Assessment of strict autumn-winter emission controls on air quality in the Beijing-Tianjin-Hebei region

Lu, Gongda; Marais, Eloise A.; Vu, Tuan V.; Xu, Jingsha; Shi, Zongbo; Lee, James D.; Zhang, Qiang; Shen, Lu; Luo, Gan; Yu, Fangqun

doi:https://doi.org/10.5194/acp-2021-428

Preprints

https://doi.org/10.5194/acp-2021-428

Preprints

02 Aug 2021

| 02 Aug 2021

Status: this preprint was under review for the journal ACP but the revision was not accepted.

Assessment of strict autumn-winter emission controls on air quality in the Beijing-Tianjin-Hebei region

Gongda Lu, Eloise A. Marais, Tuan V. Vu, Jingsha Xu, Zongbo Shi, James D. Lee, Qiang Zhang, Lu Shen, Gan Luo, and Fangqun Yu

Abstract. Strict seasonal emission controls are a popular measure in China for addressing severe air pollution, in particular fine particulate matter (PM_2.5). Here we evaluate the efficacy of these measures, with a particular focus on the strict emission controls imposed on pollution sources in 28 cities in and around the Beijing-Tianjin-Hebei region (BTH) in autumn-winter 2017/2018. For this we use the GEOS-Chem chemical transport model and air pollutant measurements from the national and Beijing local monitoring networks, after evaluating the network data with independent measurements and correcting large biases in the bottom-up emissions inventory. The network measurements are temporally consistent (r > 0.9 for PM_2.5 and r > 0.7 for gases) with the independent measurements, though with systematic differences of 5–17 % for nitrogen dioxide (NO₂) and 16–28 % for carbon monoxide (CO). The average decrease in monitoring network PM_2.5 in BTH in autumn-winter 2017/2018 relative to the previous year is 27 %, declining from 103 to 75 µg m⁻³. The regional decline in PM_2.5 in the model is 20 %, exceeding the regional target of 15 %. According to the model, pollution control measures led to decline in PM_2.5 precursor emissions of 0.27 Tg NO_x (as NO), 0.66 Tg sulfur dioxide (SO₂), 70 Gg organic carbon (OC), and 50 Gg black carbon (BC). We find though that these alone only lead to an 8 % decline in PM_2.5 and that interannual variability in meteorology accounts for more than half (57 %) the decline. This demonstrates that year-on-year comparisons are misleading for assessing the efficacy of air pollution measures and should be taken into consideration when extending such measures beyond BTH.

Received: 23 May 2021 – Discussion started: 02 Aug 2021

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Gongda Lu, Eloise A. Marais, Tuan V. Vu, Jingsha Xu, Zongbo Shi, James D. Lee, Qiang Zhang, Lu Shen, Gan Luo, and Fangqun Yu

Status: closed

RC1:
'Comment on acp-2021-428', Anonymous Referee #1, 17 Aug 2021
The manuscript evaluated emission controls on air quality in the BTH region with observation products and CTM simulations. The topic is of great importance to environmental policymakers. However, several issues should be addressed properly in the manuscript for publication in ACP.

General comments for the modification

The objectives and motivations do not seem clear in the manuscript (e.g., Estimating the emissions reduction for mitigation measures, reproducing the AW2017 case, or determining contributions of parameters to the air quality). The authors had better make your objectives and motivations articulate and explicit in the manuscript. The manuscript is also lacking in the implication of this work to report to the readers or scientific community. Accordingly, the results should be are congruent with the objectives and the implication of the study.

Second, validation is crucial for evaluating the effects of emission controls on air quality. The manuscript did not discuss the validation of NO_x, SO₂, CO precursors for the AW2017 simulation, although there are some comparisons of particulate matters. Thus, the authors need to compare the simulated gaseous species with observations (e.g., in-situ ground or satellite data).

Lastly, the authors had better reorganize the manuscript to strengthen the methodology (i.e., Adding a method section).

