Cooch Behar College

The increasing demand for water has made rainwater harvesting a crucial strategy to supplement existing surface and groundwater resources, particularly in semi-arid regions. This study evaluates potential surface rainwater harvesting (SRWH) zones in the upper Dwarakeshwar River basin using machine learning methods, including Support Vector Machines (SVM), Random Forests (RF), and Neural Networks (NN). Twelve thematic layers were integrated to develop SRWH suitability maps, categorising the basin into four classes: unsuitable, less suitable, moderately suitable, and most suitable. SVM exhibited the highest classification accuracy (AUC-ROC: 0.815), followed by RF (0.812) and NN (0.75). The SVM model identified 17.62% of the basin as unsuitable and 24.62% as most suitable, resulting in a more balanced and reliable classification than the RF and NN models. Highly suitable zones are primarily associated with gentle slopes, impermeable geology, and adequate precipitation, while Pearson correlation analysis highlights land use and land cover (0.72), drainage density (0.52), and lithology (0.27) as dominant controlling factors. The study also proposes strategic implementation of SRWH structures, including check dams, earthen dams, percolation tanks, farm ponds, and gully plugs, to enhance water retention and groundwater recharge. Validation of predictive models confirmed that SVM was the most effective approach, providing a reliable framework for sustainable water resource management. The findings provide a scalable, cost-effective, and data-efficient decision-support framework for SRWH planning, with future scope for incorporating climate variability to improve water security in arid and semi-arid regions.

This study assesses landslide susceptibility in the Rudraprayag to Badrinath region of the North Western Himalayas, Uttarakhand, India, using spatial statistics and machine learning techniques. A dataset of 268 landslide and 268 non-landslide points was analyzed to explore the spatial distribution of landslides. To achieve this, advanced spatial analysis methods including Average Nearest Neighbor (ANN), Moran’s I, K Function Analysis, and Getis-Ord Gi* Hotspot Analysis were integrated with Random Forest modeling. This integration aims to improve landslide susceptibility mapping. As a result of these analyses, the study found notable clustering of landslides impacted by environmental and geomorphological elements such as slope, elevation, seismic activity, and river proximity. Furthermore, cold spots were identified in more stable locations, while hotspot analysis highlighted high-risk areas, mostly in regions with steep slopes, weak lithology, and proximity to river systems. The Random Forest model further revealed elevation, slope, seismic activity, urbanization, and rainfall as key contributing factors in landslide susceptibility. Based on these findings, the study highlights the necessity of focused mitigation methods in high landslide prone locations and clearly identifies zones of varied susceptibility. In summary, the findings provide valuable insights into landslide dynamics and contribute to more effective landslide risk management in the region.

This research was focused on assessing the groundwater potentiality of Eastern Doars region using the ENTROPY, SWARA and its’ integrated weightage technique. Delineation of Ground Water Potential Zone (GWPZ) was done using the PROMITHEE approach. Total 12 indicators such as, the Topographic Wetness Index (TWI), Curvature, Elevation, Drainage Density, Distance from River, Rainfall, Slope, Normalized Difference Vegetation Index (NDVI), Topographic Roughness Index (TRI), Aspect, Land-Use Land Cover (LULC) and Topographic Positioning Index (TPI) were used in this research. The Ground Water Potential Zone (GWPZ) was classified into five zones such as Very High, High, Moderate, Low and Very Low. 89.255 Sq. Km, area was identified as the Very High GWPZ; 187.416 Sq. Km. area was marked as the High GWPZ (45.3

Water is a vital resource for sustaining life, and its quality is of utmost importance for the well-being of both humans and ecosystems. Rivers serve as significant sources of freshwater, providing drinking water, irrigation, and supporting various aquatic habitats. An attempt has been made in this paper to show the spatio - temporal variation of water quality along the Torsa River at Cooch Behar Town, its adjacent areas and outskirts, West Bengal, India. The field work was conducted during Pre-Monsoon, Monsoon, Post-Monsoon and Winter in the year of 2022 and 2023. To carry out the study, water samples have been collected from 5 stations, one station from urban environment, two from adjacent to urban environment, and two from outskirts. The collected samples have been tested in the laboratory by titration, gravimetric and coloration method. CCME WQI was applied using thirteen water quality parameters namely Temperature, pH, Conductivity, Turbidity, Total Hardness, Total Dissolved Solids, Total Soluble Solids, Dissolved oxygen, Biological Oxygen Demand, Chloride, Ferrous, Nitrate and Phosphate. Based on the results obtained from the index, the water quality of Torsa River ranged between 71.13 to 90.06 which indicate that river has the Fair to Good quality due to effect of various rural and urban pollutant sources. The temperature, turbidity, DO, BOD, Chloride and Phosphate do not meet the standards in different sampling stations especially in Monsoon. This makes the water unsuitable for drinking purposes without proper treatment. The work confirms the need to take an action for monitoring the river for proper management. Therefore, there is a need of intensive study leading to a contamination zone mapping to river water quality management.

Landslides happen often and cause serious problems in the Teesta River Basin, threatening buildings, communities, and the environment. This study uses a mapping system called GIS combined with a decision-making method called AHP to evaluate where landslides are likely by combining different types of information into a risk map. The AHP method gives importance to ten factors that affect landslides: slope, rainfall, shape of the land, distance from roads, distance from streams, vegetation health (NDVI), stream power (SPI), direction the land faces, land use, and geology. Several spatial tests, like Nearest Neighbour Analysis, Moran’s I, and hotspot analysis, are used to study how landslide areas are spread out. The model’s accuracy is checked using a test called the ROC curve. The results show that slope, rainfall, and land shape are the most important factors, while land use and geology have less effect. The risk map divides the area into five zones: very low, low, moderate, high, and very high risk. About 18.6