Modeling analysis of the seasonal characteristics of haze formation in Beijing

The air quality modeling system RAMS-CMAQ (Regional Atmospheric Modeling System–Community Multiscale Air Quality), coupled with an aerosol optical property scheme, was applied to simulate the meteorological field, major aerosol components (sulfate, nitrate, ammonium, black carbon, organic carbon, dust, and sea salt), and surface visibility over the North China Plain (NCP) in 2011. The modeled results in February and July 2011 were selected and analyzed to obtain an in-depth understanding of the haze formation mechanism in Beijing for different seasons. The simulation results showed that the visibility was below 10 km for most regions of the NCP, and dropped to less than 5 km over the megacities of Beijing and Tianjin, the whole of Hebei Province, and the northwest part of Shandong Province during pollution episodes in February and July. The heavy mass concentration of PM 2.5 ranged from 120 to 300 μg m−3 and was concentrated in the areas with low visibility. The haze formation mechanism in Beijing in winter was different from that in summer. The mass concentration of PM2.5 was higher, and the components more complicated, in winter. While the mass concentration of PM 2.5 in summer was lower than that in winter, the mass concentrations of hygroscopic inorganic salts were comparable with those in winter, and the relative humidity was, as expected, higher. Therefore, the water uptake of hygroscopic aerosols played a key role in summer. Moreover, the analysis showed that the influence of the PM 2.5 mass burden on visibility was very weak when its value was larger than 100 μg m −3. Only when the mass burden of PM 2.5 decreased to a certain threshold interval did the visibility increase rapidly. This indicates that, when emission reduction measures are taken to control haze occurrence, the mass burden of PM 2.5 must be cut to below this threshold interval. The relationship between the threshold of haze occurrence and the relative humidity in Beijing was fitted by an exponential function, and the resulting fitting curves could provide a new theoretical basis to understand and control haze formation in Beijing.


General Comments:
This paper presents a model study upon the haze formation in Beijing, China.It concluded that high PM2.5 loading was the main cause of haze events in Beijing, and that water uptake by aerosols resulted in the frequent formation of haze in Beijing, particularly during summertime.In general, this paper is well organized except some technical defects.However, my major concern is that this paper did not provide new concept or scientific findings relevant to haze.Coupling RAMS-CMAQ with an aerosol optical scheme is not a new idea as a similar study from the same group has been published in another Journal (Atmospheric Environment, 2013, 72: 177-191).Moreover, It is well known in atmospheric physics that high levels of aerosol concentration will resulted in cases of low visibility, and hygroscopic growth of aerosols will enhance the light scat-C12696 tering capability, or mass-specific light extinction efficiency of aerosol particles.off the formation of a specific compound in the model to investigate the corresponding effects?In that case, there could be some bias in the results.For instance, if you turn off the formation of ammonium sulfate then the ammonia will go to nitrate and change the partition and fate of N-containing species in the atmosphere.
7. Sec 4.3: Regarding the case study of size distribution, the mass fraction of accumulation mode was still ∼80% despite the increases in Aitken mode.Thus the changes in the cross section should be rather limited.I'm not convinced that the spike of "mass threshold" was due to increases of Aitken mode aerosols.Moreover, in terms of size distribution, I think that the cases of high coarse mode fraction also worth to be investigated further.
Technical Corrections: 1. Figure 4 contains two identical plots for Baotou, obviously one of them should be for Taishan.
2. In many cases the "diffusion" mentioned in the article should be "dispersion".
3. The caption of Figure 8 is inconsistent with the plots.(circle?line?...) 4. As talking about "pollutant scavenging" I think you are actually talking about "dilution" or "dispersion".
Interactive comment on Atmos.Chem.Phys. Discuss., 13, 30575, 2013.C12698 The case of Beijing is interesting because the microphysical properties of aerosols could be different from those observed in US or Europe.Unfortunately, the authors stopped at a general description of the phenomena of haze formation and did not advance fur-ther into the details of aerosol chemistry and/or physics.Therefore, I suggest reject this paper from ACP because lack of scientific merits.Specific Comments: 1. Method Sec: Calculation of light extinction coefficient of aerosols is the key compo-nent of this task.In addition to citing references, it is worth a detailed description in this Sec, so that readers know what parameters were used in the model and thereby can make judgment.2.Model evaluation: it was indicated that the model performed well as shown in the fig-ures.However, there were indeed some cases where the model value was inconsistent with the observation.To perform a model validation, I suggest make the comparison in terms of statistics and refer to Eder and Yu(AE, 2006)and Appel et al.(AE, 2012).3. Sec 4.1: It was indicated that "the heavy mass burden of PM2.5 was mainly con-centrated in four urban areas. ..".However, the urban hot spots were not shown in the figures.Actually, the pattern shown in those figures are more likely caused by a regional pollution event.4. Sec 4.1: It was indicated that "the distribution patterns of visibility broadly followed those of PM2.5. ..". Don't you think this is a result as expected and is determined by the calculation of visibility in model (i.e.EQ1)? 5. Sec 4.2: Decline in pollution caused by the enhanced vertical convection is a clas-sical case in PBL dynamics.I suggest move forward to investigate factors that were controlling the convection and, in turn, influencing air quality.6. Sec 4.3: The method for "contribution ratio" calculation is unclear.Are you turning C12697