Research area
Characterizing the dynamics of flow, particles, heat, and scalar transport in complex environments at the interface of the lower atmosphere and Earth’s surface.
Atmospheric stability changes the vertical structure of smoke plumes and the rate at which they entrain and mix with surrounding air. Daytime unstable conditions promote vigorous vertical and lateral dispersion, while nighttime stable conditions constrain plume rise and can maintain higher smoke concentrations closer to the surface. Neutral conditions provide an intermediate reference for assessing these competing effects.
Fuel moisture content modifies fire heat release and therefore the buoyant forcing that drives plume development. Increasing moisture changes plume rise, shape, concentration, propagation, and interaction with the surrounding flow; the lowest-moisture case produces the strongest heat flux, while the highest-moisture case produces the weakest.
Large-eddy simulations determine how urban form and terrain reshape turbulent transport, airborne trajectories, and the landing distributions of firebrands. Resolving the heterogeneous flow around buildings reveals preferential pathways and accumulation zones that are not captured by spatially uniform wind assumptions.
Large-eddy simulations track 50,000 firebrands with diameters from 2 to 5 mm after release over idealized urban regions. The resolved near-surface flow and terrain configuration alter where firebrands first contact the ground and where they finally settle after rolling, revealing deposition pathways that cannot be inferred from airborne trajectories alone.
This work examines how buoyancy, atmospheric structure, and built-environment geometry control plume rise, entrainment, intermittency, and downstream smoke transport.
Direct numerical simulations show how heating modifies the wake and aerodynamic forces acting on an idealized spherical firebrand. Buoyancy reorganizes the wake into elongated thermal plumes, raises the drag coefficient, and produces a sustained negative lift coefficient; incorporating these transient force changes improves predictions of flight trajectory and landing location.
A physics-based thermodynamic model predicts local dead-fuel moisture and exchanges heat and moisture with the fire-dynamics solver in both directions. The coupled approach captures spatial and temporal variation in drying, ignition, fire spread, heat-release rate, and plume forcing more realistically than a uniform or prescribed fuel-moisture assumption.
Large-eddy simulations are used to study how unresolved turbulence influences inertial particles settling through an anisotropic boundary layer. A structural subgrid-scale model partially restores kinetic energy lost through spatial filtering and improves predictions of particle acceleration, velocity fluctuations, settling, dispersion, clustering, and deposition, with the strongest effects occurring for lower-inertia particles.
Coupled surface-energy-balance and large-eddy simulations reveal how nonuniform diurnal heating, roof materials, and terrain geometry produce local stable and unstable layers. These thermal gradients modify pressure, turbulence, ventilation routes, and the removal of heat and passive scalars across morning, afternoon, and evening conditions.
Read the 2022 Boundary-Layer Meteorology paper · Read the 2014 Boundary-Layer Meteorology paper
The lab studies how flow, heat transfer, environmental forcing, and geometry interact to shape the function of natural and bio-inspired systems.
Large-eddy simulations of a bioinspired deep cavity isolate how daytime and nighttime stratification alter internal circulation and scalar exchange. Ventilation depends jointly on the entrance vortex and stability within the lower cavity: unstable conditions strengthen suction and buoyant transport, while stable conditions suppress vertical exchange.
Coupling a three-dimensional urban surface-energy model with large-eddy simulation resolves realistic, spatially nonuniform surface temperatures. As solar exposure and shadowing evolve, the resulting buoyancy and pressure gradients reorganize street-canyon turbulence, near-surface ventilation, and heat transport.
Large-eddy simulation of atmospheric flow is combined with local canopy geometry and trajectory modeling to predict transport through complex vegetation. In the golf-shot application, tree-induced turbulence and wind-direction changes produce substantial sensitivity in the ball’s flight path and landing position.
