Professional Documents
Culture Documents
Untitled
Untitled
Manuscript Number:
Order of Authors: Jun Zou, Ph.D; Jianning Sun; Te Li; Xiaomen Han; Zixuan
Xiang; Jie Sha
Please find our manuscript entitled “Observed interaction among haze, fog and atmospheric
boundary layer during a haze-fog episode in the Yangtze River Delta region, eastern China”,
which we would like to submit for publication as a Research Paper in Atmospheric Environment.
This work has not been submitted elsewhere for publication, in whole or in part, and all the
This study focuses on the interaction among haze, fog and atmospheric boundary layer during
a haze-fog episode based on observations at the SORPES site in Nanjing, which is located in
northeast part of YRD region. We find that fog induces a potential temperature jump (i.e., sharp
increase of potential temperature) at the top of fog layer, and this special thermal structure
maintains to the end of fog period, which significantly suppresses the daytime development of
planetary boundary layer after fog dissipation. The fog-induced reduction of PBLH further
increases the PM2.5 mass concentration. We also find that the wet deposition of fog on PM2.5 is
negligibly small. The scavenging effect of fog on aerosols only acts during fog period. When fog
dissipates, the aerosols come back from the fog droplets to the atmosphere.
Our results suggest that fog plays an important role in the enhancement of air pollution in the
haze-fog episode. This study is of interest to researchers in the issue of interaction among haze,
fog and atmospheric boundary layer. This study gives an observational result that fog plays an
important role in the formation of heavy air pollution in moist areas. Therefore, we feel this
manuscript is particularly suitable for Atmospheric Environment and of great interest to its readers.
We deeply appreciate your consideration of our manuscript, and we look forward to receiving
Best regards
Jun Zou, Jianning Sun, Te Li, Xiaomen Han, Zixuan Xiang, and Jie Sha
*Declaration of Interest Statement
Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:
FIG1.TIFF
Click here to download high resolution image
FIG2.TIFF
Click here to download high resolution image
FIG3.TIFF
Click here to download high resolution image
FIG4.TIFF
Click here to download high resolution image
FIG5.TIFF
Click here to download high resolution image
*Manuscript
Click here to download Manuscript: manuscript.docx Click here to view linked References
5 Jun Zoua,b,c,*, Jianning Suna,b,c,**, Te Lia,b, Xiaomen Hana,b, Zixuan Xianga,b, Jie Shaa,b
a
6 Joint International Research Laboratory of Atmospheric and Earth System Sciences, Nanjing
7 University, Nanjing, China
b
8 School of Atmospheric Sciences, Nanjing University, Nanjing, China
c
9 Collaborative Innovation Center of Climate Change, Jiangsu Province, China
10
11 *Corresponding author
12 Tel.: +86 18551752048; fax: +86 02589681162; E-mail address: junzou@nju.edu.cn (J. Zou)
13 Postal address: No. 163 Xianlin Avenue, Qixia District, Nanjing 210023, China.
14 ** Corresponding author
15 Tel.: +86 13851723266; fax: +86 02589681162; E-mail address: jnsun@nju.edu.cn (J. Sun)
16 Postal address: No. 163 Xianlin Avenue, Qixia District, Nanjing 210023, China.
17
18 Highlights:
19 Potential temperature jump at fog-top suppresses PBLH after fog dissipation.
20 Fog-induced reduction of PBLH increases the PM2.5 mass concentration.
21 The scavenging effect of fog on aerosols only acts during fog period.
22
23 Abstract: A severe haze-fog episode occurred in the Yangtze River Delta region in eastern China
24 during 22 – 30 November 2018. In this period, the PM2.5 mass concentration and meteorological
25 parameters at surface were collected at the Station for Observing Regional Processes of the Earth
26 System (SORPES) site in Nanjing. The vertical distribution of PM2.5, humidity and potential
27 temperature below 500 m were observed simultaneously by an Unmanned Aerial Vehicle, and the
28 profile of potential temperature at 1400 LST in each day was also observed by radiosonde at the
29 same site. These observations allow us to investigate the interaction among haze, fog and
30 atmospheric boundary layer. In the former four days, the PM2.5 mass concentration increases, the
31 daytime maximum planetary boundary layer height (PBLH) decreases, and the air humidity
32 increases. These are favorable conditions for fog formation. In the latter five days, fog forms in
33 four days, the PBLHs in the fog days are much lower than those in the no-fog days, and the PM2.5
34 mass concentration increases further. We find that fog induces a potential temperature jump (i.e.,
35 sharp increase of potential temperature) at the top of fog layer, and this special thermal structure
36 maintains to the end of fog period, which significantly suppresses the daytime development of
37 planetary boundary layer after fog dissipation. The fog-induced reduction of PBLH further
38 increases the PM2.5 mass concentration. We also find that the wet deposition of fog on PM2.5 is
39 negligibly small. The scavenging effect of fog on aerosols only acts during fog period. When fog
40 dissipates, the aerosols come back from the fog droplets to the atmosphere. Our observations
41 suggest that fog plays an important role in the enhancement of air pollution in the haze-fog
42 episode.
