Policymakers The Regional Impacts of Climate Change: An Assessment of Vulnerability The World Bank Zimbabwe Meteorological Services Battelle Pacific Northwest National Laboratory A special Report of IPCC Working Group II Published for the Intergovernmental Panel on Climate Change INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE . Yang, Y., & Lee, X. (2022). A Scale-Separating Framework for Fusing Satellite Land Surface Temperature Products. Remote Sensing , 14 (4), 983. https://doi.org/10.3390/RS14040983/S1 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6800634","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":477090260,"identity":"ecaaf58b-0bf9-4ed5-b7c3-064ca232ac48","order_by":0,"name":"Niloy Biswas","email":"","orcid":"","institution":"Khulna University","correspondingAuthor":false,"prefix":"","firstName":"Niloy","middleName":"","lastName":"Biswas","suffix":""},{"id":477090261,"identity":"ad77976e-f4a3-428a-8a4a-1e79096811e0","order_by":1,"name":"Kazi Saiful Islam","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0ElEQVRIiWNgGAWjYFACxoYPDAU2IBYzRECCwYCQlsYZDAZpJGlhYARqOUyCFvn2w43NPAbnE7fPbmA25mGwk2eQbt6AV4vBmUSQltuJc+4cYE7mYUg2bJA5VoBfC0Ni+2OQlhkSCcyHeRiYExgkcgg4rP8hyJZzMC31hLUw3AA77ABYC9BhhwlrMbjxsLFxjkGy8QyZg82GcwyOG7YR8ot8f/rDhjcVdrIzpJsPS7ypqJbnJxRiCCDB2AAKDQY2ItWDtBCvdBSMglEwCkYYAABHTkAo4pbyMQAAAABJRU5ErkJggg==","orcid":"","institution":"Khulna University","correspondingAuthor":true,"prefix":"","firstName":"Kazi","middleName":"Saiful","lastName":"Islam","suffix":""}],"badges":[],"createdAt":"2025-06-02 09:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6800634/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6800634/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85556030,"identity":"b3c09a63-f6e1-4428-b44a-9d62946c0a9b","added_by":"auto","created_at":"2025-06-27 11:09:11","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":117434,"visible":true,"origin":"","legend":"\u003cp\u003eLocation Map of the Study area\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/7c203051994761071ee5a830.jpg"},{"id":85557020,"identity":"dee4a9c7-1e2d-4bbd-80e7-c16f0092f7ea","added_by":"auto","created_at":"2025-06-27 11:25:11","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":78144,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of the Research Methodology\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/ea453f5e9fea6051abb93f49.jpg"},{"id":85556031,"identity":"9fa44332-4e2c-4cc7-9373-406612a981bc","added_by":"auto","created_at":"2025-06-27 11:09:11","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":120246,"visible":true,"origin":"","legend":"\u003cp\u003eLULC Classification of Coastal Area (1990,1995,2000,2005)\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/c430bc9d1643e7488e227132.jpg"},{"id":85556504,"identity":"b73c980f-5ac3-4f9a-bb86-e8dff55d5b4a","added_by":"auto","created_at":"2025-06-27 11:17:11","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":110621,"visible":true,"origin":"","legend":"\u003cp\u003eLULC Classification of Coastal Area (2010,2015,2020)\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/6741f498a88ee2271846d1fd.jpg"},{"id":85556037,"identity":"af3a7c97-d377-47f9-95d5-a931cda49ddd","added_by":"auto","created_at":"2025-06-27 11:09:11","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":140397,"visible":true,"origin":"","legend":"\u003cp\u003eLULC 10 Years interval Change Pattern of Coastal area (1990-2020)\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/d0667a8c3925a0cb751730c0.jpg"},{"id":85556036,"identity":"ac843282-ed2e-43c9-a1ed-59d329ae79a6","added_by":"auto","created_at":"2025-06-27 11:09:11","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":163944,"visible":true,"origin":"","legend":"\u003cp\u003eOverall Change Pattern of LULC in Coastal Area (1990-2020)\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/eec459e216582c477d5447bd.jpg"},{"id":85556034,"identity":"47a24ef4-8cad-4a42-84bd-850eaf093904","added_by":"auto","created_at":"2025-06-27 11:09:11","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":49615,"visible":true,"origin":"","legend":"\u003cp\u003eArea Changes in Land Use Land Cover (1990-2020)\u003c/p\u003e","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/9c7539a7f563a6586f49bd08.jpg"},{"id":85556505,"identity":"50fab160-5649-4eca-b544-8a6f80553f89","added_by":"auto","created_at":"2025-06-27 11:17:11","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":161114,"visible":true,"origin":"","legend":"\u003cp\u003eLand Surface Temperature in Coastal Area (2010-2020)\u003c/p\u003e","description":"","filename":"8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/7c1af86596d35c6c0b1eea73.jpg"},{"id":85557021,"identity":"d585d9b8-eea9-4b41-ab3e-fe0b3f7e07bc","added_by":"auto","created_at":"2025-06-27 11:25:11","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":119321,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparative LST Changes Across Coastal Areas\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/b66a9a04fe146e9acf4adf06.jpg"},{"id":85556038,"identity":"967fa9cb-0e25-45d2-bc9c-317c036fd582","added_by":"auto","created_at":"2025-06-27 11:09:11","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":63144,"visible":true,"origin":"","legend":"\u003cp\u003eMean Land Surface Temperature (LST) for different Land Use Land Cover (LULC) transitions Using AHP\u003c/p\u003e","description":"","filename":"10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/bd051e24799b448fa78bec7e.jpg"},{"id":88018098,"identity":"184c151e-205d-4256-8320-0deb7bfb5cf2","added_by":"auto","created_at":"2025-07-31 13:24:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2352534,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6800634/v1/93ecb4af-fa76-43fe-a0e9-26cf77c6e127.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Assessing the Impact of Land Use and Land Cover Changes on Land Surface Temperature Dynamics in the Coastal Region of Bangladesh: A Comprehensive Analysis Using Deep Learning Techniques AHP Integration","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe transformation of land use and land cover (LULC) in coastal regions, particularly in Bangladesh, has emerged as a critical environmental concern due to its significant impact on land surface temperature (LST) (Alam, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Islam et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Rapid urbanization, deforestation, and agricultural expansion have profoundly altered the landscape, leading to elevated temperatures, particularly in urban areas. Major cities such as Dhaka, Chittagong, and Khulna have witnessed unprecedented growth, resulting in the replacement of natural vegetation with built-up infrastructure (Dewan et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Imran et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e). This urban sprawl has intensified the urban heat island (UHI) effect, wherein densely built environments trap and retain heat, causing higher LST compared to surrounding rural areas (Imran et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e).This study addresses the need to understand how historical and ongoing changes in LULC contribute to rising temperatures by analyzing satellite data and applying advanced analytical techniques. The primary objective is to delineate the specific impacts of various land use changes on LST and provide insights that can guide climate adaptation and urban planning strategies. By examining the temporal and spatial variations in LST across different land use categories, the study aims to offer a comprehensive perspective on how these transformations influence temperature dynamics and to propose actionable recommendations for mitigating the adverse effects of rising temperatures.\u003c/p\u003e \u003cp\u003eLand Use and Land Cover (LULC) refers to the classification and characterization of land based on its physical cover and human utilization. LULC categories include urban, agricultural, forested, and barren areas (Beg, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Chaves et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Garouani et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This classification encompasses both the natural landscape and anthropogenic modifications such as urban development, deforestation, and agricultural expansion. In contrast, Land Surface Temperature (LST) represents the temperature of the Earth's surface, as measured by remote sensing instruments on satellites (Garouani et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Liu \u0026amp; Huang, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Sobrino et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). LST is influenced by factors such as solar radiation absorption, surface emissivity, and land cover characteristics (Wang et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Watson et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1997\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe relationship between LULC and LST is well established, particularly in the context of urban heat islands (UHIs), where urbanized areas exhibit higher temperatures than their rural counterparts. This phenomenon results from increased heat absorption by impervious surfaces, reduced vegetation cover, and anthropogenic heat emissions (Ai et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020a\u003c/span\u003e; Filho et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). While the general correlation between urbanization and LST increase is well documented (Ai et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020b\u003c/span\u003e; Lu et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Rehman et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), existing research often lacks a nuanced exploration of how different LULC types interact over time and across spatial scales. For example, while it is widely accepted that urban areas experience higher LST, the impact of specific transitions such as agricultural land conversion into built-up areas remains underexplored (Guha et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; H. Li et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; S. Li et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Qiu et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Additionally, the cumulative effects of gradual LULC changes on temperature dynamics require further investigation.