Specific comments for the modification

Lines 142-144: The manuscript did not mention what was utilized for NO₂ observation during the APHH campaign. Was it different with the Chemiluminescence detection system? If both measurements are not based on the same principle, the differences can be caused by instrumental sensitivity (as the authors mentioned). However, the different local sources at both network sites are also an important issue that cannot be ignored. There is ~3 km distance between them. The authors need to discuss it.

Lines 210-213: It is an important part of the methodology. The authors used scale factors of 1.5 for NO_x, 2.4 for CO, and 2.1-6.8 for SO₂ to conduct the CTM simulation for the AW2016 case. Were the spatially same factor applied? Also, no matter which (ground or satellite) observation data is used for the emission estimations, there are two crucial issues of i) nonlinearity between emissions and concentration of a species (e.g., NO₂) and ii) transfer between adjacent grid cells in the calculation. The authors need to clarify how the scale factors are derived (i.e., procedure). Furthermore, in particular, for the scale factor of NO_x, the authors need to explain how to treat the relation between observed NO₂ and the NO_x emissions (usually emitted as NO).

Lines 219 -222: It is well known that CO is a final product of NMVOC oxidations in many textbooks. So, it is not easy to agree that modeled CO is relatively unaffected by NMVOC emissions. The authors need to explain some reasons in the manuscript in terms of the lifetime of NMVOCs and their chemical evolution during long-range transport. The enhanced levels of CO would occur in other remote areas other than BTH regions through long-range transport.

Lines 231-235: I think there is a more important reason for the inconsistency. That is interference (e.g., HNO₃ and PANs) in the NO₂ chemiluminescence detection instrument equipped with a molybdenum converter, which converts NO₂ to NO. Here, the molybdenum converter also oxides NO_z (≈ HNO₃ + PANs) to NO under typically operational temperature 300 – 350 °C (refer to Winer et al., 1974 and Dunlea et al., 2007). Dunlea et al. reported the interference in the chemiluminescence detection accounting for up to 50% of ambient NO₂ Considering this issue, the correlation between the simulated and observed NO₂ would be better. In other words, the data points of NO₂ in Fig. 4 would shift to the left, and the intercept would decrease. The authors had better discuss and/or reanalyze it.

Lines 242 -243 & Figure 4: Although the scale factors of 2-7 were applied to grid cells somewhere (which was not specified in the manuscript, but probably around Shanxi province) in the MEIC SO₂ emission, the SO₂ concentrations were still significantly under-predicted. The under-predicted SO₂ concentrations can influence SO₂ and PM₅ in the BTH areas via the atmospheric chemical and physical processes (e.g., secondary aerosol formation and the transport to the BTH) because SO₂ has ~ 5 days lifetime. Accordingly, the estimation of the emission changes for the AW2017 simulation is probably hampered by low simulated SO₂. The authors had better discuss how to treat this issue in your estimate. Also, the authors need to present the results for the AW2017 case, similar to Fig. 4.

Lines 270-272 and Fig. 5: It is not easy to agree that the errors in the boundary layer dynamics are related to the overestimation of nitrate alone. The issue should also apply to sulfate and others. Therefore, the errors in the boundary layer would not be the main reason for the overestimation. It is reasonable to discuss the overestimation of nitrate in terms of understanding like a relationship between SO₂ and sulfate (as the authors mentioned). However, as shown in Figs. 4 and 5, the modeled NO₂ concentration (a precursor of nitrate) is underestimated while nitrate is overestimated. It is a logical contradiction. Thus, the authors need to re-examine the overestimation of nitrate, considering the 4^th comment pointed out by this reviewer.

Lines 285 – 298: The authors need to discuss a clear description of how to estimate the emissions fluxes for AW2017. It is also required to explain how to treat the nonlinearity between emissions and concentration in the estimation.

