Development of a recurrent spatiotemporal deep-learning method coupled with data fusion for correction of hourly ozone forecasts
Copyright © 2023 Elsevier Ltd. All rights reserved..
Ambient ozone (O3) predictions can be very challenging mainly due to the highly nonlinear photochemistry among its precursors, and meteorological conditions and regional transport can further complicate the O3 formation processes. The emission-based chemical transport models (CTM) are broadly used to predict O3 formation, but they may deviate from observations due to input uncertainties such as emissions and meteorological data, in addition to the treatment of O3 nonlinear chemistry. In this study, an innovative recurrent spatiotemporal deep-learning (RSDL) method with model-monitor coupled convolutional recurrent neural networks (ConvRNN) has been developed to improve O3 predictions of CTM. The RSDL method was first used to build the ConvRNN within a 24-h scale to characterize the spatiotemporal relationships between the monitored O3 data and CTM simulations, and then incorporated the recurrent pattern to achieve 72-h multi-site forecasts based on a pilot study over the Pearl River Delta (PRD) region of China. The results showed that the RSDL method predicted O3 with high accuracy over this case study, with an increase of 27.54% in the correlation coefficient (R) average for all sites as well as an increase in R of 0.14-0.21 for all cities compared to CTM. Moreover, the regional distribution of CTM was further improved by the RSDL predictions with the data fusion technique, which greatly reduced the underpredictions of O3 concentrations, particularly in high O3-level areas (concentrations >160 μg/m3), with a 33.55% reduction in the mean absolute error (MAE).
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:335 |
|---|---|
| Enthalten in: |
Environmental pollution (Barking, Essex : 1987) - 335(2023) vom: 15. Okt., Seite 122291 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Li, Jie [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
66H7ZZK23N |
|---|
| Anmerkungen: |
Date Completed 18.09.2023 Date Revised 18.09.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1016/j.envpol.2023.122291 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM360275338 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM360275338 | ||
| 003 | DE-627 | ||
| 005 | 20231226082652.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.envpol.2023.122291 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1200.xml |
| 035 | |a (DE-627)NLM360275338 | ||
| 035 | |a (NLM)37527757 | ||
| 035 | |a (PII)S0269-7491(23)01293-9 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Li, Jie |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Development of a recurrent spatiotemporal deep-learning method coupled with data fusion for correction of hourly ozone forecasts |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 18.09.2023 | ||
| 500 | |a Date Revised 18.09.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Copyright © 2023 Elsevier Ltd. All rights reserved. | ||
| 520 | |a Ambient ozone (O3) predictions can be very challenging mainly due to the highly nonlinear photochemistry among its precursors, and meteorological conditions and regional transport can further complicate the O3 formation processes. The emission-based chemical transport models (CTM) are broadly used to predict O3 formation, but they may deviate from observations due to input uncertainties such as emissions and meteorological data, in addition to the treatment of O3 nonlinear chemistry. In this study, an innovative recurrent spatiotemporal deep-learning (RSDL) method with model-monitor coupled convolutional recurrent neural networks (ConvRNN) has been developed to improve O3 predictions of CTM. The RSDL method was first used to build the ConvRNN within a 24-h scale to characterize the spatiotemporal relationships between the monitored O3 data and CTM simulations, and then incorporated the recurrent pattern to achieve 72-h multi-site forecasts based on a pilot study over the Pearl River Delta (PRD) region of China. The results showed that the RSDL method predicted O3 with high accuracy over this case study, with an increase of 27.54% in the correlation coefficient (R) average for all sites as well as an increase in R of 0.14-0.21 for all cities compared to CTM. Moreover, the regional distribution of CTM was further improved by the RSDL predictions with the data fusion technique, which greatly reduced the underpredictions of O3 concentrations, particularly in high O3-level areas (concentrations >160 μg/m3), with a 33.55% reduction in the mean absolute error (MAE) | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Air quality modeling | |
| 650 | 4 | |a Data fusion | |
| 650 | 4 | |a Deep learning | |
| 650 | 4 | |a Ozone photochemistry | |
| 650 | 4 | |a Ozone predictions | |
| 650 | 7 | |a Ozone |2 NLM | |
| 650 | 7 | |a 66H7ZZK23N |2 NLM | |
| 650 | 7 | |a Air Pollutants |2 NLM | |
| 700 | 1 | |a Jang, Ji-Cheng |e verfasserin |4 aut | |
| 700 | 1 | |a Zhu, Yun |e verfasserin |4 aut | |
| 700 | 1 | |a Lin, Che-Jen |e verfasserin |4 aut | |
| 700 | 1 | |a Wang, Shuxiao |e verfasserin |4 aut | |
| 700 | 1 | |a Xing, Jia |e verfasserin |4 aut | |
| 700 | 1 | |a Dong, Xinyi |e verfasserin |4 aut | |
| 700 | 1 | |a Li, Jinying |e verfasserin |4 aut | |
| 700 | 1 | |a Zhao, Bin |e verfasserin |4 aut | |
| 700 | 1 | |a Zhang, Bingyao |e verfasserin |4 aut | |
| 700 | 1 | |a Yuan, Yingzhi |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Environmental pollution (Barking, Essex : 1987) |d 1987 |g 335(2023) vom: 15. Okt., Seite 122291 |w (DE-627)NLM087741504 |x 1873-6424 |7 nnns |
| 773 | 1 | 8 | |g volume:335 |g year:2023 |g day:15 |g month:10 |g pages:122291 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.envpol.2023.122291 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 335 |j 2023 |b 15 |c 10 |h 122291 | ||