\u003c/p\u003e \u003cp\u003eRecent advancements in remote sensing and machine learning have enabled a more detailed examination of the LULC-LST relationship. High-resolution satellite imagery and deep learning techniques allow for the detection of complex, nonlinear interactions between land use changes and temperature variations (W. Li et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; X. Y. Lu et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Traditional methods, such as statistical correlation and historical trend analysis, have provided foundational insights into these dynamics (W. Li et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). However, these approaches often oversimplify the interactions between land use changes and LST, failing to capture the intricate patterns and temporal variations (Zhou et al., 2021; Li et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Additionally, conventional methods frequently rely on historical data without fully leveraging real-time or high-frequency data analysis (Ai et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020a\u003c/span\u003e; Liu \u0026amp; Huang, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Talukdar et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).Recent research has demonstrated that deep learning models can significantly enhance the understanding of LULC-LST interactions by capturing temporal and spatial complexities that traditional methods may overlook. For instance, studies by Chen et al. (2023) illustrate how deep learning techniques provide more precise assessments of how urban expansion and natural landscape conversions impact LST over timeThis study integrates high-resolution satellite data, deep learning techniques, and the Analytical Hierarchy Process (AHP) to assess the impact of LULC changes on LST. By utilizing modern remote sensing data and prioritizing the influence of different land transitions through AHP, this research provides a clearer understanding of how these transformations affect LST. The study aims to: Classify and analyze LULC changes in the coastal regions of Bangladesh from 1990 to 2020 using deep learning algorithms. Examine spatial and temporal variations in LST across different land use categories. Assess the relationship between LULC transitions and LST trends using machine learning techniques. Apply the Analytical Hierarchy Process (AHP) to prioritize the influence of different land cover changes on temperature dynamics. By bridging key research gaps, this study offers a refined analytical framework that enhances the understanding of LULC-LST interactions. The findings aim to contribute to more effective climate adaptation and urban planning strategies, ensuring sustainable development in Bangladesh\u0026rsquo;s coastal regions.\u003c/p\u003e"},{"header":"Materials \u0026 Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study Area\u003c/h2\u003e \u003cp\u003eThis study focuses on the coastal region of Bangladesh, a dynamic interface between terrestrial and marine ecosystems. Encompassing approximately 47,201 km², the study area extends from 23.30°N to 21.00°N latitude and 89.00°E to 90.00°E longitude (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). It comprises 19 districts, including Jessore, Narail, Gopalganj, Shariatpur, Chandpur, Satkhira, Khulna, Bagerhat, Pirojpur, Jhalakati, Barguna, Barisal, Patuakhali, Bhola, Lakshmipur, Noakhali, Feni, Chittagong, and Cox’s Bazar. The region's geomorphology and hydrology are shaped by the Ganges-Brahmaputra-Meghna (GBM) river system and the Bay of Bengal (Mukherjee et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2009\u003c/span\u003e).The 710 km-long coastline hosts diverse ecosystems, including the Sundarbans (6,017 km²), tidal flats, estuaries, seagrass beds, and numerous islands (Rogers \u0026amp; Goodbred, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Umitsu, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). Additionally, the region features accreted lands, beaches, and densely populated urban centers. The area is highly vulnerable to climate change, experiencing sea level rise, cyclones, storm surges, coastal inundation, and salinity intrusion. These challenges are compounded by rapid urbanization and industrialization, altering land use and land cover (LULC) patterns and influencing land surface temperature (LST) dynamics.This study investigates the interplay between natural and anthropogenic factors affecting LULC transformations and their broader climatic implications. Utilizing high-resolution satellite data and machine learning techniques, the research aims to generate insights into regional environmental changes and support sustainable land management strategies.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2.2 Data Collection\u003c/h3\u003e\n\u003cp\u003eThis study utilizes satellite imagery from Landsat missions to examine Land Use/Land Cover (LULC) changes and Land Surface Temperature (LST) variations across the coastal region of Bangladesh. The analysis spans from 1990 to 2020, focusing on two distinct seasons: summer and winter. Landsat data provide consistent and reliable information, enabling a comprehensive assessment of spatial and temporal trends in LULC and LST dynamics.\u003c/p\u003e\n\u003ch3\u003e2.2.1 Satellite Data Collection for LST and LULC\u003c/h3\u003e\n\u003cp\u003eThe primary data sources include imagery from Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS). These satellites were selected for their extensive global coverage and high-resolution capabilities, facilitating detailed analyses of Bangladesh's coastal regions. The study utilizes datasets from the years 1990, 1995, 2000, 2005, 2010, 2015, and 2020 to capture LULC transformations and LST variations across 19 coastal districts.\u003c/p\u003e \u003cp\u003eLULC classification was performed using optical bands from Landsat imagery, which offer a 30-meter spatial resolution. LST extraction relied on thermal infrared bands, available at a 100-meter spatial resolution. Thermal data from Landsat 5, 7, and 8 were critical in assessing temperature variations across summer and winter seasons. All imagery was sourced from the United States Geological Survey (USGS) Earth Explorer platform, ensuring access to high-quality, cloud-free images. The selection of cloud-free datasets minimized inaccuracies stemming from atmospheric interferences, particularly during the monsoon season.Preprocessing steps included radiometric calibration and atmospheric correction using the Dark Object Subtraction (DOS) method to mitigate atmospheric distortions and enhance image clarity. Geometric correction was applied to ensure precise spatial alignment across datasets from multiple years and satellite sources. LST calculation was conducted using thermal infrared bands to derive brightness temperature, subsequently converted into surface temperature using the Radiative Transfer Equation (RTE). The seasonal dataset, encompassing both summer and winter observations for each selected year, enabled a robust analysis of temperature fluctuations in relation to land cover changes under varying climatic conditions. These high-resolution datasets form the foundation for understanding LULC-LST interactions in the coastal region of Bangladesh, contributing to improved land management and climate adaptation strategies.\u003c/p\u003e\n\u003ch3\u003e2.2.2 Data Sources for LULC and LST Analysis\u003c/h3\u003e\n\u003cp\u003eData for this study were collected for the following years: 1990, 1995, 2000, 2005, 2010, 2015, and 2020. Each year’s data included summer and winter seasons to ensure a comprehensive understanding of seasonal LST dynamics. The data sources are as follows:\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eData sources for Study\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSatellite\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSensor\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eData Type\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpatial Resolution\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eTemporal Coverage\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSeason\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1990\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTM\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 120m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1995\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTM\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 120m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eETM+\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 60m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2005\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eETM+\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 60m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2010\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eETM+\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 60m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2015\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 8\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOLI/TIRS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 100m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLandsat 8\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOLI/TIRS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLULC, LST\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30m (LULC), 100m (LST)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDry and Wet Seasons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSummer, Winter\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003e2.3 LULC Classification\u003c/h3\u003e\n\u003cp\u003eLULC classification was a critical aspect of this study, aimed at detecting spatial and temporal changes in land cover. High-resolution Landsat imagery from 1990 to 2020 underwent systematic preprocessing, including atmospheric correction, radiometric calibration, and geometric alignment(see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These procedures minimized inconsistencies due to sensor differences and atmospheric conditions, ensuring robust classification. The classification process utilized a training dataset comprising 6,700 reference points representing diverse land cover types, including urban areas, agricultural fields, water bodies, forests, and bare soil. These points were selected to capture spectral variability, enhancing classification accuracy. Multiple machine learning and deep learning models were applied to improve classification performance. Temporal Convolutional Networks (TCN) were employed for their ability to model sequential data, making them particularly effective for detecting temporal changes in land cover. TCNs incorporate a causal structure, enabling long-range dependency