Lines 330-340: Zhang et al. (2010) mentioned “NH₃ emission varied greatly from city to city from HS1617 (AW2016 in this study) to HS1718 (AW2017). In some cities, NH₃ emissions were largely reduced, such as in Beijing (6.4%), Taiyuan (33%), and Zhengzhou (19.6%), while the NH₃ emissions showed increases in some other cities, such as Tianjin (5.0%), Shijiazhuang (0.2%) and Jinan (35.2%)”. These variations are not marginal. Also, some studies reported that the SO₂ and NO₂ emissions have a decreasing trend while atmospheric NH₃ experienced a significant increasing trend (Xia et al., 2016; Ge et al., 2019). If NH₃ emissions increase in your simulation for the AW2017 case, what change would be expected in the concentration of PM_2.5?

Minor comments for the modification

1: Provide information on the number of data in Figure 1.

Line 142: “<10%”. Clarify it, as for example, 0-10%, ~10%, or ~%.

The authors mentioned several grid points, for example, “seven grid squares” (Lines 213), “2 grid points” (Line 242), “13 grids” (Line 305), and “14 model grids” (Line 318). Clarify or leave out because readers cannot find out such information in the manuscript.

References

Dunlea, E. J., Herndon, S. C., Nelson, D. D., Volkamer, R. M., San Martini, F., Sheehy, P. M., Zahniser, M. S., Shorter, J. H., Wormhoudt, J. C., Lamb, B. K., Allwine, E. J., Gaffney, J. S., Marley, N. A., Grutter, M., Marquez, C., Blanco, S., Cardenas, B., Retama, A., Ramos Villegas, C. R., Kolb, C. E., Molina, L. T., and Molina, M. J.: Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban environment, Atmos. Chem. Phys., 7, 2691–2704, https://doi.org/10.5194/acp-7-2691-2007, 2007.

Ge, B., Xu, X., et al.: Role of ammonia on the feedback between AWC and inorganic aerosol formation during heavy pollution in the North China Plain, Earth and Space Science, 6, 1675-1693, 2019.

Xia, Y. M., Zhao, Y., & Nielsen, C. P.: Beneï¬ts of of China's efforts in gaseous pollutant control indicated by the bottomâup emissions and satellite observations 2000â2014. Atmos. Environ., 136, 43–53, 2016.

Winer, A. M., Peters, J. W., Smith, J. P., Pitts Jr., J. N.: Response of commercial chemiluminescence NO-NO₂ analyzers to other nitrogen containing compounds, Environ. Sci. Technol., 8, 1118-1121, 1974.
Citation: https://doi.org/10.5194/acp-2021-428-RC1
RC2: 'Comment on acp-2021-428', Anonymous Referee #2, 24 Aug 2021

This manuscript evaluates the impact of emission reduction policies in China on changes in winter PM2.5 concentrations. In particular, the authors evaluate the emission reduction and PM2.5 concentration change in the BTH area during 2017AW using the chemical transport model and various observational data. I expect that such an attempt will greatly help policymakers assess the impact of policy implementation not only in China but also in many polluted regions. However, in order for this manuscript to be published on ACP, various issues must be resolved. Some detailed comments are below:

- Lines 27-28: The observed PM2.5 concentration was significantly reduced compared to the expected target values (15%). It seems better to emphasize this part in comparison with observations rather than models.

- Lines 30-31: It is difficult to easily determine the effect of the emission reduction policy and the inter-annual variability of the meteorological field, respectively. Among the PM2.5 concentration reductions derived from the model (20%), if the effect of the emission reduction policy is 8%, does the remaining 12% mean the effect of the meteorological field?

- Lines 65-76: If MEE implemented a strong emission reduction policy during 2017AW, is there any data on the amount of emission reduction estimated by MEE? If any, it should be compared and discussed with the emission reductions assumed in this study.

- Lines 89-90: Authors should address the biases of bottom-up emissions inventories in more detail in the introduction part.

- Lines 108 – 120: It is recommended to organize the station information in a table and mention only those that require detailed explanation.

- Line 168: Different from the values ââshown in Figure 3. Are the emission control and target areas different?

- Line 199: Is emission reduction necessary during the two-month spin-up period before the reduction policy is implemented? Emissions during the spin-up period should also be mentioned.

