https://ar.iub.edu.bd/handle/11348/486 DESM 2026-09-10T13:52:59Z https://ar.iub.edu.bd/handle/11348/1566 Delineation of fresh groundwater zone in the coastal aquifers of Bangladesh: a multi-modal approach by remote sensing and machine learning algorithm attribution Shuvo, Ragib Mahmood Groundwater Salinity Intrusion in the coastal belts of Bangladesh poses a severe threat to the livability and livelihood of more than 20 million populations due to scarcity of freshwater. Climate change and anthropogenic activities is further exacerbating the salinity intrusion risks across these regions. This research aims to delineate the fresh groundwater potential zones in the six districts of the southwestern part of the coastal areas of Bangladesh with the help of remote sensing and machine learning approaches. 15 parameters were primarily selected to prepare the spatial layers which represent the topographic, meteorological, hydro-chemical and land use characteristics of the study area. Among these, 10 parameters were finally retained by multiple iterations of the Variable Inflation Factor (VIF) analysis at a threshold of 10 in order to exclude multicollinearity of the parameters. 117 groundwater samples for wet and dry season were collected from Bangladesh Water Development Board (BWDB) to classify the training and testing dataset. Synthetic Minority Oversampling Technique (SMOTE) was applied to the training dataset in order to address the imbalance of fresh and saline groundwater sample size. Five Machine Learning algorithms namely, Random Forest (RF), Gradient Boosting Machine (GBM), Extreme Gradient Boosting (xgboost), Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) were applied to delineate the fresh groundwater potential zones. The accuracy of the ML models was validated by using ROC/AUC, Precision-Recall, Sensitivity, Specificity, F-1 Score, Kappa accuracy. RF and xgboost delineated the potential zones with the most accuracy, having an overall accuracy of 95.83%, AUC of 0.99, an average precision of 0.917, and zero false negatives on the testing dataset. The potential zones were divided into five classes and approximately 8-10% of the study area were delineated as very high potential zone for fresh groundwater with majority of the study area falling under very low potential areas. The inability to delineate similar outputs by other ML models highlighted the limitations in collecting, managing and storing spatially distributed groundwater quality data across the region and the severe heterogeneity of fresh and non-fresh groundwater quality within the sample dataset. The findings provide an emphasis on combining hydro-chemical data in groundwater potential zone mapping, especially in the coastal areas where further increase in climate change is going to shift the dynamics of freshwater resources. Therefore, similar research can drive anticipatory actions to conserve groundwater quality and establish a sustainable groundwater governance in similar vulnerable areas. This thesis is submitted in partial fulfilment of the requirements for the degree of Master of Science (M. Sc.) in Climate Change And Development 2026. 2026-09-01T00:00:00Z https://ar.iub.edu.bd/handle/11348/1074 Impact of Urbanization on Urban Climate: A GIS and Remote Sensing Approach on Dhaka City (2000-2025) Iqbal, Iqbal Unplanned and swift urbanization from 2000-2025 has changed Dhaka’s land use and climate significantly. GIS and remote sensing were utilized in this thesis to quantify land cover change and the impact of the urban heat island (UHI) effect in Dhaka City. Four land cover categories (Built-up, Vegetation, Open Land, Water,) were classified through using multi-date Landsat satellite images (2000, 2010, 2020, 2025) to assess spatiotemporal transformations. To retrieve land surface temperatures (LST) for each period, thermal infrared data were processed which enabled analysis of how urban expansion affects and increases heating of the surface. Methodology included change detection, supervised classification, and LST retrieval, validated with field data and literature. Findings of a sharp increase in built-up area (from 29% to 41% of the area) and loss of vegetated land (from 28% to 11%) between 2000 and 2025 were recorded. Higher mean and maximum LST in urbanized zones were correlated with these land cover shifts, emphasizing the increasing UHI effect. Hottest surfaces were discovered in developed bare lands recently and also in dense urban areas. Water bodies and green spaces remained cooler on average. Some irregularity in LST patterns were observed due to seasonal differences in image acquisition, highlighting the need for interpretation cautiously. Dhaka’s urban expansion has led to greater heat retention and spatially increased UHI hotspots according to the findings of this research. Academic contributions of this thesis was by exploring updated, spatially explicit evidence of relation between land cover change and urban climate in Dhaka. These insights are significant for urban environmental management, illustrating the critical importance of sustainable planning interventions (e.g. preserving green infrastructure) to mitigate rising urban heat and improve livability in megacities like Dhaka. 2026-01-01T00:00:00Z https://ar.iub.edu.bd/handle/11348/489 Predictive assessment on landscape and coastal erosion of Bangladesh using geospatial techniques Islam, Md. Mozahidul; Rahman, Md. Saifur; Kabir, Md. Alamgir; Islam, Md. Nazrul; Chowdhury, Ruhul Mohaiman Coastal erosion, land use and land cover (LULC) changes analysis using remote sensing is a dynamic, relatively low cost based precise method using now a day. Coastal districts of Bangladesh occupied by naturally grown mangrove forest which are susceptible to rapid land cover (LC) changes and natural erosion. Barguna and Patuakhali districts of Bangladesh deserve special attention for conserving coastal mangrove forest named Tengragiri Wildlife Sanctuary and variety of human forces income. The core objective of this research is to analyze the LULC change along with coastal erosion analysis from 2000 to 2017. Combination of four years Landsat satellite image analysis, primary field data, geo-tag photography, secondary information, utilization of forest carbon inventory 2015 data, and semi-structured questionnaire are the key approaches adopted in the study. K-means cluster based unsupervised and maximum likelihood supervised classification by using ERDAS Imagine 2014 found the total study area is 33,361 ha. Random sampling (40 points/class) based accuracy assessment and verification by google earth pro 7.1 found overall accuracy 88.15% and Kappa coefficient is 0.867. Python coding program and overlay operation tested for conversion analysis any found weighted overlay provide best results. An intensive RS analysis of 33,564 ha mangrove forest and community landscapes generated six (6) distinct land cover class and sub-classes, e.g. Forest, agriculture & grassland, plantation, sandbar, settlement and waterbody. During 2000–2017, agriculture and grassland were decreasing 23 ha/year. Out of 11,831 ha (in 2000) Agri-grass land 9,326 ha remained intact while remaining 2,246 ha converted to settlement mixed with homestead plantation class. This study also presents the landscape erosion-accretion due to natural, quasi-natural and anthropogenic interventions which shows that, along the river flow and at the confluence at the Nishanbaria Union (local name Khouttar Char & Fakir hat) to lower side of the Tengragiri WS locations are susceptible to high trend of land erosion whereas accretions are prominent on the reverse sides named Baliatali Union, Barabagi Union and so on. These results of the study and developed maps will be helpful for the community people, line departments, national and international policy maker and the researchers’ community for monitoring coastal geomorphology including erosion and accretion of this landmass. 2020-09-20T00:00:00Z