I focus on developing reduced-order models and intelligent monitoring frameworks for complex multiphase systems, with a primary application in maritime hazard mitigation and environmental physics.
Following are the key research areas:

Maritime Oil Spill Prediction (The "Digital Twin" Approach): A core part of my work involves bridging the "detection-prediction" gap in ocean monitoring. I utilize the GNOME Lagrangian model to predict the trajectory of oil spills, specifically optimizing the physics by implementing length-scale dependent diffusion coefficients. This approach has been benchmarked against significant real-world events, including the Mt. Sanchi and Kerch Strait incidents.

Intelligent Remote Sensing & Interfacial Classification: To enhance real-time monitoring, I develop automated classification pipelines using Synthetic Aperture Radar (SAR) imagery. By applying Sobel-based edge detection and training Support Vector Machines (SVM) and Artificial Neural Networks (ANN) on interfacial gradient features, my research achieves high-precision discrimination between mineral oil and oceanographic "look-alikes," such as biogenic films.

Ocean Stratification & Vertical Mixing: I am currently expanding my research to account for the three-dimensional nature of oceanic transport. This involves studying vertical ocean mixing and the role of turbulence in energy and mass transport across stratified layers, aiming to improve the predictive skill of Earth System models.