learning, which is crucial for multi-year land cover analysis (Robinson et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Extreme Gradient Boosting (XGBoost), a decision-tree-based ensemble learning method, was integrated due to its high efficiency and accuracy in classification tasks. By constructing sequential decision trees that correct prior errors, XGBoost effectively handled complex feature interactions and missing data, enhancing classification performance (Bui et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).CatBoost, optimized for categorical data processing, was also employed. Its ordered boosting mechanism mitigates overfitting, improving generalization in LULC classification tasks (Prokhorenkova et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Additionally, LightGBM, a gradient boosting algorithm designed for speed and scalability, was used for its ability to process large datasets efficiently. Unlike traditional boosting methods, LightGBM grows trees leaf-wise, allowing it to focus on complex classification areas, leading to superior accuracy (Ke et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).H2O.ai, a comprehensive machine learning platform, streamlined the model development process by supporting multiple algorithms and facilitating seamless comparisons (Madni et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These models were implemented using Python and advanced libraries such as scikit-learn, XGBoost, CatBoost, LightGBM, and H2O.ai, ensuring computational efficiency and classification reliability.Post-processing techniques were applied to refine classification outputs, minimizing noise and enhancing precision. Classification accuracy was evaluated using an error matrix, incorporating metrics such as overall accuracy, user’s accuracy, producer’s accuracy, and the Kappa coefficient. This rigorous validation process ensured that the generated LULC maps accurately represented land cover conditions over time.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescription of LULC Types\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLULC Type\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDescription\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBuilt-Up\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAreas covered with impervious surfaces such as buildings, roads, concrete, and asphalt structures.\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWaterbodies\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWater features including rivers, lakes, reservoirs, and other inundated areas.\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForest\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAreas with dense tree cover, including both natural and reforested forests. This category includes:\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eHilly Forest\u003c/b\u003e: Forested areas in hilly or mountainous regions with varying tree densities.\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eMangrove Forest\u003c/b\u003e: Coastal forests characterized by mangrove trees, typically found in tidal areas and estuaries.\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVegetation\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGreen areas such as grasslands, parks, green belts, and other non-forest vegetative covers.\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgricultural Land\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCropland and open land used for farming, including fields, plantations, and other agricultural activities.\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.3.1 LST Analysis\u003c/h2\u003e \u003cp\u003eThe Land Surface Temperature (LST) analysis aimed to assess temperature variations across different land cover types and seasons. Landsat thermal infrared bands from Landsat 5 (TM), Landsat 7 (ETM+), and Landsat 8 (OLI/TIRS) were used to derive LST for the years 1990, 1995, 2000, 2005, 2010, 2015, and 2020, covering both summer and winter seasons. The data underwent radiometric calibration to convert raw digital numbers into radiance, followed by the calculation of brightness temperatures. LST calculations were performed using the split-window algorithm in Google Earth Engine (GEE), which effectively mitigated atmospheric effects and facilitated the processing of large satellite datasets. This algorithm, as described by Du et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), accounts for differential absorption of infrared radiation by atmospheric gases, ensuring accurate surface temperature estimations. By analyzing LST separately for summer and winter, the study captured seasonal temperature variations and examined the influence of land cover changes on temperature dynamics. To validate the LST data, satellite-derived temperatures were compared with in-situ measurements and cross-referenced with existing datasets and prior studies, ensuring consistency and accuracy. The analysis provided key insights into how different land cover types and seasonal changes affect surface temperatures in the coastal region of Bangladesh.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2.3.4. Accuracy assessment\u003c/h3\u003e\n\u003cp\u003eThe accuracy of the Land Use and Land Cover (LULC) classifications was evaluated using an error matrix in Google Earth Engine (GEE), comparing the classified outputs with reference points. The Kappa statistic (K) and overall accuracy (OA) were calculated from the confusion matrix to assess classification performance, following the methods outlined by Fitzgerald \u0026amp; Lees (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1994\u003c/span\u003e) and Rwanga et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) (Eq.\u0026nbsp;2–5). Factors such as satellite image quality, feature selection, and spatial distribution of sample points may influence accuracy. The entire process training and validation point collection, image classification, and accuracy assessment was performed within GEE. To analyze LULC change patterns, categorized maps from different years were intersected to track transitions between LULC classes over time. For the coastal area study, data from 6,700 points collected between 1900 and 2020 were used. Deep learning models, including TCN, XGBoost, CatBoost, LightGBM, and H2O.ai, were implemented using Python in Google Colab to analyze and model the data.\u003c/p\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\text{O}\\text{v}\\text{e}\\text{r}\\text{a}\\text{l}\\text{l}\\:\\text{a}\\text{c}\\text{c}\\text{u}\\text{r}\\text{a}\\text{c}\\text{y}=\\:\\frac{\\text{T}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{c}\\text{o}\\text{r}\\:\\text{r}\\text{e}\\text{c}\\text{t}\\text{e}\\text{d}\\:\\text{c}\\text{l}\\text{a}\\text{s}\\text{s}\\text{i}\\text{f}\\text{i}\\text{e}\\text{d}\\:\\text{p}\\text{i}\\text{x}\\text{e}\\text{l}\\text{x}\\:\\left(\\text{d}\\text{i}\\text{a}\\text{g}\\text{o}\\text{n}\\text{a}\\text{l}\\right)}{\\text{t}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{r}\\text{e}\\text{f}\\text{e}\\text{r}\\text{e}\\text{n}\\text{c}\\text{e}\\:\\text{p}\\text{i}\\text{x}\\text{e}\\text{l}\\text{s}}\\:*100\\:\\:\\:\\:\\:\\:\\:\\:\\left(2\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:\\text{U}\\text{s}\\text{e}\\text{r}\\:\\text{A}\\text{c}\\text{c}\\text{u}\\text{r}\\text{a}\\text{c}\\text{y}=\\:\\frac{\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{c}\\text{o}\\text{r}\\:\\text{r}\\text{e}\\text{c}\\text{t}\\text{l}\\text{y}\\:\\text{c}\\text{l}\\text{a}\\text{s}\\text{s}\\text{i}\\text{f}\\text{i}\\text{e}\\text{d}\\:\\text{p}\\text{i}\\text{x}\\text{e}\\text{l}\\text{x}\\text{s}\\:\\text{i}\\text{n}\\:\\text{e}\\text{a}\\text{c}\\text{h}\\:\\text{c}\\text{a}\\text{t}\\text{a}\\text{g}\\text{o}\\text{r}\\text{y}\\:(\\text{d}\\text{i}\\text{a}\\text{g}\\text{o}\\text{n}\\text{a}\\text{l}}{\\text{t}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{r}\\text{e}\\text{f}\\text{e}\\text{r}\\text{e}\\text{n}\\text{c}\\text{e}\\:\\text{p}\\text{i}\\text{x}\\text{e}\\text{l}\\text{s}\\:\\text{i}\\text{n}\\:\\text{e}\\text{a}\\text{c}\\text{h}\\:\\text{c}\\text{a}\\text{t}\\text{e}\\text{g}\\text{o}\\text{r}\\text{y}\\:\\left(\\text{r}\\text{o}\\text{w}\\:\\text{t}\\text{o}\\text{t}\\text{a}\\text{l}\\right)}\\:*100\\:\\:\\:\\:\\left(3\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:\\text{P}.\\text{A}=\\:\\frac{\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{c}\\text{o}\\text{r}\\:\\text{r}\\text{e}\\text{c}\\text{t}\\text{l}\\text{y}\\:\\text{c}\\text{l}\\text{a}\\text{s}\\text{s}\\text{i}\\text{f}\\text{i}\\text{e}\\text{d}\\:\\text{p}\\text{i}\\text{x}\\text{e}\\text{l}\\text{x}\\text{s}\\:\\text{i}\\text{n}\\:\\text{e}\\text{a}\\text{c}\\text{h}\\:\\text{c}\\text{a}\\text{t}\\text{e}\\text{g}\\text{o}\\text{r}\\text{y}\\:\\left(\\text{d}\\text{i}\\text{a}\\text{g}\\text{o}\\text{n}\\text{a}\\text{l}\\right)}{\\text{t}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{r}\\text{e}\\text{f}\\text{e}\\text{r}\\text{e}\\text{n}\\text{c}\\text{e}\\:\\text{p}\\text{i}\\text{x}\\text{e}\\text{l}\\text{s}\\:\\text{i}\\text{n}\\:\\text{e}\\text{a}\\text{c}\\text{h}\\:\\text{c}\\text{a}\\text{t}\\text{e}\\text{g}\\text{o}\\text{r}\\text{y}\\:\\left(\\text{c}\\text{o}\\text{l}\\text{u}\\text{m}\\text{n}\\:\\text{t}\\text{o}\\text{t}\\text{a}\\text{l}\\right)}\\:*100\\:\\:\\:\\:\\left(4\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$$\\:\\text{K}.\\text{C}\\left(\\text{T}\\right)=\\frac{\\text{T}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{S}\\text{a}\\text{m}\\text{p}\\text{l}\\text{e}\\text{*}\\text{T}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{N}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{C}\\text{o}\\text{r}\\:\\text{r}\\text{e}\\text{c}\\text{t}\\text{e}\\text{d}\\:\\text{S}\\text{a}\\text{m}\\text{p}\\text{l}\\text{e}\\:-\\:\\sum\\:\\left(\\text{c}\\text{o}\\text{l}.\\text{t}\\text{o}\\text{t}\\:\\text{*}\\:\\text{r}\\text{o}\\text{w}\\:\\text{t}\\text{o}\\text{t}\\right)}{\\left(\\text{T}\\text{o}\\text{t}\\text{a}\\text{l}\\:\\text{n}\\text{u}\\text{m}\\text{b}\\text{e}\\text{r}\\:\\text{o}\\text{f}\\:\\text{S}\\text{a}\\text{m}\\text{p}\\text{l}\\text{e}\\right)2\\:-\\:\\sum\\:\\left(\\text{c}\\text{o}\\text{l}.\\text{t}\\text{o}\\text{t}\\:\\text{*}\\:\\text{r}\\text{o}\\text{w}\\:\\text{t}\\text{o}\\text{t}\\right)}\\:*100\\:\\:\\:\\:\\:\\left(5\\right)\\:\\:\\:$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e \u003cp\u003eHere P.A = Producer Accuracy\u003c/p\u003e \u003cp\u003eK.C (T) = Kappa Coefficient\u003c/p\u003e \u003cp\u003eFor the Land Surface Temperature (LST) analysis, which covered both summer and winter seasons from 1990 to 2020, Landsat satellite data were used to extract temperature values. The accuracy of the LST estimates was assessed by comparing the derived values with in-situ temperature data from local meteorological stations. Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) were calculated to quantify estimation errors. This accuracy assessment highlighted the strengths and limitations of the models, confirming that the LST analysis and classification models are reliable and effectively capture the spatiotemporal dynamics of land use and temperature changes. These results provide a robust basis for the study’s conclusions regarding climate adaptation and urban planning strategies in the coastal region.\u003c/p\u003e\n\u003ch3\u003e2.4 Integration of AHP for LULC Transition Analysis\u003c/h3\u003e\n\u003cp\u003eTo assess the impacts of LULC transitions on LST, this study incorporated the Analytical Hierarchy Process (AHP) to evaluate the relative importance of different LULC transitions in influencing temperature changes. The LULC categories considered include Forest, Waterbodies, Vegetation, Agricultural Land, and Urban (Built-up). A pairwise comparison matrix was developed to assign relative importance to transitions between these categories, using a scale of 1 to 9, where 1 indicates equal importance and 9 represents the highest level of importance.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eImportance of Land use Categories\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFrom-To\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eForest\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWaterbodies\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVegetation\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAgricultural Land\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eUrban (Built-up)\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eForest\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eWaterbodies\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1/3\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVegetation\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1/5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1/5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAgricultural Land\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1/7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1/7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1/5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eUrban (Built-up)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1/9\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1/9\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1/9\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1/9\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.4.1 Normalization and Priority Weight Calculation\u003c/h2\u003e \u003cp\u003eEach element of the matrix was normalized by dividing it by the sum of its respective column. The normalized values were used to calculate priority weights, which represent the relative influence of each LULC transition on LST.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePriority of Land use Categories\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLULC Transition\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePriority Weight\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForest\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.396\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWaterbodies\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.250\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVegetation\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgricultural Land\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.082\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUrban (Built-up)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.4.2 Consistency Check\u003c/h2\u003e \u003cp\u003eTo ensure the reliability of the pairwise comparison matrix, a consistency ratio (CR) was calculated using the Equation. The CR value was found to be less than 0.1, confirming that the matrix is consistent and the derived weights are valid.\u003c/p\u003e\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$$\\:CI=\\frac{{\\lambda\\:}_{m}-n}{n-1}$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.4.3 Spatial Representation\u003c/h2\u003e \u003cp\u003eThe calculated priority weights were applied to LULC change maps to create spatial representations of the impacts of these transitions on LST. These maps identified hotspots of temperature rise, highlighting areas where specific LULC transitions, such as deforestation and urban expansion, had the most significant effects.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003ch2\u003e3.1 Land Use and Land Cover Classification\u003c/h2\u003e\u003cp\u003eThe Land Use and Land Cover (LULC) classification of the coastal region of Bangladesh, shown in 3, provides a comprehensive spatial distribution of land cover types in 1990. The classification includes five primary categories: agriculture, forest, vegetation, water bodies, and built-up areas. The temporal analysis, presented in 5, outlines the changes in these categories from 1990 to 2020, facilitating a deeper understanding of land cover dynamics over three decades.In 1990, agricultural land was the dominant category, covering 35% of the coastal zone. This included cropland and other open areas used for agricultural purposes. Vegetation, comprising forests, grasslands, and green belts, accounted for 30% of the land cover. Water bodies, including rivers, lakes, and wetlands, constituted 10% of the region, emphasizing the importance of water features within the coastal ecosystem. Built-up areas, encompassing urban settlements, infrastructure, and ports, occupied 5% of the land.This classification serves as a baseline for understanding land cover changes and provides essential context for analyzing the implications of these shifts on environmental and urban dynamics\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eYear-wise Distribution of LULC\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"11\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVegetation (km²)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eVegetation (%)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eForest (km²)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eForest (%)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWaterbodies (km²)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eWaterbodies (%)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBuilt-Up (km²)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eBuilt-Up (%)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAgricultural Land (km²)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eAgricultural Land (%)\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1990\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14,160\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9,440\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4,720\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e2,360\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e5.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e16,520\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e35.00\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1995\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13,620\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e28.80\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9,440\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5,000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10.60\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e2,720\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e5.75\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e16,880\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e35.73\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13,260\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e28.13\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9,440\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4,720\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e3,000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e6.35\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e17,580\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e37.27\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2005\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12,860\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27.20\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9,440\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5,000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10.60\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e3,080\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e6.52\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e18,080\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e38.30\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2010\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12,600\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e26.74\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9,600\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.30\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4,800\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10.20\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e3,400\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e7.20\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e18,600\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e39.49\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2015\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12,340\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e26.12\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9,700\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20.50\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4,000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e8.50\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e3,800\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e8.05\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e19,000\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e40.25\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e9,440\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4,720\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e10.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3,776\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e8.