- Lines 211-213: The authors scaled up MEIC emissions from observations and models for the 2016 AW period. However, the authors do not specifically mention the criteria for increasing NOx and CO emissions by 1.5 and 2.4 times, respectively (Line 203 does not provide such information). Although the authors uniformly increased NOx and CO, do the differences in model and observation appear uniformly across the entire domain? Limitations on this should be mentioned. Moreover, although SO2 concentrations would be underestimated in most regions, the increase in emissions was applied to only 7 model grids. Also, how is the number 2.1-6.8 times calculated?

- Lines 236, 244: The model underestimates despite the increase in emissions. The authors mention the positive bias of the monitoring network as one of the causes. Although there is a bias of two points compared to APHH in Fig. 2, it is not clear whether the value can represent the bias in the entire domain. Even SO2 is inconsistent. In addition, APHH and both observation points are located in urban, so they are greatly affected by mobile sources. Therefore, we cannot be sure that the difference between APHH and the two points represents the bias of CNEMN and BJENM.

- Lines 285-298: Are 2017AW emissions scaled up in the same way as 2016AW? It is not clearly described in the manuscript.

- Line 289 and Lines 315-319: The authors scaled up emissions outside of BTH, but did not change emissions outside of the area shown in Figure 3. Is there any clear reason for that? Are MEIC emissions underestimated only in BTH and its vicinity but are assumed to be similar elsewhere? As the authors mentioned in the manuscript, PM2.5 affects different regions through long-range transport, so the impact of fixed emissions outside the domain should be mentioned.

- Line 340: Since the influence of the meteorological field is also an important part of this study, it is recommended to present the changed PM2.5 concentration field as a 2-D map due to the interannual variability of the meteorological field. This will allow us to evaluate in more detail the impact of interannual variability in regional meteorological fields.

- Lines 346-348: The authors should highlight the significant differences between Zhang et al. (2021) and this study.

- Line 345: Zhang et al. (2019) evaluated the difference between the meteorological fields in December 2017 and December 2016. Therefore, the authors should also compare for the same period (December). This is because, even in winter, the meteorological field can have large fluctuations from month to month.

- Line 364: “PM2.5 in BTH decreased by 28% from 103 µg m-3 to 75 µg m-3 in the control period relative to the previous year”. However, in the abstract, it is described as “PM2.5 in BTH in autumn-winter 2017/2018 relative to the previous year is 27%, declining from 103 to 75 µg m-3”. The same value should be used for the same content.

- Lines 368-370: The authors used observations and models to evaluate emission reductions in the BTH region. However, there may be large errors in the scale-up assumed by the authors. Although there are difficulties to be derived only with limited data, the authors should evaluate the reliability of the presented numbers and make important comments about the effect of uncertainty.

- Line 371: How 8% is derived should be presented in more detail in Session 4.

- Line 737: Does "n" mean NMB?

Citation: https://doi.org/10.5194/acp-2021-428-RC2
AC1: 'AC1', Gongda Lu, 30 Oct 2021

Please find attached the point-by-point responses to reviewer comments, the manuscript with changes tracked and a supplement for the manuscript.

Citation: https://doi.org/10.5194/acp-2021-428-AC1

Status: closed

RC1:
'Comment on acp-2021-428', Anonymous Referee #1, 17 Aug 2021
The manuscript evaluated emission controls on air quality in the BTH region with observation products and CTM simulations. The topic is of great importance to environmental policymakers. However, several issues should be addressed properly in the manuscript for publication in ACP.

General comments for the modification

The objectives and motivations do not seem clear in the manuscript (e.g., Estimating the emissions reduction for mitigation measures, reproducing the AW2017 case, or determining contributions of parameters to the air quality). The authors had better make your objectives and motivations articulate and explicit in the manuscript. The manuscript is also lacking in the implication of this work to report to the readers or scientific community. Accordingly, the results should be are congruent with the objectives and the implication of the study.

Second, validation is crucial for evaluating the effects of emission controls on air quality. The manuscript did not discuss the validation of NO_x, SO₂, CO precursors for the AW2017 simulation, although there are some comparisons of particulate matters. Thus, the authors need to compare the simulated gaseous species with observations (e.g., in-situ ground or satellite data).