43 Key words: haze-fog episode, interaction, potential temperature jump, planetary boundary layer
45
46 1. Introduction
48 and industrialization, many regions of China, especially megacities and city clusters in the coastal
49 eastern China, have experienced heavy air pollution (He et al., 2002; Chen et al., 2008; Tie and
50 Cao, 2009; Ding et al., 2016a). In these regions the fine particular matter, PM2.5 (particles of 2.5
51 microns or less in aerodynamic diameter in the ambient air), often show extremely high
52 concentrations during the winter haze period (Ding et al. 2013a; Huang et al., 2014; Guo et al.,
53 2014; R. Zhang, 2014). Heavy air pollution can significantly modify the weather by changing air
54 temperature, cloud and even precipitation (Ding et al., 2013b, Huang et al., 2016). The increase in
55 atmospheric aerosols over China is shown to heat the atmosphere and generates a cyclonic
56 circulation anomaly that weakens the East Asian winter monsoon (F. Niu et al., 2010). The
57 weakened East Asian winter monsoon circulation brings less cold and dry air to the region, and
58 favor the formation of haze and fog over eastern China (F. Niu et al., 2010; R. H. Zhang et al.,
59 2014).
60 The high PM2.5 levels induce a decrease of surface air temperature by shadowing effect of
61 aerosols and an increase of air temperature in the upper planetary boundary layer (PBL) by
62 heating effect of absorbing aerosols, leading to a stabilized PBL. The enhanced PBL stability
63 suppresses the vertical mixing and dispersion of air pollutants, resulting in higher PM2.5
64 concentrations (Wang et al., 2014; Gao et al., 2015). Actually, the scattering and absorbing effects
65 of aerosols can reduce surface heat flux, leading to the decrease of PBL height (PBLH) and
66 consequently the increase of PM2.5 concentrations (Zou et al., 2017). Ding et al. (2016b)
67 demonstrate that black carbon (BC) aerosols induce heating in the PBL, particularly in the upper
68 PBL, and the resulting stabilized stratification and decreased surface heat flux substantially
69 depress the development of PBL and consequently enhance the occurrences of extreme haze
70 pollution episodes. They define this process as “dome effect” of BC and suggest an urgent need
71 for reducing BC emissions as an efficient way to mitigate the extreme haze pollution in megacities
72 of China.
73 Owing to heterogeneous nucleation, aerosols can promote fog formation (Sachweh and
74 Koepke, 1995; Mohan and Payra, 2006, 2009) to such an extent that supersaturation or saturation
75 is not even needed for fog to form in the polluted air (Kumala et al., 1997). Therefore in moist
76 areas heavy air pollution often induces more fog events. Observations show that the frequencies of
77 fog events in wintertime over eastern-central China have been doubled over the past three decades
78 (F. Niu et al., 2010). Yangtze River Delta (YRD) is one of the regions of high fog occurrence in
79 China (S. Niu et al., 2010). In the YRD region, a common type of fog is advection fog, which
80 forms when quite warm, moist air from East China Sea comes across this region (Shi et al., 2010;
81 Lin et al., 2017). Ding et al., (2019) find that BC plays an important role in advection-radiation
82 fog formation in the YRD region by inducing a mesoscale cyclonic wind anomaly, which can
83 transport moister air from sea into this region and help to form heavy fog at night.
84 The fog water can provide a medium for aqueous phase reactions and transform primary
85 pollutants into secondary aerosols (Graedel et al., 1985; Dall'Osto et al., 2009). Wet scavenging
86 effect of fog can also significantly reduce aerosol concentration in the atmosphere (Yuskiewicz et
87 al., 1998; S. Niu et al., 2010; Han et al., 2018). The interrelation between fog and aerosols is
88 crucial in the understanding of the formation and evolution of both haze and fog episodes, which
89 are associated with the PBL processes (Ding et al., 2019). However, the interaction among haze,
90 fog and PBL has been rarely reported in previous studies. During 22 – 30 November 2018, the
91 YRD region suffered from a severe haze-fog episode with high PM2.5 mass concentration and
92 multi-day fog occurrence. In this study, we focus on the interaction among haze, fog and PBL
94
95 2. Observation and data processing
96 During 22 – 30 November 2018, a severe haze-fog event occurred in the YRD region. PM2.5
97 mass concentration, surface heat flux, and meteorological data (including wind, temperature,
98 humidity and precipitation) in this period that were collected at the Station for Observing Regional
99 Processes of the Earth System (SORPES) site (118º57′10″E, 32º7′14″N) in Nanjing (Ding et al.,
100 2013a, b; Zou et al., 2017) are used in this study. The PM2.5 mass concentration was measured by
101 a mass analyzer (SHARP-5030, Thermo Fisher Scientific, Massachusetts, USA), which was
102 housed on the top floor of a laboratory building. The surface sensible and latent heat fluxes were
103 measured by an eddy covariance system (CSAT 3A, Campbell Scientific, Utah, USA), which was
104 mounted in the observation yard at the height of 2.2 m above ground level (AGL). The signals for
105 the eddy covariance calculations were sampled at 10 Hz, and the data were subjected to spike
106 detection and rejection algorithms (Zou et al., 2017). The sensible heat flux (Hs) ware computed
107 at half-hour interval, which is the widely-accepted duration for calculating the turbulent fluxes
108 (Nordbo et al., 2013). Corrections were applied for sonic virtual temperature (Schotanus et al.,
109 1983). The data of wind, temperature and humidity ware measured by sensors installed at the
110 height of 9 m AGL on the 75 m-tower (Actually, there are 6 level measurements on this tower.