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e9,440\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e20.00\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e19,824\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e42.00\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003eOver the decades, the distribution of land cover types in the coastal region of Bangladesh has undergone significant changes, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. By 1995, agricultural land had increased slightly to 35.73%, reflecting a gradual expansion of cropland and agricultural activities. In contrast, vegetation decreased to 28.80%, indicating a reduction in natural green spaces. Water bodies rose to 10.60%, and built-up areas increased to 5.75%, marking the early stages of urban expansion. These trends continued into the early 2000s, with agricultural land growing further to 37.27%, while vegetation declined to 28.13%. Built-up areas increased to 6.35%, reflecting the expansion of urban and industrial development, while the extent of water bodies remained stable, indicating consistent water features in the region.\u003c/p\u003e\u003cp\u003eFrom 2005 to 2010, agricultural land further increased to 38.30%, and built-up areas expanded to 6.52%. Vegetation continued to decrease, reaching 27.20%, while water bodies remained stable at 10.60%, showing minor fluctuations in the region’s water distribution. These changes reflect the ongoing influence of urbanization and environmental pressures on land cover.The period from 2010 to 2020 saw even more pronounced shifts. By 2010, agricultural land had reached 39.49%, and built-up areas expanded to 7.20%. Vegetation continued to decline, dropping to 26.74%, while water bodies slightly reduced to 10.20%. By 2015, agricultural land rose to 40.25%, with built-up areas increasing to 8.05%. Vegetation declined further to 26.12%, and water bodies decreased to 8.50%. By 2020, agricultural land reached 42.00%, and built-up areas surged dramatically to 20.00%. Vegetation fell significantly to 20.00%, while water bodies decreased to 8.00%. Throughout the study period, forest land, including both hilly and mangrove forests, remained relatively stable, indicating some resilience in these ecosystems. These trends highlight the complex interplay between natural processes and human activities in Bangladesh’s coastal zone. The expansion of agricultural and built-up areas, coupled with the decline in vegetation and water bodies, underscores the need for effective land management strategies. Sustainable practices are essential to balance development with environmental preservation, especially in a region as vulnerable and dynamic as Bangladesh’s coastal zone.\u003c/p\u003e\u003ch2\u003e4.2 Accuracy Assessment\u003c/h2\u003e\u003cp\u003eAccuracy assessment is essential in land use and land cover (LULC) classification, as it evaluates the reliability of classified images by comparing them with an accurate reference dataset or ground truth. This study used five deep learning algorithms Temporal Convolutional Network (TCN), XGBoost, CatBoost, LightGBM, and H2O.ai to assess LULC classifications for the years 1990, 1995, 2000, 2005, 2010, 2015, and 2020. The performance of each algorithm was evaluated based on accuracy and Kappa values, as summarized in Table \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.TCN showed gradual improvement over time, with accuracy increasing from 0.62 and Kappa from 0.61 in 1995 to 0.71 and 0.70, respectively, by 2020. TCN consistently outperformed XGBoost, achieving a peak accuracy of 0.73 and a Kappa value of 0.69 in 2020. In comparison, CatBoost and LightGBM performed slightly lower, with CatBoost reaching a maximum accuracy of 0.67 and LightGBM 0.65 in 2020. H2O.ai demonstrated competitive results, with accuracy and Kappa values of 0.69 and 0.67, respectively, in 2020.\u003c/p\u003e\u003cp\u003eOverall, TCN and H2O.ai were identified as the top-performing algorithms, with TCN showing substantial improvement over the study period.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c14\" colnum=\"14\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c15\" colnum=\"15\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparative Assessment of Supervised Models\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"15\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgorithm\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003cp\u003eAc\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2020 ka\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2015 Ac\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2015 Ka\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2010 Ac\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2010 Ka\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e2005 Ac\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2005 Ka\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003e2000 Ac\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003e2000 Ka\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003e1995 Ac\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003e1995 Ka\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c14\"\u003e \u003cp\u003e1990 Ac\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c15\"\u003e \u003cp\u003e1990 Ka\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTCN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c14\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c14\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCatBoost\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c14\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLightGBM\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e0.59\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e0.59\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c14\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e0.57\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH2O.ai\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c14\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"15\"\u003eAc = Accuracy, Ka = Kappa\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003ch2\u003eLand Cover Changes from 1990 to 2020\u003c/h2\u003e\u003cp\u003eThe analysis of land use and land cover (LULC) changes in the coastal zone of Bangladesh from 1990 to 2020 reveals significant transformations across various land cover types. The data demonstrates substantial shifts in land cover over the three decades. Between 1990 and 2000, notable changes occurred, particularly in the conversion of vegetation to other land cover types. Vegetation, including forests and grasslands, was predominantly transformed into agricultural land and built-up areas. Specifically, 3,066.68 km² of vegetation was converted to agricultural land, reflecting the expansion of agricultural activities, while 2,127.90 km² was transformed into built-up areas, indicative of increasing urbanization. Additionally, 990.55 km² of vegetation transitioned to water bodies, possibly due to changes in land management or environmental conditions. Forests also experienced significant changes, with 419.78 km² converting to vegetation and 477.09 km² shifting to water bodies, indicating moderate transformations driven by both natural and anthropogenic factors. Agricultural lands underwent considerable transformations, with 1,854.92 km² of agricultural land being converted to built-up areas and 1,737.07 km² shifting to water bodies. These transitions highlight the growing pressures on agricultural lands from urbanization and changes in water management.\u003c/p\u003e\u003cp\u003eFrom 2000 to 2010, significant land cover changes continued (see in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e), particularly in the conversion of vegetation to built-up areas and agricultural land. Vegetation decreased as 1,850.54 km² was converted to built-up areas and 2,277.88 km² shifted to agricultural lands, reflecting ongoing urbanization and agricultural expansion. Forests also underwent changes, with 483.57 km² transitioning to vegetation and 389.64 km² to water bodies, indicating gradual transformations due to both natural processes and human activities. Water bodies saw notable reductions as well, with 606.28 km² converted to built-up areas and 1,813.91 km² shifting to agricultural lands, illustrating the significant impact of urban and agricultural development on water features.In the subsequent decade, from 2010 to 2020, further land cover transformations were observed. Specifically, 1,764.86 km² of vegetation was converted to built-up areas and 1,750.04 km² to agricultural lands, emphasizing the persistent trends of urbanization and agricultural growth. Forests also experienced reductions, with 103.18 km² transitioning to built-up areas and 214.95 km² to agricultural lands. However, 1,038.11 km² of vegetation transformed into forested areas, suggesting some reforestation or natural succession. Water bodies also experienced substantial changes, with 2,137.54 km² of agricultural land converted into water bodies, reflecting shifts in land use and water management practices.\u003c/p\u003e\u003ch2\u003e4.3 Change of Land Use and Land Cover\u003c/h2\u003e\u003cp\u003eOver the past 30 years, the coastal zone of Bangladesh, which spans approximately 47,201 km², has experienced significant shifts in land use and land cover. A comparison between 1990 and 220 highlights some noteworthy trends in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eThe coastal zone of Bangladesh has undergone significant changes in land cover from 1990 to 2020. One of the most notable transformations is the substantial decrease in vegetation see in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e. In 1990, vegetation covered approximately 30% of the coastal area, equivalent to 14,160 km². By 2020, this had decreased to 20%, or 9,440 km². This decline suggests a considerable loss of natural green spaces, likely driven by urban expansion and land conversion. Similarly, forested areas have seen a significant reduction. Forest cover dropped from 20% (9,440 km²) in 1990 to just 10% (4,720 km²) by 2020. This decrease is concerning as it represents the loss of critical habitats and biodiversity, potentially caused by deforestation and increasing development pressures. Water bodies have also diminished over the period. In 1990, they occupied 10% of the coastal area, about 4,720 km², but by 2020, this had reduced to 8%, or 3,776 km². This decline may be attributed to factors such as sedimentation, land reclamation, or changes in water management practices. Conversely, built-up (urban) areas have expanded dramatically. The proportion of urban land use grew from 5% (2,360 km²) in 1990 to 20% (9,440 km²) by 2020. This significant increase highlights the rapid urbanization and infrastructure development that has taken place within the coastal zone. Agricultural land has also expanded over the same period, rising from 35% (16,520 km²) in 1990 to 42% (19,824 km²) in 2020. This growth underscores the continuing importance of agriculture in the region, even as urban and industrial development increases.