Interactive Atmospheric Composition Emulation for NASA GISS Earth System Models
Status: Active
Smart NINT: A machine learning approach that emulates interactive atmospheric composition in Earth System Models, reducing computational costs while maintaining real-time feedback between aerosols and climate processes.
View Project →Stochastic Fluvial Flood Inundation Mapping
Status: Active
A stochastic rating curve approach using copula-based Bayesian frameworks to characterize channel geometry-discharge relationships, generating probabilistic flood inundation maps that capture uncertainties in stage-discharge variations.
View Project →Multiscale Fire Prediction for Reinsurance Risk Assessment
Status: Active
Novel integration of generative AI with Earth system modeling to produce high-resolution fire risk metrics for reinsurance stakeholders, addressing the $394-893 billion annual costs of wildfire activity in the U.S.
View Project →Predicting Drought, Preventing Famine: Advanced Soil Moisture Analysis for Climate Resilience
Status: Pending
A two-phase methodology combining temporal Graph Neural Networks (GNNs) with physics-based data assimilation to create continuous global soil moisture coverage from satellite missions, addressing critical gaps in drought monitoring and food security applications.
View Project →