Deep Learning for Prediction of the Air Quality Response to Emission Changes

Xing, Jia; Zheng, Shuxin; Ding, Dian; Kelly, James T.; Wang, Yafei; Li, Siwei; Qin, Tao; Ma, Mingyuan; Dong, Zhaoxin; Jang, Carey; Zhu, Yun; Zheng, Haotian; Ren, Lu; Liu, Tie-Yan; Hao, Jiming

doi:10.1021/acs.est.0c02923

Cited by 82 publications

(49 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, deep-learning technology was used to fit response surfaces for the 3 months in 2019 and 2020 using CMAQ simulations for baseline and zero-out emissions conditions (see Fig. 2 in Xing et al, 2020b). The response surfaces were developed using yearspecific meteorology based on WRF simulations to account for differences in meteorological conditions between 2019 and 2020.…”

Section: Response Model To Estimate the Actual Emissions From Observementioning

confidence: 99%

Quantifying the emission changes and associated air quality impacts during the COVID-19 pandemic on the North China Plain: a response modeling study

Xing

Jiang

et al. 2020

Atmos. Chem. Phys.

Self Cite

View full text Add to dashboard Cite

Abstract. Quantification of emission changes is a prerequisite for the assessment of control effectiveness in improving air quality. However, the traditional bottom-up method for characterizing emissions requires detailed investigation of emissions data (e.g., activity and other emission parameters) that usually takes months to perform and limits timely assessments. Here we propose a novel method to address this issue by using a response model that provides real-time estimation of emission changes based on air quality observations in combination with emission-concentration response functions derived from chemical transport modeling. We applied the new method to quantify the emission changes on the North China Plain (NCP) due to the COVID-19 pandemic shutdown, which overlapped the Spring Festival (also known as Chinese New Year) holiday. Results suggest that the anthropogenic emissions of NO2, SO2, volatile organic compound (VOC) and primary PM2.5 on the NCP were reduced by 51 %, 28 %, 67 % and 63 %, respectively, due to the COVID-19 shutdown, indicating longer and stronger shutdown effects in 2020 compared to the previous Spring Festival holiday. The reductions of VOC and primary PM2.5 emissions are generally effective in reducing O3 and PM2.5 concentrations. However, such air quality improvements are largely offset by reductions in NOx emissions. NOx emission reductions lead to increases in O3 and PM2.5 concentrations on the NCP due to the strongly VOC-limited conditions in winter. A strong NH3-rich condition is also suggested from the air quality response to the substantial NOx emission reduction. Well-designed control strategies are recommended based on the air quality response associated with the unexpected emission changes during the COVID-19 period. In addition, our results demonstrate that the new response-based inversion model can well capture emission changes based on variations in ambient concentrations and thereby illustrate the great potential for improving the accuracy and efficiency of bottom-up emission inventory methods.

show abstract

Section: Response Model To Estimate the Actual Emissions From Observementioning

confidence: 99%

Quantifying the emission changes and associated air quality impacts during the COVID-19 pandemic on the North China Plain: a response modeling study

Xing

Jiang

et al. 2020

Atmos. Chem. Phys.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The framework of the response model is illustrated in Figure 1. We conduct chemical transport model simulations using prior emissions to get the original simulated concentrations of six pollutants (i.e., NO2; O3; SO2; PM2.5; sulfate, SO4 2-; and nitrate, NO3 -), as well as the response functions derived from the RSM (Xing et al, 2011;Xing et al, 2017;2018).…”

Section: Response Model To Estimate the Actual Emissions From Observementioning

confidence: 99%

Quantifying the emission changes and associated air quality impacts during the COVID-19 pandemic in North China Plain: a response modeling study

