Deep Learning Methods for Inferring Industrial CO<sub>2</sub> Hotspots from Co-Emitted NO<sub>2</sub> Plumes

The “top-down” global stocktake (GST) requires the processing of vast volumes of hyperspectral data to derive emission information, placing greater demands on data processing efficiency. Deep learning, leveraging its strengths in the automated and rapid analysis of image datasets, holds significant...

Full description

Saved in:
Bibliographic Details
Main Authors: Erchang Sun, Shichao Wu, Xianhua Wang, Hanhan Ye, Hailiang Shi, Yuan An, Chao Li
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/7/1167
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The “top-down” global stocktake (GST) requires the processing of vast volumes of hyperspectral data to derive emission information, placing greater demands on data processing efficiency. Deep learning, leveraging its strengths in the automated and rapid analysis of image datasets, holds significant potential to enhance the efficiency and effectiveness of data processing in the GST. This paper develops a method for detecting carbon dioxide (CO<sub>2</sub>) emission hotspots using a convolutional neural network (CNN) with short-lived and co-emitted nitrogen dioxide (NO<sub>2</sub>) as a proxy. To address the data gaps in model parameter training, we constructed a dataset comprising over 210,000 samples of NO<sub>2</sub> plumes and emissions based on atmospheric dispersion models. The trained model performed well on the test set, with most samples achieving an identification accuracy above 80% and more than half exceeding 94%. The trained model was also applied to the NO<sub>2</sub> column data from the TROPOspheric Monitoring Instrument (TROPOMI) for hotspot detection, and the detections were compared with the MEIC inventory. The results demonstrate that in high-emission areas, the proposed method successfully identifies emission hotspots with an average accuracy of over 80%, showing a high degree of consistency with the emission inventory. In areas with multiple observations from TROPOMI, we observed a high degree of consistency between high NO<sub>2</sub> emission areas and high CO<sub>2</sub> emission areas from the Global Open-Source Data Inventory for Anthropogenic CO<sub>2</sub> (ODIAC), indicating that high NO<sub>2</sub> emission hotspots can also indicate CO<sub>2</sub> emission hotspots. In the future, as hyperspectral and high spatial resolution remote sensing data for CO<sub>2</sub> and NO<sub>2</sub> continue to grow, our methods will play an increasingly important role in global data preprocessing and global emission estimation.
ISSN:2072-4292