the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Analysis of new particle nucleation events and comparisons to simulations of particle number concentrations based on GEOS-Chem/APM in Beijing, China
- 1Key Laboratory for Aerosol-Cloud-Precipitation of China Meteorological Administration, Nanjing University of Information Science & Technology, Nanjing 210044, China
- 2Atmospheric Sciences Research Center, University at Albany, Albany, NY, USA
- 1Key Laboratory for Aerosol-Cloud-Precipitation of China Meteorological Administration, Nanjing University of Information Science & Technology, Nanjing 210044, China
- 2Atmospheric Sciences Research Center, University at Albany, Albany, NY, USA
Abstract. Aerosol particles play important roles in air quality and global climate change. In this study, we analyzed the measurements of particle size distribution from March 12th to April 6th, 2016 in Beijing to characterize new particle formation (NPF) by using the observational data of sulfuric acid, meteorological parameters, solar radiation, and PM2.5 mass concentration. During this 26-day campaign, 11 new particle formation events were identified with obvious bursts of sub-3 nm particle number concentrations and subsequent growth of these nucleated particles. It is found that sulfuric acid concentration in Beijing did not have a significant difference between NPF and non-event days. Although the temperature during NPF days in Beijing was slightly higher than that on non-event days, temperature was not necessarily the key factor to determine NPF because higher solar radiation intensity usually increases the temperature. Low relative humidity (RH) and high daily total solar radiation appeared to be favorable to the occurrence of NPF events, which was more obvious in this campaign. A quantitative analysis indicated that more than 90 % of NPF events occur when the daily total solar radiation was greater than 19 MJ/m2/day and RH was less than 26.5 %. The PM2.5 mass concentration can also be used as a rough and simple criterion to predict the occurrence of NPF events. In addition, the simulations using four nucleation schemes, i.e., H2SO4-H2O binary homogeneous nucleation (BHN), H2SO4-H2O-NH3 ternary homogeneous nucleation (THN), H2SO4-H2O-ion binary ion-mediated nucleation (BIMN), and H2SO4-H2O-NH3-ion ternary ion-mediated nucleation (TIMN), based on a global chemistry transport model (GEOS-Chem) coupled with an advanced particle microphysics (APM) model, were conducted to study the particle number concentrations and new particle formation process. Our comparisons between measurements and simulations indicate that BHN scheme and BIMN scheme significantly underestimated the observed particle number concentrations, and the THN scheme captured well the total particle number concentration on most NPF event days but failed to capture the noticeable increase in particle number concentrations on March 18th and April 1st. TIMN scheme had obvious improvement in terms of total and sub-3 nm particle number concentrations and nucleation rates. This study provides a basis for further understanding of new particle nucleation mechanism in Beijing.
Kun Wang et al.
Status: final response (author comments only)
-
RC1: 'Comment on acp-2022-797', Anonymous Referee #1, 05 Dec 2022
This study reanalyzed an NPF dataset reported in previous literature and explored possible nucleation mechanisms by contrasting measurements to simulations. It is interesting to see an attempt to reproduce urban NPF events with a global chemistry transport model. The topic of this study fits the scope of Atmospheric Chemistry and Physics. I recommend the authors take advantage of the review process to improve the manuscript substantially, such that it can meet the quality for publication.
Major comments:
1. The TIMN scheme was found to be able to "overall well simulate the total and sub-3 nm particle number concentrations and nucleation rates in Beijing", with significantly higher values than BHN, BIHN, and THN. However, the nucleation rates in Fig. 10 seem to be lower than typical values in polluted megacities. For instance, the formation rates of freshly nucleated particles were usually higher than 10 cm-3 s-1 in Shanghai (Yao et al., 2018) and Beijing (Yan et al., 10.1029/2020GL091944). Since the rate of ion-mediated nucleation may be limited by the ion production rate, as well as the high sink, does this indicate that ion-mediated nucleation may not able to produce those high formation rates?
Besides, it will be more convincing to show the measured particle formation rate in Fig. 10.2. There might be a large room for improvement in the manuscript when addressing the current knowledge of NPF in terms of both nucleation mechanisms and their roles in the atmosphere. Some important advances in the last decade are missing from discussions. There are quite many places where the discussions are confusing and some are even at the risk of self-contradictory. Two examples are given below and some are given in minor comments, and I encourage the authors to improve the manuscript thoroughly.
- Amines can be a key base for sulfuric acid nucleation in polluted megacities for their much higher efficiency in stabilizing clusters than ammonia and ions, as has been discussed in Yao et al. (2018) and many other studies. The authors have cited Yao et al. (2018) but why not address the roles of amines in the simulation? Is it possible that sulfuric acid-amine nucleation can produce a comparable or higher nucleation rate than TIMN?
- The authors stated that LVOCs play an important role in NPF, which is plausibly true and consistent with previous findings in Beijing. However, it is also stated that "TIMN scheme has a good simulation performance on the growth, condensation, coagulation and other processes after the nucleation process." Have the LVOCs been accounted for in TIMN? If not, does this indicates either a negligible contribution from LVOCs or a bias in the simulation results?3. Most of the findings in the measurement part have been discussed in previous literature, which can also be seen from the discussions in the main text. The authors may need to clearly show the advances of this study compared to previous studies, including but not limited to the source of the dataset used in this study (Cai et al., 2017). Shortening the discussions, figures, and conclusions based on the measurement results can be an alternative way, and this will help emphasize the results based on simulations.
