Integrating Remote Sensing and Machine Learning to Assess Climate‑Driven Yield Dynamics and Food Security in Bangladesh

Integrating Remote Sensing and Machine Learning to Assess Climate‑Driven Yield Dynamics and Food Security in Bangladesh | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Integrating Remote Sensing and Machine Learning to Assess Climate‑Driven Yield Dynamics and Food Security in Bangladesh Avrodip Biswas, Tanmoy Chakraborty This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6962219/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 The agriculture sector is responsible for the majority of food production in Bangladesh. However, rapid urbanization and various anthropogenic factors have accelerated the rate of climate change, posing a significant threat to food security. This study utilizes a remote sensing-driven methodology to assess the potential impacts of climate change on food security in Bangladesh, with a specific focus on rice production. High-resolution Sentinel-1 imagery was used within the Google Earth Engine (GEE) platform to classify rice yield patterns, focusing on major growing seasons (Aman, Aus, Boro) for the years 2018, 2020, and 2022. For classification, the Random Forest algorithm was employed due to its high precision and reliability. Subsequently, an Artificial Neural Network model (Multi-Layer Perceptron) was used within MOLUSCE to predict future yield dynamics for the years 2026 and 2030. Among the climatic variables, precipitation, evapotranspiration, soil moisture, sunshine duration, and cloud cover were integrated with three topographic variables: DEM, slope, and aspect, to assess their influence on rice productivity. The rice yield classification achieved a high degree of precision (AUC = 0.968). The analysis reveals a significant decline in rice cultivation area, from 519,318 hectares in 2018 to 442,902 hectares in 2022, with projected reductions to 421,697 hectares by 2026 and 357,145 hectares by 2030. Correlation analysis indicated a strong positive association between rice yield and sunshine (r = 0.70), a weaker positive correlation with precipitation (r = 0.26), and a moderate negative relationship with evapotranspiration (r = -0.32), while the remaining variables showed insignificant correlations. This study highlights the increasing vulnerability of rice production to climate change and emphasizes the need for acknowledging these effects. The developed method can contribute to improved crop mapping and early prediction of food security situations in the South Asian region. Geographic Information Systems Food Security Artificial Neural Network Climate Change Crop Mapping Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 1. Introduction Agriculture contributes to approximately 90% of the calories in our food and around 80% of the proteins and fats, primarily through livestock production (Viana & Rocha, 2020 ). Besides supplying food, agricultural production is also responsible for delivering a variety of ecosystem services (Zulfiqar et al., 2019 ). Therefore, agriculture is critical for achieving food security and is fundamental to the success of attaining the Sustainable Development Goals (Viana et al., 2022 ). Despite decades of dedicated efforts to achieve global food security, it still remain as a major problem, affecting approximately 10% of people around the globe (FAO, 2020 ; Müller et al., 2020 ). To meet the demands of an ever growing population, a number of studies have estimated that global production of cereal will need to double by 2050 (Tilman et al., 2011 ). According to the Intergovernmental Panel on Climate Change (IPCC) Climate Change and Land report, changes in climate such as massive rainfall, temperature fluctuations, and water scarcity have negative impacts on agricultural productivity (Vos et al., 2020 ). Climate change, driven by human activities such as deforestation and greenhouse gas emissions, is a global issue that affects the Earth's atmosphere, raises temperatures, and impacts human health, air quality, water supply and food security (Fahad et al., 2023 ). Also, climate change is particularly affecting the flora and fauna, as well as natural systems that are essential for human survival (Saini, 2023 ). In regions with fragile agricultural systems, climate change will adversely affect crop yields, ultimately compromising the supply of food and accessibility (Sishodia et al., 2020 ). Anticipating the state of food insecurity is vital for enabling timely actions, particularly by human efforts, to address and mitigate potential crises (Westerveld et al., 2021 ). In order to minimize the challenges associated with climate change impacts on food supply, it is crucial to invest and develop new technologies for data acquisition techniques like remote sensing as well as create reliable and validated models derived from multiple sources (Weiss et al., 2020 ). Remote sensing is vital for forecasting and addressing climate change by supplying long-term data to study Earth's climate system across different scales, supporting research and impact evaluation (Wang, 2023 ). Advanced tools such as remote sensing, GPS, Geographic Information Systems (GIS), Big Data analysis, and artificial intelligence (AI) are efficient in optimizing agricultural practices and inputs to boost production while minimizing losses (Delgado et al., 2019 ; Saleem et al., 2021 ). Remote sensing enhances agriculture by assessing crop moisture status, increasing irrigation efficiency with indices such as NDWI, and facilitating precision agriculture for optimized water resource management (Singh et al., 2023 ). Remote sensing techniques also enables the collection of information on the biophysical condition of vegetation across extensive areas with frequent revisits (Anderson et al., 2020 ). Conventional methods for estimating crop yields, such as agronomic forecasting, crop-growth models, and meteorological statistical approaches, face spatial and temporal limitations when compared to remote sensing techniques. These methods are not only time consuming but also inconvenient when applied in the large scales. (Khanian et al., 2018 ; Cheng et al., 2022 ). Remote sensing indices such as NDVI, EVI, SAVI, NDWI are used in precision agriculture for crop management, helping to estimate agricultural production from field preparation to harvest (Gao, 1996 ; A. R. Huete, 1988 ; A. R. Huete et al., 1997 ; Kumawat et al., 2023 ). Still some indices have inherent limitations, such as NDVI being susceptible to atmospheric conditions, and NDWI being affected by seasonal variations (A. R. Huete et al., 2002 ; Xu, 2006 ). Studies on food security conducted in Sub-Saharan Africa indicate that climate hazards, such as droughts and floods, adversely impact agricultural productivity (Mwongera et al., 2017 ). Similar research related to food security has also been conducted in South Asia and Latin America, demonstrating comparable impacts of climate change on agricultural productivity (Bandara & Cai, 2014 ; Magrin et al., 2014 ). The conventionally used remote sensing indices might be useful for a lot of regions but inconsistent data quality across different regions poses significant challenges to monitor the food security conditions (Justice et al., 1985 ; Roy et al., 2014 ). LULC is the most commonly used technique that helps identifying agricultural areas and assessing their productivity, as changes in LULC can indicate shifts in agricultural practices, environmental degradation, making accurate LULC data essential for planning and implementing food security strategies (Lambin et al., 2003 ; Foley et al., 2005 ). Still different regions may use various classification systems and inconsistent sensor technologies and data processing methods, making comparisons difficult and affecting data reliability. Therefore, standardization of LULC data collection and processing is necessary for accurate monitoring (Herold et al., 2006 ; Giri, 2012 ). Machine learning (ML) is a fast-evolving technical field that sits at the crossroads of computer science and statistics, forming a crucial part of artificial intelligence and data science (Jordan & Mitchell, 2015 ). ML algorithms can process a large amount of remote sensing data to assess crop health and predict yields, enhancing decision-making by providing precise and timely information to farmers and policymakers (Kamilaris & Prenafeta-Boldú, 2018 ; Liakos et al., 2018 ). Predictive models forecast crop yields, ensuring better resource allocation and planning, while ML can predict the impact of climate change on agricultural productivity (Lobell & Asseng, 2017 ). GEE is a cloud-based geospatial processing service that utilizes Google's infrastructure, offering access to a vast data archive and advanced analytical tools (Gorelick et al., 2017 ). Conventional GIS software often has difficulty managing the large and complex datasets needed for precise agricultural analysis. On the contrary, GEE's cloud-based platform efficiently handles extensive datasets, including high-resolution satellite imagery and historical climate data, which are essential for accurate crop yield forecasting (Hansen et al., 2013 ). An extensive literature review indicates that there have been a few studies on crop yield prediction in certain study areas (Rudiyanto et al., 2019; Cao et al., 2021 ; Cheng et al., 2022 ). Several studies have been conducted in Iran, Nepal, and India to assess the impact of climate change on food security. (Ahmad et al., 2011 ; Bocchiola et al., 2019 ; Kazemi Garajeh et al., 2023 ). While random forest algorithms have been used in past yield estimation studies for various crops, the assessment of food security considering related climatic variables remained unexamined for Bangladesh. Our study combines climatic and geographic variables alongside ANN and ML algorithms to identify potential future food security risks focusing on rice. In addition to estimating national yield for rice, the study correlates it with relevant variables to show the current situation and forecast probable future trends. This study aims to identify the present food security situation from 2018 to 2022, determine the degree of relation between food security and climate change using climatic and geospatial variables, and predict the food security situation for the years 2026 and 2030. 2. Materials & Methods 2.1 Study Area Bangladesh's unique geographical and geophysical attributes render it exceptionally susceptible to environmental hazards, a vulnerability exacerbated by the projected impacts of anthropogenic climate change (Md. N. Islam et al., 2021 ). Moreover, rice is the main food for the vast majority (70–75%) of Bangladeshis, contributing up to 70% of their daily calorie consumption (Kamruzzaman et al., 2025 ). Consequently, Bangladesh presents a salient case study for the development and application of predictive models designed to assess the influence of climate change on food security. 2.2 Data Collection & Processing Table 2.1 Characteristics of all the variables used in assessment of the impact of climate change on food security (Coordinate System: WGS 1984) Variable Spatial Resolution Temporal Resolution Number of Images (Per year) Source Type Satellite Image 10m 6-day ~ 61 Sentinel-1 SAR Raster LST 1km 8-day ~ 46 MOD11A2 V6.1 Raster Precipitation 5.5km Daily 365 CHIRPS (Daily) Raster Evapotranspiration 500m 8-day ~ 46 MOD16A2GF.061 Raster Soil Moisture 11km 1-day 365 ERA5 Raster DEM 30m N/A N/A NASA SRTM Digital Elevation Raster Slope 30m N/A N/A NASA SRTM Digital Elevation Raster Aspect 30m N/A N/A NASA SRTM Digital Elevation Raster Sunny Days N/A N/A N/A BARC Vector Cloud Coverage N/A N/A N/A BARC Vector Several climatic and geographical factors were considered to assess their influence on crop yields in Bangladesh using Google Earth Engine (GEE) to analyze data from 2018, 2020, 2022 and 2024. Land surface temperature (LST), a crucial factor for plant growth (Majumder et al., 2020 ), was derived from the MODIS MOD11A2 dataset ('LST_Day_1km' band) and converted to Celsius. Precipitation, vital for crop productivity (Njenga, 2013 ), was assessed using daily rainfall data from the UCSB-CHG CHIRPS dataset, with annual average precipitation calculated to determine mean daily precipitation (mm/day). Evapotranspiration (ET), directly linked to crop production efficiency (Hussain et al., 2025 ), was analyzed using the MODIS MOD16A2GF.061 daily ET data ('ET' band, kg/m²/8-day) to understand annual variations. Finally, soil moisture, significant for soil fertility and agroecosystem productivity (P. Zhang et al., 2025 ), was examined using the ERA5 dataset by extracting the surface soil moisture (ssm) band and computing monthly means. Topographic factors, including elevation, slope, and aspect, significantly impact crop yield by influencing water flow, soil moisture, and nutrient distribution (Jiang & Thelen, 2004 ). To analyze these variables, a 30-meter resolution Digital Elevation Model (DEM) was obtained from NASA's Shuttle Radar Topography Mission (SRTM) dataset. Subsequently, slope and aspect rasters, representing the rate of elevation change, were generated from the DEM using terrain analysis functions in Google Earth Engine. Sunshine hours and cloud coverage, critical for crop development (Song & Jin, 2020 ), were sourced as station-wise point data from the Bangladesh Agricultural Research Council (BARC) for the years 2018, 2020, 2022 and 2024. These point data were then converted into continuous raster across Bangladesh using the Inverse Distance Weighting (IDW) interpolation technique to estimate values all over the region. 2.3 Spatiotemporal Analysis for Climatic and Geographical Variables For the spatiotemporal analysis, raster datasets were first generated using a range of methodologies explained in the data collection section. Each variable of interest was then individually processed within ArcMap 10.8 to produce separate thematic maps. Specifically, three distinct maps were generated for each variable corresponding to the years 2018, 2020, 2022 and 2024 in order to capture and compare temporal dynamics. It is important to note that for variables such as the digital elevation model (DEM), slope, and aspect which remain constant within a defined study area and a single map was produced for each of them. After map generation, a comparison of the high and low values was conducted using their respective units. This method helped clearly illustrate the spatial and temporal trends in the study area. 2.4 Crop Classification Using Machine Learning based Algorithm In a study examining the genetic structures of rice in Bangladesh, the predominant rice cultivation seasons identified are Aus, Amon, and Boro, each representing critical periods within the agricultural calendar of the region (Parsons et al., 1999 ). Consequently, this study specifically focused on the major rice-growing seasons, namely Aus, Amon, and Boro, to ensure comprehensive spatiotemporal coverage of rice cultivation across different periods of the year. The classification of paddy or rice as the primary crop was undertaken using Sentinel-1 Synthetic Aperture Radar (SAR) data within the Google Earth Engine (GEE) platform. The temporal extent for image acquisition was carefully aligned with the rice transplanting seasons, wherein Sentinel-1 images were collected from June to August for Aus, August to October for Amon, and February to May for Boro (Alam et al., 2021 ). For the classification process, Sentinel-1 SAR images were acquired in the Interferometric Wide Swath (IW) imaging mode using the VH polarization. The selection of VH polarization was based on its higher sensitivity to water, which separates rice fields from other land cover types (Nguyen et al., 2016 ). Given the unique backscatter behavior of rice, the images were processed to generate multiple temporal stacks, which were subsequently combined to produce a composite image representing an annual overview of rice cultivation. This compositing approach allowed for the integration of multi-temporal SAR data, thereby improving classification accuracy by capturing the temporal dynamics of rice growth. The characteristic backscatter response of rice pixels in the composite imagery was observed in the form of cyan and magenta coloration, indicating the presence of rice fields due to their water retention properties and distinct phenological variations over time. To develop a robust classification model, approximately 350 samples were collected, representing three distinct land cover classes: rice yield, urban areas, and water bodies. These samples were digitized as polygons to ensure spatial accuracy and sufficient representation across diverse geographic regions. The dataset was subsequently divided into training and testing subsets, following an 80:20 ratio, where 80% of the samples were used to train the model, and the remaining 20% were reserved for accuracy assessment. A Random Forest (RF) algorithm was employed for classification due to its well-documented efficacy in handling high-dimensional remote sensing data, as well as its ability to mitigate overfitting by aggregating multiple decision trees. The RF classifier was trained using the selected input features derived from SAR backscatter values, and the resulting classification output was validated to assess its accuracy. Although three land cover classes: rice fields, urban areas, and water bodies were initially classified, only the rice yield raster was retained as the final output. Additionally, the urban and water body classes were excluded from the final map. Finally, the classified raster, delineating the spatial distribution of rice fields, was exported for further analysis and interpretation. This methodology ensured a robust and replicable approach to paddy classification using SAR-based remote sensing techniques. 