111 Here we only use the lowest level data). The wind speed (WS) and direction (WD) ware measured
112 by an anemometer, including a 010C wind speed sensor (Campbell, USA) and a 020C wind
113 direction sensor (Campbell, USA). The temperature (T), relative humidity (RH) and specific
114 humidity (q) ware measured by a HMP155A Temperature and Relative Humidity Probe (Campbell,
115 USA). The precipitation was measured by a TE525WS-L Rain Gage (Campbell, USA) in the
117 In order to know the maximum PBLH in each day during this haze-fog event, radiosonde
118 measurements were carried out at 1400 local standard time (LST) at the SORPES site. The iMet-4
119 radiosonde (International Met Systems, USA) was used to measure the temperature, pressure,
120 humidity and GPS information, and the sounding data is received by a 3050A receiver
121 (International Met Systems, USA). The GPS data was used to calculate the wind speed and
122 direction, supposing that the radiosonde moves with the wind. To obtain the vertical profiles of
124 Unmanned Aerial Vehicle (UAV). The UAV is a M600 Pro hexacopter (DJI, China). The
125 maximum flight height is 500 m AGL, and the flight time for going down with the payload is
126 16-18 min (the moving speed is about 1 m s–1). An iMet-4 radiosonde was mounted on the UAV to
127 measure the air temperature and humidity, and an AM520 PM2.5 sensor (TSI, USA) was used to
128 monitoring PM2.5 concentration. The profiles measured in the UAV rising process are sued in this
130 The PBLH can be derived from the radiosonde data, as well as the UAV observation data
131 (when the PBLH is below 500 m). In this study, when there is no fog, we determine the daytime
132 PBLH as the height at which the vertical gradient of potential temperature has the maximum value
133 (Stull, 1988; Batchvarova and Gryning, 1991), and determine the nighttime PBLH as the height of
134 near-surface inversion layer top (Stull, 1988). However, when there exists fog, we define the
135 PBLH as the height at the top of fog layer, above which an elevated inversion or stable layer can
136 be identified, or the air moisture reduces significantly (Seibert et al., 2000).
Time 11/22 11/23 11/24 11/25 11/26 11/27 11/28 11/29 11/30
0800 × × × × × × × × ×
0830 × × × × × × × ×
0900 × × × × × × × × ×
0930 × × × ×
1000 × × × × × ×
1030 × × × ×
1100
1130
1200 ×
1230 ×
1300 × × ×
1330 × × × × × ×
1400
1430 × × ×
1500 × × × × × ×
1530 × × × × ×
1600 × × × × ×
1630
1700
1730 ×
1800 ×
1830 × × ×
1900 × × × × × ×
1930 × × × × × × ×
2000 × × × × × × ×
137 Table 1. The UAV measurement time in the fog and haze episode during 22 – 30 November, 2018.
139
141 The time series of wind (WS and WD), humidity (q and RH), temperature (T) and surface
142 sensible heat flux (Hs), PBLH, PM2.5 mass concentration, and precipitation in the haze-fog event
143 during 22 – 30 November 2018 are shown in Figure 1, in which the periods of fog occurrence are
144 shaded. In the air pollution area, the relative humidity with the value of 90% is used to distinguish
145 light fog from haze (Schichtel et al., 2001; Doyle and Dorling, 2002). We use this criterion to
146 separate fog from haze and determine the onset time of fog. In the nine days the WS is often less
147 than 2 m s–1 and rarely reaches 3 m s–1, while the WD covers almost all the orientations and is
148 mostly in northeastern or southeastern. Small wind speed is a favorable condition for the
149 formation of haze-fog episodes. In the former four days there is no fog, and the relative humidity
150 RH appears diurnal variation, showing the lower values in the daytime and higher values in the
151 nighttime, due to the diurnal change of temperature. However, the specific humidity q increases
152 steadily from 4 g kg–1 to about 7 g kg–1 in these four days (and an obvious increasing trend of RH
153 can also be seen), suggesting that the air moisture accumulates in this period. This is a favorable
154 condition for fog formation. In the latter five days fog occurs in each day, although the beginning
155 time and the duration are different. On 26 November, the specific humidity q drops back to 4 g
156 kg–1 (a relatively low level of air moisture). The reason is likely that precipitation occurs in the
157 night before and remove some water vapor from the near-surface atmosphere, although the
158 precipitation is light. However, in the following days the air moisture accumulates again, and thus
159 fog can still occur. The near surface air temperature shows diurnal variation between 6 °C (279 K)
160 and 20 °C (293 K). The minimum temperature at night is not very low. The fog can occur due to
161 relatively high level of air moisture. It can be seen that, when the air moisture is not very high, fog
162 occurs later, because a longer radiative cooling time is needed so that a lower temperature can
163 result in a nearly saturated air moisture. For example, fog occurs at about 0800 LST on 28
164 November, the time of sunrise at which the near-surface air temperature reaches the minimum
165 value. The maximum PBLHs at 1400 LST in the former four days are systematically higher than
166 those in the latter five fog days. It can also be seen that in the former four days the PBLH has a
167 decreasing trend while PM2.5 mass concentration increases from 50 μg m–3 to approximately 150
168 μg m–3. Although PM2.5 mass concentration drops back to about 50 μg m–3 on 26 November due to
169 wet deposition by the light precipitation in the night before, it increases rapidly in the following
170 days to more than 200 μg m–3, suggesting that the lower PBLH is a favorable condition for
171 accumulation of air pollutants, as well as air humidity, in the lower atmosphere.
172
173 Figure 1. The time series of (a) wind speed (WS) and wind direction (WD), (b) specific humidity
174 (q) and relative humidity (RH), (c) temperature (T) and surface sensible heat flux (Hs), (d) PBLH,
175 and (e) PM2.5 mass concentration and precipitation in the haze-fog event during 22 – 30 November
176 2018.
