1 November 2024 in Physics of Fluids
A physics-informed machine learning model is proposed in this paper to reconstruct the high-fidelity three-dimensional boundary layer wind field of tropical cyclones. The governing equations of the wind field, which incorporate a spatially varying eddy diffusivity coefficient, are derived and embedded within the model's loss function. This integration allows the model to learn the underlying physics of the boundary layer wind field. The model is applied to reconstruct two tropical cyclone events in different oceanic basins. A wide range of observational data from satellite, dropsonde, and Doppler radar records are assimilated into the model. The model's performance is evaluated by comparing its results with observations and a classic linear model. The findings demonstrate that the model's accuracy improves with an increased amount of real data and the introduction of spatially varying eddy diffusivity. Furthermore, the proposed model does not require strict boundary conditions to reconstruct the wind field, offering greater flexibility compared to traditional numerical models. With the assimilation of observational data, the proposed model accurately reconstructs the horizontal, radial, and vertical distributions of the wind field. Compared with the linear model, the proposed model more effectively captures the nonlinearities and asymmetries of the wind field, thus presents more realistic outcomes.
25 October 2024 in Journal of Wind Engineering and Industrial Aerodynamics
19 September 2026 in Journal of Geophysical Research: Atmospheres
Boundary layer turbulence and vertical mixing exert great control over the structure and evolution of tropical cyclones (TCs). However, scaling and parameterizing them remains challenging due to their inherent complexity and limited observations. Based on a 356‐m‐tall meteorological tower and wind profilers at three sites, this study analyzes the structure of wind and turbulence in the boundary layer of Super Typhoon Ragasa and Typhoon Wipha during landfall. Wind profile observations suggest an asymmetric structure with more significant inflow in the right‐front quadrant than right‐rear. Local scaling works for describing the dimensionless wind velocity variances and turbulent kinetic energy but is subject to a high degree of uncertainty. Friction velocity‐based non‐local parameterization performs better than wind velocity variance‐based local schemes in reproducing the observed vertical eddy diffusivity for momentum, while the larger‐than‐unity scaling parameter in the commonly used K‐profile parameterization scheme indicates higher‐than‐parameterized vertical turbulent mixing. These observational results suggest that directly applying similarity laws and parameterization schemes developed for homogeneous, straight‐line winds to landfalling TCs may introduce significant uncertainties. Accounting for the influences from the inhomogeneous features, such as rolls, mesoscale vortices, and underlying surface heterogeneity, may represent a future direction of improvement.