Machine Learning and Deep Learning Approaches to Map Earthquake Susceptibility: A Study of South Asia
Developed an ML- and DL-based framework to map earthquake susceptibility across South Asia using 17 predictive factors.
GIS • Remote Sensing • GeoAI
Research Projects & Applications
A portfolio of geospatial research applying Earth Observation, machine learning, deep learning, and GeoAI to understand environmental change, natural hazards, climate risks, and sustainable resource management.
Research Focus
Geospatial science for real-world environmental challenges.
My work combines GIS, satellite remote sensing, Earth Observation, GeoAI, machine learning, deep learning, Explainable AI, and uncertainty assessment to build data-driven environmental and disaster-risk solutions.
06
Featured
Projects
Projects
21+
GeoAI
Factors
Factors
AI
Data-Driven
Methods
Methods
Selected Work
Research Projects
Selected applications of GeoAI, remote sensing, and geospatial modelling across natural hazards and environmental monitoring.
Satellite-Based Air Quality Monitoring Using Google Earth Engine
An Earth Observation workflow using Google Earth Engine and satellite-derived PM2.5, PM10, NO2, SO2, CO, and AOD data to assess spatial and temporal air-pollution patterns.
Landslide Detection Using Deep Learning and Sentinel-2 Imagery
Developed a deep learning approach for landslide detection using Sentinel-2 imagery to automate landslide mapping and support disaster-management applications.
GeoAI-Based Drought Susceptibility Mapping, Hotspot Analysis & Uncertainty Assessment Across South Asia
Designed a GeoAI framework using 21 environmental factors, multiple ML/DL models, Explainable AI, hotspot analysis, and uncertainty quantification.
GeoAI-Based Flood Susceptibility Mapping, Hotspot Analysis & Uncertainty Assessment
Developed a GeoAI framework using 22 flood-conditioning factors and multiple ML/DL models, with hotspot analysis, Explainable AI, and uncertainty quantification.
GeoAI-Based Landslide Susceptibility Mapping, Hotspot Analysis & Uncertainty Assessment
Developed a GeoAI-based framework using 16 landslide- conditioning factors and multiple ML/DL models, with hotspot analysis, Explainable AI, and uncertainty quantification.
Research Areas
Methods, themes & applications
My research brings together geospatial data, computational methods, and environmental science to develop interpretable and decision- relevant solutions for complex Earth-system problems.
01
GeoAI & Machine Learning
ML, deep learning, Explainable AI, ensemble modelling, and geospatial prediction.
02
Earth Observation
Satellite remote sensing, Google Earth Engine, environmental datasets, and spatial analysis.
03
Natural Hazards
Flood, drought, landslide, and earthquake susceptibility and risk assessment.
04
Uncertainty & Decision Support
Uncertainty assessment, hotspot analysis, spatial modelling, and environmental decision support.
Research Portfolio
Interested in the research?
Explore the complete collection of projects, maps, analyses, and geospatial applications.
View Complete Portfolio →