GenDA: Generative Data Assimilation on Complex Urban Areas via Classifier-Free Diffusion Guidance

TL;DR

GenDA is a novel generative data assimilation framework that reconstructs high-resolution urban wind fields from sparse sensor data using a graph-based diffusion model, enabling obstacle-aware predictions and generalization across unseen geometries.

Contribution

The paper introduces GenDA, a new generative framework employing classifier-free diffusion guidance for urban wind field reconstruction, capable of handling complex geometries and sparse data without retraining.

Findings

Reduces RRMSE by 25-57% compared to baselines

Increases SSIM by 23-33% over classical methods

Demonstrates effective reconstruction on real urban CFD data

Abstract

Urban wind flow reconstruction is essential for assessing air quality, heat dispersion, and pedestrian comfort, yet remains challenging when only sparse sensor data are available. We propose GenDA, a generative data assimilation framework that reconstructs high-resolution wind fields on unstructured meshes from limited observations. The model employs a multiscale graph-based diffusion architecture trained on computational fluid dynamics (CFD) simulations and interprets classifier-free guidance as a learned posterior reconstruction mechanism: the unconditional branch learns a geometry-aware flow prior, while the sensor-conditioned branch injects observational constraints during sampling. This formulation enables obstacle-aware reconstruction and generalization across unseen geometries, wind directions, and mesh resolutions without retraining. We consider both sparse fixed sensors and trajectory-based observations using the same reconstruction procedure. When evaluated against supervised graph neural network (GNN) baselines and classical reduced-order data assimilation methods, GenDA reduces the relative root-mean-square error (RRMSE) by 25-57% and increases the structural similarity index (SSIM) by 23-33% across the tested meshes. Experiments are conducted on Reynolds-averaged Navier-Stokes (RANS) simulations of a real urban neighbourhood in Bristol, United Kingdom, at a characteristic Reynolds number of $\mathrm{Re}\approx2\times10^{7}$, featuring complex building geometry and irregular terrain. The proposed framework provides a scalable path toward generative, geometry-aware data assimilation for environmental monitoring in complex domains.