Lastly, the authors had better reorganize the manuscript to strengthen the methodology (i.e., Adding a method section).

Specific comments for the modification

Lines 142-144: The manuscript did not mention what was utilized for NO₂ observation during the APHH campaign. Was it different with the Chemiluminescence detection system? If both measurements are not based on the same principle, the differences can be caused by instrumental sensitivity (as the authors mentioned). However, the different local sources at both network sites are also an important issue that cannot be ignored. There is ~3 km distance between them. The authors need to discuss it.

Lines 210-213: It is an important part of the methodology. The authors used scale factors of 1.5 for NO_x, 2.4 for CO, and 2.1-6.8 for SO₂ to conduct the CTM simulation for the AW2016 case. Were the spatially same factor applied? Also, no matter which (ground or satellite) observation data is used for the emission estimations, there are two crucial issues of i) nonlinearity between emissions and concentration of a species (e.g., NO₂) and ii) transfer between adjacent grid cells in the calculation. The authors need to clarify how the scale factors are derived (i.e., procedure). Furthermore, in particular, for the scale factor of NO_x, the authors need to explain how to treat the relation between observed NO₂ and the NO_x emissions (usually emitted as NO).

Lines 219 -222: It is well known that CO is a final product of NMVOC oxidations in many textbooks. So, it is not easy to agree that modeled CO is relatively unaffected by NMVOC emissions. The authors need to explain some reasons in the manuscript in terms of the lifetime of NMVOCs and their chemical evolution during long-range transport. The enhanced levels of CO would occur in other remote areas other than BTH regions through long-range transport.

Lines 231-235: I think there is a more important reason for the inconsistency. That is interference (e.g., HNO₃ and PANs) in the NO₂ chemiluminescence detection instrument equipped with a molybdenum converter, which converts NO₂ to NO. Here, the molybdenum converter also oxides NO_z (≈ HNO₃ + PANs) to NO under typically operational temperature 300 – 350 °C (refer to Winer et al., 1974 and Dunlea et al., 2007). Dunlea et al. reported the interference in the chemiluminescence detection accounting for up to 50% of ambient NO₂ Considering this issue, the correlation between the simulated and observed NO₂ would be better. In other words, the data points of NO₂ in Fig. 4 would shift to the left, and the intercept would decrease. The authors had better discuss and/or reanalyze it.

Lines 242 -243 & Figure 4: Although the scale factors of 2-7 were applied to grid cells somewhere (which was not specified in the manuscript, but probably around Shanxi province) in the MEIC SO₂ emission, the SO₂ concentrations were still significantly under-predicted. The under-predicted SO₂ concentrations can influence SO₂ and PM₅ in the BTH areas via the atmospheric chemical and physical processes (e.g., secondary aerosol formation and the transport to the BTH) because SO₂ has ~ 5 days lifetime. Accordingly, the estimation of the emission changes for the AW2017 simulation is probably hampered by low simulated SO₂. The authors had better discuss how to treat this issue in your estimate. Also, the authors need to present the results for the AW2017 case, similar to Fig. 4.

Lines 270-272 and Fig. 5: It is not easy to agree that the errors in the boundary layer dynamics are related to the overestimation of nitrate alone. The issue should also apply to sulfate and others. Therefore, the errors in the boundary layer would not be the main reason for the overestimation. It is reasonable to discuss the overestimation of nitrate in terms of understanding like a relationship between SO₂ and sulfate (as the authors mentioned). However, as shown in Figs. 4 and 5, the modeled NO₂ concentration (a precursor of nitrate) is underestimated while nitrate is overestimated. It is a logical contradiction. Thus, the authors need to re-examine the overestimation of nitrate, considering the 4^th comment pointed out by this reviewer.

Lines 285 – 298: The authors need to discuss a clear description of how to estimate the emissions fluxes for AW2017. It is also required to explain how to treat the nonlinearity between emissions and concentration in the estimation.

