Back to Search
Start Over
Estimation of daily NO2 with explainable machine learning model in China, 2007–2020.
- Source :
-
Atmospheric Environment . Dec2023, Vol. 314, pN.PAG-N.PAG. 1p. - Publication Year :
- 2023
-
Abstract
- Surface nitrogen dioxide (NO 2) is an effective indicator of anthropogenic combustion and is associated with regional burden of disease. Though satellite-borne column NO 2 is widely used to acquire surface concentration through the integration of sophisticated models, long-term and full-coverage estimation is hindered by the incomplete retrieval of satellite data. Moreover, the mechanical relationship between surface and tropospheric NO 2 is often ignored in the context of machine learning (ML) approach. Here we develop a gap-filling method to obtain full-coverage column NO 2 by fusing satellite data from different sources. The surface NO 2 is then estimated during 2007–2020 in China using the XGBoost model, with daily out-of-sample cross-validation (CV) R2 of 0.75 and root-mean-square error (RMSE) of 9.11 μg/m3. The back-extrapolation performance is verified through by-year CV (daily R2 = 0.60 and RMSE = 11.46 μg/m3) and external estimations in Taiwan before 2013 (daily R2 = 0.69 and RMSE = 8.59 μg/m3). We explore the variable impacts in three hotspots of eastern China through SHAP (Shapley additive explanation) values. We find the driving contributions of column NO 2 to the variation of ground pollution during 2007–2020 (average SHAP = 5.09 μg/m3 compared with the baseline concentration of 33.39 μg/m3). The estimated effect is also compared with ordinary least squares (OLS) model to provide a straightforward understanding. We demonstrate the employment of explainable ML model is beneficial to comprehend the coupled relationship in surface NO 2 change. • A gap-filling method is proposed to obtain full-coverage column NO 2. • Multi-source data are employed to extrapolate surface NO 2 in China during 2007–2020. • Column NO 2 is the major contributor to surface NO 2 variation in high-pollution areas. • The explainable machine learning model is favorable to pointing out the variable contribution at local scale. [ABSTRACT FROM AUTHOR]
- Subjects :
- *MACHINE learning
*NITROGEN dioxide
Subjects
Details
- Language :
- English
- ISSN :
- 13522310
- Volume :
- 314
- Database :
- Academic Search Index
- Journal :
- Atmospheric Environment
- Publication Type :
- Academic Journal
- Accession number :
- 173032116
- Full Text :
- https://doi.org/10.1016/j.atmosenv.2023.120111