DeepSAT4D: Deep learning empowers four-dimensional atmospheric chemical concentration and emission retrieval from satellite

The high spectral resolution afforded by Hyperspectral Imaging (HSI) sensors is poised to bring unprecedented advancements to signature characterization applications. Thus far, much of the research in the machine learning field devoted to HSI applications has focused on a few specific tasks like land-use/land-cover classification. In land classification tasks, spatial information is very important, and model architectures are often designed to leverage spatial contexts. However, it is unclear how well these spatially-tuned models will translate to tasks where spectral information is critical, like the detection and characterization of chemicals. In this work, we compare spectral models (inputs are 1D spectra) and spatial-spectral models (inputs are 3D cubes) in the context of predicting chemical concentration maps. We find that spatial-spectral models perform the best, though we find a wide range in performance across the different architectures tested. Additionally, we find that model performance is impacted by the availability of training data, particularly in scenarios where the training data doesn’t fully capture the true variance of real-world conditions. We find that data augmentation can help mitigate sparse coverage of observed parameter space (e.g., seasonal or geographic variability in ground cover), and present augmentation strategies that are tailored to hyperspectral data.

show abstract

Remote Sensing of Tropospheric Ozone from Space: Progress and Challenges

Xu,

Zhang,

Rao

et al. 2024

J Remote Sens

View full text Add to dashboard Cite

Ozone stands out as a crucial trace gas within the Earth’s atmosphere, exerting a substantial influence on climate change and air pollution. Tropospheric ozone plays an important role in the formation of photochemical smog, and its variations are associated with human activities. The utilization of satellite remote sensing technology for tropospheric ozone monitoring enables a quantitative analysis of its global and regional spatiotemporal characteristics. It also facilitates the investigation of the mechanisms involved in ozone formation within the troposphere. The significant progress in product accuracy and spatiotemporal resolution of ozone remote sensing products, including total ozone and vertical profiles, can be attributed to the extensive development of satellite remote sensing techniques. Nevertheless, the precision of tropospheric ozone products remains inadequate for contemporary scientific purposes, primarily because of faint signals in the lower atmosphere, the intricate nature of the underlying surface, and the existence of clouds and aerosols. This study places emphasis on the satellite remote sensing of tropospheric ozone, encompassing a comprehensive review of the advancements in satellite sensors and the characteristics and suitability of various retrieval algorithms. Moreover, this research delves into the possible utilization of satellite remote sensing for the provision of reliable tropospheric ozone observation data on a global and regional level.

show abstract

DeepSAT4D: Deep learning empowers four-dimensional atmospheric chemical concentration and emission retrieval from satellite

Cited by 5 publications

References 60 publications

Artificial intelligence for geoscience: Progress, challenges, and perspectives

Artificial intelligence for geoscience: Progress, challenges, and perspectives

Chemical signature characterization with hyperspectral imagery: novel deep learning model architectures and physically-motivated data augmentation techniques

Remote Sensing of Tropospheric Ozone from Space: Progress and Challenges

Contact Info

Product

Resources

About