Lines 330-340: Zhang et al. (2010) mentioned “NH₃ emission varied greatly from city to city from HS1617 (AW2016 in this study) to HS1718 (AW2017). In some cities, NH₃ emissions were largely reduced, such as in Beijing (6.4%), Taiyuan (33%), and Zhengzhou (19.6%), while the NH₃ emissions showed increases in some other cities, such as Tianjin (5.0%), Shijiazhuang (0.2%) and Jinan (35.2%)”. These variations are not marginal. Also, some studies reported that the SO₂ and NO₂ emissions have a decreasing trend while atmospheric NH₃ experienced a significant increasing trend (Xia et al., 2016; Ge et al., 2019). If NH₃ emissions increase in your simulation for the AW2017 case, what change would be expected in the concentration of PM_2.5?

Minor comments for the modification

1: Provide information on the number of data in Figure 1.

Line 142: “<10%”. Clarify it, as for example, 0-10%, ~10%, or ~%.

The authors mentioned several grid points, for example, “seven grid squares” (Lines 213), “2 grid points” (Line 242), “13 grids” (Line 305), and “14 model grids” (Line 318). Clarify or leave out because readers cannot find out such information in the manuscript.

References

Dunlea, E. J., Herndon, S. C., Nelson, D. D., Volkamer, R. M., San Martini, F., Sheehy, P. M., Zahniser, M. S., Shorter, J. H., Wormhoudt, J. C., Lamb, B. K., Allwine, E. J., Gaffney, J. S., Marley, N. A., Grutter, M., Marquez, C., Blanco, S., Cardenas, B., Retama, A., Ramos Villegas, C. R., Kolb, C. E., Molina, L. T., and Molina, M. J.: Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban environment, Atmos. Chem. Phys., 7, 2691–2704, https://doi.org/10.5194/acp-7-2691-2007, 2007.

Ge, B., Xu, X., et al.: Role of ammonia on the feedback between AWC and inorganic aerosol formation during heavy pollution in the North China Plain, Earth and Space Science, 6, 1675-1693, 2019.

Xia, Y. M., Zhao, Y., & Nielsen, C. P.: Beneï¬ts of of China's efforts in gaseous pollutant control indicated by the bottomâup emissions and satellite observations 2000â2014. Atmos. Environ., 136, 43–53, 2016.

Winer, A. M., Peters, J. W., Smith, J. P., Pitts Jr., J. N.: Response of commercial chemiluminescence NO-NO₂ analyzers to other nitrogen containing compounds, Environ. Sci. Technol., 8, 1118-1121, 1974.
Citation: https://doi.org/10.5194/acp-2021-428-RC1
RC2: 'Comment on acp-2021-428', Anonymous Referee #2, 24 Aug 2021

This manuscript evaluates the impact of emission reduction policies in China on changes in winter PM2.5 concentrations. In particular, the authors evaluate the emission reduction and PM2.5 concentration change in the BTH area during 2017AW using the chemical transport model and various observational data. I expect that such an attempt will greatly help policymakers assess the impact of policy implementation not only in China but also in many polluted regions. However, in order for this manuscript to be published on ACP, various issues must be resolved. Some detailed comments are below:

- Lines 27-28: The observed PM2.5 concentration was significantly reduced compared to the expected target values (15%). It seems better to emphasize this part in comparison with observations rather than models.

- Lines 30-31: It is difficult to easily determine the effect of the emission reduction policy and the inter-annual variability of the meteorological field, respectively. Among the PM2.5 concentration reductions derived from the model (20%), if the effect of the emission reduction policy is 8%, does the remaining 12% mean the effect of the meteorological field?

- Lines 65-76: If MEE implemented a strong emission reduction policy during 2017AW, is there any data on the amount of emission reduction estimated by MEE? If any, it should be compared and discussed with the emission reductions assumed in this study.

- Lines 89-90: Authors should address the biases of bottom-up emissions inventories in more detail in the introduction part.

- Lines 108 – 120: It is recommended to organize the station information in a table and mention only those that require detailed explanation.

- Line 168: Different from the values ââshown in Figure 3. Are the emission control and target areas different?