2.5 ANN Model for Rice Yield Prediction A Multi-Layer Perceptron (MLP), a type of ANN, was employed to develop a neural network sample capable of learning complex, non-linear relationships between the spatial variables and rice yield outcomes. To predict rice yield in Bangladesh, MOLUSCE plugin was used within QGIS 2.18.0, offering a robust framework for analyzing agricultural productivity (Kamaraj & Rangarajan, 2022 ). The ANN model was trained using a 1-pixel neighborhood to capture yield patterns. A learning rate of 0.1 was applied to ensure efficient convergence, while a momentum of 0.05 helped stabilize weight updates. To allow the model to learn complex relationships between input variables and crop yield transitions, 25 hidden layers were used. The model utilized rice classification rasters from 2018 as the initial state and 2020 as the final state, providing temporal benchmarks for the prediction process. These were complemented by a set of spatial variables influencing rice yield, including Land Surface Temperature (LST), Precipitation, Evapotranspiration, Soil Moisture, Sunshine, Cloud Coverage, Digital Elevation Model (DEM), Slope, and Aspect, which jointly captured the environmental and topographical conditions critical to rice production. This was followed by the application of Cellular Automata Simulation (Santé et al., 2010 ) to project rice yield for 2022, incorporating spatial dynamics and temporal progression into the prediction. The predicted yield for 2022 was validated against actual rice classification data for the same year, enabling an accuracy assessment to confirm the model's reliability. Following successful validation, the model was extended to forecast rice yields for 2026 and 2030 through multiple iterations, refining the predictions and providing insights into future agricultural productivity under the specified conditions. 2.6 Correlation Analysis for Variables To assess the relationship between rice yield and various environmental and topographical factors, a comprehensive correlation analysis was conducted. The analysis aimed to identify the influence of key variables, including Land Surface Temperature (LST), Precipitation, Evapotranspiration, Soil Moisture, Sunshine, Cloud Coverage, Digital Elevation Model (DEM), Slope, and Aspect, on rice cultivation patterns. The workflow initiated with the importation of all raster datasets into ArcGIS software to underpin spatial analysis. A fishnet polygon grid was then constructed. This grid structure facilitated the orderly retrieval of spatial data points, ensuring even sampling throughout the study area. Once the fishnet was created, values from all raster layers corresponding to the selected variables were extracted at each grid point. Following data extraction, the dataset was processed in Microsoft Excel for further refinement. The extracted values were meticulously filtered to remove any anomalies or inconsistencies, ensuring data integrity. Min-Max Normalization technique was applied to the dataset (Jayalakshmi & Santhakumaran, 2011 ), making it suitable for correlation analysis. The following equation was used for performing the normalization. $$\:{z}_{i}=\frac{{x}_{i}-\text{min}\left(x\right)}{\text{max}\left(x\right)\:-\:\text{m}\text{i}\text{n}\left(x\right)}$$ 2 …………………. The correlation analysis was then performed to determine the statistical relationship between rice yield and the selected environmental and topographical factors. Pearson’s correlation coefficient was computed between rice yield and each of the variables to quantify the degree and direction of their association. LST and Precipitation were analyzed to assess their impact on rice growth, considering the thermal and hydrological conditions essential for crop development. Evapotranspiration and Soil Moisture were evaluated to understand their role in water availability and plant water stress, both of which significantly influence rice production. Sunshine and Cloud Coverage were examined to capture the effects of solar radiation and atmospheric conditions on photosynthesis and crop growth. DEM, Slope, and Aspect were incorporated into the analysis to account for terrain influences, which can affect water drainage, soil erosion, and overall land suitability for rice cultivation. 3. Results 3.1 Delineation of Rice Yield Classification Spatial representations illustrate rice yield distribution across Bangladesh for the years 2018, 2020, and 2022. Corresponding cultivated areas were recorded as 519,318 hectares in 2018, 484,271 hectares in 2020, and 442,902.1 hectares in 2022. These data indicate a progressive decline in rice cultivation over the specified period. 3.2 Spatiotemporal Delineation of All the Predisposing Variables DEM, Slope & Aspect: The DEM map shows elevation ranging from -34 m to 1052 m, with the highest elevations located in the Chittagong Hill Tracts in the southeast. The slope map depicts terrain steepness, with values from 0° to 63.83°. The aspect map indicates slope direction (0°–360°), with varied orientations in the hilly southeast and little variation in the flat lowlands. Precipitation: The maps show the spatial distribution of precipitation (in mm/day) across Bangladesh for the years 2018, 2020, 2022 and 2024. The color gradient from pink to blue represents decreasing rainfall intensity, with pink indicating higher values. In 2018, the mean daily precipitation was 7.96 mm, showing moderate rainfall across the country. In 2020, the intensity increased notably, with a mean of 10.62 mm, especially concentrated in the northeastern and southeastern regions. By 2022, the average declined slightly to 9.49 mm/day, yet remained higher than in 2018. And, in 2024 the average rainfall was 7.08 which is the least among all these years. LST: The series of maps illustrates the spatial distribution of Land Surface Temperature (LST) across Bangladesh for the years 2018, 2020, 2022 and 2024. Numerically, the mean LST values were 27.67°C in 2018, 27.65°C in 2020, 28.49°C in 2022 and 26.57 in 2024. While there was a slight dip in the average LST between 2018 and 2020, a marked increase occurred by 2022 which fell again in 2024. Evapotranspiration: The series of maps in Figure 3.5 illustrate the spatial distribution of evapotranspiration (ET) across Bangladesh for the years 2018, 2020, 2022 and 2024, as derived from remote sensing data. For the years, 2018, 2020, 2022 and 2024 the average value of evapotranspiration was 184.16, 201.68, 187.90 and 186.40 respectively. It is clearly visible from the numbers that during those periods, evapotranspiration remained stable except for 2020 when it was the highest. Soil Moisture: The maps in Figure 3.6 depict the spatial distribution of soil moisture across Bangladesh for the years 2018, 2020, 2022 and 2024 in centimeter unit. The mean soil moisture remained almost the same from 2018 to 2022 which was 0.3cm approximately. In 2024 it decreased slightly to 0.29cm. Overall, the soil moisture remained mostly stable during 2018 to 2024. Sunshine: The spatial distribution maps of sunshine hours across Bangladesh for the years 2018, 2020, and 2022 (Figure 3.7) reveal a consistent pattern with regional variation. High sunshine duration is predominantly observed in the southeastern and southwestern coastal regions, while the central and northern parts exhibit relatively lower sunshine intensity. Over the years, a slight increase in sunshine duration is observed, as supported by the average daily sunshine data, rising from 5.83 hours in 2018 to 6.12 hours in 2024. Apart from that, there is insignificant variation in the middle years (2020 and 2022). Cloud Coverage: The cloud coverage maps of Bangladesh for the years 2018, 2020, 2022 and 2024 (Figure 3.8) illustrate spatial and temporal variations in atmospheric cloud distribution. Higher cloud density is consistently concentrated in the southern coastal and southeastern hilly regions, while the central and northwestern areas generally experience lower coverage. Over the observed period, a slight increase in average cloud cover is noted from 1.19 octas in 2018 to 1.38 octas in 2020, followed by a marginal decrease to 1.35 octas in 2022 and remained same in 2024 as well. 3.3 Correlation between Rice Yield and Pre-disposing variables This heatmap presents the pairwise Pearson correlation coefficients between six climatic variables: Land Surface Temperature (LST), Precipitation, Soil Moisture, Evapotranspiration, Sunshine hours, and Cloud Cover, derived from data spanning the years (2018, 2020, 2022, 2024). The color gradient and numerical values illustrate the strength and direction of linear relationships between these variables. Notably, strong positive correlations are observed between LST and Sunshine, and between Precipitation and Cloud Cover, indicating co-variation. Conversely, a moderate negative correlation exists between Evapotranspiration and LST/Soil Moisture. This visualization facilitates the rapid identification of inter-variable dependencies and provides a basis for understanding potential climate change impacts by highlighting how shifts in one climatic factor may influence others within the system. Table 3.1: Correlation between Rice Yield and Climatic Variables Agricultural Land Climatic Variables Evapotranspiration -0.32 Sunshine 0.70 Cloud Coverage 0.05 LST -0.04 Precipitation 0.26 Soil Moisture -0.01 Table 3.1 presents the Pearson correlation coefficients between rice yield and six key climatic variables: Evapotranspiration, Sunny Days, Cloud Coverage, Land Surface Temperature (LST), Precipitation, and Soil Moisture. The table reveals that sunshine possesses the strongest positive correlation (r = 0.70) with rice yield, suggesting a substantial direct relationship where increased sunshine is associated with higher yield. Precipitation shows a moderate positive correlation (r = 0.26), indicating a positive but weak association with rice yield. Evapotranspiration displays a moderate negative correlation (r = -0.32), implying an inverse relationship with rice yield. Cloud Coverage, LST, and Soil Moisture show negligible correlations (r = 0.05, -0.04, and -0.01, respectively), indicating their insignificant roles in rice yields. Overall, the table highlights sunshine as a dominant factor influencing rice yield, while the effects of other climatic variables are comparatively weaker or negligible. These correlations provide a quantitative basis for understanding the relationship between rice yield and climatic factors, offering insights for agricultural management and climate change adaptation strategies. 3.4 Prediction of Food Security (Rice Yield) for 2026 and 2030 Table 3.2: Prediction of Rice Yield for 2026 and 2030 Year Rice Cultivation Area (Ha) 2018 519318.2 2020 484271.1 2022 442902.1 2026 421697.3 2030 357145.4 Table 3.4 presents the predicted rice cultivation area (in hectares) for the years 2026 and 2030, alongside the recorded areas for 2018, 2020, and 2022. This table illustrates a projected decline in rice cultivation area over the coming decade. The data reveals a consistent decreasing trend from 519,318.2 Ha in 2018 to 357,145.4 Ha predicted for 2030. Specifically, the observed decline between 2018 and 2022 suggests an ongoing contraction of rice cultivation, which is further extrapolated to predict a continued reduction in 2026 and 2030. 3.5 Accuracy Assessment Rice Classification Accuracy The ROC curve (Figure 3.10) and the corresponding AUC of 0.968 demonstrate excellent discriminative ability of the rice classification model. This high AUC value indicates a strong likelihood (96.8%) that the model will correctly rank a randomly chosen rice pixel higher than a non-rice pixel. The curve's shape, characterized by a rapid ascent to a high True Positive Rate at a low False Positive Rate, further supports the model's robust performance in distinguishing between rice and non-rice areas within the dataset. These results suggest the model's suitability for accurate rice cultivation mapping and subsequent spatial analysis. Rice Prediction Accuracy The predicted rice cultivation area for the year 2022, derived from ANN model in MOLUSCE, was validated against the independently classified rice map of 2022. The spatial agreement between the predicted and classified extents yielded an accuracy of 74%. This indicates a substantial agreement beyond that expected by chance, suggesting a moderate to high level of accuracy in the model's predictive capability for rice cultivation extent. 4. Discussion & Conclusion 4.1 Discussion Over the past three decades, global rice production has come under mounting pressure from climate variability, with each 1°C rise in mean growing‑season temperature linked to a ~ 10% yield reduction (Zhao et al., 2017 ) and elevated night‑time temperatures alone causing similar losses (Peng et al., 2004 ). In Bangladesh, rice cultivation area contracted from 3.81% of the landscape in 2002 to 3.43% in 2021, mirroring broader trends in food insecurity. After two decades of decline, global hunger began to rise in the early 1990s, driven largely by surges in food prices; the 2008 financial crisis then dampened agricultural markets, reducing trade and market exchanges. Today, an estimated one billion people subsist on less than USD 1.25 per day, three‑quarters of whom depend directly on agriculture for their livelihoods that underscore the sector’s centrality to rural poverty alleviation and national economic development (FAO, 2005 ). The spatiotemporal analysis reveals a marked contraction in rice cultivation area in Bangladesh, from 519,318 ha in 2018 to 442,902 ha in 2022, with predictive modeling indicating further declines to approximately 421,697 ha by 2026 and 357,145 ha by 2030. This ongoing reduction of a 32% loss over twelve years actually raises significant food‑security concerns for a nation where rice contributes to both caloric intake and rural livelihoods. Moreover, country‑specific simulations forecast a potential 33% decline in Bangladesh rice yields by 2100 under high‑emission trajectories (Karim et al., 2012 ). These converging lines of evidence underscore the urgency of both mitigation and adaptive responses to safeguard future rice production. The Pearson correlation analysis identified sunny days as the strongest climatic driver of rice yield (r = 0.70), signifying the critical role of solar radiation in photosynthetic carbon assimilation and biomass accumulation. Empirical studies corroborate this: mean daily solar radiation during the crop growth period exhibits significant positive correlations with rice yield and harvest index (Wei et al., 2021 ); similarly, radiation interception and utilization metrics have been shown to explain a substantial proportion of yield variability in field trials (Huang et al., 2016 ). Conversely, evapotranspiration demonstrated a moderate negative association with yield (r = − 0.32), suggesting that elevated evaporative demand may exacerbate plant‑water stress and reduce grain filling. Such inverse relationships between ET and rice productivity have been documented in both upland and paddy systems (Maina et al., 2014 ), highlighting the importance of optimized irrigation scheduling and water‑use efficiency interventions. The rice classification model achieved an AUC of 0.968, indicating robust discrimination between rice and non‑rice pixels. While ground verification in this study was necessarily limited to Google Earth Pro due to funding constraints, integrating dedicated field‑survey methods in future work would markedly enhance classification accuracy and model calibration. Building upon the remote‑sensing framework established here, survey‑based ground truthing could yield richer, higher‑resolution insights into rice heterogeneity and the environmental drivers identified. Furthermore, incorporating socio‑economic datasets such as farm management practices, market accessibility, and household livelihood indicators would provide a more holistic perspective on yield dynamics, enabling models that better account for both biophysical and human dimensions of food‑security change. The prediction of rice cultivation area through 2030 highlights a potentially severe contraction in arable rice land. These projections align with broader modeling by the International Food Policy Research Institute, which anticipates up to a 15% global decline in rice yield by mid‑century under unmitigated climate change scenarios (A. R. Md. T. Islam et al., 2020 ). To counteract these trends, Bangladesh has begun deploying adaptation measures most notably, the development and dissemination of heat‑tolerant rice varieties by IRRI in partnership with BRRI, which demonstrate that the newly developed rice varieties produce 1.5–2 metric tons per hectare higher yields than conventional strains, even when exposed to elevated nighttime temperatures (IRRI, 2024). Additionally, farmers are adopting resilient agronomic practices such as alternate wetting and drying irrigation, diversified cropping patterns, and ecosystem‑based adaptation strategies (Li, 2023 ). Continued expansion of such initiatives, coupled with enhanced monitoring and modeling, will be critical for sustaining rice output in a changing climate. 