We develop advanced sensing systems to observe and predict environmental processes across scales.

• Hyperspectral imaging for sediment, soil, algae, and water quality monitoring: Wang et al. (2024)
• Satellite, 5G, and optical-fiber sensing for extreme weather and urban microclimate observation.
• Quantum gravimetry for detecting subsurface and coastal hydrodynamic changes.
• Impact of sensor failures and disruption on forecasting accuracy: Wu & Wang (2025), Golparvar et al. (2024)

These efforts build a connected environmental observation network to enhance predictive Earth system models.

My research explores the physics of environmental flows—from greenhouse gas transport to sediment and ice–water interactions—through advanced modeling and AI integration.

• AI–CFD modeling for greenhouse gas emission inversion using satellite imagery and autonomous underwater robots.
• Multiphase flow studies of sediment transport and the ice–water boundary layer.
• Renewable energy impacts on marine and coastal systems, including turbine–animal interaction, sediment disturbance, noise propagation, and flooding or erosion risks: Asok and Wang (2025)

These studies reveal how natural and engineered systems interact across scales, advancing predictive models for environmental management and climate resilience.

AI-driven Disaster Resilience

Coastal and urban regions face rising flood risks that disrupt infrastructure, damage property, threaten public health, and impose mounting economic and ecological costs. Our work develops AI-enhanced tools to observe, interpret, and predict these emerging flood hazards.

• Social media–based flood observation using NLP and crowdsourced imagery to expand situational awareness: Wang et al. (2018)
• Automated monoplotting and visual data mining to recover water extent and depth from everyday photos and cameras: Golparvar & Wang (2025)
• Real-time computer vision systems for street-level flood detection and monitoring: Wang et al. (2025), Wang & Ding (2023)
• Flood hazard simulation to create realistic flood scenarios and assess potential impacts: Bazzett et al. (2024)
• LLM-based explanation and decision support to translate flood model outputs into clear, actionable information for planners and emergency managers. Martelo et al. (2025)

These tools combine physical modeling with AI and public data streams to build more adaptive and equitable flood preparedness systems.

Fluid–Structure–Particle Interaction in Microfluidics

Microfluidic devices offer powerful new ways to isolate and concentrate biological targets, from bacteria to algae. Their performance, however, is governed by complex interactions among fluid flow, deformable structures, and suspended particles—interactions that remain challenging to predict and optimize.

My research develops advanced theoretical and computational models of fluid–structure–particle dynamics and integrates them with laboratory experiments to improve microfluidic device design and operation.

• Modeling of particle capture and transport for isolating bacteria and other microorganisms from complex fluids.
• Fluid–structure interaction analysis to optimize deformable microchannels and membranes.
• Design of next-generation microfluidic systems for biomedical diagnostics, environmental monitoring, and algae-based applications: Chen et al. (2025)

This work advances both the fundamental physics of microscale flows and the development of practical devices for health, biotechnology, and environmental sensing.