An Improved Hybrid GC-LSTM Framework for Hourly Nowcasting of Ground-Level NO2 Concentrations over Beijing-Tianjin-Hebei Region

• Published in: IEEE transactions on geoscience and remote sensing 2024-12, p.1-1
• Main Authors: Han, Zongfu, Fan, Meng, Song, Shipeng, Liang, Xiaoxia, Song, Meina, He, Guangyan, Tao, Jinhua, Chen, Liangfu
• Format: Article
• Language: English
• Subjects: Accuracy; Air pollution; Atmospheric modeling; Forecasting; gcForest; Graphical models; Ground-level NO<sub xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">2; Long short term memory; LSTM; Monitoring; nowcasting; Predictive models; Satellites; Spatial resolution; TROPOMI
• ISSN: 0196-2892
• DOI: 10.1109/TGRS.2024.3514158

Nitrogen dioxide (NO 2 ) is a critical air pollutant with significant health and environmental implications, particularly in urban areas where high levels of emissions are prevalent. Accurate nowcasting of ground-level NO 2 concentrations is essential for effective air quality management and timely public health interventions. Traditional methods often struggle with balancing the spatial accuracy of ensemble learning models and the temporal forecasting strengths of time-series models like Long Short-Term Memory (LSTM) networks. In this study, we propose an improved hybrid framework, GC-LSTM, to nowcast regional ground-level NO 2 concentrations on an hourly scale based on satellite-derived NO 2 Vertical Column Densities (VCDs), meteorological data, and on-site observations. GC-LSTM integrates the spatial learning capabilities of gcForest with the temporal prediction strengths of LSTM networks, leveraging the strengths of both spatial inference and time-series prediction. This study focuses on the Beijing-Tianjin-Hebei (BTH) region, one of China's most polluted areas, as a case study. Our results indicate that the GC-LSTM framework performs a strong correlation between predicted and observed ground-level NO 2 concentrations, with an R 2 of 0.746 and a Mean Absolute Percentage Error (MAPE) of 18.4% at a 1-hour prediction interval. Even as the prediction intervals extended to 2 and 3 hours, the GC-LSTM consistently outperforms the gcForest model across all evaluated metrics, with R 2 values higher by 0.097 and 0.117, and Root Mean Square Error (RMSE) values lower of 0.666 and 1.76μg/ m 3 than those nowcasted by using the standalone gcForest model, respectively, highlighting its robustness and adaptability. Furthermore, the capacity of GC-LSTM framework for continual learning and adaptation ensures its effectiveness in dynamic environments, making it a valuable tool for real-time air quality forecasting and environmental management.