Machine Learning Applications in Hydrogeology

Machine Learning Applications in Hydrogeology is a practice-oriented course for graduate students and professionals who want to apply modern machine learning methods to hydrogeological questions. The course introduces participants to the full workflow of reproducible data-driven analysis, beginning with computing environment setup and Python fundamentals, and progressing toward hands-on hydrogeological case studies. Through browser-based Google Colab notebooks, learners can run code immediately, explore datasets interactively, and adapt examples for their own research or professional applications. All materials are shared online, and users may also copy the notebooks to their own Google Drive for private editing, experimentation, and further development.

The course places strong emphasis on real hydrogeological applications of artificial intelligence, including groundwater-level forecasting, spatio-temporal modelling, isotope estimation from routine hydrochemical data, groundwater risk or prospectivity mapping, and analysis of water-storage anomalies. In addition to technical implementation, it highlights reproducible scientific workflows, data engineering, quality control, model validation, interpretability, and transparent documentation, reflecting current best practices in hydrogeological machine learning. The article describing the course notes that it was developed as an open, continuously updated, reproducible resource for university students and professional hydrogeologists, with documented case studies and browser-based materials intended to lower the barrier to applying AI methods in groundwater research and practice.

This course is also used as part of the curriculum in Vilnius University graduate geology studies, supporting advanced training in hydrogeology, spatial data analysis, and applied artificial intelligence. Its pedagogical approach connects theory with practical, project-based learning and helps students develop the skills needed to work responsibly with spatial, temporal, and spatio-temporal environmental data. The course is maintained by the Department of Hydrogeology and Engineering Geology at Vilnius University together with collaborators at the San Diego Supercomputer Center, and it continues to grow as new case studies and teaching materials are added.

Article describing the course: Artificial intelligence in hydrogeology: applications and an open, reproducible machine learning course

Spatial Data Science and Deep Learning

Deep learning and neural networks enable data scientists to apply machine learning techniques to a growing range of real world problems—many fueled by massive repositories of GIS data. A unique course at UC San Diego has become a national exemplar.

Browse this gallery of StoryMaps, web mapping apps, Python notebooks, journal articles, or other digital resources supporting the science presented in this chapter of the Esri Press book GIS for Science: Maps for Saving the Planet, Volume 3.

Learn more.