Naimur Rahman
GeoAI for Climate, Environmental & Multi-Hazard Risk Assessment
I develop GeoAI and Earth Observation-based methods for environmental and natural-hazard modeling, integrating machine learning, deep learning, explainable AI, uncertainty quantification, and spatially robust validation.
Research Focus
Current research directionsGeoAI & Earth Observation
Geospatial AI, satellite observations and large-scale environmental modeling.
Climate & Natural Hazards
Flood, drought, wildfire and multi-hazard risk assessment.
Explainable & Uncertainty-Aware AI
XAI, ensemble modeling and uncertainty quantification.
Spatially Robust Validation
Spatial cross-validation and generalization of geospatial models.
About Me
Researching Environmental Systems Through GeoAI
I am a researcher in Geography and Environmental Studies at the University of Rajshahi, Bangladesh, working at the intersection of GeoAI, Earth Observation, machine learning and environmental risk assessment.
My current research focuses on developing interpretable and uncertainty-aware approaches for multi-hazard susceptibility modeling, with particular emphasis on spatial validation and the generalization of geospatial prediction models.
I am currently exploring PhD opportunities in GeoAI, remote sensing, environmental modeling, climate risk and related Earth-system research.
Academic Profiles
Selected Research
Research Projects
My research applies GeoAI, Earth Observation, machine learning and deep learning to understand and model environmental and natural-hazard risks across different spatial scales.
A GeoAI-Based Multi-Hazard Susceptibility Assessment Framework
A comprehensive GeoAI framework integrating machine learning, deep learning, explainable artificial intelligence, uncertainty quantification and Earth Observation data to investigate the spatial susceptibility of multiple natural hazards across South Asia.
Hazards Investigated
Flood Susceptibility Modeling
GeoAI-based flood susceptibility assessment using environmental, topographic, hydrological and anthropogenic conditioning factors across the study region.
Methods: ML · DL · Earth Observation · Ensemble Modeling · Spatial Validation
Drought Susceptibility Modeling
Spatial assessment of drought susceptibility using climate, vegetation, environmental and hydro-climatic indicators integrated within a GeoAI modeling framework.
Methods: ML · DL · Climate Data · Remote Sensing · XAI
Earthquake Susceptibility Modeling
GeoAI-based spatial modeling of earthquake susceptibility using geological, tectonic, topographic and environmental conditioning variables.
Methods: ML · DL · Geospatial Modeling · Ensemble Learning · XAI
Landslide Susceptibility Modeling
Spatial landslide susceptibility assessment using terrain, geological, environmental and hydro-climatic factors within a machine learning and deep learning framework.
Methods: ML · DL · Remote Sensing · Terrain Analysis · Spatial Modeling
Multi-Hazard GeoAI Framework
An integrated framework combining hazard-specific susceptibility models with explainable AI, ensemble approaches, uncertainty quantification and spatially robust validation.
Methods: GeoAI · XAI · Ensemble Modeling · Uncertainty Quantification · Spatial Validation
Climate-Driven Winter Wildfire Susceptibility
Machine learning and deep learning approaches for mapping rare winter wildfire susceptibility in Los Angeles, California using climatic, topographic, vegetation and anthropogenic factors.
Methods: ML · DL · Remote Sensing · Climate Variables · Feature Importance
Cross-cutting methodological interests
Research Themes
Areas of Research Interest
GeoAI & Earth Observation
Geospatial artificial intelligence, satellite observations and large-scale environmental modeling.
Climate & Natural Hazards
Flood, drought, wildfire and climate-related environmental risks.
Explainable & Uncertainty-Aware AI
Explainable machine learning, ensemble approaches and uncertainty quantification.
Spatial Modeling & Validation
Spatial cross-validation, grid-based validation and model generalization.
Future Research
Toward Reliable GeoAI for Environmental Decision-Making
My prospective research interests include large-scale environmental modeling, climate-risk assessment, hydrological and hazard modeling, and the development of interpretable and uncertainty-aware GeoAI methods for understanding environmental systems and supporting resilient decision-making.
Academic Collaboration
Interested in My Research?
I am interested in connecting with researchers and prospective PhD supervisors working in GeoAI, remote sensing, environmental modeling, climate risk and natural-hazard research.