Ensemble flow reconstruction in the atmospheric boundary layer from spatially limited measurements through latent diffusion models

• DOI: 10.48550/arxiv.2303.00836
• Publication Date: 2023-01-01
• Authors: Alex Rybchuk; Malik Hassanaly; Nicholas Hamilton; Paula Doubrawa; Mitchell J. Fulton; Luis A. Martínez‐...
• Keywords: Large-Eddy Simulation; Eddy diffusion

Due to costs and practical constraints, field campaigns in the atmospheric boundary layer typically only measure a fraction of volume interest. Machine learning techniques have previously successfully reconstructed unobserved regions flow canonical fluid mechanics problems two-dimensional geophysical flows, but these not yet been demonstrated three-dimensional layer. Here, we conduct numerical analogue campaign with spatially limited measurements using large-eddy simulation. We pose reconstruction as an inpainting problem, reconstruct realistic samples turbulent, use latent diffusion model. The model generates physically plausible turbulent structures on larger spatial scales, even when input observations cover less than 1% volume. Through combination qualitative visualization quantitative assessment, demonstrate that meaningfully diverse conditioned just one observation. These serve initial conditions for simulation code. find models show promise potential other applications problems.