177
178 On 29 November there is no fog before sunrise. Therefore the daytime PBLH develops
179 immediately after sunrise, which is similar to the situation in the former four no-fog days. Here we
180 attribute 29 November to a no-fog day, since in the daytime and the whole night before there is no
181 fog. However, on 26-28 and 30 November, fog lasts after sunrise (the fog just begins at sunrise on
182 28 November), and it is inevitable that the development of daytime PBLH is influenced by the fog
183 in these days. We call these four days the fog days. Figure 2 shows the maximum PBLH (at 1400
184 LST) in each day during 22-30 November, which is plotted against the daytime mean PM2.5 mass
185 concentration. The maximum PBLHs in the no-fog days are significantly higher than those in the
186 fog days, suggesting that the development of daytime PBL is suppressed by fog. As shown in
187 Figure 1, the fog lasts to later morning and even afternoon, which must delay the development of
188 daytime PBL and consequently reduce the maximum PBLH. The decrease of PBLH is a favorable
189 condition for air pollutants to accumulate in the PBL, which can lead to heavy air pollution. From
190 27 to 30 November the PM2.5 mass concentration experiences an accumulating process and
191 increases from 50 μg m–3 to more than 200 μg m–3 at the end of this haze-fog episode (Fig. 1e).
192 Figure 2 also shows that the maximum PBLH decreases with PM2.5 mass concentration at the
193 surface in both no-fog and fog days. In the no-fog days, the shadowing effect of aerosols reduces
194 the surface sensible heat flux, and consequently deceases the PBLH (Barbaro et al., 2013; Zou et
195 al., 2017). Our observed results of the daytime maximum PBLH are consistent with the previous
196 studies. However, the situation in the fog days may be more complex. The development of PBL
197 forced by surface heating begins after fog dissipation. On the one hand, the maximum PBLH is
198 influenced by the dissipation time of fog. The later the fog dissipates in the morning, the later the
199 PBL develops, and thus the lower the maximum PBLH can reach (of course, if the fog can last to
200 the afternoon, as on 26 November, the daytime PBL cannot develop, and the PBLH is almost as
201 the same as the depth of the fog layer). On the other hand, it can be expected that the shadowing
202 effect still acts in the delayed developing process of convective PBL, which may further decrease
203 the maximum PBLH. Our observed results in the fog days show a decreasing trend of the
204 maximum PBLH with increasing PM2.5 mass concentration, which seems to support this
205 argument.
206
207 Figure 2. The maximum PBLHs at 1400 LST in fog days (black dots) and no-fog days (circles)
208 during 22-30 November 2018.
209
210 When neglecting large-scale synoptic forcing, the daytime PBL grows mainly driven by
211 surface heating and upper-PBL entrainment (Stull, 1988; Sühring et al., 2014). In a simplified
212 mixed-layer model, the PBLH can be predicted when the surface heat flux and the profile of
213 potential temperature are known (Stull, 1988; van Zanten et al., 1999; Fedorovich et al., 2004; Sun,
214 2009). The growth rate of the PBLH can then be characterized by a simplified parameterization
215 from Sun (2009), in which only the surface heat flux and the background stratification are needed.
217 (1)
218 where is the PBLH, is the entrainment flux ratio (the ratio of the heat flux at the top of
219 convective PBL to that at the surface), which is approximately a constant of 0.2 (Stull, 1976;
220 Deardorff et al., 1980), is the surface kinematic heat flux, is the lapse rate of potential
221 temperature in the free atmosphere above the convective PBL, and is a parameter, which is
222 approximately a constant of 0.9 for a purely buoyancy-driven convective PBL (Deardorff et al.,
223 1980; Fedorovich et al., 2004; Sun, 2009). Equation (1) shows that the growth rate of PBLH is
224 inversely proportional to the lapse rate of potential temperature in the free atmosphere above the
225 convective PBL. Therefore the profile of potential temperature plays an important role in the
228 profiles of potential temperature at 0800 LST and 1400 LST in each day and at the end of fog
229 period during 22-30 November 2018. These profiles are plotted in Figure 3. The upper panels
230 show that in the no-fog days at 0800 LST the near-surface atmosphere becomes more stable when
231 the PM2.5 mass concentration increases to a high level, while the profiles of potential temperature
232 at 1400 LST present a decreasing trend of the daytime maximum PBLH. This situation is more
233 evident on 29 November, a heavy air pollution day, in which the lapse rate of potential
234 temperature in the near-surface layer at 0800 LST is significantly larger, while the PBLH at 1400
235 is significantly lower than those in the previous no-fog days. A larger lapse rate of potential
236 temperature in the near-surface layer at 0800 LST means that after sunrise the PBL will develop in
237 a more stably stratified background atmosphere and the growth rate of PBLH will be suppressed.
238 The result that the near-surface atmosphere at the end of night becomes more stable in a heavier
239 air pollution day is consistent with the long term observations presented in Zou et al. (2017).
240 Therefore air pollution can impact the PBLH by different processes. One the one hand, high level
241 of aerosol loading in the PBL is apt to form a more stable nocturnal PBL, which will suppress the
242 growth rate of PBLH in the morning. On the other hand, high PM2.5 mass concentration increases
243 the shadowing effect of aerosols on solar radiation and reduces the surface sensible heat flux,
244 which will lead to a lower daytime PBLH. Actually, the heating effect of aerosols also influence
245 the development of daytime PBL. The absorbing aerosols induce solar heating in the atmosphere,
246 and the heating effect is stronger in the upper part of PBL (Barbaro et al., 2013; Ding et al.,
247 2016b). To our knowledge for convective PBL, the stabilized upper part of PBL can suppress the
248 overshooting of thermals and weaken the entrainment process, which means that the entrainment
249 flux ratio is reduced. According to Eq. (1), a smaller leads to a smaller growth rate of
250 PBLH.
251
252 Figure 3. Profiles of potential temperature at 0800 LST (blue curves), 1400 LST (red curves) and
253 the moment of fog dissipation (black curves) in the fog days (lower panels) and the no fog days
254 (upper panels) during 22-30 November. The fog dissipates at 1513 LST on 26 Nov., 1214 LST on
255 27 Nov., 1048 LST on 28 Nov., and 1027 LST on 30 Nov., respectively. The mean mass
256 concentration of PM2.5 in each day is given below each panel.
