1. School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China 2. Shanghai Jinshan Environmental Monitoring Station, Shanghai 201500, China 3. School of Economics, Shanghai University, Shanghai 200237, China
● Used a double-stage attention mechanism model to predict ozone.
● The model can autonomously select the appropriate time series for forecasting.
● The model outperforms other machine learning models and WRF-CMAQ.
● We used the model to analyze the driving factors of VOCs that cause ozone pollution.
Ozone is becoming a significant air pollutant in some regions, and VOCs are essential for ozone prediction as necessary ozone precursors. In this study, we proposed a recurrent neural network based on a double-stage attention mechanism model to predict ozone, selected an appropriate time series for prediction through the input attention and temporal attention mechanisms, and analyzed the cause of ozone generation according to the contribution of feature parameters. The experimental data show that our model had an RMSE of 7.71 μg/m3 and a mean absolute error of 5.97 μg/m3 for 1-h predictions. The DA-RNN model predicted ozone closer to observations than the other models. Based on the importance of the characteristics, we found that the ozone pollution in the Jinshan Industrial Zone mainly comes from the emissions of petrochemical enterprises, and the good generalization performance of the model is proved through testing multiple stations. Our experimental results demonstrate the validity and promising application of the DA-RNN model in predicting atmospheric pollutants and investigating their causes.
R J Avery. (2006). Reactivity-based VOC control for solvent products: more efficient ozone reduction strategies. Environmental Science & Technology, 40( 16): 4845– 4850 https://doi.org/10.1021/es060296u
pmid: 16955876
3
M D Bas, J Ortiz, L Ballesteros, S Martorell. (2017). Evaluation of a multiple linear regression model and SARIMA model in forecasting 7Be air concentrations. Chemosphere, 177 : 326– 333 https://doi.org/10.1016/j.chemosphere.2017.03.029
pmid: 28319886
4
Y Bengio, P Simard, P Frasconi. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5( 2): 157– 166 https://doi.org/10.1109/72.279181
pmid: 18267787
5
D Byun, K L Schere. (2006). Review of the governing equations, computational algorithms, and other components of the Models-3 community multiscale air quality (CMAQ) modeling system. Applied Mechanics Reviews, 59( 2): 51– 77 https://doi.org/10.1115/1.2128636
6
Y Chen, V Phonevilay, J Tao, X Chen, R Xia, Q Zhang, K Yang, J Xiong, J Xie. (2021). The face image super-resolution algorithm based on combined representation learning. Multimedia Tools and Applications, 80( 20): 30839– 30861 https://doi.org/10.1007/s11042-020-09969-1
7
M Chin, P Ginoux, S Kinne, O Torres, B N Holben, B N Duncan, R V Martin, J A Logan, A Higurashi, T Nakajima. (2002). Tropospheric aerosol optical thickness from the GOCART model and comparisons with satellite and Sun photometer measurements. Journal of the Atmospheric Sciences, 59( 3): 461– 483 https://doi.org/10.1175/1520-0469(2002)059<0461:TAOTFT>2.0.CO;2
8
L Delle Monache, T Nipen, X X Deng, Y M Zhou, R Stull. (2006). Ozone ensemble forecasts: 2. A Kalman filter predictor bias correction. Journal of Geophysical Research, 111( D5): D05308 https://doi.org/10.1029/2005JD006311
9
Y Ding, Z Chen, W Lu, X Wang. (2021). A CatBoost approach with wavelet decomposition to improve satellite-derived high-resolution PM2.5 estimates in Beijing-Tianjin-Hebei. Atmospheric Environment, 249 : 118212 https://doi.org/10.1016/j.atmosenv.2021.118212
10
