Deep Learning Methods for Inferring Industrial CO2 Hotspots from Co-Emitted NO2 Plumes

Abstract

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 (CO2) emission hotspots using a convolutional neural network (CNN) with short-lived and co-emitted nitrogen dioxide (NO2) as a proxy. To address the data gaps in model parameter training, we constructed a dataset comprising over 210,000 samples of NO2 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 NO2 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 NO2 emission areas and high CO2 emission areas from the Global Open-Source Data Inventory for Anthropogenic CO2 (ODIAC), indicating that high NO2 emission hotspots can also indicate CO2 emission hotspots. In the future, as hyperspectral and high spatial resolution remote sensing data for CO2 and NO2 continue to grow, our methods will play an increasingly important role in global data preprocessing and global emission estimation.