257
258 The lower panels in Figure 3 show the profiles of potential temperature at different time in
259 the fog days. 26 November is somewhat special, because the fog lasts to afternoon and dissipates
260 at about 1530 LST in this day. When the fog dissipates, the profile of potential temperature (the
261 black line) shows that the PBLH is about 350 m. At 1400 LST, the profile of potential temperature
262 from radiosonde (the red line) indicates that the PBLH is 400 m (here we choose the middle height
263 of the lowest inversion layer as the daytime PBLH, when the profile of potential temperature
264 appears more than one inversion layer). Figure 1 shows the time series of PBLH in this day. The
265 PBLH is actually just the depth of fog layer during the fog period. Since the fog dissipates in the
266 later afternoon, the PBLH decreases slightly in the following afternoon time. Therefore the fog
267 layer depth determines the daytime PBLH in this day. On 27 November, the profile at 0800 LST
268 (the blue line) shows a large potential temperature jump at the height of about 150 m, where is the
269 top of fog layer. At about 1200 LST the fog dissipates, but the potential temperature jump at this
270 height still exists. The fog layer depth is shallow and potential temperature jump becomes smaller
271 at this moment. Additionally, the surface sensible heat flux increases rapidly to about 200 W m–3
272 after fog dissipation (Figure 1). Thus in the following time the surface heating is relatively strong,
273 and the potential temperature jump is easy to be consumed. The potential temperature profile (the
274 red line) shows that at 1400 LST the PBLH reaches a relatively high altitude of about 800 m. The
275 similar situations appear on 28 and 30 November. In the two days, at 0800 LST there is a strong
276 capping inversion layer above the fog layer, and at the time of fog dissipation a step-like potential
277 temperature jump still exists at the top of fog layer. Since the fog layer depth (about 350 m) in
278 these two days is thicker and the potential temperature jump at the moment of fog dissipation is
279 somewhat larger than that on 27 November, surface heating after fog dissipation is to increase air
280 temperature and maintain turbulent motion in the whole mixed layer, but the BPLH is not changed
281 until the potential temperature jump is consumed. This situation is similar to that in many
282 large-eddy simulation studies of convective PBL, in which a large potential temperature jump is
283 set at a chosen height so that the top of mixed layer is approximately fixed at the chosen height
284 during the simulation. To our knowledge, Eq. (1) is not suitable for this situation, because the
285 potential temperature jump cuts off the interaction between the mixed layer and the free
286 atmosphere. Here we distinguish the potential temperature jump from the potential temperature
287 difference across the entrainment zone for a convective PBL that is developing in the equilibrium
288 state. The former corresponds to a step-like shape of potential temperature profile, however, the
289 latter corresponds to a smooth curve of potential temperature profile. Therefore Eq. (1) is only
290 applicable when the convective PBL reaches an equilibrium state, in which the interaction
291 between the mixed layer and the free atmosphere forms a stable interfacial layer, i.e., the
293 The potential temperature profile at 1400 LST on 28 November shows that the step-like
294 shape disappears but the PBLH almost does not change, suggesting that the potential temperature
295 jump is consumed just before this moment. On 30 November, the potential temperature profile at
296 1400 LST shows that the PBLH increases only about 150 m from the moment of fog dissipation,
297 implying that most time between 1030 LST and 1400 LST is used to consume the potential
298 temperature jump. Therefore our observations indicate that the potential temperature jump formed
299 at the fog layer top will delay the PBL development after fog dissipation, because a time period is
300 needed for increasing the air temperature in the whole mixed layer by surface heating so that the
301 potential temperature jump is eliminated, and then the PBL begins to develop. Why can a potential
302 temperature jump form at the fog layer top? The most possible reason is that longwave radiative
303 cooling occurs at fog layer top generates well-mixed fog with a well-defined top edge and this
304 sharp top concentrates the radiative divergence closer to that region, which reinforces vertical
305 mixing in the fog layer (Stull, 1988). Therefore a well-mixed fog appears more uniform in the
306 vertical direction and has a well-defined top edge, which forms a sharp potential temperature jump
307 at the fog layer top. Our observations show that the step-like jump corresponds to 3-5 K
308 temperature increase over a thin layer with its depth only about 20 m. Such a large lapse rate
309 cannot be resolved by the weather and climate models, and the growth rate of PBLH after fog
311
312 Figure 4. The mean lapse rate of potential temperature at the fog layer top in each temperature
313 profile measured by UAV. The shaded areas represent the daytime duration of fog. The green stars
314 denote the lapse rate in the layer with a depth of 20 m, while the blue stars denote the lapse rate in
315 the layer with a depth of 100 m.
