Ciencias Exactas y Ciencias de la Salud
Permanent URI for this collectionhttps://hdl.handle.net/11285/551014
Pertenecen a esta colección Tesis y Trabajos de grado de los Doctorados correspondientes a las Escuelas de Ingeniería y Ciencias así como a Medicina y Ciencias de la Salud.
Browse
Search Results
- Seawater intrusion pattern recognition supported by unsupervised learning(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2025-06-13) Narváez Montoya, Christian Felipe; Mahlknecht, Jürgen; emimmayorquin; Mora Polanco, Abrahan Rafael; Bertrand, Guillaume; Bonasia, Rosanna; School of Engineering and Sciences; Campus Monterrey; Torrés Martínez, Juan AntonioClimate change and anthropogenic activities have negatively affected the world's water resources in the last 200 years. Seawater intrusion, one of the leading causes of groundwater contamination, particularly affects coastal systems. Coastal aquifers are naturally connected to seawater, with the saltwater forming a wedge beneath the freshwater due to the difference in densities. Seawater migrates further inland when the aquifer hydraulic head decreases concerning the sea level, and is driven by groundwater overexploitation or sea level rise. The resulting salty water is extracted for public water supply, irrigation, or industrial purposes. Thus, monitoring and understanding this process is essential for developing appropriate water management strategies. Environmental tracers, such as the major ions (Mg2+, Ca2+, Na+, K+, SO4 2-, HCO3-, NO3-, and Cl-), are recognized as valuable tools for identifying seawater intrusion and other salinization sources. In this context, unsupervised learning has supported the multivariate analysis to characterize the variability and range of major ions at different spatial and temporal scales. However, complex case studies with multiple processes triggering high salinity concentrations make pattern recognition of salinization sources challenging with traditional unsupervised techniques. This research identifies seawater intrusion and triggering factors for two complex case studies: the Caplina/Concordia aquifer system in the hyper-arid Atacama Desert and the Yucatan Peninsula, one of the world's largest coastal karst lowland areas. For this, novel network-based clustering was applied to major ions water quality datasets sourced from governmental institutions. The outcomes show that the Caplina/Concordia water samples associated with seawater intrusion were identified up to 5.5 km inland in zones with hydraulic heads less than 6 m.a.s.l. These findings align with a developed hydrogeological model and underscore overexploitation as a key driver for seawater intrusion. On the other hand, in the Yucatan Peninsula, water samples were indicated to be associated with seawater intrusion in zones with hydraulic heads less than 5 m.a.s.l. The natural extensive seawater wedge is the product of low hydraulic gradients, facilitating the extraction of seawater-groundwater mixture (upconing) to more than 100 km from the coast. Furthermore, gypsum dissolution and nitrate pollution are also critical concerns for water quality in the peninsula. Additionally, the thesis advocates for improved open science in water research, urging journals and researchers to make raw data publicly available.