Search papers, labs, and topics across Lattice.
This study introduces a station-guided framework for downscaling coarse atmospheric PM$_{2.5}$ fields to high-resolution (approximately 1 km) levels by integrating various heterogeneous data sources without temporal sequence modeling. The method employs a time-agnostic propagation strategy that utilizes spatial Gaussian blending of sparse ground-truth observations to effectively supervise a multi-scale transformer network. Results show significant improvements in recovering fine-grained spatial structures and reducing biases in CAMS data across Europe, highlighting the framework's efficacy in addressing the challenges of atmospheric data resolution.
Bridging the gap between coarse atmospheric models and local PM$_{2.5}$ variations, this framework achieves a 40x super-resolution without relying on temporal data.
Super-resolving coarse atmospheric fields to local PM$_{2.5}$ 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 PM$_{2.5}$ 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 ($\times 40$, $\approx$ 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.
