Shugar, Daniel HKshetri, Tek Bahadur2025-01-092025-01-092025-01-08Kshetri, T. B. (2025). Monitoring sand mining in Nepalese rivers using deep learning and Earth observation (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.https://hdl.handle.net/1880/120411Sand mining has become one of the major environmental issues impacting river systems worldwide, including in Nepal. This thesis focuses on monitoring sand mining activities in Nepalese rivers using deep learning techniques for detecting sand processing plants (SPPs) and assessing the geomorphological consequences of mining over several decades. The study examines five major rivers: East Rapti, Indrawati, Kamala, Tinau, and West Rapti, using multi-temporal satellite imagery from Landsat, Sentinel-2, PlanetScope, and other remote sensing sources to track the changes in river morphology (between 1973 and 2023) and sand mining operations (between 1990 and 2023). The deep learning model was specifically trained to detect SPPs, providing a comprehensive timeline of sand mining trends. Findings showed that sand mining began to escalate significantly after the year 2000, coinciding with increased infrastructure development in Nepal. The number of detected SPPs peaked between 2015 and 2020, particularly along the Kamal, Ratu (102 locations in 2020) and West Rapti (37 in 2019) rivers, with a notable growth in mining intensity during this period. However, from 2020 onwards, a declining trend was observed, potentially influenced by stricter regulations. This temporal analysis offers key insights into the trajectory of sand mining activities, underscoring the need for continuous monitoring to assess the effectiveness of policy interventions. In parallel, the geomorphological impacts of human intervention were quantified by calculating two key indices—the Braiding Index (BI) and Sinuosity Index (SI)—to evaluate changes in river morphology. Significant geomorphological changes were detected across all studied rivers mainly due to human intervention (sand and gravel mining, river training, and urban encroachment), with both BI and SI showing sharper declines in mining-affected areas than either up- or downstream. In the Tinau River, the BI dropped by over 70% near Butwal mainly due to urban encroachment, indicating a loss of channel complexity. Similarly, the SI in the East Rapti River decreased by more than half, reflecting a transition toward more straightened channels primarily due to urban expansion and mining. Other rivers, such as the Kamala and West Rapti, also exhibited clear trends of reduced channel braiding and sinuosity, with human intervention (mainly due to sand and gravel mining) seriously altering natural river dynamics. This thesis highlights the dual approach of employing deep learning for automated detection of sand processing plants and remote sensing techniques for detailed geomorphological analysis. The integration of these methodologies provides a powerful tool for monitoring and managing sand mining activities across Nepal’s diverse river systems. By offering both a temporal and spatial understanding of the scope and consequences of sand mining, the research underscores the urgent need for sustainable resource management practices. The findings suggest that without stronger regulation and continuous monitoring, sand mining will continue to pose severe risks to river health, biodiversity, and local communities.enUniversity of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.Sand MiningRiver MorphologyDeep LearningConvolution Neural NetworkRemote SensingSatellite ImagerySand Processing Plants (SPP) DetectionBraiding IndexSinuosity IndexGeomorphological ImpactsNepalease RiversRiver DynamicsSustainable Resource ManagementSand Extraction TrendsMining RegulationHydrologyGeographyArtificial IntelligenceEngineering--EnvironmentalEngineering--MiningComputer ScienceRemote Sensingmaster thesis