316
317 In order to know the exact strength of potential temperature jump at the fog layer top, we
318 calculate the lapse rate in the thin layer with a depth of 20 m from each temperature profile
319 measured by UAV. The results are shown in Figure 4 and denoted by green starts. On 26
320 November the potential temperature jump is relatively small (as shown in Fig. 3), but the value of
321 lapse rate at the fog layer top is larger than 50 K km–1 and can reach 100 K km–1. In the following
322 three fog days, the potential temperature jump is relatively strong, and the lapse rate at the fog
323 layer top is usually larger than 100 K km–1 and can even reach 250 K km–1. Observations show
324 that a very strong stable layer is formed in a thin layer at the fog top, although the lapse rate in this
325 layer varies with time and the mean strength is different in different case. For the high resolution
326 numerical models, the vertical grid distance is about 100 m. To know the strength of stratification
327 in the layer with a depth of 100 m at the fog layer top, we also calculate the mean lapse rate of
328 potential temperature in this layer and plot the results in Figure 4 as blue starts. On 26 November
329 the lapse rate is about 25 K km–1, while in the following three fog days the lapse rate is about 50 K
330 km–1. Such a large lapse rate maintains to the end of fog period. It means that the convective PBL
331 will develop in a much stably stratified background atmosphere after fog dissipation and the
332 growth rate of PBLH is significantly suppressed. In this situation, the overshooting of rising
333 thermals from mixed layer is inhibited by the strongly stabilized layer above the mixed layer top,
334 which eliminates the entrainment process. Therefore we can use the encroachment model (Boers
335 et al. 1984) to estimate PBLH growth rate. It reads , where is the lapse
336 rate of potential temperature immediately above the mixed layer. This model neglects the
337 entrainment process, implying that the mixed layer top erodes into the strongly stable layer due to
338 the increase of air temperature in the mixed layer. Assuming that K km–1 (considering the
339 situation in the later three fog days), K m s–1 (the corresponding the surface heat
340 flux is 180 W m–2, and this value is estimated according to our measurements at the noontime),
341 and a initial PBLH of 350 m at 1030 LST (according to observed fog layer depth), calculation
342 shows that it needs about 3.7 h for the PBLH to increases to 450 m. This estimation agrees with
343 the observations in the latter two fog days. Our observations on 28 and 30 November indicate that
344 it spends about 3.5 h for the mixed layer top to penetrate this stable layer. Therefore fog can
345 significantly suppress the daytime PBL development by inducing a strongly stabilized layer at the
353 Fog droplets interact with aerosol particles and soluble gases in the atmosphere. Thus fogs
354 can affect pollutant formation, transformation, and removal. In this study we observed the vertical
355 distribution of PM2.5 mass concentration in the lower atmosphere both during fog period and after
356 fog dissipation. Figure 5 shows the profiles of PM2.5 mass concentration and relative humidity at
357 different time in the fog days. On 26 November, the profiles of relative humidity at 0830 LST and
358 1000 LST show that the fog layer top reaches about 300 m AGL. Meanwhile, the profiles of PM2.5
359 mass concentration at the two moments show that in the upper part of fog layer the PM2.5 mass
360 concentration in the atmosphere is very small (only about 25 μg m–3), although it increases when
361 approaching to the ground level. However, at 1430 LST, the profiles show that in the fog layer the
362 PM2.5 mass concentration increases to about 100 μg m–3, while the relative humidity in this layer
363 decreases to about 90%, suggesting that at this moment the fog still exists, but evaporation of fog
364 droplets releases a lot of PM2.5 back to the atmosphere. At 1530 LST, the fog dissipates and
365 evaporation ends at this moment, and the PM2.5 mass concentration further increases to about 120
366 μg m–3 that is almost the same level before fog formation, implying that almost all the PM2.5 in the
367 fog droplets comes back to the atmosphere. Figure 5 show that the same process repeats in the
368 following fog days. We note that on 27 November at the moment of fog dissipation the PM2.5 mass
369 concentration reduces to a lower level in comparison with the value before fog formation, which is
370 evidently caused by the light precipitation during fog period (Figure 1). However, in the
371 no-precipitation fog days the PM2.5 mass concentration at the moment of fog dissipation comes
372 back to the level before fog formation. The scavenging effect of fog can significantly reduce
373 aerosol concentration in the atmosphere (Yuskiewicz et al., 1998; S. Niu et al., 2010; Han et al.,
374 2018). Our observations suggest that the wet deposition effect of fog on PM2.5 is negligibly small
375 unless precipitation can occur during fog period. It means that wet scavenging effect of fog on
376 aerosols can only act during fog period. Once the fog dissipates, the aerosols will come back from
377 the fog droplets to the atmosphere. The fog droplets provide a liquid phase reaction environment
378 for the PM2.5 with different chemical compositions and the soluble gases. But this problem is
379 beyond the scope of this study. We will investigate this problem based on observations in our
381
383 A severe haze-fog event occurred in the YRD region, eastern China during 22 – 30
384 November 2018. In this study we focuses on the interaction among haze, fog and atmospheric
385 boundary layer during the haze-fog episode based on observations at the SORPES site in Nanjing,
386 which is located in northeast part of YRD region. Observations show that during this period the
387 PM2.5 mass concentration increases from about 50 μg m–3 to more than 200 μg m–3 while the
388 daytime maximum PBLH decreases from about 1500 m to 500 m and even lower height. During
389 the former half period, i.e., in the clear-sky days, PM2.5 accumulates due to small wind speed, and
390 the air humidity increases to a high level. During the latter half period, fog occurs almost every
391 day, the PM2.5 mass concentration increases, and the daytime maximum PBLHs in the fog days are
392 significantly lower than those in the clear-sky days. These observational results suggest that the
393 interaction among haze, fog and atmospheric boundary layer enhances air pollution. On the one
394 hand, the increase of aerosol loading reduces PBLH by decreasing the daytime surface heating and
395 stabilizing the atmospheric boundary layer, and the aerosol-induced reduction of PBLH limits the
396 aerosols in a smaller space and results in higher PM2.5 mass concentration. This is a well-known
397 interaction process in the clear-sky days. On the other hand, the increase of aerosol loading and air
398 humidity provides favorable conditions for fog formation, and the daytime PBLH is significantly
399 reduced after fog dissipation, which can further increase PM2.5 mass concentration and lead to
401 Observations show that the daytime maximum PBLHs in the fog days are much lower than
402 those in the no-fog days, suggesting that the occurrence of fog significantly suppresses the
403 development of daytime PBL. One the one hand, the convective PBL develops after sunrise in
404 clear-sky days, whereas it develops after fog dissipation in the fog days. Therefore fog delays the
405 development of convective PBL. On the other hand, the existence of fog results in a step-like
406 potential temperature jump at the top of fog layer by longwave radiative cooling and remains it to
407 the moment of fog dissipation. Thereafter the convective PBL will firstly spends a few hours on
408 increasing the air temperature in the mixed layer by surface heating to eliminate the step-like
409 potential temperature jump, and then it can develop freely in the equilibrium state. This process
410 further delays the development of convective PBL, which leads to a much lower daytime PBLH
411 and consequently a high level of PM2.5 mass concentration in the fog days.