A Donnelly, B Misstear, B Broderick. (2015). Real time air quality forecasting using integrated parametric and non-parametric regression techniques. Atmospheric Environment, 103 : 53– 65 https://doi.org/10.1016/j.atmosenv.2014.12.011
11
G A Efstathiou, N M Zoumakis, D Melas, C J Lolis, P Kassomenos ( 2013). Sensitivity of WRF to boundary layer parameterizations in simulating a heavy rainfall event using different microphysical schemes : e ffect on large-scale processes. Atmospheric Research, 132– 133: 125– 143 https://doi.org/10.1016/j.atmosres.2013.05.004
12
M B Ek, K E Mitchell, Y Lin, E Rogers, P Grunmann, V Koren, G Gayno, J D Tarpley. (2003). Implementation of Noah land surface model advances in the national centers for environmental prediction operational mesoscale Eta model. Journal of Geophysical Research, 108( D22): 2002JD003296 https://doi.org/10.1029/2002JD003296
D Garaud, V Mallet. (2011). Automatic calibration of an ensemble for uncertainty estimation and probabilistic forecast: application to air quality. Journal of Geophysical Research, 116( D19): D19304 https://doi.org/10.1029/2011JD015780
15
B Z Ge, Z F Wang, X B Xu, J B Wu, X L Yu, J Li. (2014). Wet deposition of acidifying substances in different regions of China and the rest of East Asia: modeling with updated NAQPMS. Environmental Pollution, 187 : 10– 21 https://doi.org/10.1016/j.envpol.2013.12.014
pmid: 24418974
16
S Y Hong, Y Noh, J Dudhia. (2006). A new vertical diffusion package with an explicit treatment of entrainment processes. Monthly Weather Review, 134( 9): 2318– 2341 https://doi.org/10.1175/MWR3199.1
17
X Hu, J H Belle, X Meng, A Wildani, L A Waller, M J Strickland, Y Liu. (2017). Estimating PM2.5 concentrations in the conterminous United States using the random forest approach. Environmental Science & Technology, 51( 12): 6936– 6944 https://doi.org/10.1021/acs.est.7b01210
pmid: 28534414
18
L Hui, X Liu, Q Tan, M Feng, J An, Y Qu, Y Zhang, M Jiang. (2018). Characteristics, source apportionment and contribution of VOCs to ozone formation in Wuhan, Central China. Atmospheric Environment, 192 : 55– 71 https://doi.org/10.1016/j.atmosenv.2018.08.042
19
P Jia, N Cao, S Yang. (2021). Real-time hourly ozone prediction system for Yangtze River Delta area using attention based on a sequence to sequence model. Atmospheric Environment, 244 : 117917 https://doi.org/10.1016/j.atmosenv.2020.117917
C Liu, H Zhang, Z Cheng, J Shen, J Zhao, Y Wang, S Wang, Y Cheng. (2021). Emulation of an atmospheric gas-phase chemistry solver through deep learning: case study of Chinese mainland. Atmospheric Pollution Research, 12( 6): 101079 https://doi.org/10.1016/j.apr.2021.101079
23
E J Mlawer, S J Taubman, P D Brown, M J Iacono, S A Clough. (1997). Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. Journal of Geophysical Research, 102( D14): 16663– 16682 https://doi.org/10.1029/97JD00237
24
N I Molina-Gómez, J L Diaz-Arevalo, P A Lopez-Jimenez. (2021). Air quality and urban sustainable development: the application of machine learning tools. International Journal of Environmental Science and Technology, 18( 4): 1029– 1046 https://doi.org/10.1007/s13762-020-02896-6
25
J Ou, J Zheng, R Li, X Huang, Z Zhong, L Zhong, H Lin ( 2015). Speciated OVOC and VOC emission inventories and their implications for reactivity-based ozone control strategy in the Pearl River Delta region, China. Sci ence of the Total Environ ment, 530– 531: 393– 402 https://doi.org/10.1016/j.scitotenv.2015.05.062
pmid: 26057544
26
Y Qin, D Song, H Cheng, W Cheng, G Jiang, G W Cottrell. (2017). A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction. Melbourne, Australia : 2627– 2633 https://doi.org/10.48550/arXiv.1704.02971
27