4.2 Conclusion By integrating remote sensing, GEE, and machine learning, this research evaluated and forecast how climate change will affect rice production and food security. To summarize, this work sought to: (1) understand the spatial variation of rice yields, (2) explain the primary climatic controls on rice production, and (3) anticipate how climate change will alter the areas where rice is cultivated. According to study, rice cultivation in the region has experienced a notable decline over recent years, shrinking considerably between 2018 and 2024. This trend highlights a significant shift in agricultural land use. Among the various climatic factors examined, the amount of sunshine emerged as the most significant, showing a strong positive relationship with rice yield. On the contrary, evapotranspiration exhibited a moderate negative association, suggesting it can hinder yield. Other climatic elements appeared to have only minor connections to rice production. The effectiveness of the rice-classification model in predicting outcomes was high, and its forecasts for 2022 aligned well with actual observations. Looking ahead, projections indicate a continued decrease in rice cultivation area through 2026 and further by 2030. These findings underscore the critical need for adaptive management strategies to address the evolving landscape of rice farming. The clear decline in cultivation area, strong dependence on solar radiation, and projected contractions by 2030 underscore the multi‑faceted threats to food security in Bangladesh. While the classification model exhibits excellent discrimination, further refinement through higher‑resolution and more frequent satellite data could improve mapping accuracy. This research offers a valuable tool for policymakers in Bangladesh, enabling the proactive utilization of early mapping and predictive modeling to mitigate the adverse impacts of climatic hazards. Consequently, it is anticipated that this study will foster an increased awareness of the imperative to transition towards contemporary agricultural technologies, such as precision agriculture. While the significance of climatic variables in shaping agricultural output is acknowledged, it is crucial to recognize the substantial influence of other intrinsic factors, including soil quality, cultivation practices, and fertilizer efficacy. Future scholarly investigations could fruitfully integrate these elements to provide a more holistic understanding. Furthermore, subsequent research endeavors should consider incorporating socio-economic data and management variables to provide a more nuanced perspective. Exploring causal modeling frameworks would enable a deeper understanding of the complex interrelationships at play. Finally, evaluating the effectiveness of emergent adaptation strategies, encompassing precision irrigation, precision agriculture methodologies, and the deployment of genetically modified crop species, warrants rigorous investigation to inform evidence-based policy decisions. Policymakers and stakeholders should prioritize investments in climate‑smart agriculture, targeted breeding programs, and real‑time monitoring systems to mitigate yield losses and ensure the resilience of Bangladesh’s rice sector. References Ahmad, J., Alam, D., & Haseen, M. S. (2011). Impact of climate change on agriculture and food security in India. International Journal of Agriculture, Environment and Biotechnology , 4 (2), 129–137. Alam, M. J., Al-Mahmud, A.-, Islam, M. A., Hossain, M. F., Ali, M. A., Dessoky, E. S., El-Hallous, E. I., Hassan, M. M., Begum, N., & Hossain, A. (2021). Crop Diversification in Rice—Based Cropping Systems Improves the System Productivity, Profitability and Sustainability. Sustainability , 13 (11), Article 11. https://doi.org/10.3390/su13116288 Anderson, R., Bayer, P. E., & Edwards, D. (2020). Climate change and the need for agricultural adaptation. Current Opinion in Plant Biology , 56 , 197–202. https://doi.org/10.1016/j.pbi.2019.12.006 Bandara, J. S., & Cai, Y. (2014). The impact of climate change on food crop productivity, food prices and food security in South Asia. Economic Analysis and Policy , 44 (4), 451–465. Bocchiola, D., Brunetti, L., Soncini, A., Polinelli, F., & Gianinetto, M. (2019). Impact of climate change on agricultural productivity and food security in the Himalayas: A case study in Nepal. Agricultural Systems , 171 , 113–125. https://doi.org/10.1016/j.agsy.2019.01.008 Cao, J., Zhang, Z., Tao, F., Zhang, L., Luo, Y., Zhang, J., Han, J., & Xie, J. (2021). Integrating Multi-Source Data for Rice Yield Prediction across China using Machine Learning and Deep Learning Approaches. Agricultural and Forest Meteorology , 297 , 108275. https://doi.org/10.1016/j.agrformet.2020.108275 Cheng, E., Zhang, B., Peng, D., Zhong, L., Yu, L., Liu, Y., Xiao, C., Li, C., Li, X., Chen, Y., Ye, H., Wang, H., Yu, R., Hu, J., & Yang, S. (2022). Wheat yield estimation using remote sensing data based on machine learning approaches. Frontiers in Plant Science , 13 , 1090970. https://doi.org/10.3389/fpls.2022.1090970 Delgado, J. A., Short, N. M., Roberts, D. P., & Vandenberg, B. (2019). Big Data Analysis for Sustainable Agriculture on a Geospatial Cloud Framework. Frontiers in Sustainable Food Systems , 3 , 54. https://doi.org/10.3389/fsufs.2019.00054 Fahad, S., Adnan, M., Saud, S., & Nie, L. (Eds.). (2023). Climate change and ecosystems: Challenges to sustainable development (First edition). CRC Press. FAO. (2005). Agricultural trade and poverty: Can trade work for the poor? Food and Agriculture Organization of the United Nations. FAO. (2020). The State of Agricultural Commodity Markets 2020 . FAO ; https://openknowledge.fao.org/handle/20.500.14283/cb0665en Foley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., Chapin, F. S., Coe, M. T., Daily, G. C., Gibbs, H. K., Helkowski, J. H., Holloway, T., Howard, E. A., Kucharik, C. J., Monfreda, C., Patz, J. A., Prentice, I. C., Ramankutty, N., & Snyder, P. K. (2005). Global Consequences of Land Use. Science , 309 (5734), 570–574. https://doi.org/10.1126/science.1111772 Gao, B.-C. (1996). NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment , 58 (3), 257–266. Giri, C. P. (2012). Remote sensing of land use and land cover: Principles and applications . CRC press. Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment , 202 , 18–27. https://doi.org/10.1016/j.rse.2017.06.031 Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A., Tyukavina, A., Thau, D., Stehman, S. V., Goetz, S. J., Loveland, T. R., Kommareddy, A., Egorov, A., Chini, L., Justice, C. O., & Townshend, J. R. G. (2013). High-Resolution Global Maps of 21st-Century Forest Cover Change. Science , 342 (6160), 850–853. https://doi.org/10.1126/science.1244693 Herold, M., Woodcock, C. E., Antonio Di Gregorio, Mayaux, P., Belward, A. S., Latham, J., & Schmullius, C. C. (2006). A joint initiative for harmonization and validation of land cover datasets. IEEE Transactions on Geoscience and Remote Sensing , 44 (7), 1719–1727. https://doi.org/10.1109/TGRS.2006.871219 Huang, M., Shan, S., Cao, F., & Zou, Y. (2016). The solar radiation-related determinants of rice yield variation across a wide range of regions. NJAS: Wageningen Journal of Life Sciences , 78 (1), 123–128. https://doi.org/10.1016/j.njas.2016.05.004 Huete, A., Didan, K., Miura, T., Rodriguez, E. P., Gao, X., & Ferreira, L. G. (2002). Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment , 83 (1–2), 195–213. Huete, A. R. (1988). A soil-adjusted vegetation index (SAVI). Remote Sensing of Environment , 25 (3), 295–309. Huete, A. R., Liu, H. Q., Batchily, K. V., & Van Leeuwen, W. (1997). A comparison of vegetation indices over a global set of TM images for EOS-MODIS. Remote Sensing of Environment , 59 (3), 440–451. Hussain, M., Zeeshan, M., Mahmood, M. H., Farwa, F., Siddiqa, S., Ahmad, Hussain, Gill, A., & Nasir, A. (2025). Advanced Applications of AquaCrop for Field Management and Climate Impact Assessment. Asian Journal of Research in Crop Science , 10 , 1–17. https://doi.org/10.9734/ajrcs/2025/v10i1327 IRRI. (2024, July 2). Heat-Tolerant Rice Varieties Promises Increased Yields in Bangladesh . International Rice Research Institute. https://www.irri.org/news-and-events/news/heat-tolerant-rice-varieties-promises-increased-yields-bangladesh Islam, A. R. Md. T., Shen, S., Yang, S., Hu, Z., & Atiqur Rahman, M. (2020). Spatiotemporal rice yield variations and potential agro-adaptation strategies in Bangladesh: A biophysical modeling approach. Sustainable Production and Consumption , 24 , 121–138. https://doi.org/10.1016/j.spc.2020.07.005 Islam, Md. N., Tamanna, S., van Amstel, A., Noman, Md., Ali, Md. S. S., Aparajita, D. M., Roy, P., Tanha, S. R., Sarkar, N., Ashiquzzaman, Md., Ghosh, S. K., Hasnat, S. R., Bhuiyan, Md. F., Sayef, Md., Saha, P., Hasan, G. R., & Ghosh, A. (2021). Climate Change Impact and Comprehensive Disaster Management Approach in Bangladesh: A Review. In Md. N. Islam & A. van Amstel (Eds.), Bangladesh II: Climate Change Impacts, Mitigation and Adaptation in Developing Countries (pp. 1–39). Springer International Publishing. https://doi.org/10.1007/978-3-030-71950-0_1 Jayalakshmi, T., & Santhakumaran, A. (2011). Statistical normalization and back propagation for classification. International Journal of Computer Theory and Engineering , 3 (1), 1793–8201. Jiang, P., & Thelen, K. D. (2004). Effect of Soil and Topographic Properties on Crop Yield in a North‐Central Corn–Soybean Cropping System. Agronomy Journal , 96 (1), 252–258. https://doi.org/10.2134/agronj2004.0252 Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. Science , 349 (6245), 255–260. https://doi.org/10.1126/science.aaa8415 Justice, C. O., Townshend, J. R. G., Holben, B. N., & Tucker, C. J. (1985). Analysis of the phenology of global vegetation using meteorological satellite data. International Journal of Remote Sensing , 6 (8), 1271–1318. https://doi.org/10.1080/01431168508948281 Kamaraj, M., & Rangarajan, S. (2022). Predicting the future land use and land cover changes for Bhavani basin, Tamil Nadu, India, using QGIS MOLUSCE plugin. Environmental Science and Pollution Research , 29 (57), 86337–86348. https://doi.org/10.1007/s11356-021-17904-6 Kamilaris, A., & Prenafeta-Boldú, F. X. (2018). Deep learning in agriculture: A survey. Computers and Electronics in Agriculture , 147 , 70–90. Kamruzzaman, M., Shozib, H. B., Kader, Md. A., Iftekharuddaula, K. M., Rahman, M. A., Lipi, L. F., Khan, Md. A. R., & Kabir, Md. S. (2025). Progress of healthier rice development in Bangladesh: A review. Journal of Food Composition and Analysis , 139 , 107082. https://doi.org/10.1016/j.jfca.2024.107082 Karim, M. R., Ishikawa, M., Ikeda, M., & Islam, Md. T. (2012). Climate change model predicts 33 % rice yield decrease in 2100 in Bangladesh. Agronomy for Sustainable Development , 32 (4), 821–830. https://doi.org/10.1007/s13593-012-0096-7 Kazemi Garajeh, M., Salmani, B., Zare Naghadehi, S., Valipoori Goodarzi, H., & Khasraei, A. (2023). An integrated approach of remote sensing and geospatial analysis for modeling and predicting the impacts of climate change on food security. Scientific Reports , 13 (1), 1057. https://doi.org/10.1038/s41598-023-28244-5 Khanian, M., Marshall, N., Zakerhaghighi, K., Salimi, M., & Naghdi, A. (2018). Transforming agriculture to climate change in Famenin County, West Iran through a focus on environmental, economic and social factors. Weather and Climate Extremes , 21 , 52–64. https://doi.org/10.1016/j.wace.2018.05.006 Kumawat, R. K., Tiwari, G., Ramakrishnan, R. S., Bhayal, D., Debnath, S., Thakur, S., & Bhayal, L. (2023). Remote Sensing Related Tools and their Spectral Indices Applications for Crop Management in Precision Agriculture. International Journal of Environment and Climate Change , 171–188. https://doi.org/10.9734/ijecc/2023/v13i11665 Lambin, E. F., Geist, H. J., & Lepers, E. (2003). Dynamics of land-use and land-cover change in tropical regions. Annual Review of Environment and Resources , 28 (1), 205–241. Li, M. (2023). Adaptation to expected and unexpected weather fluctuations: Evidence from Bangladeshi smallholder farmers. World Development , 161 , 106066. https://doi.org/10.1016/j.worlddev.2022.106066 Liakos, K. G., Busato, P., Moshou, D., Pearson, S., & Bochtis, D. (2018). Machine learning in agriculture: A review. Sensors , 18 (8), 2674. Lobell, D. B., & Asseng, S. (2017). Comparing estimates of climate change impacts from process-based and statistical crop models. Environmental Research Letters , 12 (1), 015001. Magrin, G. O., Marengo, J. A., Boulanger, J. P., Buckeridge, M. S., Castellanos López, E. J., Poveda, G., Scarano, F. R., Vicuña, S., Alfaro Martínez, E. J., & Anthelme, F. (2014). Central and south america . https://www.kerwa.ucr.ac.cr/handle/10669/89608 Maina, M., Amin, M., Rowshon, M., Wayayok, A., Abd Aziz, S., & Yazid, M. A. (2014). Effects of crop evapotranspiration estimation techniques and weather parameters on rice crop water requirement. Australian Journal of Crop Science , 8 , 495–501. Majumder, A., Kingra, P. K., Setia, R., Singh, S. P., & Pateriya, B. (2020). Influence of land use/land cover changes on surface temperature and its effect on crop yield in different agro-climatic regions of Indian Punjab. Geocarto International , 35 (6), 663–686. https://doi.org/10.1080/10106049.2018.1520927 Müller, B., Hoffmann, F., Heckelei, T., Müller, C., Hertel, T. W., Polhill, J. G., Van Wijk, M., Achterbosch, T., Alexander, P., & Brown, C. (2020). Modelling food security: Bridging the gap between the micro and the macro scale. Global Environmental Change , 63 , 102085. Mwongera, C., Shikuku, K. M., Twyman, J., Läderach, P., Ampaire, E., Van Asten, P., Twomlow, S., & Winowiecki, L. A. (2017). Climate smart agriculture rapid appraisal (CSA-RA): A tool for prioritizing context-specific climate smart agriculture technologies. Agricultural Systems , 151 , 192–203. Nguyen, D. B., Gruber ,Alexander, & and Wagner, W. (2016). Mapping rice extent and cropping scheme in the Mekong Delta using Sentinel-1A data. Remote Sensing Letters , 7 (12), 1209–1218. https://doi.org/10.1080/2150704X.2016.1225172 Njenga, N. W. (2013). Vulnerability to Climate Change and Adaptation Strategies for Small Scale Farmers in North Kinangop Location,Nyandarua South District,Kenya . http://ir-library.ku.ac.ke/handle/123456789/6545 Pang, J., Zhang, R., Yu, B., Liao, M., Lv, J., Xie, L., Li, S., & Zhan, J. (2021). Pixel-level rice planting information monitoring in Fujin City based on time-series SAR imagery. International Journal of Applied Earth Observation and Geoinformation , 104 , 102551. https://doi.org/10.1016/j.jag.2021.102551 Parsons, B. J., Newbury, H. J., Jackson, M. T., & Ford-Lloyd, B. V. (1999). The genetic structure and conservation of aus, aman and boro rices from Bangladesh. Genetic Resources and Crop Evolution , 46 (6), 587–598. https://doi.org/10.1023/A:1008749532171 Peng, S., Huang, J., Sheehy, J. E., Laza, R. C., Visperas, R. M., Zhong, X., Centeno, G. S., Khush, G. S., & Cassman, K. G. (2004). Rice yields decline with higher night temperature from global warming. Proceedings of the National Academy of Sciences of the United States of America , 101 (27), 9971–9975. https://doi.org/10.1073/pnas.0403720101 Roy, D. P., Wulder, M. A., Loveland, T. R., C.E., W., Allen, R. G., Anderson, M. C., Helder, D., Irons, J. R., Johnson, D. M., Kennedy, R., Scambos, T. A., Schaaf, C. B., Schott, J. R., Sheng, Y., Vermote, E. F., Belward, A. S., Bindschadler, R., Cohen, W. B., Gao, F., … Zhu, Z. (2014). Landsat-8: Science and product vision for terrestrial global change research. Remote Sensing of Environment , 145 , 154–172. https://doi.org/10.1016/j.rse.2014.02.001 Rudiyanto, Minasny, B., Shah, R., Che Soh, N., Arif, C., & Indra Setiawan, B. (2019). Automated Near-Real-Time Mapping and Monitoring of Rice Extent, Cropping Patterns, and Growth Stages in Southeast Asia Using Sentinel-1 Time Series on a Google Earth Engine Platform. Remote Sensing , 11 (14), 1666. https://doi.org/10.3390/rs11141666 