- Line 199: Is emission reduction necessary during the two-month spin-up period before the reduction policy is implemented? Emissions during the spin-up period should also be mentioned.

- Lines 211-213: The authors scaled up MEIC emissions from observations and models for the 2016 AW period. However, the authors do not specifically mention the criteria for increasing NOx and CO emissions by 1.5 and 2.4 times, respectively (Line 203 does not provide such information). Although the authors uniformly increased NOx and CO, do the differences in model and observation appear uniformly across the entire domain? Limitations on this should be mentioned. Moreover, although SO2 concentrations would be underestimated in most regions, the increase in emissions was applied to only 7 model grids. Also, how is the number 2.1-6.8 times calculated?

- Lines 236, 244: The model underestimates despite the increase in emissions. The authors mention the positive bias of the monitoring network as one of the causes. Although there is a bias of two points compared to APHH in Fig. 2, it is not clear whether the value can represent the bias in the entire domain. Even SO2 is inconsistent. In addition, APHH and both observation points are located in urban, so they are greatly affected by mobile sources. Therefore, we cannot be sure that the difference between APHH and the two points represents the bias of CNEMN and BJENM.

- Lines 285-298: Are 2017AW emissions scaled up in the same way as 2016AW? It is not clearly described in the manuscript.

- Line 289 and Lines 315-319: The authors scaled up emissions outside of BTH, but did not change emissions outside of the area shown in Figure 3. Is there any clear reason for that? Are MEIC emissions underestimated only in BTH and its vicinity but are assumed to be similar elsewhere? As the authors mentioned in the manuscript, PM2.5 affects different regions through long-range transport, so the impact of fixed emissions outside the domain should be mentioned.

- Line 340: Since the influence of the meteorological field is also an important part of this study, it is recommended to present the changed PM2.5 concentration field as a 2-D map due to the interannual variability of the meteorological field. This will allow us to evaluate in more detail the impact of interannual variability in regional meteorological fields.

- Lines 346-348: The authors should highlight the significant differences between Zhang et al. (2021) and this study.

- Line 345: Zhang et al. (2019) evaluated the difference between the meteorological fields in December 2017 and December 2016. Therefore, the authors should also compare for the same period (December). This is because, even in winter, the meteorological field can have large fluctuations from month to month.

- Line 364: “PM2.5 in BTH decreased by 28% from 103 µg m-3 to 75 µg m-3 in the control period relative to the previous year”. However, in the abstract, it is described as “PM2.5 in BTH in autumn-winter 2017/2018 relative to the previous year is 27%, declining from 103 to 75 µg m-3”. The same value should be used for the same content.

- Lines 368-370: The authors used observations and models to evaluate emission reductions in the BTH region. However, there may be large errors in the scale-up assumed by the authors. Although there are difficulties to be derived only with limited data, the authors should evaluate the reliability of the presented numbers and make important comments about the effect of uncertainty.

- Line 371: How 8% is derived should be presented in more detail in Session 4.

- Line 737: Does "n" mean NMB?

Citation: https://doi.org/10.5194/acp-2021-428-RC2
AC1: 'AC1', Gongda Lu, 30 Oct 2021

Please find attached the point-by-point responses to reviewer comments, the manuscript with changes tracked and a supplement for the manuscript.

Citation: https://doi.org/10.5194/acp-2021-428-AC1

Gongda Lu, Eloise A. Marais, Tuan V. Vu, Jingsha Xu, Zongbo Shi, James D. Lee, Qiang Zhang, Lu Shen, Gan Luo, and Fangqun Yu

Data sets

GEOS-Chem model outputs used in Lu et al.: Assessment of strict autumn-winter emission controls on air quality in the Beijing-Tianjin-Hebei region Gongda Lu https://github.com/GongdaLu/BTH_emission_control

Gongda Lu, Eloise A. Marais, Tuan V. Vu, Jingsha Xu, Zongbo Shi, James D. Lee, Qiang Zhang, Lu Shen, Gan Luo, and Fangqun Yu

Viewed

Total article views: 1,571 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
1,127	385	59	1,571	43	50