412 Observations also show that the wet deposition effect of fog on PM2.5 is negligibly small if
413 the fog cannot form precipitation. Previous studies point out that the scavenging effect of fog can
414 significantly reduce aerosol concentration in the atmosphere. Our observations suggest that this
415 effect only acts during the fog period. When the fog dissipates, the aerosols will come back from
416 the fog droplets to the atmosphere. Along with the effect of fog on the thermal structure of PBL,
417 which can significantly reduce PBLH, fog plays an important role in the formation of heavy air
418 pollution in moist areas, such as the YRD region, eastern China.
419
420 Acknowledgements The work is supported by Ministry of Science and Technology of China
422
423 References
424 Barbaro, E., J. de Arellano, V.-G., Krol, M. C., Holtslag, A. A. M., 2013. Impacts of aerosol
426 boundary layer. Bound. Layer Meteorol. 148, 31-49. doi: 10.1007/s10546-013-9800-7.
427 Batchavarova, E., Gryning, S. E., 1991. Applied model for the growth of the daytime mixed layer.
429 Boers, R., Eloranta, E. W., Coulter, R. L., 1984. Lidar observations of mixed layer dynamics: Tests
430 of parameterized entrainment models of mixed layer growth rate. J. Climate Appl. Meteor. 23,
432 Dall'Osto, M., Harrison, R. M., Coe, H., Williams, P., 2009. Real-time secondary aerosol
433 formation during a fog event in London. Atmos. Chem. Phys. 9, 2459-2469. doi:
434 10.5194/acp-9-2459-2009.
435 Deardorff, J. W., Willis, G. E., Stockton, B. H., 1980. Laboratory studies of the entrainment zone
437 10.1017/S0022112080001000.
438 Ding. A. J., Fu, C. B., Yang, X. Q., Sun, J. N., Zheng, L. F., Xie, Y. N., Herrmann, E., Nie, W.,
439 Petäjä, T., Kerminen, V. M., Kulmala, M., 2013a. Ozone and fine particle in the western
440 Yangtze River Delta: an overview of 1 yr data at the SORPES station. Atmos. Chem. Phys.
442 Ding, A. J., Fu, C. B., Yang, X. Q., Sun, J. N., Petäjä, T., Kerminen, V. M., Wang, T., Xie, Y.,
443 Herrmann, E., Zheng, L. F., Nie, W., Liu, Q., Wei, X. L., and Kulmala, M., 2013b. Intense
444 atmospheric pollution modifies weather: a case of mixed biomass burning with fossil fuel
445 combustion pollution in eastern China. Atmos. Chem. Phys. 13, 10545-10554. doi:
446 10.5194/acp-13-10545-2013.
447 Ding, A,, Nie, W., Huang, X., Chi, X., Sun, J., Kerminen, V. M., Xu, Z., Guo, W., Petäjä, T., Yang,
448 X., Kulmala, M., Fu, C., 2016a. Long-term observation of air pollution-weather/climate
449 interactions at the SORPES station: a review and outlook. Front. Environ. Sci. Eng. 10, 15.
451 Ding, A. J., Huang, X., Nie, W., Sun, J. N., Kerminen, V. M., Petäjä, T., Su, H., Cheng, Y. F., Yang,
452 X. Q., Wang, M. H., Chi, X. G., Wang, J. P., Virkkula, A., Guo, W. D., Yuan, J., Wang, S. Y.,
453 Zhang, R. J., Wu, Y. F., Song, Y., Zhu, T., Zilitinkevich, S., Kulmala, M., Fu, C. B., 2016b.
454 Enhanced haze pollution by black carbon in megacities in China. Geophys. Res. Lett. 43,
456 Ding, Q. J., Sun, J. N., Huang, X., Ding, A. J., Zou, J., Yang, X.-Q., and Fu, C. B., 2019. Impacts
457 of black carbon on the formation of advection-radiation fog during a haze pollution episode
458 in eastern China. Atmos. Chem. Phys. 19, 7759–7774. doi: 10.5194/acp-19-7759-2019.
459 Doyle, M., Dorling, S., 2001. Visibility trends in the UK 1950–1997. Atmos. Environ. 36,
461 Fedorovich, E., Conzemius, R., Mironov, D., 2004. Convective entrainment into a shear-Free,
462 linearly stratified atmosphere: Bulk models reevaluated through large eddy simulations. J.