\u003c/p\u003e\u003ch2\u003e4.4 Land Surface Temperature\u003c/h2\u003e\u003cp\u003eThe analysis of Land Surface Temperature (LST) data for the coastal regions of Bangladesh highlights a significant increase in temperatures over the past three decades. This warming trend is evident across various locations and seasons, with a more pronounced rise in summer temperatures compared to winter.\u003c/p\u003e\u003cp\u003eThe data clearly shows (see Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e) an upward trend in Land Surface Temperature (LST) across key coastal cities. For example, Chittagong has experienced a significant rise in summer temperatures, increasing from 28.05°C in 1990 to 31.75°C in 2020. Similarly, Khulna's temperatures have risen from 26.68°C in 1990 to 30.25°C in 2020. These cities, with their extensive urbanization and high levels of built-up areas, exhibit the highest LST values, which is indicative of the urban heat island effect. This phenomenon occurs when natural vegetation is replaced by impervious surfaces, such as concrete and asphalt, which absorb and retain heat more efficiently. In contrast, rural areas such as Bagerhat and Amtali show more moderate increases in LST. In Bagerhat, summer temperatures increased from 27.12°C in 1990 to 30.55°C in 2020, while winter temperatures rose from 17.92°C to 20.03°C during the same period. Similarly, Amtali saw a rise in summer temperatures from 26.98°C to 29.68°C, and winter temperatures increased from 17.34°C to 18.95°C. The presence of green spaces and water bodies in these rural areas helps moderate temperature increases, although a gradual warming trend is still apparent.\u003c/p\u003e\u003cp\u003eDeforestation and agricultural expansion have exacerbated the rise in Land Surface Temperature (LST) in the region. Forest cover, which was 4.96% of the area in 1990, has decreased to 3.85% by 2020 in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e, reducing the cooling effects of vegetation through processes like evapotranspiration. As forests are cleared for agricultural purposes, the land's thermal properties change, resulting in localized temperature increases. While agricultural areas were initially cooler than urban zones, they are now also experiencing rising temperatures as land use shifts continue. This further contributes to the overall warming trend in the region\u003c/p\u003e\u003ch2\u003e4.5 Impact of Land Use Land Cover Transitions on Surface Temperature from AHP Analysis\u003c/h2\u003e\u003cp\u003eThe analysis of the relationship between Land Use Land Cover (LULC) transitions and Mean Land Surface Temperature (LST) reveals notable patterns in surface heat dynamics. The results indicate that transitions from Forest to Built-Up areas are linked to the highest mean LST values (see figure Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e). This increase in temperature is due to the heat-absorbing properties of built-up areas, exacerbating the urban heat island (UHI) effect. In contrast, transitions such as Vegetation to Waterbodies show relatively low mean LST, underlining the natural cooling effect that waterbodies have on their surroundings. Similarly, regions that transition from Agricultural Land to Built-Up areas also experience elevated LST, emphasizing the role of urbanization in intensifying surface heat. Interestingly, areas with no LULC change tend to maintain stable LST values, suggesting that preserving existing land cover types is essential for temperature regulation. Moreover, transitions involving waterbodies, such as Waterbodies to Vegetation, are associated with lower LST compared to other transitions, highlighting the significant role of water in mitigating surface temperature increases.\u003c/p\u003e\u003cp\u003eto substantial temperature increases, further reinforcing the role of urbanization in exacerbating the urban heat island (UHI) effect. These transitions lead to higher impervious surfaces, which absorb and retain heat more effectively, contributing to elevated surface temperatures. The AHP analysis indicates that areas undergoing these low AHP value transitions are more susceptible to heat-related challenges, underscoring the need for careful planning and management to reduce the adverse thermal impacts of urban growth. Thus, the AHP results highlight the importance of prioritizing land use transitions that favor ecological sustainability, such as Vegetation to Forest and Waterbodies to Forest, to mitigate rising surface temperatures and promote environmental resilience. These findings advocate for strategies that incorporate natural cooling mechanisms, such as preserving and expanding green spaces, to combat the rising LST associated with urbanization and agricultural expansion.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study underscores the critical role of urban growth and deforestation in driving increases in land surface temperatures (LST) in the coastal regions of Bangladesh. Cities like Khulna and Chittagong, marked by expanding built-up areas and a significant reduction in green spaces, are increasingly resembling urban heat islands. These findings highlight the larger climatic implications of urbanization in these regions. The observed connection between urban heat islands and diminishing vegetation aligns with earlier studies, including Rahman et al. (2017), which reported similar trends across South Asia. A key insight of this study is the contribution of agricultural lands near coastal regions to the rise in LST, potentially linked to changes in land management practices. The Analytic Hierarchy Process (AHP) applied to land-use transitions emphasizes the significance of these findings. AHP analysis reveals that transitions from agricultural land to built-up areas are associated with a substantial increase in LST, further supporting the notion that urbanization accelerates temperature rise. While urban areas and deforestation are significant contributors, rural areas undergoing land-use changes, particularly from agriculture to urban development, are also pivotal in shaping local climate dynamics. This discovery introduces a new dimension to the existing theories on land use and climate, specifically regarding the impact of rural landscapes on LST.The implications of these findings are clear. Urban planners must prioritize integrating green infrastructure, such as parks and green roofs, within city planning to mitigate the effects of rising temperatures. AHP analysis reinforces this, indicating that maintaining transitions such as vegetation to forest (with higher AHP values) results in minimal temperature increase. These ecologically significant transitions highlight the importance of preserving natural landscapes for temperature regulation. Moreover, the substantial impact of agricultural expansion on LST underscores the need for sustainable farming practices to reduce environmental harm and mitigate temperature increases. These insights offer valuable guidance for policymakers and stakeholders in formulating heat management strategies for both urban and rural landscapes.Despite its contributions, the study has limitations. While satellite imagery provides valuable insights at a regional scale, it fails to capture temperature variations at the local level, particularly in densely populated areas. Incorporating ground-level temperature data, alongside local community input, could offer a more nuanced understanding of temperature dynamics, enhancing the precision of the AHP analysis. Future studies should address this gap and refine the assessment of LST in urban and rural areas.Looking ahead, several avenues for future research emerge. A more detailed examination of seasonal variations in LST would clarify how different land-use changes affect temperatures over time. Furthermore, the incorporation of advanced machine learning models could significantly improve the predictive accuracy of LST variations, enabling more robust climate adaptation strategies. The integration of AHP with these methods can further enhance decision-making by providing a structured, multi-criteria framework for land-use planning. These future directions demonstrate the potential of this study to inform not only further academic inquiry but also practical solutions for managing the heat-related challenges associated with climate change\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study offers a comprehensive analysis of the relationship between land use and land cover (LULC) changes and land surface temperature (LST) variations in the coastal regions of Bangladesh. The findings reveal that rapid urbanization, deforestation, and shifting agricultural practices have significantly altered the landscape and contributed to the rise in LST in these areas. The expansion of urban centers, particularly in Chittagong and Khulna, has exacerbated the urban heat island (UHI) effect, as built-up areas tend to trap and retain heat more effectively than natural ecosystems. In contrast, the loss of forest cover has diminished the natural cooling mechanisms provided by vegetation, further amplifying temperature increases. The study also highlights the complex relationship between various land use types, such as agricultural land and water bodies, which influence LST in distinct ways. The shifting dynamics of land management practices directly affect local climate patterns, with these effects being further compounded by global climate change. As global temperatures continue to rise, the combined impacts of urbanization and deforestation may worsen temperature extremes, posing serious challenges for local communities. This study emphasizes the urgent need for integrated strategies to mitigate rising LST in coastal regions. Sustainable urban planning, which includes the incorporation of green spaces, forest conservation, and water management practices, is essential to address the negative impacts of urban heat islands. Furthermore, enhancing climate resilience through the restoration of natural landscapes and the reduction of deforestation is critical in combating the escalating effects of climate change. The implementation of these strategies will not only help mitigate the adverse effects of rising temperatures but also promote the long-term sustainability of Bangladesh\u0026rsquo;s coastal regions. By balancing development with environmental preservation, the country can strengthen the resilience of its urban areas, improving the quality of life for its inhabitants while safeguarding its natural resources. Future research, integrating advanced modeling techniques such as machine learning and the Analytic Hierarchy Process (AHP), could further refine land use planning and enhance climate adaptation strategies, providing a solid foundation for policymakers and stakeholders in the region.