Minor comments:
4. Lines 55-56. This sentence is confusing because whether ions, specifically, charged particles measured by NAIS herein, grow faster than neutral particles or not is not directly relevant to the formation of the critical nucleus.
5. Lines 60-64. The mechanism proposed by Wu et al. (2020) is not a nucleation mechanism. It will be better to address it elsewhere, e.g., in lines 230-240.
6. Line 213, "7 % higher". It can be questionable to conclude the importance of temperature on NPF based on a 7 % difference.
7. Lines 296-297. Better to explain why and how a nucleation scheme can simulate the processes after nucleation.Technical comments:
8. Lines 40-41. It is worth double-checking whether Huang et al. (2020) and Li et al. (2021) concluded that "new particles derived from NPF played a significant role in the formation of haze".
9. Line 130. Better to use steady-state or quasi-steady-state. NPF cannot reach an equilibrium.
10. Line 138, "banana shape". This is perhaps not necessary since NPF events in urban environments may not be banana-type events.
11. Line 342. Please use "TIMN scheme" instead of "TIMN nucleation scheme", as N is for nucleation. -
RC2: 'Comment on acp-2022-797', Anonymous Referee #2, 26 Dec 2022
This study simulates NPF events in Beijing by applying GEOS-Chem/APM model, considering four nucleation mechanisms. It improved our understanding on nucleation and influence by meteorological factors. The TIMN nucleation scheme can predict nucleation well, however, more direct measurement data needed for validation the modeling results. The effect of meteorological conditions and precursors on nucleation should be further discussed in details. This paper is well organized and written. I recommend it can be accepted after the following major revisions.
Major issues:
- Line 113, The nucleation mechanism used in APM model is IMN parameterization scheme, which is developed based on the measurement and laboratory data elsewhere (Yu, 2006b). However, whether the parameterization is applicable in urban Beijing, with the high pollution level and unclear role that organics take place? Can you talk about the uncertainties or bias of the simulation result due to the four parameterizations?
- Figure 2, can you explain why there no clear difference of sulfuric acid concentration between NPF and non-NPF days. Table 1 is not necessary as only two numbers are given. It can be given in the text, and better to give the mean±standard deviation. In addition, as the authors mentioned, the sulfuric acid reported in this study is lower than the other studies, can you give the concentration level given by other studies? As SO2 decreased recent year in Beijing, the comparison should be conducted at the recent years, and also differed by seasons.
- Line 214, some studies reported that temperature can influence the NH3 stabilizing with sulfuric acid, which finally affect the nucleation rate. However, in this work, it can not be concluded the roles of temperature. In Beijing, NPF occurs more frequent in spring, winter than summer. The higher temperature on NPF days probably related with stronger solar radiation on clear days. It is difficult to evaluate the roles of temperature, as temperature, RH and solar radiation correlated under the similar synoptic conditions. As well as in line 228, I don’t think a simple metrological factor, RH or solar radiation, can explain the NPF reasonably (such as, high RH usually occurs under cloudy days with low solar radiation). The meteorological factors have systematically influence on NPF.
- Figure 7, can you explain why modeled RH is much lower than the observed value? It this reasonable with the model uncertainties? For example, on March 26 and 27, the bias can be 10 K.
- Figure 10, can you calculate the observed nucleation rate, as compared with the simulated nucleation rate. Figure 11, the vertical distribution of nucleation rate has large uncertainties, and even no vertical data can validate the model result. I don’t think it is robust confidence to represent the nucleation rates in the upper boundary layer in line 297.
- Can you model the sulfuric acid and validate the results by the measurement data? This can improve the confidence of model results.
Minor problems:
- Line 31-34, First, “new particle nucleation” is a repetitive phrase, normally we call this phenomenon “new particle formation”, which includes nucleation and growth process. For the second sentence, the nucleated particles undergone condensation and coagulation processes and grow into larger sizes. However, water absorption is an independent process that characterize the particle hygroscopicity, which should not be included in the new particle formation process.
- Line 99 and 101, for the data sources from website, the latest access time should be given.
- Figure 1, the contour plot of PNSD near the detection limit below 5 nm looks wired, it seems only data of NPF was given. How is the data on other days? It is zero or has been excluded from the dataset? The author should provide the details about how to handle the PNSD data.
- Line 213, 8 to 16:00 UTC or Local time, o’clock is not a formal written language.
- Table 2, the restricted conditions (RH and solar radiation) proposed for identifying NPF only based on one-month measurement data is not robust. Even at the same location, the criterion can be changed due to seasonal variation of meteorological factors.
- Figure 13, APM model can not capture the peaks of PM5 and PM10, especially the severe pollution episode from March 14 to 18, is this due to the model spatial resolution or the emission inventory uncertainty?
Kun Wang et al.
Kun Wang et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 243 | 62 | 10 | 315 | 4 | 5 |
- HTML: 243
- PDF: 62
- XML: 10
- Total: 315
- BibTeX: 4
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1