Manuscript ID: acp-2021-455 TITLE: Long-term trends and drivers of aerosol pH in eastern China

Comments to the Author Major Comments 1. The estimates ALWCo seem unreasonably small (lines 120 127)? How was organic aerosol measured? Was it PM2.5 as well, or was it PM1? Response: The concentration of organic aerosol was estimated by multiplying the measured concentration of organic carbon by a factor of 1.6 (Turpin and Lim, 2001). A Thermal/Optical Carbon Aerosol Analyzer (model RT-4, Sunset laboratory Inc.) equipped with a PM2.5 cyclone was used for the organic carbon measurement. The annual concentrations of organic carbon in Shanghai were 5.6– 10.6μg/m from 2011 to 2019, and the relative humidity were 69-75%. ALWCo was calculated by the following equation (Guo et al., 2015).

where is the mass concentration of organic aerosol, is the density of water ( =1.0g/cm 3 ), is the mean density of organics assumed to be 1.4g/cm 3 ) , and is the hygroscopicity parameter of organic aerosol ( = 0.087) (Li et al., 2016). Adopting these values, we that NVCs had gone down. However, it took far too much time to interpret and is still not easily understandable even after spending much time on it. The convention used by Tao and Murphy (2021) is much clearer -I suggest edits to follow their approach.
Response: Thanks for the comment. To study the driving factors of aerosol pH, different sensitivity analysis methods have been used in previous studies (Ding et al., 2019;Tao and Murphy, 2021;Zheng et al., 2020). The convention used in Tao and Murphy (2021) defined the base scenario as the average condition, aiming at illustrating the contribution of different factors to the deviation from the base scenario. However, the base scenario can change with the analysis time periods. In comparison, our bar plot here aimed at showing the factor contribution of the ΔpH between two adjacent scenarios (i.e., two continuous years or two continuous hour periods), and is not subject to change in the average conditions.
That is, our plots emphasized differently with that used in Tao and Murphy (2021). We've clarified this in the revised figure captions. In addition, to provide more viewpoints, we've added the figures with Tao and Murphy's approach in the supplement following the reviewer's suggestion.

Changes in manuscript:
(1) Line 204-206: " Figure 1b shows the contributions of individual factors to the ΔpH from 2011 to 2019. Here the bar plots indicate the factors contributing to the ΔpH between two adjacent scenarios, e.g., 2011 to 2013. See Fig. S9a for the factor contribution to the variation from average conditions." (2) Line 241-243: " Figure 3 shows the contributions of individual factors to the ΔpH across the four seasons. Here the bar plots indicate the factors contributing to the ΔpH between two adjacent seasons, e.g., spring (MAM) to summer (JJA). See Fig. S9b for the factor contribution to the variation from average conditions." (3) Line 273-275: " Figure 5 shows the effects of individual factors to the ΔpH between day and night.
Here the bar plots indicate the factors contributing to the ΔpH between two adjacent hour periods, e.g., 0:00 to 6:00. See Fig. S9c for the factor contribution to the variation from average conditions." We've revised 3. Discussion about the limited effects of future emissions control measures on haze pollution (e.g., line 35-36, 298-299) is just wrong. Although the partitioning of NH3 and HNO3 may shift towards the particulate phase in the future, it does not mean their total PM concentration has increased. If the total concentration (i.e., NH3 + NH4 + ) decreased enough, then a shift in partitioning towards the particle phase could still occur with a decrease in the aerosol NH4 + . This discussion would be much better with associated predictions of the PM2.5, NH4 + , SO4 2-, and NO3-aerosol concentrations.
Response: Thanks for the comment. We agree that the precursor decrease will finally lead to a PM decrease. Here we are discussing about the efficiency of PM reduction concentrations against the precursor reduction concentrations. To further clarify our points, we've revised the corresponding manuscript and figures with more detailed explanations. In addition, we've added the prediction of the changes in major chemical components (NH4 + , SO4 2-, NO3and Cl -) as Fig. 6g-i following the reviewer's suggestion. See detailed modifications below.

