Selected projects spanning physics-informed AI, compound flooding, tropical-cyclone extremes, coastal water levels, probabilistic engineering, nature-based resilience, and hazard communication.
A physics-informed neural emulator for compound flooding that incorporates continuity and both momentum equations into spatiotemporal learning. The project targets fast prediction without discarding hydrodynamic structure.
Research spanning how compound floods are represented in modeling frameworks, how forcing and parameter uncertainty propagate through simulations, and how flood information can be translated into infrastructure decisions such as coastal-road closure criteria.
Research on how anomalously warm ocean conditions alter tropical-cyclone intensification, predictability, and destructive potential, from the Gulf of Mexico to global analyses.
Research connecting long-record tidal dynamics with changing coastal extremes, including global co-variability among major tidal constituents and nonstationary extreme-value behavior under long-term environmental and climate variability.
Probabilistic coastal-design research addressing joint wave–water-level extremes, dependence among environmental variables, copula-model choice, and nonstationary extreme-wave behavior in reliability assessment and design under changing climate conditions.
A systematic review of process-based modeling for nature-based coastal flood mitigation, mapping model choices, spatial scales, solution types, and research-to-operation gaps.
The “Angles” method represents flood drivers as vectors in geometric space, supporting clearer communication of dependence and nonstationarity to academic and non-academic audiences.