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors express their sincere gratitude to the Ministry of Science and Technology, Government of Bangladesh, for their generous financial support through the Special Research Grant (SRG-234554), titled \u003cem\u003e\u0026ldquo;A Deep Learning Based Geospatial Analysis of the Effects of Changing Land Use Land Cover (LULC) on the Dynamics of Land Surface Temperature Along the Bangladeshi Coast.\u0026rdquo;\u003c/em\u003e This research would not have been possible without their continued investment in scientific innovation, particularly in climate-vulnerable regions.\u003c/p\u003e\n\u003cp\u003eWe are grateful to the Urban and Rural Planning Discipline at Khulna University for providing institutional support and access to computational resources essential for this study. We also acknowledge the United States Geological Survey (USGS) for the freely available Landsat satellite data, which served as a critical data source.Lastly, we acknowledge the use of \u003cstrong\u003eGPT-4o mini\u003c/strong\u003e, which significantly contributed to language refinement, content structuring, and technical writing throughout the manuscript preparation process.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the Ministry of Science and Technology, Government of Bangladesh, under the Special Research Grant (SRG-234554).\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eThe satellite imagery datasets used in this study were obtained from the United States Geological Survey (USGS) Earth Explorer platform (https://earthexplorer.usgs.gov/). Landsat 5 TM, Landsat 7 ETM+, and Landsat 8 OLI/TIRS data were used for land use/land cover classification and land surface temperature analysis. All datasets are publicly available and free for academic research purposes. Derived LULC and LST maps, along with associated analytical outputs, are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNiloy Biswas (NB):\u003c/strong\u003e Methodology, Software, Validation, Formal Analysis, Data Curation, Visualization, Writing \u0026ndash; Original Draft.\u003cbr\u003e\u003cstrong\u003eKazi Saiful Islam (KSI):\u003c/strong\u003e Conceptualization, Supervision, Funding Acquisition, Project Administration, Methodology, Resources, Writing \u0026ndash; Review \u0026amp; Editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRole Details:\u003c/strong\u003e\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cem\u003eConceptualization:\u003c/em\u003e NB and KSI jointly developed the research objectives and defined the study scope.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eMethodology:\u003c/em\u003e NB designed the deep learning framework (TCN, XGBoost) and integrated AHP; KSI contributed to the geospatial analysis methodology.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eSoftware:\u003c/em\u003e NB implemented all models using Python and Google Earth Engine.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eValidation:\u003c/em\u003e NB conducted accuracy assessments (Kappa, RMSE) and validated results with reference data.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eFormal Analysis:\u003c/em\u003e NB analyzed LULC transitions and LST trends; KSI provided interpretation of climatic implications.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eData Curation:\u003c/em\u003e NB managed Landsat data preprocessing and integration.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eVisualization:\u003c/em\u003e NB created all maps, figures, and AHP-based spatial analyses.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eWriting \u0026ndash; Original Draft:\u003c/em\u003e NB drafted the manuscript.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eSupervision \u0026amp; Funding Acquisition:\u003c/em\u003e KSI oversaw the project and secured funding.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eResources:\u003c/em\u003e KSI arranged access to institutional and technical resources.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eWriting \u0026ndash; Review \u0026amp; Editing:\u003c/em\u003e KSI reviewed and refined the manuscript for technical and contextual accuracy.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eBoth authors have read and approved the final version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAi, J., Zhang, C., Chen, L., \u0026amp; Li, D. (2020a). Mapping Annual Land Use and Land Cover Changes in the Yangtze Estuary Region Using an Object-Based Classification Framework and Landsat Time Series Data. \u003cem\u003eSustainability 2020, Vol. 12, Page 659\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(2), 659. https://doi.org/10.3390/SU12020659\u003c/li\u003e\n\u003cli\u003eAi, J., Zhang, C., Chen, L., \u0026amp; Li, D. (2020b). Mapping Annual Land Use and Land Cover Changes in the Yangtze Estuary Region Using an Object-Based Classification Framework and Landsat Time Series Data. \u003cem\u003eSustainability 2020, Vol. 12, Page 659\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(2), 659. https://doi.org/10.3390/SU12020659\u003c/li\u003e\n\u003cli\u003eAlam, M. S. (2013). Climate Change and Land Use Change in the Eastern Coastal Belt of Bangladesh, Elucidated by Analyzing Rice Production Area in the Past and Future. \u003cem\u003eJournal of Life and Earth Science\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e, 83\u0026ndash;92. https://doi.org/10.3329/JLES.V8I0.20151\u003c/li\u003e\n\u003cli\u003eBeg, A. (2018). Assessment of land surface temperature variation over Rusafa side of Baghdad city, Iraq. \u003cem\u003eMATEC Web of Conferences\u003c/em\u003e, \u003cem\u003e162\u003c/em\u003e, 03032. https://doi.org/10.1051/MATECCONF/201816203032\u003c/li\u003e\n\u003cli\u003eBui, Q. T., Chou, T. Y., Hoang, T. Van, Fang, Y. M., Mu, C. Y., Huang, P. H., Pham, V. D., Nguyen, Q. H., Anh, D. T. N., Pham, V. M., \u0026amp; Meadows, M. E. (2021). Gradient Boosting Machine and Object-Based CNN for Land Cover Classification. \u003cem\u003eRemote Sensing 2021, Vol. 13, Page 2709\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(14), 2709. https://doi.org/10.3390/RS13142709\u003c/li\u003e\n\u003cli\u003eChaves, M. E. D., Picoli, M. C. A., \u0026amp; Sanches, I. D. (2020). Recent Applications of Landsat 8/OLI and Sentinel-2/MSI for Land Use and Land Cover Mapping: A Systematic Review. \u003cem\u003eRemote Sensing 2020, Vol. 12, Page 3062\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(18), 3062. https://doi.org/10.3390/RS12183062\u003c/li\u003e\n\u003cli\u003eDewan, A. M., Kabir, M. H., Nahar, K., \u0026amp; Rahman, M. Z. (2012). Urbanisation and environmental degradation in Dhaka metropolitan area of Bangladesh. \u003cem\u003eInternational Journal of Environment and Sustainable Development\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(2), 118\u0026ndash;147. https://doi.org/10.1504/IJESD.2012.049178\u003c/li\u003e\n\u003cli\u003eDu, C., Ren, H., Qin, Q., Meng, J., \u0026amp; Zhao, S. (2015). A Practical Split-Window Algorithm for Estimating Land Surface Temperature from Landsat 8 Data. \u003cem\u003eRemote Sensing 2015, Vol. 7, Pages 647-665\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(1), 647\u0026ndash;665. https://doi.org/10.3390/RS70100647\u003c/li\u003e\n\u003cli\u003eFilho, W. L., Icaza, L. E., Emanche, V. O., \u0026amp; Al-Amin, A. Q. (2017). An Evidence-Based Review of Impacts, Strategies and Tools to Mitigate Urban Heat Islands. \u003cem\u003eInternational Journal of Environmental Research and Public Health 2017, Vol. 14, Page 1600\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(12), 1600. https://doi.org/10.3390/IJERPH14121600\u003c/li\u003e\n\u003cli\u003eFitzgerald, R. W., \u0026amp; Lees, B. G. (1994). Assessing the classification accuracy of multisource remote sensing data. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e47\u003c/em\u003e(3), 362\u0026ndash;368. https://doi.org/10.1016/0034-4257(94)90103-1\u003c/li\u003e\n\u003cli\u003eGarouani, M. El, Amyay, M., Lahrach, A., \u0026amp; Oulidi, H. J. (2021). Land Surface Temperature in Response to Land Use/Cover Change Based on Remote Sensing Data and GIS Techniques: Application to Sa\u0026iuml;ss Plain, Morocco. \u003cem\u003eJournal of Ecological Engineering\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(7), 100\u0026ndash;112. https://doi.org/10.12911/22998993/139065\u003c/li\u003e\n\u003cli\u003eGuha, S., Govil, H., Dey, A., \u0026amp; Gill, N. (2020). A case study on the relationship between land surface temperature and land surface indices in Raipur City, India. \u003cem\u003eGeografisk Tidsskrift-Danish Journal of Geography\u003c/em\u003e, \u003cem\u003e120\u003c/em\u003e(1), 35\u0026ndash;50. https://doi.org/10.1080/00167223.2020.1752272\u003c/li\u003e\n\u003cli\u003eImran, H. M., Hossain, A., Islam, A. K. M. S., Rahman, A., Bhuiyan, M. A. E., Paul, S., \u0026amp; Alam, A. (2021a). Impact of Land Cover Changes on Land Surface Temperature and Human Thermal Comfort in Dhaka City of Bangladesh. \u003cem\u003eEarth Systems and Environment\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(3), 667\u0026ndash;693. https://doi.org/10.1007/S41748-021-00243-4/FIGURES/10\u003c/li\u003e\n\u003cli\u003eImran, H. M., Hossain, A., Islam, A. K. M. S., Rahman, A., Bhuiyan, M. A. E., Paul, S., \u0026amp; Alam, A. (2021b). Impact of Land Cover Changes on Land Surface Temperature and Human Thermal Comfort in Dhaka City of Bangladesh. \u003cem\u003eEarth Systems and Environment\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(3), 667\u0026ndash;693. https://doi.org/10.1007/S41748-021-00243-4/FIGURES/10\u003c/li\u003e\n\u003cli\u003eIslam, M. S., Uddin, M. A., \u0026amp; Hossain, M. A. (2021). Assessing the dynamics of land cover and shoreline changes of Nijhum Dwip (Island) of Bangladesh using remote sensing and GIS techniques. \u003cem\u003eRegional Studies in Marine Science\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e, 101578. https://doi.org/10.1016/J.RSMA.2020.101578\u003c/li\u003e\n\u003cli\u003eKe, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., \u0026amp; Liu, T.