Changes in manuscript:
(1) Line 34-38: We've revised the statement into: "The corresponding aerosol pH in eastern China is estimated to increase by ~0.9, and the reduction in particle phase NO3and NH4 + is less than the reduced amount of total HNO3 and total NH3. This suggests a reduced benefit of NH3 and NOx emission control in mitigating haze pollution in eastern China." (2) Discussions in section 3.4: We've revised Fig. 6 and the corresponding discussions into (Line 320-366 in the revised section 3.4): "Under the reference scenario of SSP3-70-BAU with weak control policy (blue lines in Fig. 6 a-f), SO2 and NOx are predicted to increase, while the NHx is relatively stable. Correspondingly, both SO4 2and NO3will increase, and NH4 + will also increase in response (Fig. 6g). Considering the stable NHx, NH4 + partition ratio (NH4 + / (NH4 + + NH3)) will increase. In comparison, there is little change in aerosol pH and the predicted NO3partition ratio (NO3 -/ (NO3 -+ HNO3)).
Under the moderate control policy (SSP2-45-ECP), the emissions of SO2, NOx, and NH3 in 2050 will be reduced by 62.7%, 49.0% and 25.0%, respectively. Correspondingly, SO 4 2-, NO 3and NH4 + will all decrease (Fig. 6h), with a total PM reduction of ~14.4 μg m -3 . Moreover, the predicted pH will increase by ~0.5, and the NO3and NH4 + partition ratios will decrease by 0.14 and 0.23, respectively (green lines in Fig. 6d-f). That is, more nitrate and ammonium will exist in the gas phase as HNO3 and NH3, thus the reduced NH4 + and NO3is higher than the reduced NHx and TNO3, which is a control bonus in terms of reduced PM per reduced emissions for this scenario.
With the strict control policy (SSP1-26-BHE), the emissions of SO2, NOx and NH3 in 2050 will decrease by 86.9%, 74.9% and 41.7%, respectively. Its effect on PM reductions resembles that of the moderate one (SSP2-45-ECP) before 2040. Afterwards, however, the NO3partition ratio increased despite the increasing pH, and reached near 1 in 2050 ( Fig. 6 d, e). On second check, we found this pattern is due to the sharp decrease in SO4 2and constant NVCs. After 2040, there will be a major anion deficit considering the non-volatile species only (sulfate and Ca 2+ , K + , Mg 2+ ), and therefore more NO 3will be captured by the NVCs to the particle phase. As a result, NO 3 − partition ratio even increased from 0.92 in 2015 to 1.00 in 2050. Although NH4 + partition ratio showed a continuous decrease, in 2050 both the reduced NH 4 + and NO 3is smaller than the reduced NHx and TNO3 (Fig. 6i). That is in contrast with the effect of the moderate one (SSP2-45-ECP).
Correspondingly, the total reduced PM is only slightly larger for the strict SSP1-26-BHE policy

The Conclusions section needs substantial revision. A brief recap is ok, but Section 4 is mostly
redundant with the prior section. Rather than just reiterating what has already been said, more discussion of the significance of the work is warranted.
Response: Thanks for the comment. We rewrote the conclusions section. Please see the following changes.

Changes in manuscript:
Line 369 Finally, to explore the effects of China's future anthropogenic emission control pathways on aerosol pH and compositions, we chose three different emission reduction scenarios proposed by Tong et al.(2020) for future haze mitigation, naming SSP3-70-BAU, SSP2-45-ECP and SSP1-26-BHE as case studies. We estimated that the future trend of aerosol pH and NO3partition ratio will change little under the weak control policy (SSP3-70-BAU), while SO4 2-, NO3and NH4 + will increase substantially. The results also demonstrate that future aerosol pH will increase under both strict control policy (SSP1-26-BHE) and moderate control policy (SSP2-45-ECP), but more drastically under former scenario. The significant increase in aerosol pH with the strict control policy will lead to the reduced aerosol NH4 + and NO3is smaller than the reduced amount of total NH3 and total HNO3, which is in contrast with effect of the moderate control policy. This suggests that a reduced efficiency in terms of PM controls in responses to the emission controls with the strict control policy. These results highlight the importance of proportional reductions in precursors and follow-up variations in aerosol pH in future pollution control policy." 5. Finally, the entire manuscript needs to be edited for language consistency -specifically, verb tenses change within and between paragraphs. There are too many instances to list here.

Response:
Thanks for the comment. The language consistency in the manuscript has been polished, please see the modifications in the revised manuscript.
Technical/Minor Comments may have minor effects on ion balance. Meanwhile, we also find that the average equivalent ratios of cation/anion(C/A) were close to unity in many cities of China (Huang et al., 2014;Shen et al., 2010;Sun et al., 2006;Zhang et al., 2018). In our study, the modelled and measured NH3 and NH4 + concentrations were in good agreement based on observed aerosol composition, further indicating that the measurement of the ions was accurate. We rewrote this sentence as: "The correlation between cation and anion was strong (R 2 =0.94), with a slope of 1.00, indicating that these ion species were charge balanced and well represented major components in PM2.5." (16) Line 314-315: We rewrote this sentence into: "SSP3-70-BAU is a reference scenario that without additional efforts to constrain emissions." (17) Line 330-333: We edited this sentence into: "Moreover, the predicted pH will increase by ~0.5, and the NO3and NH4 + partition ratios will decrease by 0.14 and 0.23, respectively (green lines in Fig.   6d-f)."