GeoAI-Driven Groundwater Potential Zone Mapping in Devanahalli, Bengaluru Rural, Karnataka, India
Chandrashekharappa Agasnalli *
Department of Civil Engineering, BMS Institute of Technology & Management, Bengaluru - 560119, Karnataka, India.
Anupkumar G. Ekbote
Department of Civil Engineering, BMS Institute of Technology & Management, Bengaluru - 560119, Karnataka, India.
Dhanashree Nerlikar
Department of Civil Engineering, Vijaya Vittala Institute of Technology, Bengaluru - 560077, Karnataka, India.
*Author to whom correspondence should be addressed.
Abstract
Aims: To develop an integrated GeoAI framework for mapping groundwater potential zones in Devanahalli, Bengaluru Rural, and to interpret the spatial influence of major hydrogeological and terrain controls for groundwater exploration and recharge planning.
Study Design: Geospatial groundwater-potential assessment integrating remote sensing, GIS-based thematic analysis and a machine-learning-oriented decision framework.
Place and Duration of Study: Devanahalli Taluk, Bengaluru Rural District, Karnataka, India. The analysis uses the spatial datasets and thematic products assembled for the 2026 study.
Methodology: Groundwater-conditioning layers representing elevation, slope, drainage, lineaments, land use/land cover, soil, geology, rainfall and related terrain characteristics were prepared and standardised in a GIS environment. Their hydrogeological significance was evaluated and integrated in a GeoAI workflow that links spatial preprocessing, feature extraction, model development, groundwater-potential prediction, zonation and validation. The final potential classes were interpreted together with observed terrain, drainage and structural patterns.
Results: The analysis indicates that groundwater favourability in Devanahalli is spatially heterogeneous. Lower-relief and gently sloping zones favour infiltration, whereas steeper terrain and densely drained sectors promote runoff. Structural discontinuities represented by lineaments provide potential pathways for groundwater storage and movement in the hard-rock setting. Agricultural, vegetated and less-impervious surfaces are comparatively supportive of recharge, while expanding built-up areas can restrict direct infiltration. The integrated groundwater-potential map delineates very high, high, moderate, low and very low potential zones and provides a spatial basis for prioritising groundwater investigation and recharge interventions.
Conclusion: Integrating geospatial evidence with GeoAI offers a reproducible framework for groundwater-potential assessment in rapidly urbanising hard-rock terrains. The resulting zonation can support borewell investigation, artificial recharge, rainwater harvesting, land-use planning and long-term groundwater management in Devanahalli.
Keywords: GeoAI, groundwater potential zonation, remote sensing, GIS, machine learning, artificial recharge, hard-rock aquifer, Devanahalli