Saini, N. (2023). Climate Change: Its Impact on Equity, Diversity, and Livelihood of the Rural Population. In S. Bala & P. Singhal (Eds.), Advances in Human Resources Management and Organizational Development (pp. 168–193). IGI Global. https://doi.org/10.4018/978-1-6684-6878-4.ch011 Saleem, M. H., Potgieter, J., & Arif, K. M. (2021). Automation in Agriculture by Machine and Deep Learning Techniques: A Review of Recent Developments. Precision Agriculture , 22 (6), 2053–2091. https://doi.org/10.1007/s11119-021-09806-x Santé, I., García, A. M., Miranda, D., & Crecente, R. (2010). Cellular automata models for the simulation of real-world urban processes: A review and analysis. Landscape and Urban Planning , 96 (2), 108–122. https://doi.org/10.1016/j.landurbplan.2010.03.001 Singh, M., Karada, M. S., Rai, R. K., Pratap, D., Agnihotri, D., Singh, A. K., & Singh, B. K. (2023). A Review on Remote Sensing as a Tool for Irrigation Monitoring and Management. International Journal of Environment and Climate Change , 13 (6), 203–211. https://doi.org/10.9734/ijecc/2023/v13i61817 Sishodia, R. P., Ray, R. L., & Singh, S. K. (2020). Applications of remote sensing in precision agriculture: A review. Remote Sensing , 12 (19), 3136. Song, L., & Jin, J. (2020). Effects of Sunshine Hours and Daily Maximum Temperature Declines and Cultivar Replacements on Maize Growth and Yields. Agronomy , 10 (12), 1862. https://doi.org/10.3390/agronomy10121862 Tilman, D., Balzer, C., Hill, J., & Befort, B. L. (2011). Global food demand and the sustainable intensification of agriculture. Proceedings of the National Academy of Sciences , 108 (50), 20260–20264. https://doi.org/10.1073/pnas.1116437108 Viana, C. M., Freire, D., Abrantes, P., Rocha, J., & Pereira, P. (2022). Agricultural land systems importance for supporting food security and sustainable development goals: A systematic review. Science of The Total Environment , 806 , 150718. https://doi.org/10.1016/j.scitotenv.2021.150718 Viana, C. M., & Rocha, J. (2020). Evaluating Dominant Land Use/Land Cover Changes and Predicting Future Scenario in a Rural Region Using a Memoryless Stochastic Method. Sustainability , 12 (10), Article 10. https://doi.org/10.3390/su12104332 Vos, R., Jackson, J., James, S., & Sánchez, M. V. (2020). Refugees and Conflict-Affected People: Integrating Displaced Communities into Food Systems (0 ed.). International Food Policy Research Institute. https://doi.org/10.2499/9780896293670_05 Wang, X. (2023). Remote Sensing Applications to Climate Change. Remote Sensing , 15 (3), 747. https://doi.org/10.3390/rs15030747 Wei, S., Peng, A., Liu, J., Wei, D., & Chen, C. (2021). Asymmetric Differences in the Effects of Average Air Temperature and Solar Radiation on Early Rice and Late Rice Yield. Atmosphere , 12 (12), 1541. https://doi.org/10.3390/atmos12121541 Weiss, M., Jacob, F., & Duveiller, G. (2020). Remote sensing for agricultural applications: A meta-review. Remote Sensing of Environment , 236 , 111402. https://doi.org/10.1016/j.rse.2019.111402 Westerveld, J. J. L., van den Homberg, M. J. C., Nobre, G. G., van den Berg, D. L. J., Teklesadik, A. D., & Stuit, S. M. (2021). Forecasting transitions in the state of food security with machine learning using transferable features. Science of The Total Environment , 786 , 147366. https://doi.org/10.1016/j.scitotenv.2021.147366 Xu, H. (2006). Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. International Journal of Remote Sensing , 27 (14), 3025–3033. https://doi.org/10.1080/01431160600589179 Zhang, M., Lin, H., Wang, G., Sun, H., & Fu, J. (2018). Mapping Paddy Rice Using a Convolutional Neural Network (CNN) with Landsat 8 Datasets in the Dongting Lake Area, China. Remote Sensing , 10 (11), 1840. https://doi.org/10.3390/rs10111840 Zhang, P., Li, L., Fu, Q., Du, C., Yang, A., Sun, N., Wang, L., & Li, M. (2025). Balancing soil carbon emissions and productivity in maize agroecosystems through nitrogen, biochar, and straw regulation. Industrial Crops and Products , 224 , 120442. https://doi.org/10.1016/j.indcrop.2024.120442 Zhao, C., Liu, B., Piao, S., Wang, X., Lobell, D. B., Huang, Y., Huang, M., Yao, Y., Bassu, S., Ciais, P., Durand, J.-L., Elliott, J., Ewert, F., Janssens, I. A., Li, T., Lin, E., Liu, Q., Martre, P., Müller, C., … Asseng, S. (2017). Temperature increase reduces global yields of major crops in four independent estimates. Proceedings of the National Academy of Sciences of the United States of America , 114 (35), 9326–9331. https://doi.org/10.1073/pnas.1701762114 Zulfiqar, F., Navarro, M., Ashraf, M., Akram, N. A., & Munné-Bosch, S. (2019). Nanofertilizer use for sustainable agriculture: Advantages and limitations. Plant Science , 289 , 110270. https://doi.org/10.1016/j.plantsci.2019.110270 Additional Declarations The authors declare no competing interests. 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-6962219","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":475558034,"identity":"fd603780-ffca-4c40-a1cb-5f6a14e9118f","order_by":0,"name":"Avrodip Biswas","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0ElEQVRIiWNgGAWjYBACAwgpIccPohMKiNZSYWEs2QDSYkC0ljMViRsOILj4gTn7GbMPH9skGDefX5344YEBgzy/2AH8Wix7coxnzmyTYDa78XazBNBhhjNnJxBw2IG0ZGbeNgk2sxtnN4C0JBjcJqTl/DOwFh7jGWc3/yBOy43kw8w8ZyQkDPh7txFni+WMx4cZZ1RIGEjc4N1mkWAgQdgv5vyJzQwfDOrq+/vPbr75o8JGnl+agBYEkACrlCBWOQjwHyBF9SgYBaNgFIwkAAAwjEKeapZ8NQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-9469-3148","institution":"Khulna University of Engineering and Technology","correspondingAuthor":true,"prefix":"","firstName":"Avrodip","middleName":"","lastName":"Biswas","suffix":""},{"id":475558035,"identity":"344f51ca-7f2d-4ccd-8bf5-3142545899d6","order_by":1,"name":"Tanmoy Chakraborty","email":"","orcid":"","institution":"Clark University","correspondingAuthor":false,"prefix":"","firstName":"Tanmoy","middleName":"","lastName":"Chakraborty","suffix":""}],"badges":[],"createdAt":"2025-06-24 06:37:05","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6962219/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6962219/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85354648,"identity":"0849d0d4-f856-4132-8012-fcbfba2d2e34","added_by":"auto","created_at":"2025-06-25 04:21:57","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":58615,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 2.1: Study Area\u003c/p\u003e","description":"","filename":"2.1.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/a50a09c862adefa6119466f1.png"},{"id":85354651,"identity":"fb05beb3-1f45-467b-920f-9c926713eeb4","added_by":"auto","created_at":"2025-06-25 04:21:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":149058,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.1(a): Rice Yield Map of Bangladesh for 2018; 3.1(b):Rice Yield Map of Bangladesh for 2020; 3.1(c): Rice Yield Map of Bangladesh for 2022\u003c/p\u003e","description":"","filename":"3.1.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/f30c0ec7b7779ea13a79e54b.png"},{"id":85355297,"identity":"3dfa4fc5-9597-4da1-aa1c-a28da3f567ca","added_by":"auto","created_at":"2025-06-25 04:37:57","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":317909,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.2(a): DEM; 3.2(b): Slope; 3.2 (c): Aspect\u003c/p\u003e","description":"","filename":"3.2.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/4bff659f95dc6fe06f90fd77.png"},{"id":85355063,"identity":"5f898969-f370-42b2-a017-f8186559a929","added_by":"auto","created_at":"2025-06-25 04:29:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":163862,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.3(a): Precipitation 2018; 3.3(b): Precipitation 2020; 3.3(c): Precipitation 2022; 3.3(d): Precipitation 2024\u003c/p\u003e","description":"","filename":"3.3.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/59e3cbbd5c4a2d0d3c9213b1.png"},{"id":85354653,"identity":"eda21f16-dba8-40e8-84a2-2f2712940d47","added_by":"auto","created_at":"2025-06-25 04:21:57","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":175976,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.4(a): LST 2018; 3.4(b): LST 2020; 3.4(c): LST 2022; 3.4(d): LST 2024\u003c/p\u003e","description":"","filename":"3.4.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/ff58d84ffeaa122ea0ba3a3a.png"},{"id":85354658,"identity":"38b7a8bf-5265-4021-a31f-b115e215d275","added_by":"auto","created_at":"2025-06-25 04:21:57","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":258023,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.5(a): ET 2018; 3.5(b): ET 2020; 3.5(c): ET 2022; 3.5(d): ET 2024\u003c/p\u003e","description":"","filename":"3.5.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/ab7562efb08132010454c541.png"},{"id":85355069,"identity":"26e8d8c2-777e-40ca-ba0b-9a5ce3e75cd1","added_by":"auto","created_at":"2025-06-25 04:29:57","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":144372,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.6(a): Soil Moisture 2018; 3.6(b): Soil Moisture 2020; 3.6(c): Soil Moisture 2022; 3.6(d): Soil Moisture 2024\u003c/p\u003e","description":"","filename":"3.6.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/a13c54d34b1958a577a5f2bc.png"},{"id":85355301,"identity":"ae9adc78-6d5e-42fb-93ed-2e3db07f2498","added_by":"auto","created_at":"2025-06-25 04:37:57","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":200698,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.7(a): Sunshine 2018; 3.7(b): Sunshine 2020; 3.7(c): Sunshine 2022; 3.7(d): Sunshine 2024\u003c/p\u003e","description":"","filename":"3.7.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/f4140eb209ffac148c7d45da.png"},{"id":85354673,"identity":"e6b08732-5e31-4120-b27a-2ec7294bb324","added_by":"auto","created_at":"2025-06-25 04:21:58","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":192606,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.8(a): Cloud Coverage 2018; 3.8(b): Cloud Coverage 2020; 3.8(c): Cloud Coverage 2022; 3.8(d): Cloud Coverage 2022\u003c/p\u003e","description":"","filename":"3.8.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/ea88873649f3ea2d5af8d643.png"},{"id":85355300,"identity":"8fbd100b-59ca-4a02-a061-2fcfbfafbc18","added_by":"auto","created_at":"2025-06-25 04:37:57","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":65608,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.9: Correlation Heatmap of the Climatic Variables between 2018-2024\u003c/p\u003e","description":"","filename":"3.9.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/b5a01a5f9736a5f9c6f2c6d4.png"},{"id":85354657,"identity":"24cca6ca-9e64-4bf5-9496-75627b9039e9","added_by":"auto","created_at":"2025-06-25 04:21:57","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":265042,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.10(a): Rice Yield Prediction Map of Bangladesh for 2026; \u0026nbsp;3.10(b): Rice Yield Prediction Map of Bangladesh for 2030;\u003c/p\u003e","description":"","filename":"3.10.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/48121aa423b3eb8ac65152cb.png"},{"id":85354661,"identity":"5cd952ef-ae32-4b5d-8791-9bc9618b1aa4","added_by":"auto","created_at":"2025-06-25 04:21:57","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":26895,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3.10: ROC curve for rice yield classification\u003c/p\u003e","description":"","filename":"3.11.png","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/5054ba6758313d1080908323.png"},{"id":85356170,"identity":"ca2f1cd4-9581-4089-8fa0-c4d2035282a3","added_by":"auto","created_at":"2025-06-25 04:53:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2860525,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6962219/v1/86688624-d980-42d7-ae14-b2f248114752.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eIntegrating Remote Sensing and Machine Learning to Assess Climate‑Driven Yield Dynamics and Food Security in Bangladesh\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAgriculture contributes to approximately 90% of the calories in our food and around 80% of the proteins and fats, primarily through livestock production (Viana \u0026amp; Rocha, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Besides supplying food, agricultural production is also responsible for delivering a variety of ecosystem services (Zulfiqar et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Therefore, agriculture is critical for achieving food security and is fundamental to the success of attaining the Sustainable Development Goals (Viana et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Despite decades of dedicated efforts to achieve global food security, it still remain as a major problem, affecting approximately 10% of people around the globe (FAO, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; M\u0026uuml;ller et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). To meet the demands of an ever growing population, a number of studies have estimated that global production of cereal will need to double by 2050 (Tilman et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). According to the Intergovernmental Panel on Climate Change (IPCC) Climate Change and Land report, changes in climate such as massive rainfall, temperature fluctuations, and water scarcity have negative impacts on agricultural productivity (Vos et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eClimate change, driven by human activities such as deforestation and greenhouse gas emissions, is a global issue that affects the Earth's atmosphere, raises temperatures, and impacts human health, air quality, water supply and food security (Fahad et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Also, climate change is particularly affecting the flora and fauna, as well as natural systems that are essential for human survival (Saini, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In regions with fragile agricultural systems, climate change will adversely affect crop yields, ultimately compromising the supply of food and accessibility (Sishodia et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAnticipating the state of food insecurity is vital for enabling timely actions, particularly by human efforts, to address and mitigate potential crises (Westerveld et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In order to minimize the challenges associated with climate change impacts on food supply, it is crucial to invest and develop new technologies for data acquisition techniques like remote sensing as well as create reliable and validated models derived from multiple sources (Weiss et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Remote sensing is vital for forecasting and addressing climate change by supplying long-term data to study Earth's climate system across different scales, supporting research and impact evaluation (Wang, \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAdvanced tools such as remote sensing, GPS, Geographic Information Systems (GIS), Big Data analysis, and artificial intelligence (AI) are efficient in optimizing agricultural practices and inputs to boost production while minimizing losses (Delgado et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Saleem et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Remote sensing enhances agriculture by assessing crop moisture status, increasing irrigation efficiency with indices such as NDWI, and facilitating precision agriculture for optimized water resource management (Singh et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Remote sensing techniques also enables the collection of information on the biophysical condition of vegetation across extensive areas with frequent revisits (Anderson et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eConventional methods for estimating crop yields, such as agronomic forecasting, crop-growth models, and meteorological statistical approaches, face spatial and temporal limitations when compared to remote sensing techniques. These methods are not only time consuming but also inconvenient when applied in the large scales. (Khanian et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Cheng et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Remote sensing indices such as NDVI, EVI, SAVI, NDWI are used in precision agriculture for crop management, helping to estimate agricultural production from field preparation to harvest (Gao, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; A. R. Huete, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1988\u003c/span\u003e; A. R. Huete et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Kumawat et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Still some indices have inherent limitations, such as NDVI being susceptible to atmospheric conditions, and NDWI being affected by seasonal variations (A. R. Huete et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Xu, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eStudies on food security conducted in Sub-Saharan Africa indicate that climate hazards, such as droughts and