HTML: 1,127
PDF: 385
XML: 59
Total: 1,571
BibTeX: 43
EndNote: 50

Views and downloads (calculated since 02 Aug 2021)

Month	HTML	PDF	XML	Total
Aug 2021	302	83	10	395
Sep 2021	61	17	0	78
Oct 2021	70	23	2	95
Nov 2021	49	20	0	69
Dec 2021	33	17	1	51
Jan 2022	24	4	3	31
Feb 2022	19	10	2	31
Mar 2022	12	11	1	24
Apr 2022	33	11	0	44
May 2022	32	7	1	40
Jun 2022	35	3	1	39
Jul 2022	26	4	0	30
Aug 2022	12	7	0	19
Sep 2022	9	3	0	12
Oct 2022	8	3	1	12
Nov 2022	7	9	0	16
Dec 2022	15	6	0	21
Jan 2023	9	14	0	23
Feb 2023	11	7	2	20
Mar 2023	16	1	2	19
Apr 2023	7	6	0	13
May 2023	12	5	2	19
Jun 2023	5	5	0	10
Jul 2023	14	11	1	26
Aug 2023	13	5	0	18
Sep 2023	20	8	1	29
Oct 2023	11	4	1	16
Nov 2023	10	4	3	17
Dec 2023	17	8	2	27
Jan 2024	33	5	1	39
Feb 2024	17	6	1	24
Mar 2024	17	13	2	32
Apr 2024	23	6	6	35
May 2024	18	7	4	29
Jun 2024	40	5	3	48
Jul 2024	21	1	2	24
Aug 2024	16	3	2	21
Sep 2024	13	5	1	19
Oct 2024	14	9	0	23
Nov 2024	15	6	1	22
Dec 2024	8	3	0	11

Cumulative views and downloads (calculated since 02 Aug 2021)

Month	HTML	PDF	XML	Total
Aug 2021	302	83	10	395
Sep 2021	61	17	0	78
Oct 2021	70	23	2	95
Nov 2021	49	20	0	69
Dec 2021	33	17	1	51
Jan 2022	24	4	3	31
Feb 2022	19	10	2	31
Mar 2022	12	11	1	24
Apr 2022	33	11	0	44
May 2022	32	7	1	40
Jun 2022	35	3	1	39
Jul 2022	26	4	0	30
Aug 2022	12	7	0	19
Sep 2022	9	3	0	12
Oct 2022	8	3	1	12
Nov 2022	7	9	0	16
Dec 2022	15	6	0	21
Jan 2023	9	14	0	23
Feb 2023	11	7	2	20
Mar 2023	16	1	2	19
Apr 2023	7	6	0	13
May 2023	12	5	2	19
Jun 2023	5	5	0	10
Jul 2023	14	11	1	26
Aug 2023	13	5	0	18
Sep 2023	20	8	1	29
Oct 2023	11	4	1	16
Nov 2023	10	4	3	17
Dec 2023	17	8	2	27
Jan 2024	33	5	1	39
Feb 2024	17	6	1	24
Mar 2024	17	13	2	32
Apr 2024	23	6	6	35
May 2024	18	7	4	29
Jun 2024	40	5	3	48
Jul 2024	21	1	2	24
Aug 2024	16	3	2	21
Sep 2024	13	5	1	19
Oct 2024	14	9	0	23
Nov 2024	15	6	1	22
Dec 2024	8	3	0	11

Viewed (geographical distribution)

Total article views: 1,593 (including HTML, PDF, and XML) Thereof 1,593 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 13 Dec 2024

Short summary

Emission controls were imposed in Beijing-Tianjin-Hebei in northern China in autumn-winter 2017. We find that regional PM_2.5 targets (15 % decrease relative to previous year) were exceeded. Our analysis shows that decline in precursor emissions only leads to less than half (43 %) the improved air quality. Most of the change (57 %) is due to interannual variability in meteorology. Stricter emission controls may be necessary in years with unfavourable meteorology.


Total:	0
HTML:	0
PDF:	0
XML:	0