Super-Resolution of BVOC Maps by Adapting Deep Learning Methods

TL;DR

This paper explores enhancing low-resolution BVOC emission maps using adapted deep learning super-resolution techniques, addressing challenges like dynamic range and outliers, with considerations for real-world temporal and geographical constraints.

Contribution

It adapts state-of-the-art image super-resolution neural networks to improve BVOC map resolution, accounting for emission variability and outliers, and discusses future generalization strategies.

Findings

Super-resolution methods improve BVOC map detail.

Adapted models handle large dynamic range and outliers.

Consideration of real-world temporal and geographical constraints.

Abstract

Biogenic Volatile Organic Compounds (BVOCs) play a critical role in biosphere-atmosphere interactions, being a key factor in the physical and chemical properties of the atmosphere and climate. Acquiring large and fine-grained BVOC emission maps is expensive and time-consuming, so most available BVOC data are obtained on a loose and sparse sampling grid or on small regions. However, high-resolution BVOC data are desirable in many applications, such as air quality, atmospheric chemistry, and climate monitoring. In this work, we investigate the possibility of enhancing BVOC acquisitions, further explaining the relationships between the environment and these compounds. We do so by comparing the performances of several state-of-the-art neural networks proposed for image Super-Resolution (SR), adapting them to overcome the challenges posed by the large dynamic range of the emission and reduce the impact of outliers in the prediction. Moreover, we also consider realistic scenarios, considering both temporal and geographical constraints. Finally, we present possible future developments regarding SR generalization, considering the scale-invariance property and super-resolving emissions from unseen compounds.