floods, adversely impact agricultural productivity (Mwongera et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Similar research related to food security has also been conducted in South Asia and Latin America, demonstrating comparable impacts of climate change on agricultural productivity (Bandara \u0026amp; Cai, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Magrin et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). The conventionally used remote sensing indices might be useful for a lot of regions but inconsistent data quality across different regions poses significant challenges to monitor the food security conditions (Justice et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1985\u003c/span\u003e; Roy et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). LULC is the most commonly used technique that helps identifying agricultural areas and assessing their productivity, as changes in LULC can indicate shifts in agricultural practices, environmental degradation, making accurate LULC data essential for planning and implementing food security strategies (Lambin et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Foley et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Still different regions may use various classification systems and inconsistent sensor technologies and data processing methods, making comparisons difficult and affecting data reliability. Therefore, standardization of LULC data collection and processing is necessary for accurate monitoring (Herold et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Giri, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2012\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMachine learning (ML) is a fast-evolving technical field that sits at the crossroads of computer science and statistics, forming a crucial part of artificial intelligence and data science (Jordan \u0026amp; Mitchell, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). ML algorithms can process a large amount of remote sensing data to assess crop health and predict yields, enhancing decision-making by providing precise and timely information to farmers and policymakers (Kamilaris \u0026amp; Prenafeta-Bold\u0026uacute;, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Liakos et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Predictive models forecast crop yields, ensuring better resource allocation and planning, while ML can predict the impact of climate change on agricultural productivity (Lobell \u0026amp; Asseng, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGEE is a cloud-based geospatial processing service that utilizes Google's infrastructure, offering access to a vast data archive and advanced analytical tools (Gorelick et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Conventional GIS software often has difficulty managing the large and complex datasets needed for precise agricultural analysis. On the contrary, GEE's cloud-based platform efficiently handles extensive datasets, including high-resolution satellite imagery and historical climate data, which are essential for accurate crop yield forecasting (Hansen et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAn extensive literature review indicates that there have been a few studies on crop yield prediction in certain study areas (Rudiyanto et al., 2019; Cao et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Cheng et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Several studies have been conducted in Iran, Nepal, and India to assess the impact of climate change on food security. (Ahmad et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Bocchiola et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Kazemi Garajeh et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). While random forest algorithms have been used in past yield estimation studies for various crops, the assessment of food security considering related climatic variables remained unexamined for Bangladesh. Our study combines climatic and geographic variables alongside ANN and ML algorithms to identify potential future food security risks focusing on rice. In addition to estimating national yield for rice, the study correlates it with relevant variables to show the current situation and forecast probable future trends. This study aims to identify the present food security situation from 2018 to 2022, determine the degree of relation between food security and climate change using climatic and geospatial variables, and predict the food security situation for the years 2026 and 2030.\u003c/p\u003e"},{"header":"2. Materials \u0026 Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study Area\u003c/h2\u003e \u003cp\u003eBangladesh's unique geographical and geophysical attributes render it exceptionally susceptible to environmental hazards, a vulnerability exacerbated by the projected impacts of anthropogenic climate change (Md. N. Islam et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Moreover, rice is the main food for the vast majority (70\u0026ndash;75%) of Bangladeshis, contributing up to 70% of their daily calorie consumption (Kamruzzaman et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Consequently, Bangladesh presents a salient case study for the development and application of predictive models designed to assess the influence of climate change on food security.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data Collection \u0026amp; Processing\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2.1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCharacteristics of all the variables used in assessment of the impact of climate change on food security (Coordinate System: WGS 1984)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpatial\u003c/p\u003e \u003cp\u003eResolution\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTemporal Resolution\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNumber\u003c/p\u003e \u003cp\u003eof Images\u003c/p\u003e \u003cp\u003e(Per year)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eType\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSatellite Image\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6-day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e~\u0026thinsp;61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSentinel-1 SAR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLST\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1km\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8-day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e~\u0026thinsp;46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMOD11A2 V6.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrecipitation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.5km\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDaily\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e365\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCHIRPS (Daily)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEvapotranspiration\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e500m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8-day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e~ 46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMOD16A2GF.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSoil Moisture\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11km\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1-day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e365\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eERA5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDEM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNASA SRTM Digital Elevation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSlope\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNASA SRTM Digital Elevation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAspect\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNASA SRTM Digital Elevation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRaster\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSunny Days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBARC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eVector\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCloud Coverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBARC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eVector\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eSeveral climatic and geographical factors were considered to assess their influence on crop yields in Bangladesh using Google Earth Engine (GEE) to analyze data from 2018, 2020, 2022 and 2024. Land surface temperature (LST), a crucial factor for plant growth (Majumder et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), was derived from the MODIS MOD11A2 dataset ('LST_Day_1km' band) and converted to Celsius. Precipitation, vital for crop productivity (Njenga, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), was assessed using daily rainfall data from the UCSB-CHG CHIRPS dataset, with annual average precipitation calculated to determine mean daily precipitation (mm/day). Evapotranspiration (ET), directly linked to crop production efficiency (Hussain et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), was analyzed using the MODIS MOD16A2GF.061 daily ET data ('ET' band, kg/m\u0026sup2;/8-day) to understand annual variations. Finally, soil moisture, significant for soil fertility and agroecosystem productivity (P. Zhang et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), was examined using the ERA5 dataset by extracting the surface soil moisture (ssm) band and computing monthly means.\u003c/p\u003e \u003cp\u003eTopographic factors, including elevation, slope, and aspect, significantly impact crop yield by influencing water flow, soil moisture, and nutrient distribution (Jiang \u0026amp; Thelen, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). To analyze these variables, a 30-meter resolution Digital Elevation Model (DEM) was obtained from NASA's Shuttle Radar Topography Mission (SRTM) dataset. Subsequently, slope and aspect rasters, representing the rate of elevation change, were generated from the DEM using terrain analysis functions in Google Earth Engine. Sunshine hours and cloud coverage, critical for crop development (Song \u0026amp; Jin, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), were sourced as station-wise point data from the Bangladesh Agricultural Research Council (BARC) for the years 2018, 2020, 2022 and 2024. These point data were then converted into continuous raster across Bangladesh using the Inverse Distance Weighting (IDW) interpolation technique to estimate values all over the region.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Spatiotemporal Analysis for Climatic and Geographical Variables\u003c/h2\u003e \u003cp\u003eFor the spatiotemporal analysis, raster datasets were first generated using a range of methodologies explained in the data collection section. Each variable of interest was then individually processed within ArcMap 10.8 to produce separate thematic maps. Specifically, three distinct maps were generated for each variable corresponding to the years 2018, 2020, 2022 and 2024 in order to capture and compare temporal dynamics. It is important to note that for variables such as the digital elevation model (DEM), slope, and aspect which remain constant within a defined study area and a single map was produced for each of them. After map generation, a comparison of the high and low values was conducted using their respective units. This method helped clearly illustrate the spatial and temporal trends in the study area.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Crop Classification Using Machine Learning based Algorithm\u003c/h2\u003e \u003cp\u003eIn a study examining the genetic structures of rice in Bangladesh, the predominant rice cultivation seasons identified are Aus, Amon, and Boro, each representing critical periods within the agricultural calendar of the region (Parsons et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Consequently, this study specifically focused on the major rice-growing seasons, namely Aus, Amon, and Boro, to ensure comprehensive spatiotemporal coverage of rice cultivation across different periods of the year.\u003c/p\u003e \u003cp\u003eThe classification of paddy or rice as the primary crop was undertaken using Sentinel-1 Synthetic Aperture Radar (SAR) data within the Google Earth Engine (GEE) platform. The temporal extent for image acquisition was carefully aligned with the rice transplanting seasons, wherein Sentinel-1 images were collected from June to August for Aus, August to October for Amon, and February to May for Boro (Alam et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor the classification process, Sentinel-1 SAR images were acquired in the Interferometric Wide Swath (IW) imaging mode using the VH polarization. The selection of VH polarization was based on its higher sensitivity to water, which separates rice fields from other land cover types (Nguyen et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Given the unique backscatter behavior of rice, the images were processed to generate multiple temporal stacks, which were subsequently combined to produce a composite image representing an annual overview of rice cultivation. This compositing approach allowed for the integration of multi-temporal SAR data, thereby improving classification accuracy by capturing the temporal dynamics of rice growth. The characteristic backscatter response of rice pixels in the composite imagery was observed in the form of cyan and magenta coloration, indicating the presence of rice fields due to their water retention properties and distinct phenological variations over time.\u003c/p\u003e \u003cp\u003eTo develop a robust classification model, approximately 350 samples were collected, representing three distinct land cover classes: rice yield, urban areas, and water bodies. These samples were digitized as polygons to ensure spatial accuracy and sufficient representation across diverse geographic regions. The dataset was subsequently divided into training and testing subsets, following an 80:20 ratio, where 80% of the samples were used to train the model, and the remaining 20% were reserved for accuracy assessment. A Random Forest (RF) algorithm was employed for classification due to its well-documented efficacy in handling high-dimensional remote sensing data, as well as its ability to mitigate overfitting by aggregating multiple decision trees. The RF classifier was trained using the selected input features derived from SAR backscatter values, and the resulting classification output was validated to assess its accuracy. Although three land cover classes: rice fields, urban areas, and water bodies were initially classified, only the rice yield raster was retained as the final output. Additionally, the urban and water body classes were excluded from the final map. Finally, the classified raster, delineating the spatial distribution of rice fields, was exported for further analysis and interpretation. This methodology ensured a robust and replicable approach to paddy classification using SAR-based remote sensing techniques.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 ANN Model for Rice Yield Prediction\u003c/h2\u003e \u003cp\u003eA Multi-Layer Perceptron (MLP), a type of ANN, was employed to develop a neural network sample capable of learning complex, non-linear relationships between the spatial variables and rice yield outcomes. To predict rice yield in Bangladesh, MOLUSCE plugin was used within QGIS 2.18.0, offering a robust framework for analyzing agricultural productivity (Kamaraj \u0026amp; Rangarajan, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The ANN model was trained using a 1-pixel neighborhood to capture yield patterns. A learning rate of 0.1 was applied to ensure efficient convergence, while a momentum of 0.05 helped stabilize weight updates. To allow the model to learn complex relationships between input variables and crop yield transitions, 25 hidden layers were used. The model utilized rice classification rasters from 2018 as the initial state and 2020 as the final state, providing temporal benchmarks for the prediction process. These were complemented by a set of spatial variables influencing rice yield, including Land Surface Temperature (LST), Precipitation, Evapotranspiration, Soil Moisture, Sunshine, Cloud Coverage, Digital Elevation Model (DEM), Slope, and Aspect, which jointly captured the environmental and topographical conditions critical to rice production.