Xing¹,

Li²,

Jiang³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. Quantification of emission changes is a prerequisite for the assessment of control effectiveness in improving air quality. However, the traditional bottom-up method for characterizing emissions requires detailed investigation of emissions data (e.g., activity and other emission parameters) that usually takes months to perform and limits timely assessments. Here we propose a novel method to address this issue by using a response model that provides real-time estimation of emission changes based on air quality observations in combination with emission-concentration response functions derived from chemical transport modeling. We applied the new method to quantify the emission changes in the North China Plain (NCP) due to the COVID-19 pandemic shutdown, which overlapped the Spring Festival holiday. Results suggest that the anthropogenic emissions of NO2, SO2, VOC, and primary PM2.5 in NCP were reduced by 51 %, 28 %, 67 % and 63 %, respectively, due to the COVID-19 shutdown, indicating longer and stronger shutdown effects in 2020 compared to the previous Spring Festival holiday. The reductions of VOC and primary PM2.5 emissions are generally effective in reducing O3 and PM2.5 concentrations. However, such air quality improvements are largely offset by reductions in NOx emissions. NOx emission reductions lead to increases in O3 and PM2.5 concentrations in NCP due to the strongly VOC-limited conditions in winter. A strong NH3-rich condition is also suggested from the air quality response to the substantial NOx emission reduction. Well-designed control strategies are recommended based on the air quality response associated with the unexpected emission changes during the COVID-19 period. In addition, our results demonstrate that the new response-based inversion model can well capture emission changes based on variations in ambient concentrations, and thereby illustrate the great potential for improving the accuracy and efficiency of bottom-up emission inventory methods.

show abstract

“…Advanced machine learning techniques enable its fast application across any time period and spatial location [17]. Different from inversion modeling, the RSM modifies anthropogenic emissions of five pollutants at the regionally aggregated level (by city in this study) based on the assumption that the spatial distribution of emissions is relatively accurate compared to emission magnitudes.…”

Section: Introductionmentioning

confidence: 99%

“…The RSM can identify the emission control factors needed to meet air quality targets, and thus provides information on the changes in emissions of multiple pollutants needed to improve air quality predictions against monitoring data [ 14 – 15 ]. Advanced machine learning techniques enables its fast application across any time period and spatial location [ 16 ]. Different from inversion modeling, the RSM modifies anthropogenic emissions of five pollutants at the regionally aggregated level (by city in this study) based on an assumption that the spatial distribution of emissions is relatively accurate compared with emission magnitudes.…”

Section: Introductionmentioning

confidence: 99%

Data Assimilation of Ambient Concentrations of Multiple Air Pollutants Using an Emission-Concentration Response Modeling Framework

Xing

Ding

et al. 2020

Atmosphere

Self Cite

View full text Add to dashboard Cite

Data assimilation for multiple air pollutant concentrations has become an important need for modeling air quality attainment, human exposure, and related health impacts, especially in China, which experiences both PM2.5 and O3 pollution. Traditional data assimilation or fusion methods are mainly focused on individual pollutants and thus cannot support simultaneous assimilation for both PM2.5 and O3. To fill the gap, this study proposed a novel multipollutant assimilation method by using an emission-concentration response model (noted as RSM-assimilation). The new method was successfully applied to assimilate precursors for PM2.5 and O3 in the 28 cities of the North China Plain (NCP). By adjusting emissions of five pollutants (i.e., NOx, sulfur dioxide = SO2, ammonia = NH3, VOC, and primary PM2.5) in the 28 cities through RSM-assimilation, the RMSEs (root mean square errors) of O3 and PM2.5 were reduced by about 35% and 58% from the original simulations. The RSM-assimilation results in small sensitivity to the number of observation sites due to the use of prior knowledge of the spatial distribution of emissions; however, the ability to assimilate concentrations at the edge of the control region is limited. The emission ratios of five pollutants were simultaneously adjusted during the RSM-assimilation, indicating that the emission inventory may underestimate NO2 in January, April, and October, and SO2 in April, but overestimate NH3 in April, and VOC in January and October. Primary PM2.5 emissions were also significantly underestimated, particularly in April (dust season in NCP). Future work should focus on expanding the control area and including NH3 observations to improve the RSM-assimilation performance and emission inventories.

show abstract

Deep Learning for Prediction of the Air Quality Response to Emission Changes

Cited by 82 publications

References 46 publications

Quantifying the emission changes and associated air quality impacts during the COVID-19 pandemic on the North China Plain: a response modeling study

Quantifying the emission changes and associated air quality impacts during the COVID-19 pandemic on the North China Plain: a response modeling study

Quantifying the emission changes and associated air quality impacts during the COVID-19 pandemic in North China Plain: a response modeling study

Data Assimilation of Ambient Concentrations of Multiple Air Pollutants Using an Emission-Concentration Response Modeling Framework

Contact Info

Product

Resources

About