-Y. (2017). \u003cem\u003eLightGBM: A Highly Efficient Gradient Boosting Decision Tree\u003c/em\u003e. https://hal.science/hal-03953007\u003c/li\u003e\n\u003cli\u003eLi, H., Wang, C., Zhong, C., Zhang, Z., \u0026amp; Liu, Q. (2017). Mapping Typical Urban LULC from Landsat Imagery without Training Samples or Self-Defined Parameters. \u003cem\u003eRemote Sensing 2017, Vol. 9, Page 700\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(7), 700. https://doi.org/10.3390/RS9070700\u003c/li\u003e\n\u003cli\u003eLi, S., Qin, Z., Zhao, S., Gao, M., Li, S., Liao, Q., \u0026amp; Du, W. (2022). Spatiotemporal Variation of Land Surface Temperature in Henan Province of China from 2003 to 2021. \u003cem\u003eLand 2022, Vol. 11, Page 1104\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(7), 1104. https://doi.org/10.3390/LAND11071104\u003c/li\u003e\n\u003cli\u003eLi, W., Cao, Q., Lang, K., \u0026amp; Wu, J. (2017). Linking potential heat source and sink to urban heat island: Heterogeneous effects of landscape pattern on land surface temperature. \u003cem\u003eScience of The Total Environment\u003c/em\u003e, \u003cem\u003e586\u003c/em\u003e, 457\u0026ndash;465. https://doi.org/10.1016/J.SCITOTENV.2017.01.191\u003c/li\u003e\n\u003cli\u003eLiu, S., \u0026amp; Huang, G. (2018). The Evaluation of Urban Heat Island Effect and New Town Development Based on Remote Sensing. \u003cem\u003eIOP Conference Series: Earth and Environmental Science\u003c/em\u003e, \u003cem\u003e170\u003c/em\u003e(3), 032096. https://doi.org/10.1088/1755-1315/170/3/032096\u003c/li\u003e\n\u003cli\u003eLu, X. Y., Chen, X., Zhao, X. L., Lv, D. J., \u0026amp; Zhang, Y. (2021). Assessing the impact of land surface temperature on urban net primary productivity increment based on geographically weighted regression model. \u003cem\u003eScientific Reports 2021 11:1\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 1\u0026ndash;14. https://doi.org/10.1038/s41598-021-01757-7\u003c/li\u003e\n\u003cli\u003eLu, Y., Yang, J., \u0026amp; Ma, S. (2021). Dynamic Changes of Local Climate Zones in the Guangdong\u0026ndash;Hong Kong\u0026ndash;Macao Greater Bay Area and Their Spatio-Temporal Impacts on the Surface Urban Heat Island Effect between 2005 and 2015. \u003cem\u003eSustainability 2021, Vol. 13, Page 6374\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(11), 6374. https://doi.org/10.3390/SU13116374\u003c/li\u003e\n\u003cli\u003eMadni, H. A., Umer, M., Ishaq, A., Abuzinadah, N., Saidani, O., Alsubai, S., Hamdi, M., \u0026amp; Ashraf, I. (2023). Water-Quality Prediction Based on H2O AutoML and Explainable AI Techniques. \u003cem\u003eWater 2023, Vol. 15, Page 475\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(3), 475. https://doi.org/10.3390/W15030475\u003c/li\u003e\n\u003cli\u003eMukherjee, A., Fryar, A. E., \u0026amp; Thomas, W. A. (2009). Geologic, geomorphic and hydrologic framework and evolution of the Bengal basin, India and Bangladesh. \u003cem\u003eJournal of Asian Earth Sciences\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(3), 227\u0026ndash;244. https://doi.org/10.1016/J.JSEAES.2008.05.011\u003c/li\u003e\n\u003cli\u003eProkhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., \u0026amp; Gulin, A. (2017). CatBoost: unbiased boosting with categorical features. \u003cem\u003eAdvances in Neural Information Processing Systems\u003c/em\u003e, \u003cem\u003e2018-December\u003c/em\u003e, 6638\u0026ndash;6648. https://arxiv.org/abs/1706.09516v5\u003c/li\u003e\n\u003cli\u003eQiu, B., Li, H., Chen, C., Tang, Z., Zhang, K., \u0026amp; Berry, J. (2019). Tracking spatial\u0026ndash;temporal landscape changes of impervious surface areas, bare lands, and inundation areas in China during 2001\u0026ndash;2017. \u003cem\u003eLand Degradation \u0026amp; Development\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e(15), 1802\u0026ndash;1812. https://doi.org/10.1002/LDR.3352\u003c/li\u003e\n\u003cli\u003eRehman, A., Qin, J., Pervez, A., Khan, M. S., Ullah, S., Ahmad, K., \u0026amp; Rehman, N. U. (2022). Land-Use/Land Cover Changes Contribute to Land Surface Temperature: A Case Study of the Upper Indus Basin of Pakistan. \u003cem\u003eSustainability 2022, Vol. 14, Page 934\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(2), 934. https://doi.org/10.3390/SU14020934\u003c/li\u003e\n\u003cli\u003eRobinson, C., Ortiz, A., Ferres, J. M. L., Anderson, B., \u0026amp; Ho, D. E. (2021). Temporal Cluster Matching for Change Detection of Structures from Satellite Imagery. \u003cem\u003eProceedings of 2021 4th ACM SIGCAS Conference on Computing and Sustainable Societies, COMPASS 2021\u003c/em\u003e, 138\u0026ndash;146. https://doi.org/10.1145/3460112.3471952\u003c/li\u003e\n\u003cli\u003eRogers, K. G., \u0026amp; Goodbred, S. L. (2014). The Sundarbans and Bengal Delta: The World\u0026rsquo;s Largest Tidal Mangrove and Delta System. \u003cem\u003eWorld Geomorphological Landscapes\u003c/em\u003e, 181\u0026ndash;187. https://doi.org/10.1007/978-94-017-8029-2_18\u003c/li\u003e\n\u003cli\u003eRwanga, S. S., Ndambuki, J. M., Rwanga, S. S., \u0026amp; Ndambuki, J. M. (2017). Accuracy Assessment of Land Use/Land Cover Classification Using Remote Sensing and GIS. \u003cem\u003eInternational Journal of Geosciences\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(4), 611\u0026ndash;622. https://doi.org/10.4236/IJG.2017.84033\u003c/li\u003e\n\u003cli\u003eSobrino, J. A., Julien, Y., \u0026amp; Garc\u0026iacute;a-Monteiro, S. (2020). Surface Temperature of the Planet Earth from Satellite Data. \u003cem\u003eRemote Sensing 2020, Vol. 12, Page 218\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(2), 218. https://doi.org/10.3390/RS12020218\u003c/li\u003e\n\u003cli\u003eTalukdar, S., Singha, P., Mahato, S., Shahfahad, Pal, S., Liou, Y. A., \u0026amp; Rahman, A. (2020). Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations\u0026mdash;A Review. \u003cem\u003eRemote Sensing 2020, Vol. 12, Page 1135\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(7), 1135. https://doi.org/10.3390/RS12071135\u003c/li\u003e\n\u003cli\u003eTan, J., Yu, D., Li, Q., Tan, X., \u0026amp; Zhou, W. (2020). Spatial relationship between land-use/land-cover change and land surface temperature in the Dongting Lake area, China. \u003cem\u003eScientific Reports 2020 10:1\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 1\u0026ndash;9. https://doi.org/10.1038/s41598-020-66168-6\u003c/li\u003e\n\u003cli\u003eUmitsu, M. (1997). Landforms and floods in the Ganges delta and coastal lowland of Bangladesh. \u003cem\u003eMarine Geodesy\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(1), 77\u0026ndash;87. https://doi.org/10.1080/01490419709388096\u003c/li\u003e\n\u003cli\u003eWang, Y. C., Hu, B. K. H., Myint, S. W., Feng, C. C., Chow, W. T. L., \u0026amp; Passy, P. F. (2018). Patterns of land change and their potential impacts on land surface temperature change in Yangon, Myanmar. \u003cem\u003eScience of The Total Environment\u003c/em\u003e, \u003cem\u003e643\u003c/em\u003e, 738\u0026ndash;750. https://doi.org/10.1016/J.SCITOTENV.2018.06.209\u003c/li\u003e\n\u003cli\u003eWatson, R. T., Zinyowera, M. C., Moss, R. H., \u0026amp; Dokken, D. J. (1997). \u003cem\u003eSummary for Policymakers The Regional Impacts of Climate Change: An Assessment of Vulnerability The World Bank Zimbabwe Meteorological Services Battelle Pacific Northwest National Laboratory A special Report of IPCC Working Group II Published for the Intergovernmental Panel on Climate Change INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eYang, Y., \u0026amp; Lee, X. (2022). A Scale-Separating Framework for Fusing Satellite Land Surface Temperature Products. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(4), 983. https://doi.org/10.3390/RS14040983/S1\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Land Surface Temperature (LST), Temporal Convolutional Network (TCN), Analytic Hierarchy Process (AHP), Coastal Bangladesh, Deep Learning","lastPublishedDoi":"10.21203/rs.3.rs-6800634/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6800634/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study examines the relationship between land use and land cover (LULC) changes and land surface temperature (LST) dynamics in the climate-vulnerable coastal regions of Bangladesh from 1990 to 2020. Utilizing multi-temporal Landsat imagery and advanced deep learning techniques, particularly Temporal Convolutional Networks (TCN), the research achieved a classification accuracy of 73% in 2020, outperforming other models such as XGBoost (71%). The findings reveal that rapid urbanization and deforestation are the principal drivers of increasing LST, with urban centers such as Khulna and Chittagong experiencing a temperature rise of up to 2.5\u0026deg;C over the study period. An Analytic Hierarchy Process (AHP)-based prioritization of LULC transitions identified agricultural-to-urban (weighted impact: 82%) and vegetation-to-urban conversions as the most significant contributors to LST escalation, whereas forested and water-covered areas were associated with relatively lower temperature increases. Seasonal analysis indicates a more pronounced warming during summer, with rural areas showing a mitigated rise due to residual vegetation cover. The study further underscores the compounding effects of climate change, suggesting that continued LULC transformations without adaptive measures could intensify future heat stress. To mitigate the urban heat island effect, the study recommends the implementation of green infrastructure, enforcement of forest conservation, and the promotion of climate-sensitive urban planning. By integrating deep learning with multi-criteria decision analysis, this research contributes a robust methodological framework and empirical insights to support sustainable land management and climate adaptation in coastal regions.\u003c/p\u003e","manuscriptTitle":"Assessing the Impact of Land Use and Land Cover Changes on Land Surface Temperature Dynamics in the Coastal Region of Bangladesh: A Comprehensive Analysis Using Deep Learning Techniques AHP Integration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-27 11:09:06","doi":"10.21203/rs.3.rs-6800634/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3d5627d6-3ce9-41f2-a8eb-1a2d2d986a0c","owner":[],"postedDate":"June 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-07-31T13:23:25+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-27 11:09:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6800634","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6800634","identity":"rs-6800634","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}