\u003c/p\u003e \u003cp\u003eThis was followed by the application of Cellular Automata Simulation (Sant\u0026eacute; et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) to project rice yield for 2022, incorporating spatial dynamics and temporal progression into the prediction. The predicted yield for 2022 was validated against actual rice classification data for the same year, enabling an accuracy assessment to confirm the model's reliability. Following successful validation, the model was extended to forecast rice yields for 2026 and 2030 through multiple iterations, refining the predictions and providing insights into future agricultural productivity under the specified conditions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Correlation Analysis for Variables\u003c/h2\u003e \u003cp\u003eTo assess the relationship between rice yield and various environmental and topographical factors, a comprehensive correlation analysis was conducted. The analysis aimed to identify the influence of key variables, including Land Surface Temperature (LST), Precipitation, Evapotranspiration, Soil Moisture, Sunshine, Cloud Coverage, Digital Elevation Model (DEM), Slope, and Aspect, on rice cultivation patterns.\u003c/p\u003e \u003cp\u003eThe workflow initiated with the importation of all raster datasets into ArcGIS software to underpin spatial analysis. A fishnet polygon grid was then constructed. This grid structure facilitated the orderly retrieval of spatial data points, ensuring even sampling throughout the study area. Once the fishnet was created, values from all raster layers corresponding to the selected variables were extracted at each grid point.\u003c/p\u003e \u003cp\u003eFollowing data extraction, the dataset was processed in Microsoft Excel for further refinement. The extracted values were meticulously filtered to remove any anomalies or inconsistencies, ensuring data integrity. Min-Max Normalization technique was applied to the dataset (Jayalakshmi \u0026amp; Santhakumaran, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), making it suitable for correlation analysis. The following equation was used for performing the normalization.\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{z}_{i}=\\frac{{x}_{i}-\\text{min}\\left(x\\right)}{\\text{max}\\left(x\\right)\\:-\\:\\text{m}\\text{i}\\text{n}\\left(x\\right)}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;.\u003c/p\u003e \u003cp\u003eThe correlation analysis was then performed to determine the statistical relationship between rice yield and the selected environmental and topographical factors. Pearson\u0026rsquo;s correlation coefficient was computed between rice yield and each of the variables to quantify the degree and direction of their association. LST and Precipitation were analyzed to assess their impact on rice growth, considering the thermal and hydrological conditions essential for crop development. Evapotranspiration and Soil Moisture were evaluated to understand their role in water availability and plant water stress, both of which significantly influence rice production. Sunshine and Cloud Coverage were examined to capture the effects of solar radiation and atmospheric conditions on photosynthesis and crop growth. DEM, Slope, and Aspect were incorporated into the analysis to account for terrain influences, which can affect water drainage, soil erosion, and overall land suitability for rice cultivation.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003ch2\u003e3.1 Delineation of Rice Yield Classification\u003c/h2\u003e\n\u003cp\u003eSpatial representations illustrate rice yield distribution across Bangladesh for the years 2018, 2020, and 2022. Corresponding cultivated areas were recorded as 519,318 hectares in 2018, 484,271 hectares in 2020, and 442,902.1 hectares in 2022. These data indicate a progressive decline in rice cultivation over the specified period.\u003c/p\u003e\n\u003ch2\u003e3.2 Spatiotemporal Delineation of All the Predisposing Variables\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eDEM, Slope \u0026amp; Aspect:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe DEM map shows elevation ranging from -34 m to 1052 m, with the highest elevations located in the Chittagong Hill Tracts in the southeast. The slope map depicts terrain steepness, with values from 0\u0026deg; to 63.83\u0026deg;. The aspect map indicates slope direction (0\u0026deg;\u0026ndash;360\u0026deg;), with varied orientations in the hilly southeast and little variation in the flat lowlands.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrecipitation:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe maps show the spatial distribution of precipitation (in mm/day) across Bangladesh for the years 2018, 2020, 2022 and 2024. The color gradient from pink to blue represents decreasing rainfall intensity, with pink indicating higher values.\u003c/p\u003e\n\u003cp\u003eIn 2018, the mean daily precipitation was 7.96 mm, showing moderate rainfall across the country. In 2020, the intensity increased notably, with a mean of 10.62 mm, especially concentrated in the northeastern and southeastern regions. By 2022, the average declined slightly to 9.49 mm/day, yet remained higher than in 2018. And, in 2024 the average rainfall was 7.08 which is the least among all these years.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eLST:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe series of maps illustrates the spatial distribution of Land Surface Temperature (LST) across Bangladesh for the years 2018, 2020, 2022 and 2024.\u003c/p\u003e\n\u003cp\u003eNumerically, the mean LST values were 27.67\u0026deg;C in 2018, 27.65\u0026deg;C in 2020, 28.49\u0026deg;C in 2022 and 26.57 in 2024. While there was a slight dip in the average LST between 2018 and 2020, a marked increase occurred by 2022 which fell again in 2024.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEvapotranspiration:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe series of maps in Figure 3.5 illustrate the spatial distribution of evapotranspiration (ET) across Bangladesh for the years 2018, 2020, 2022 and 2024, as derived from remote sensing data.\u003c/p\u003e\n\u003cp\u003eFor the years, 2018, 2020, 2022 and 2024 the average value of evapotranspiration was 184.16, 201.68, 187.90 and 186.40 respectively. It is clearly visible from the numbers that during those periods, evapotranspiration remained stable except for 2020 when it was the highest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSoil Moisture:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe maps in Figure 3.6 depict the spatial distribution of soil moisture across Bangladesh for the years 2018, 2020, 2022 and 2024 in centimeter unit.\u003c/p\u003e\n\u003cp\u003eThe mean soil moisture remained almost the same from 2018 to 2022 which was 0.3cm approximately. In 2024 it decreased slightly to 0.29cm. Overall, the soil moisture remained mostly stable during 2018 to 2024.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSunshine:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe spatial distribution maps of sunshine hours across Bangladesh for the years 2018, 2020, and 2022 (Figure 3.7) reveal a consistent pattern with regional variation. High sunshine duration is predominantly observed in the southeastern and southwestern coastal regions, while the central and northern parts exhibit relatively lower sunshine intensity. Over the years, a slight increase in sunshine duration is observed, as supported by the average daily sunshine data, rising from 5.83 hours in 2018 to 6.12 hours in 2024. Apart from that, there is insignificant variation in the middle years (2020 and 2022).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCloud Coverage:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe cloud coverage maps of Bangladesh for the years 2018, 2020, 2022 and 2024 (Figure 3.8) illustrate spatial and temporal variations in atmospheric cloud distribution. Higher cloud density is consistently concentrated in the southern coastal and southeastern hilly regions, while the central and northwestern areas generally experience lower coverage. Over the observed period, a slight increase in average cloud cover is noted from 1.19 octas in 2018 to 1.38 octas in 2020, followed by a marginal decrease to 1.35 octas in 2022 and remained same in 2024 as well.\u003c/p\u003e\n\u003ch2\u003e3.3 Correlation between Rice Yield and Pre-disposing variables\u003c/h2\u003e\n\u003cp\u003eThis heatmap presents the pairwise Pearson correlation coefficients between six climatic variables: Land Surface Temperature (LST), Precipitation, Soil Moisture, Evapotranspiration, Sunshine hours, and Cloud Cover, derived from data spanning the years (2018, 2020, 2022, 2024). The color gradient and numerical values illustrate the strength and direction of linear relationships between these variables. Notably, strong positive correlations are observed between LST and Sunshine, and between Precipitation and Cloud Cover, indicating co-variation. Conversely, a moderate negative correlation exists between Evapotranspiration and LST/Soil Moisture.\u003c/p\u003e\n\u003cp\u003eThis visualization facilitates the rapid identification of inter-variable dependencies and provides a basis for understanding potential climate change impacts by highlighting how shifts in one climatic factor may influence others within the system.\u003c/p\u003e\n\u003cp\u003eTable 3.1: Correlation between Rice Yield and Climatic Variables\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAgricultural Land\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClimatic Variables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003eEvapotranspiration\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e-0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003eSunshine\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003eCloud Coverage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003eLST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e-0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003ePrecipitation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003eSoil Moisture\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 312px;\"\u003e\n \u003cp\u003e-0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 3.1 presents the Pearson correlation coefficients between rice yield and six key climatic variables: Evapotranspiration, Sunny Days, Cloud Coverage, Land Surface Temperature (LST), Precipitation, and Soil Moisture.\u003c/p\u003e\n\u003cp\u003eThe table reveals that sunshine possesses the strongest positive correlation (r = 0.70) with rice yield, suggesting a substantial direct relationship where increased sunshine is associated with higher yield.\u003c/p\u003e\n\u003cp\u003ePrecipitation shows a moderate positive correlation (r = 0.26), indicating a positive but weak association with rice yield.\u003c/p\u003e\n\u003cp\u003eEvapotranspiration displays a moderate negative correlation (r = -0.32), implying an inverse relationship with rice yield.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCloud Coverage, LST, and Soil Moisture show negligible correlations (r = 0.05, -0.04, and -0.01, respectively), indicating their insignificant roles in rice yields.\u003c/p\u003e\n\u003cp\u003eOverall, the table highlights sunshine as a dominant factor influencing rice yield, while the effects of other climatic variables are comparatively weaker or negligible. \u0026nbsp;These correlations provide a quantitative basis for understanding the relationship between rice yield and climatic factors, offering insights for agricultural management and climate change adaptation strategies.\u003c/p\u003e\n\u003ch2\u003e3.4 Prediction of Food Security (Rice Yield) for 2026 and 2030\u003c/h2\u003e\n\u003cp\u003eTable 3.2: Prediction of Rice Yield for 2026 and 2030\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eYear\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 194px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 223px;\"\u003e\n \u003cp\u003eRice Cultivation Area (Ha)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2018\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 194px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 223px;\"\u003e\n \u003cp\u003e519318.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2020\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 194px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 223px;\"\u003e\n \u003cp\u003e484271.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 194px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 223px;\"\u003e\n \u003cp\u003e442902.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2026\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 194px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 223px;\"\u003e\n \u003cp\u003e421697.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2030\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 194px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 223px;\"\u003e\n \u003cp\u003e357145.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eTable 3.4 presents the predicted rice cultivation area (in hectares) for the years 2026 and 2030, alongside the recorded areas for 2018, 2020, and 2022. This table illustrates a projected decline in rice cultivation area over the coming decade.\u003c/p\u003e\n\u003cp\u003eThe data reveals a consistent decreasing trend from 519,318.2 Ha in 2018 to 357,145.4 Ha predicted for 2030. Specifically, the observed decline between 2018 and 2022 suggests an ongoing contraction of rice cultivation, which is further extrapolated to predict a continued reduction in 2026 and 2030.\u003c/p\u003e\n\u003ch2\u003e3.5 Accuracy Assessment\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eRice Classification Accuracy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe ROC curve (Figure 3.10) and the corresponding AUC of 0.968 demonstrate excellent discriminative ability of the rice classification model.\u003c/p\u003e\n\u003cp\u003eThis high AUC value indicates a strong likelihood (96.8%) that the model will correctly rank a randomly chosen rice pixel higher than a non-rice pixel. The curve\u0026apos;s shape, characterized by a rapid ascent to a high True Positive Rate at a low False Positive Rate, further supports the model\u0026apos;s robust performance in distinguishing between rice and non-rice areas within the dataset. These results suggest the model\u0026apos;s suitability for accurate rice cultivation mapping and subsequent spatial analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRice Prediction Accuracy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe predicted rice cultivation area for the year 2022, derived from ANN model in MOLUSCE, was validated against the independently classified rice map of 2022. The spatial agreement between the predicted and classified extents yielded an accuracy of 74%. This indicates a substantial agreement beyond that expected by chance, suggesting a moderate to high level of accuracy in the model\u0026apos;s predictive capability for rice cultivation extent.\u003c/p\u003e"},{"header":"4. Discussion \u0026 Conclusion","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Discussion\u003c/h2\u003e \u003cp\u003eOver the past three decades, global rice production has come under mounting pressure from climate variability, with each 1\u0026deg;C rise in mean growing‑season temperature linked to a\u0026thinsp;~\u0026thinsp;10% yield reduction (Zhao et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) and elevated night‑time temperatures alone causing similar losses (Peng et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). In Bangladesh, rice cultivation area contracted from 3.81% of the landscape in 2002 to 3.43% in 2021, mirroring broader trends in food insecurity. After two decades of decline, global hunger began to rise in the early 1990s, driven largely by surges in food prices; the 2008 financial crisis then dampened agricultural markets, reducing trade and market exchanges. Today, an estimated one billion people subsist on less than USD 1.25 per day, three‑quarters of whom depend directly on agriculture for their livelihoods that underscore the sector\u0026rsquo;s centrality to rural poverty alleviation and national economic development (FAO, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe spatiotemporal analysis reveals a marked contraction in rice cultivation area in Bangladesh, from 519,318 ha in 2018 to 442,902 ha in 2022, with predictive modeling indicating further declines to approximately 421,697 ha by 2026 and 357,145 ha by 2030. This ongoing reduction of a 32% loss over twelve years actually raises significant food‑security concerns for a nation where rice contributes to both caloric intake and rural livelihoods. Moreover, country‑specific simulations forecast a potential 33% decline in Bangladesh rice yields by 2100 under high‑emission trajectories (Karim et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). These converging lines of evidence underscore the urgency of both mitigation and adaptive responses to safeguard future rice production.