Time-aware UNet and super-resolution deep residual networks for spatial downscaling

TL;DR

This paper introduces time-aware extensions to deep learning models for spatial downscaling of satellite atmospheric data, significantly improving accuracy and convergence with minimal added complexity.

Contribution

It proposes lightweight temporal modules integrated into UNet and SRDRN architectures for enhanced spatial downscaling of ozone data.

Findings

Temporal modules improve downscaling performance

Models converge faster with temporal encoding

Minimal increase in computational complexity

Abstract

Satellite data of atmospheric pollutants are often available only at coarse spatial resolution, limiting their applicability in local-scale environmental analysis and decision-making. Spatial downscaling methods aim to transform the coarse satellite data into high-resolution fields. In this work, two widely used deep learning architectures, the super-resolution deep residual network (SRDRN) and the encoder-decoder-based UNet, are considered for spatial downscaling of tropospheric ozone. Both methods are extended with a lightweight temporal module, which encodes observation time using either sinusoidal or radial basis function (RBF) encoding, and fuses the temporal features with the spatial representations in the networks. The proposed time-aware extensions are evaluated against their baseline counterparts in a case study on ozone downscaling over Italy. The results suggest that, while only slightly increasing computational complexity, the temporal modules significantly improve downscaling performance and convergence speed.