English

Air Quality Downscaling with Station-Guided Pseudo-Supervision

Machine Learning 2026-07-06 v1 Artificial Intelligence Computer Vision and Pattern Recognition

Abstract

Super-resolving coarse atmospheric fields to local PM2.5_{2.5} variations is uniquely challenged by a mismatch in spatial support: while pixels represent regional averages, ground-truth observations are discrete, unaligned samples of a continuous spatial signal. To bridge this gap, we present a station-guided framework for high-resolution PM2.5_{2.5} downscaling over Europe. Taking coarse CAMS atmospheric composition fields alongside heterogeneous side information (i.e., human activity, land cover, elevation, satellite aerosol observations, and wind fields) our framework jointly super-resolves (×40\times 40, \approx 1 km) and bias-corrects CAMS rasters, without relying on temporal sequence modelling. To address the challenge of densely supervising our multi-scale transformer network with sparse in-situ data, we introduce a time-agnostic propagation strategy that utilises spatial Gaussian blending of interpolated OpenAQ observations. Extensive qualitative and station-level evaluations across Europe demonstrate that our model recovers fine-grained spatial structures and effectively mitigates localised CAMS biases.

Keywords

Cite

@article{arxiv.2607.05292,
  title  = {Air Quality Downscaling with Station-Guided Pseudo-Supervision},
  author = {Guorun Wang and Simone Foti and Andreas D. Demou and Leonidas Kotoulas and Theodoros Christoudias and Alexandros Koliousis and Mihalis Nicolaou and Stefanos Zafeiriou},
  journal= {arXiv preprint arXiv:2607.05292},
  year   = {2026}
}