\u003c/p\u003e \u003cp\u003eThe Pearson correlation analysis identified sunny days as the strongest climatic driver of rice yield (r\u0026thinsp;=\u0026thinsp;0.70), signifying the critical role of solar radiation in photosynthetic carbon assimilation and biomass accumulation. Empirical studies corroborate this: mean daily solar radiation during the crop growth period exhibits significant positive correlations with rice yield and harvest index (Wei et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2021\u003c/span\u003e); similarly, radiation interception and utilization metrics have been shown to explain a substantial proportion of yield variability in field trials (Huang et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Conversely, evapotranspiration demonstrated a moderate negative association with yield (r = \u0026minus;\u0026thinsp;0.32), suggesting that elevated evaporative demand may exacerbate plant‑water stress and reduce grain filling. Such inverse relationships between ET and rice productivity have been documented in both upland and paddy systems (Maina et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), highlighting the importance of optimized irrigation scheduling and water‑use efficiency interventions.\u003c/p\u003e \u003cp\u003eThe rice classification model achieved an AUC of 0.968, indicating robust discrimination between rice and non‑rice pixels. While ground verification in this study was necessarily limited to Google Earth Pro due to funding constraints, integrating dedicated field‑survey methods in future work would markedly enhance classification accuracy and model calibration. Building upon the remote‑sensing framework established here, survey‑based ground truthing could yield richer, higher‑resolution insights into rice heterogeneity and the environmental drivers identified. Furthermore, incorporating socio‑economic datasets such as farm management practices, market accessibility, and household livelihood indicators would provide a more holistic perspective on yield dynamics, enabling models that better account for both biophysical and human dimensions of food‑security change.\u003c/p\u003e \u003cp\u003eThe prediction of rice cultivation area through 2030 highlights a potentially severe contraction in arable rice land. These projections align with broader modeling by the International Food Policy Research Institute, which anticipates up to a 15% global decline in rice yield by mid‑century under unmitigated climate change scenarios (A. R. Md. T. Islam et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). To counteract these trends, Bangladesh has begun deploying adaptation measures most notably, the development and dissemination of heat‑tolerant rice varieties by IRRI in partnership with BRRI, which demonstrate that the newly developed rice varieties produce 1.5\u0026ndash;2 metric tons per hectare higher yields than conventional strains, even when exposed to elevated nighttime temperatures (IRRI, 2024). Additionally, farmers are adopting resilient agronomic practices such as alternate wetting and drying irrigation, diversified cropping patterns, and ecosystem‑based adaptation strategies (Li, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Continued expansion of such initiatives, coupled with enhanced monitoring and modeling, will be critical for sustaining rice output in a changing climate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Conclusion\u003c/h2\u003e \u003cp\u003eBy integrating remote sensing, GEE, and machine learning, this research evaluated and forecast how climate change will affect rice production and food security. To summarize, this work sought to: (1) understand the spatial variation of rice yields, (2) explain the primary climatic controls on rice production, and (3) anticipate how climate change will alter the areas where rice is cultivated.\u003c/p\u003e \u003cp\u003eAccording to study, rice cultivation in the region has experienced a notable decline over recent years, shrinking considerably between 2018 and 2024. This trend highlights a significant shift in agricultural land use. Among the various climatic factors examined, the amount of sunshine emerged as the most significant, showing a strong positive relationship with rice yield. On the contrary, evapotranspiration exhibited a moderate negative association, suggesting it can hinder yield. Other climatic elements appeared to have only minor connections to rice production.\u003c/p\u003e \u003cp\u003eThe effectiveness of the rice-classification model in predicting outcomes was high, and its forecasts for 2022 aligned well with actual observations. Looking ahead, projections indicate a continued decrease in rice cultivation area through 2026 and further by 2030. These findings underscore the critical need for adaptive management strategies to address the evolving landscape of rice farming.\u003c/p\u003e \u003cp\u003eThe clear decline in cultivation area, strong dependence on solar radiation, and projected contractions by 2030 underscore the multi‑faceted threats to food security in Bangladesh. While the classification model exhibits excellent discrimination, further refinement through higher‑resolution and more frequent satellite data could improve mapping accuracy.\u003c/p\u003e \u003cp\u003eThis research offers a valuable tool for policymakers in Bangladesh, enabling the proactive utilization of early mapping and predictive modeling to mitigate the adverse impacts of climatic hazards. Consequently, it is anticipated that this study will foster an increased awareness of the imperative to transition towards contemporary agricultural technologies, such as precision agriculture. While the significance of climatic variables in shaping agricultural output is acknowledged, it is crucial to recognize the substantial influence of other intrinsic factors, including soil quality, cultivation practices, and fertilizer efficacy. Future scholarly investigations could fruitfully integrate these elements to provide a more holistic understanding.\u003c/p\u003e \u003cp\u003eFurthermore, subsequent research endeavors should consider incorporating socio-economic data and management variables to provide a more nuanced perspective. Exploring causal modeling frameworks would enable a deeper understanding of the complex interrelationships at play. Finally, evaluating the effectiveness of emergent adaptation strategies, encompassing precision irrigation, precision agriculture methodologies, and the deployment of genetically modified crop species, warrants rigorous investigation to inform evidence-based policy decisions. Policymakers and stakeholders should prioritize investments in climate‑smart agriculture, targeted breeding programs, and real‑time monitoring systems to mitigate yield losses and ensure the resilience of Bangladesh\u0026rsquo;s rice sector.\u003c/p\u003e \u003c/div\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAhmad, J., Alam, D., \u0026amp; Haseen, M. S. (2011). Impact of climate change on agriculture and food security in India. \u003cem\u003eInternational Journal of Agriculture, Environment and Biotechnology\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(2), 129\u0026ndash;137.\u003c/li\u003e\n\u003cli\u003eAlam, M. J., Al-Mahmud, A.-, Islam, M. A., Hossain, M. F., Ali, M. A., Dessoky, E. S., El-Hallous, E. I., Hassan, M. M., Begum, N., \u0026amp; Hossain, A. (2021). Crop Diversification in Rice\u0026mdash;Based Cropping Systems Improves the System Productivity, Profitability and Sustainability. \u003cem\u003eSustainability\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(11), Article 11. https://doi.org/10.3390/su13116288\u003c/li\u003e\n\u003cli\u003eAnderson, R., Bayer, P. E., \u0026amp; Edwards, D. (2020). Climate change and the need for agricultural adaptation. \u003cem\u003eCurrent Opinion in Plant Biology\u003c/em\u003e, \u003cem\u003e56\u003c/em\u003e, 197\u0026ndash;202. https://doi.org/10.1016/j.pbi.2019.12.006\u003c/li\u003e\n\u003cli\u003eBandara, J. S., \u0026amp; Cai, Y. (2014). The impact of climate change on food crop productivity, food prices and food security in South Asia. \u003cem\u003eEconomic Analysis and Policy\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(4), 451\u0026ndash;465.\u003c/li\u003e\n\u003cli\u003eBocchiola, D., Brunetti, L., Soncini, A., Polinelli, F., \u0026amp; Gianinetto, M. (2019). Impact of climate change on agricultural productivity and food security in the Himalayas: A case study in Nepal. \u003cem\u003eAgricultural Systems\u003c/em\u003e, \u003cem\u003e171\u003c/em\u003e, 113\u0026ndash;125. https://doi.org/10.1016/j.agsy.2019.01.008\u003c/li\u003e\n\u003cli\u003eCao, J., Zhang, Z., Tao, F., Zhang, L., Luo, Y., Zhang, J., Han, J., \u0026amp; Xie, J. (2021). Integrating Multi-Source Data for Rice Yield Prediction across China using Machine Learning and Deep Learning Approaches. \u003cem\u003eAgricultural and Forest Meteorology\u003c/em\u003e, \u003cem\u003e297\u003c/em\u003e, 108275. https://doi.org/10.1016/j.agrformet.2020.108275\u003c/li\u003e\n\u003cli\u003eCheng, E., Zhang, B., Peng, D., Zhong, L., Yu, L., Liu, Y., Xiao, C., Li, C., Li, X., Chen, Y., Ye, H., Wang, H., Yu, R., Hu, J., \u0026amp; Yang, S. (2022). Wheat yield estimation using remote sensing data based on machine learning approaches. \u003cem\u003eFrontiers in Plant Science\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e, 1090970. https://doi.org/10.3389/fpls.2022.1090970\u003c/li\u003e\n\u003cli\u003eDelgado, J. A., Short, N. M., Roberts, D. P., \u0026amp; Vandenberg, B. (2019). Big Data Analysis for Sustainable Agriculture on a Geospatial Cloud Framework. \u003cem\u003eFrontiers in Sustainable Food Systems\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e, 54. https://doi.org/10.3389/fsufs.2019.00054\u003c/li\u003e\n\u003cli\u003eFahad, S., Adnan, M., Saud, S., \u0026amp; Nie, L. (Eds.). (2023). \u003cem\u003eClimate change and ecosystems: Challenges to sustainable development\u003c/em\u003e (First edition). CRC Press.\u003c/li\u003e\n\u003cli\u003eFAO. (2005). \u003cem\u003eAgricultural trade and poverty: Can trade work for the poor?\u003c/em\u003e Food and Agriculture Organization of the United Nations.\u003c/li\u003e\n\u003cli\u003eFAO. (2020). \u003cem\u003eThe State of Agricultural Commodity Markets 2020\u003c/em\u003e. FAO ; https://openknowledge.fao.org/handle/20.500.14283/cb0665en\u003c/li\u003e\n\u003cli\u003eFoley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., Chapin, F. S., Coe, M. T., Daily, G. C., Gibbs, H. K., Helkowski, J. H., Holloway, T., Howard, E. A., Kucharik, C. J., Monfreda, C., Patz, J. A., Prentice, I. C., Ramankutty, N., \u0026amp; Snyder, P. K. (2005). Global Consequences of Land Use. \u003cem\u003eScience\u003c/em\u003e, \u003cem\u003e309\u003c/em\u003e(5734), 570\u0026ndash;574. https://doi.org/10.1126/science.1111772\u003c/li\u003e\n\u003cli\u003eGao, B.-C. (1996). NDWI\u0026mdash;A normalized difference water index for remote sensing of vegetation liquid water from space. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(3), 257\u0026ndash;266.\u003c/li\u003e\n\u003cli\u003eGiri, C. P. (2012). \u003cem\u003eRemote sensing of land use and land cover: Principles and applications\u003c/em\u003e. CRC press.\u003c/li\u003e\n\u003cli\u003eGorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., \u0026amp; Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e202\u003c/em\u003e, 18\u0026ndash;27. https://doi.org/10.1016/j.rse.2017.06.031\u003c/li\u003e\n\u003cli\u003eHansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A., Tyukavina, A., Thau, D., Stehman, S. V., Goetz, S. J., Loveland, T. R., Kommareddy, A., Egorov, A., Chini, L., Justice, C. O., \u0026amp; Townshend, J. R. G. (2013). High-Resolution Global Maps of 21st-Century Forest Cover Change. \u003cem\u003eScience\u003c/em\u003e, \u003cem\u003e342\u003c/em\u003e(6160), 850\u0026ndash;853. https://doi.org/10.1126/science.1244693\u003c/li\u003e\n\u003cli\u003eHerold, M., Woodcock, C. E., Antonio Di Gregorio, Mayaux, P., Belward, A. S., Latham, J., \u0026amp; Schmullius, C. C. (2006). A joint initiative for harmonization and validation of land cover datasets. \u003cem\u003eIEEE Transactions on Geoscience and Remote Sensing\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(7), 1719\u0026ndash;1727. https://doi.org/10.1109/TGRS.2006.871219\u003c/li\u003e\n\u003cli\u003eHuang, M., Shan, S., Cao, F., \u0026amp; Zou, Y. (2016). The solar radiation-related determinants of rice yield variation across a wide range of regions. \u003cem\u003eNJAS: Wageningen Journal of Life Sciences\u003c/em\u003e, \u003cem\u003e78\u003c/em\u003e(1), 123\u0026ndash;128. https://doi.org/10.1016/j.njas.2016.05.004\u003c/li\u003e\n\u003cli\u003eHuete, A., Didan, K., Miura, T., Rodriguez, E. P., Gao, X., \u0026amp; Ferreira, L. G. (2002). Overview of the radiometric and biophysical performance of the MODIS vegetation indices. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e83\u003c/em\u003e(1\u0026ndash;2), 195\u0026ndash;213.\u003c/li\u003e\n\u003cli\u003eHuete, A. R. (1988). A soil-adjusted vegetation index (SAVI). \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(3), 295\u0026ndash;309.\u003c/li\u003e\n\u003cli\u003eHuete, A. R., Liu, H. Q., Batchily, K. V., \u0026amp; Van Leeuwen, W. (1997). A comparison of vegetation indices over a global set of TM images for EOS-MODIS. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(3), 440\u0026ndash;451.\u003c/li\u003e\n\u003cli\u003eHussain, M., Zeeshan, M., Mahmood, M. H., Farwa, F., Siddiqa, S., Ahmad, Hussain, Gill, A., \u0026amp; Nasir, A. (2025). Advanced Applications of AquaCrop for Field Management and Climate Impact Assessment. \u003cem\u003eAsian Journal of Research in Crop Science\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e, 1\u0026ndash;17. https://doi.org/10.9734/ajrcs/2025/v10i1327\u003c/li\u003e\n\u003cli\u003eIRRI. (2024, July 2). \u003cem\u003eHeat-Tolerant Rice Varieties Promises Increased Yields in Bangladesh\u003c/em\u003e. International Rice Research Institute. https://www.irri.org/news-and-events/news/heat-tolerant-rice-varieties-promises-increased-yields-bangladesh\u003c/li\u003e\n\u003cli\u003eIslam, A. R. Md. T., Shen, S., Yang, S., Hu, Z., \u0026amp; Atiqur Rahman, M. (2020). Spatiotemporal rice yield variations and potential agro-adaptation strategies in Bangladesh: A biophysical modeling approach. \u003cem\u003eSustainable Production and Consumption\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e, 121\u0026ndash;138. https://doi.org/10.1016/j.spc.2020.07.005\u003c/li\u003e\n\u003cli\u003eIslam, Md. N., Tamanna, S., van Amstel, A., Noman, Md., Ali, Md. S. S., Aparajita, D. M., Roy, P., Tanha, S. R., Sarkar, N., Ashiquzzaman, Md., Ghosh, S. K., Hasnat, S. R., Bhuiyan, Md. F., Sayef, Md., Saha, P., Hasan, G. R., \u0026amp; Ghosh, A. (2021). Climate Change Impact and Comprehensive Disaster Management Approach in Bangladesh: A Review. In Md. N. Islam \u0026amp; A. van Amstel (Eds.), \u003cem\u003eBangladesh II: Climate Change Impacts, Mitigation and Adaptation in Developing Countries\u003c/em\u003e (pp. 1\u0026ndash;39). Springer International Publishing. https://doi.org/10.1007/978-3-030-71950-0_1\u003c/li\u003e\n\u003cli\u003eJayalakshmi, T., \u0026amp; Santhakumaran, A. (2011). Statistical normalization and back propagation for classification. \u003cem\u003eInternational Journal of Computer Theory and Engineering\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1), 1793\u0026ndash;8201.\u003c/li\u003e\n\u003cli\u003eJiang, P., \u0026amp; Thelen, K. D. (2004). Effect of Soil and Topographic Properties on Crop Yield in a North‐Central Corn\u0026ndash;Soybean Cropping System. \u003cem\u003eAgronomy Journal\u003c/em\u003e, \u003cem\u003e96\u003c/em\u003e(1), 252\u0026ndash;258. https://doi.org/10.2134/agronj2004.0252\u003c/li\u003e\n\u003cli\u003eJordan, M. I., \u0026amp; Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. \u003cem\u003eScience\u003c/em\u003e, \u003cem\u003e349\u003c/em\u003e(6245), 255\u0026ndash;260. https://doi.org/10.1126/science.aaa8415\u003c/li\u003e\n\u003cli\u003eJustice, C. O., Townshend, J. R. G., Holben, B. N., \u0026amp; Tucker, C. J. (1985). Analysis of the phenology of global vegetation using meteorological satellite data. \u003cem\u003eInternational Journal of Remote Sensing\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(8), 1271\u0026ndash;1318. https://doi.org/10.1080/01431168508948281\u003c/li\u003e\n\u003cli\u003eKamaraj, M., \u0026amp; Rangarajan, S. (2022). Predicting the future land use and land cover changes for Bhavani basin, Tamil Nadu, India, using QGIS MOLUSCE plugin. \u003cem\u003eEnvironmental Science and Pollution Research\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(57), 86337\u0026ndash;86348. https://doi.org/10.1007/s11356-021-17904-6\u003c/li\u003e\n\u003cli\u003eKamilaris, A., \u0026amp; Prenafeta-Bold\u0026uacute;, F. X. (2018). Deep learning in agriculture: A survey. \u003cem\u003eComputers and Electronics in Agriculture\u003c/em\u003e, \u003cem\u003e147\u003c/em\u003e, 70\u0026ndash;90.