M Reichstein, G Camps-Valls, B Stevens, M Jung, J Denzler, N Carvalhais. (2019). Deep learning and process understanding for data-driven Earth system science. Nature, 566( 7743): 195– 204 https://doi.org/10.1038/s41586-019-0912-1
pmid: 30760912
28
M Shao, Y Zhang, L Zeng, X Tang, J Zhang, L Zhong, B Wang. (2009). Ground-level ozone in the Pearl River Delta and the roles of VOC and NOx in its production. Journal of Environmental Management, 90( 1): 512– 518 https://doi.org/10.1016/j.jenvman.2007.12.008
pmid: 18207632
29
Q Tao F Liu Y Li D Sidorov ( 2019). Air pollution forecasting using a deep learning model based on 1D convnets and bidirectional GRU. IEEE Access: Practical Innovations, Open Solutions, 7: 76690– 76698 https://doi.org/10.1109/ACCESS.2019.2921578
30
R Vautard, P H J Builtjes, P Thunis, C Cuvelier, M Bedogni, B Bessagnet, C Honore, N Moussiopoulos, G Pirovano, M Schaap, R Stern, L Tarrason, P Wind. (2007). Evaluation and intercomparison of ozone and PM10 simulations by several chemistry transport models over four European cities within the CityDelta project. Atmospheric Environment, 41( 1): 173– 188 https://doi.org/10.1016/j.atmosenv.2006.07.039
31
B Wang, Q Yuan, Q Yang, L Zhu, T Li, L Zhang. (2021). Estimate hourly PM2.5 concentrations from Himawari-8 TOA reflectance directly using geo-intelligent long short-term memory network. Environmental Pollution, 271 : 116327 https://doi.org/10.1016/j.envpol.2020.116327
pmid: 33360654
32
H Wang, Z Xiang, L Wang, S Jing, S Lou, S Tao, J Liu, M Yu, L Li, L Lin, Y Chen, A Wiedensohler, C Chen. (2018). Emissions of volatile organic compounds (VOCs) from cooking and their speciation: A case study for Shanghai with implications for China. Science of the Total Environment, 621 : 1300– 1309 https://doi.org/10.1016/j.scitotenv.2017.10.098
pmid: 29054635
33
S Wang, C Fang, H Ma, Y Wang, J Qin. (2014). Spatial differences and multi-mechanism of carbon footprint based on GWR model in provincial China. Journal of Geographical Sciences, 24( 4): 612– 630 https://doi.org/10.1007/s11442-014-1109-z
34
P Y Wong, H J Su, H Y Lee, Y C Chen, Y P Hsiao, J W Huang, T A Teo, C D Wu, J D Spengler. (2021). Using land-use machine learning models to estimate daily NO2 concentration variations in Taiwan, China. Journal of Cleaner Production, 317 : 128411 https://doi.org/10.1016/j.jclepro.2021.128411
35
R Xia, Y Chen, B Ren ( 2022). Improved anti-occlusion object tracking algorithm using Unscented Rauch-Tung-Striebel smoother and kernel correlation filter. Journal of King Saud University - Computer and Information Sciences
36
X Xu, C Zhang, Y Liang. (2021). Review of satellite-driven statistical models PM2.5 concentration estimation with comprehensive information. Atmospheric Environment, 256 : 118302 https://doi.org/10.1016/j.atmosenv.2021.118302
37
H Zhang, J C Linford, A Sandu, R Sander. (2011). Chemical mechanism solvers in air quality models. Atmosphere (Toronto), 2( 3): 510– 532 https://doi.org/10.3390/atmos2030510
38
J M Zhang, W J Feng, T Y Yuan, J Wang, A K Sangaiah. (2022a). SCSTCF: spatial-channel selection and temporal regularized correlation filters for visual tracking. Applied Soft Computing, 118 : 108485 https://doi.org/10.1016/j.asoc.2022.108485
39
J M Zhang, J Sun, J Wang, Z P Li, X Chen. (2022b). An object tracking framework with recapture based on correlation filters and Siamese networks. Computers & Electrical Engineering, 98 : 107730 https://doi.org/10.1016/j.compeleceng.2022.107730
40
Y Zhang, X Y Wen, C J Jang. (2010). Simulating chemistry-aerosol-cloud-radiation-climate feedbacks over the continental US using the online-coupled Weather Research Forecasting Model with chemistry (WRF/Chem). Atmospheric Environment, 44( 29): 3568– 3582 https://doi.org/10.1016/j.atmosenv.2010.05.056