\u003c/li\u003e\n\u003cli\u003eKamruzzaman, M., Shozib, H. B., Kader, Md. A., Iftekharuddaula, K. M., Rahman, M. A., Lipi, L. F., Khan, Md. A. R., \u0026amp; Kabir, Md. S. (2025). Progress of healthier rice development in Bangladesh: A review. \u003cem\u003eJournal of Food Composition and Analysis\u003c/em\u003e, \u003cem\u003e139\u003c/em\u003e, 107082. https://doi.org/10.1016/j.jfca.2024.107082\u003c/li\u003e\n\u003cli\u003eKarim, M. R., Ishikawa, M., Ikeda, M., \u0026amp; Islam, Md. T. (2012). Climate change model predicts 33 % rice yield decrease in 2100 in Bangladesh. \u003cem\u003eAgronomy for Sustainable Development\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(4), 821\u0026ndash;830. https://doi.org/10.1007/s13593-012-0096-7\u003c/li\u003e\n\u003cli\u003eKazemi Garajeh, M., Salmani, B., Zare Naghadehi, S., Valipoori Goodarzi, H., \u0026amp; Khasraei, A. (2023). An integrated approach of remote sensing and geospatial analysis for modeling and predicting the impacts of climate change on food security. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(1), 1057. https://doi.org/10.1038/s41598-023-28244-5\u003c/li\u003e\n\u003cli\u003eKhanian, M., Marshall, N., Zakerhaghighi, K., Salimi, M., \u0026amp; Naghdi, A. (2018). Transforming agriculture to climate change in Famenin County, West Iran through a focus on environmental, economic and social factors. \u003cem\u003eWeather and Climate Extremes\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e, 52\u0026ndash;64. https://doi.org/10.1016/j.wace.2018.05.006\u003c/li\u003e\n\u003cli\u003eKumawat, R. K., Tiwari, G., Ramakrishnan, R. S., Bhayal, D., Debnath, S., Thakur, S., \u0026amp; Bhayal, L. (2023). Remote Sensing Related Tools and their Spectral Indices Applications for Crop Management in Precision Agriculture. \u003cem\u003eInternational Journal of Environment and Climate Change\u003c/em\u003e, 171\u0026ndash;188. https://doi.org/10.9734/ijecc/2023/v13i11665\u003c/li\u003e\n\u003cli\u003eLambin, E. F., Geist, H. J., \u0026amp; Lepers, E. (2003). Dynamics of land-use and land-cover change in tropical regions. \u003cem\u003eAnnual Review of Environment and Resources\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(1), 205\u0026ndash;241.\u003c/li\u003e\n\u003cli\u003eLi, M. (2023). Adaptation to expected and unexpected weather fluctuations: Evidence from Bangladeshi smallholder farmers. \u003cem\u003eWorld Development\u003c/em\u003e, \u003cem\u003e161\u003c/em\u003e, 106066. https://doi.org/10.1016/j.worlddev.2022.106066\u003c/li\u003e\n\u003cli\u003eLiakos, K. G., Busato, P., Moshou, D., Pearson, S., \u0026amp; Bochtis, D. (2018). Machine learning in agriculture: A review. \u003cem\u003eSensors\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(8), 2674.\u003c/li\u003e\n\u003cli\u003eLobell, D. B., \u0026amp; Asseng, S. (2017). Comparing estimates of climate change impacts from process-based and statistical crop models. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(1), 015001.\u003c/li\u003e\n\u003cli\u003eMagrin, G. O., Marengo, J. A., Boulanger, J. P., Buckeridge, M. S., Castellanos L\u0026oacute;pez, E. J., Poveda, G., Scarano, F. R., Vicu\u0026ntilde;a, S., Alfaro Mart\u0026iacute;nez, E. J., \u0026amp; Anthelme, F. (2014). \u003cem\u003eCentral and south america\u003c/em\u003e. https://www.kerwa.ucr.ac.cr/handle/10669/89608\u003c/li\u003e\n\u003cli\u003eMaina, M., Amin, M., Rowshon, M., Wayayok, A., Abd Aziz, S., \u0026amp; Yazid, M. A. (2014). Effects of crop evapotranspiration estimation techniques and weather parameters on rice crop water requirement. \u003cem\u003eAustralian Journal of Crop Science\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e, 495\u0026ndash;501.\u003c/li\u003e\n\u003cli\u003eMajumder, A., Kingra, P. K., Setia, R., Singh, S. P., \u0026amp; Pateriya, B. (2020). Influence of land use/land cover changes on surface temperature and its effect on crop yield in different agro-climatic regions of Indian Punjab. \u003cem\u003eGeocarto International\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(6), 663\u0026ndash;686. https://doi.org/10.1080/10106049.2018.1520927\u003c/li\u003e\n\u003cli\u003eM\u0026uuml;ller, B., Hoffmann, F., Heckelei, T., M\u0026uuml;ller, C., Hertel, T. W., Polhill, J. G., Van Wijk, M., Achterbosch, T., Alexander, P., \u0026amp; Brown, C. (2020). Modelling food security: Bridging the gap between the micro and the macro scale. \u003cem\u003eGlobal Environmental Change\u003c/em\u003e, \u003cem\u003e63\u003c/em\u003e, 102085.\u003c/li\u003e\n\u003cli\u003eMwongera, C., Shikuku, K. M., Twyman, J., L\u0026auml;derach, P., Ampaire, E., Van Asten, P., Twomlow, S., \u0026amp; Winowiecki, L. A. (2017). Climate smart agriculture rapid appraisal (CSA-RA): A tool for prioritizing context-specific climate smart agriculture technologies. \u003cem\u003eAgricultural Systems\u003c/em\u003e, \u003cem\u003e151\u003c/em\u003e, 192\u0026ndash;203.\u003c/li\u003e\n\u003cli\u003eNguyen, D. B., Gruber ,Alexander, \u0026amp; and Wagner, W. (2016). Mapping rice extent and cropping scheme in the Mekong Delta using Sentinel-1A data. \u003cem\u003eRemote Sensing Letters\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(12), 1209\u0026ndash;1218. https://doi.org/10.1080/2150704X.2016.1225172\u003c/li\u003e\n\u003cli\u003eNjenga, N. W. (2013). \u003cem\u003eVulnerability to Climate Change and Adaptation Strategies for Small Scale Farmers in North Kinangop Location,Nyandarua South District,Kenya\u003c/em\u003e. http://ir-library.ku.ac.ke/handle/123456789/6545\u003c/li\u003e\n\u003cli\u003ePang, J., Zhang, R., Yu, B., Liao, M., Lv, J., Xie, L., Li, S., \u0026amp; Zhan, J. (2021). Pixel-level rice planting information monitoring in Fujin City based on time-series SAR imagery. \u003cem\u003eInternational Journal of Applied Earth Observation and Geoinformation\u003c/em\u003e, \u003cem\u003e104\u003c/em\u003e, 102551. https://doi.org/10.1016/j.jag.2021.102551\u003c/li\u003e\n\u003cli\u003eParsons, B. J., Newbury, H. J., Jackson, M. T., \u0026amp; Ford-Lloyd, B. V. (1999). The genetic structure and conservation of aus, aman and boro rices from Bangladesh. \u003cem\u003eGenetic Resources and Crop Evolution\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(6), 587\u0026ndash;598. https://doi.org/10.1023/A:1008749532171\u003c/li\u003e\n\u003cli\u003ePeng, S., Huang, J., Sheehy, J. E., Laza, R. C., Visperas, R. M., Zhong, X., Centeno, G. S., Khush, G. S., \u0026amp; Cassman, K. G. (2004). Rice yields decline with higher night temperature from global warming. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e101\u003c/em\u003e(27), 9971\u0026ndash;9975. https://doi.org/10.1073/pnas.0403720101\u003c/li\u003e\n\u003cli\u003eRoy, D. P., Wulder, M. A., Loveland, T. R., C.E., W., Allen, R. G., Anderson, M. C., Helder, D., Irons, J. R., Johnson, D. M., Kennedy, R., Scambos, T. A., Schaaf, C. B., Schott, J. R., Sheng, Y., Vermote, E. F., Belward, A. S., Bindschadler, R., Cohen, W. B., Gao, F., \u0026hellip; Zhu, Z. (2014). Landsat-8: Science and product vision for terrestrial global change research. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e145\u003c/em\u003e, 154\u0026ndash;172. https://doi.org/10.1016/j.rse.2014.02.001\u003c/li\u003e\n\u003cli\u003eRudiyanto, Minasny, B., Shah, R., Che Soh, N., Arif, C., \u0026amp; Indra Setiawan, B. (2019). Automated Near-Real-Time Mapping and Monitoring of Rice Extent, Cropping Patterns, and Growth Stages in Southeast Asia Using Sentinel-1 Time Series on a Google Earth Engine Platform. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(14), 1666. https://doi.org/10.3390/rs11141666\u003c/li\u003e\n\u003cli\u003eSaini, N. (2023). Climate Change: Its Impact on Equity, Diversity, and Livelihood of the Rural Population. In S. Bala \u0026amp; P. Singhal (Eds.), \u003cem\u003eAdvances in Human Resources Management and Organizational Development\u003c/em\u003e (pp. 168\u0026ndash;193). IGI Global. https://doi.org/10.4018/978-1-6684-6878-4.ch011\u003c/li\u003e\n\u003cli\u003eSaleem, M. H., Potgieter, J., \u0026amp; Arif, K. M. (2021). Automation in Agriculture by Machine and Deep Learning Techniques: A Review of Recent Developments. \u003cem\u003ePrecision Agriculture\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(6), 2053\u0026ndash;2091. https://doi.org/10.1007/s11119-021-09806-x\u003c/li\u003e\n\u003cli\u003eSant\u0026eacute;, I., Garc\u0026iacute;a, A. M., Miranda, D., \u0026amp; Crecente, R. (2010). Cellular automata models for the simulation of real-world urban processes: A review and analysis. \u003cem\u003eLandscape and Urban Planning\u003c/em\u003e, \u003cem\u003e96\u003c/em\u003e(2), 108\u0026ndash;122. https://doi.org/10.1016/j.landurbplan.2010.03.001\u003c/li\u003e\n\u003cli\u003eSingh, M., Karada, M. S., Rai, R. K., Pratap, D., Agnihotri, D., Singh, A. K., \u0026amp; Singh, B. K. (2023). A Review on Remote Sensing as a Tool for Irrigation Monitoring and Management. \u003cem\u003eInternational Journal of Environment and Climate Change\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(6), 203\u0026ndash;211. https://doi.org/10.9734/ijecc/2023/v13i61817\u003c/li\u003e\n\u003cli\u003eSishodia, R. P., Ray, R. L., \u0026amp; Singh, S. K. (2020). Applications of remote sensing in precision agriculture: A review. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(19), 3136.\u003c/li\u003e\n\u003cli\u003eSong, L., \u0026amp; Jin, J. (2020). Effects of Sunshine Hours and Daily Maximum Temperature Declines and Cultivar Replacements on Maize Growth and Yields. \u003cem\u003eAgronomy\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(12), 1862. https://doi.org/10.3390/agronomy10121862\u003c/li\u003e\n\u003cli\u003eTilman, D., Balzer, C., Hill, J., \u0026amp; Befort, B. L. (2011). Global food demand and the sustainable intensification of agriculture. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e108\u003c/em\u003e(50), 20260\u0026ndash;20264. https://doi.org/10.1073/pnas.1116437108\u003c/li\u003e\n\u003cli\u003eViana, C. M., Freire, D., Abrantes, P., Rocha, J., \u0026amp; Pereira, P. (2022). Agricultural land systems importance for supporting food security and sustainable development goals: A systematic review. \u003cem\u003eScience of The Total Environment\u003c/em\u003e, \u003cem\u003e806\u003c/em\u003e, 150718. https://doi.org/10.1016/j.scitotenv.2021.150718\u003c/li\u003e\n\u003cli\u003eViana, C. M., \u0026amp; Rocha, J. (2020). Evaluating Dominant Land Use/Land Cover Changes and Predicting Future Scenario in a Rural Region Using a Memoryless Stochastic Method. \u003cem\u003eSustainability\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(10), Article 10. https://doi.org/10.3390/su12104332\u003c/li\u003e\n\u003cli\u003eVos, R., Jackson, J., James, S., \u0026amp; S\u0026aacute;nchez, M. V. (2020). \u003cem\u003eRefugees and Conflict-Affected People: Integrating Displaced Communities into Food Systems\u003c/em\u003e (0 ed.). International Food Policy Research Institute. https://doi.org/10.2499/9780896293670_05\u003c/li\u003e\n\u003cli\u003eWang, X. (2023). Remote Sensing Applications to Climate Change. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(3), 747. https://doi.org/10.3390/rs15030747\u003c/li\u003e\n\u003cli\u003eWei, S., Peng, A., Liu, J., Wei, D., \u0026amp; Chen, C. (2021). Asymmetric Differences in the Effects of Average Air Temperature and Solar Radiation on Early Rice and Late Rice Yield. \u003cem\u003eAtmosphere\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(12), 1541. https://doi.org/10.3390/atmos12121541\u003c/li\u003e\n\u003cli\u003eWeiss, M., Jacob, F., \u0026amp; Duveiller, G. (2020). Remote sensing for agricultural applications: A meta-review. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e236\u003c/em\u003e, 111402. https://doi.org/10.1016/j.rse.2019.111402\u003c/li\u003e\n\u003cli\u003eWesterveld, J. J. L., van den Homberg, M. J. C., Nobre, G. G., van den Berg, D. L. J., Teklesadik, A. D., \u0026amp; Stuit, S. M. (2021). Forecasting transitions in the state of food security with machine learning using transferable features. \u003cem\u003eScience of The Total Environment\u003c/em\u003e, \u003cem\u003e786\u003c/em\u003e, 147366. https://doi.org/10.1016/j.scitotenv.2021.147366\u003c/li\u003e\n\u003cli\u003eXu, H. (2006). Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. \u003cem\u003eInternational Journal of Remote Sensing\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(14), 3025\u0026ndash;3033. https://doi.org/10.1080/01431160600589179\u003c/li\u003e\n\u003cli\u003eZhang, M., Lin, H., Wang, G., Sun, H., \u0026amp; Fu, J. (2018). Mapping Paddy Rice Using a Convolutional Neural Network (CNN) with Landsat 8 Datasets in the Dongting Lake Area, China. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(11), 1840. https://doi.org/10.3390/rs10111840\u003c/li\u003e\n\u003cli\u003eZhang, P., Li, L., Fu, Q., Du, C., Yang, A., Sun, N., Wang, L., \u0026amp; Li, M. (2025). Balancing soil carbon emissions and productivity in maize agroecosystems through nitrogen, biochar, and straw regulation. \u003cem\u003eIndustrial Crops and Products\u003c/em\u003e, \u003cem\u003e224\u003c/em\u003e, 120442. https://doi.org/10.1016/j.indcrop.2024.120442\u003c/li\u003e\n\u003cli\u003eZhao, C., Liu, B., Piao, S., Wang, X., Lobell, D. B., Huang, Y., Huang, M., Yao, Y., Bassu, S., Ciais, P., Durand, J.-L., Elliott, J., Ewert, F., Janssens, I. A., Li, T., Lin, E., Liu, Q., Martre, P., M\u0026uuml;ller, C., \u0026hellip; Asseng, S. (2017). Temperature increase reduces global yields of major crops in four independent estimates. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e114\u003c/em\u003e(35), 9326\u0026ndash;9331. https://doi.org/10.1073/pnas.1701762114\u003c/li\u003e\n\u003cli\u003eZulfiqar, F., Navarro, M., Ashraf, M., Akram, N. A., \u0026amp; Munn\u0026eacute;-Bosch, S. (2019). Nanofertilizer use for sustainable agriculture: Advantages and limitations. \u003cem\u003ePlant Science\u003c/em\u003e, \u003cem\u003e289\u003c/em\u003e, 110270. https://doi.org/10.1016/j.plantsci.2019.110270\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Khulna University of Engineering and Technology","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":"Food Security, Artificial Neural Network, Climate Change, Crop Mapping","lastPublishedDoi":"10.21203/rs.3.rs-6962219/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6962219/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe agriculture sector is responsible for the majority of food production in Bangladesh. However, rapid urbanization and various anthropogenic factors have accelerated the rate of climate change, posing a significant threat to food security. This study utilizes a remote sensing-driven methodology to assess the potential impacts of climate change on food security in Bangladesh, with a specific focus on rice production. High-resolution Sentinel-1 imagery was used within the Google Earth Engine (GEE) platform to classify rice yield patterns, focusing on major growing seasons (Aman, Aus, Boro) for the years 2018, 2020, and 2022. For classification, the Random Forest algorithm was employed due to its high precision and reliability. Subsequently, an Artificial Neural Network model (Multi-Layer Perceptron) was used within MOLUSCE to predict future yield dynamics for the years 2026 and 2030. Among the climatic variables, precipitation, evapotranspiration, soil moisture, sunshine duration, and cloud cover were integrated with three topographic variables: DEM, slope, and aspect, to assess their influence on rice productivity. The rice yield classification achieved a high degree of precision (AUC\u0026thinsp;=\u0026thinsp;0.968). The analysis reveals a significant decline in rice cultivation area, from 519,318 hectares in 2018 to 442,902 hectares in 2022, with projected reductions to 421,697 hectares by 2026 and 357,145 hectares by 2030. Correlation analysis indicated a strong positive association between rice yield and sunshine (r\u0026thinsp;=\u0026thinsp;0.70), a weaker positive correlation with precipitation (r\u0026thinsp;=\u0026thinsp;0.26), and a moderate negative relationship with evapotranspiration (r = -0.32), while the remaining variables showed insignificant correlations. This study highlights the increasing vulnerability of rice production to climate change and emphasizes the need for acknowledging these effects. The developed method can contribute to improved crop mapping and early prediction of food security situations in the South Asian region.\u003c/p\u003e","manuscriptTitle":"Integrating Remote Sensing and Machine Learning to Assess Climate‑Driven Yield Dynamics and Food Security in Bangladesh","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-25 04:21:52","doi":"10.21203/rs.3.rs-6962219/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":"c32c48c3-d84d-47cf-8934-0e1ec681ef0a","owner":[],"postedDate":"June 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":50498187,"name":"Geographic Information Systems"}],"tags":[],"updatedAt":"2025-06-25T04:21:52+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-25 04:21:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6962219","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6962219","identity":"rs-6962219","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}