464 Gao, Y., Zhang, M., Liu, Z., Wang, L., Wang, P., Xia, X., Tao, M., Zhu L., 2015. Modeling the
465 feedback between aerosol and meteorological variables in the atmospheric boundary layer
466 during a severe fog–haze event over the North China Plain. Atmos. Chem. Phys. 15,
468 Graedel, T. E., Weschler, C. J., Mandich, M. L., 1985. Influence of transition metal complexes on
470 Guo, S., Hu, M., Zamora, M. L., Peng, J., Shang, D., Zheng, J., Du, Z., Wu, Z., Shao, M., Zeng, L.,
471 Molina, M. J., Zhang, R., 2014. Elucidating severe urban haze formation in China. Proc. Natl.
473 Han, S. Q., Liu, J. L., Hao, T. Y., Zhang, Y. F., Li, P. Y., Yang, J. B., Wang, Q. L., Cai, Z. Y., Yao,
474 Q., Zhang, M., Wang, X. J., 2018: Boundary layer structure and scavenging effect during a
475 typical winter haze-fog episode in a core city of BTH region, China. Atmos. Environ. 179,
477 Huang, X., Ding, A., Liu, L., Liu, Q., Ding, K., Niu, X., Nie, W., Xu, Z., Chi, X., Wang, M., Sun,
478 J., Guo, W., Fu, C., 2016. Effects of aerosol–radiation interaction on precipitation during
479 biomass-burning season in East China. Atmos. Chem. Phys. 16, 10063-10082. doi:
480 10.5194/acp-16-10063-2016.
481 Huang, R., Zhang, Y., Bozzetti, C., Ho, K. F., Cao, J., Han, Y., Daellenbach, K. R., Slowik, J. G.,
482 Platt, S. M., Canonaco, F., Zotter, P., Wolf, R., Pieber, S. M., Bruns, E. A., Crippa, M.,
483 Ciarelli, G., Piazzalunga, A., Schwikowski, M., Abbaszade, G., Schnelle-Kreis, J.,
484 Zimmermann, R., An, Z., Szidat, S., Baltensperger, U., El Haddad, I., Prévôt, A. S. H., 2014.
485 High secondary aerosol contribution to particulate pollution during haze events in China.
487 Lin, C., Zhang, Z., Zhao, P. U., Wang, F., 2017. Numerical simulations of an advection fog event
488 over Shanghai Pudong international air-port with the WRF model. J. Meteorol. Res-PRC. 31,
490 Mohan, M., Payra, S., 2006. Relation between fog formation and accumulation mode aerosols in
491 urban environment of megacity Delhi. Indian Journal of Environmental Protection, 26,
492 294–300.
493 Mohan, M., and Payra, S., 2009. Influence of aerosol spectrum and air pollutants on fog formation
494 in urban environment of megacity Delhi, India. Environ. Monit. Assess. 151, 265–277. doi:
495 10.1007/s10661-008-0268-8.
496 Niu, F., Li, Z., Li, C., Lee, K.-H., Wang, M. 2010. Increase of wintertime fog in China: Potential
497 impacts of weakening of the Eastern Asian monsoon circulation and increasing aerosol
499 Niu, S., Lu, C., Yu, H., Zhao, L., Lu, J., 2010. Fog research in China: An overview. Adv. Atmos.
501 Nordbo, A., Järvi, L., Haapanala, S., Moilanen, J., Vesala, T., 2013. Intra-City Variation in Urban
502 Morphology and Turbulence Structure in Helsinki, Finland. Bound. Layer Meteorol. 146,
504 Sachweh, M., Koepke, P., 1995. Radiation fog and urban climate. Geophys. Res. Lett. 22,
505 1073–1076. doi: 10.1029/95GL00907.
506 Schichtel, B. A., Husar, R. B., Falke, S. R., Wilson, W.E., 2001. Haze trends over the United
508 Schotanus, P., Nieuwstadt, F. T. M., Bruin, H. A. R. D., 1983. Temperature measurement with a
509 sonic anemometer and its application to heat and moisture fluxes. Bound. Layer Meteorol. 26,
511 Seibert, P., Beyrich, F., Gryning, S.-E., Joffre, S., Rasmussen, A., Tercier, P., Review and
512 intercomparison of operational methods for the determination of the mixing height. Atmos.
514 Shi, C., Yang, J., Qiu, M., Hao, Z., Su, Z., Li, Z., 2010. Analysis of an extremely dense regional
515 fog event in Eastern China using a mesoscale model. Atmos. Res. 95, 428-440. doi:
516 10.1016/j.atmosres.2009.11.006.
517 Stull, R. B., 1976. The energetics of entrainment across a density interface. J. Atmos. Sci. 33,
518 1260-1267.
519 Stull, R. B., 1988. An introduction to boundary layer meteorology. Dordrecht, Kluwer Academic
521 Sun, J. N., 2009. On the parameterization of convective entrainment: Inherent relationships among
522 entrainment parameters in bulk models. Adv. Atmos. Sci. 26, 1005-1014. doi:
523 10.1007/s00376-009-7222-8.
524 Sühring, M., Maronga, B., Herbort, F., Raasch, S., 2014. On the effect of surface heat-flux
525 heterogeneities on the mixed-layer-top entrainment. Bound. Layer Meteorol. 151, 531-556.
527 van Zanten, M. C., Duynkerke, P. G., Cuijpers, J. W. M., 1999. Entrainment parameterization in
529 10.1175/1520-0469(1999)056<0813:EPICBL>2.0.CO;2.
530 Wang, J., Wang, S., Jiang, J., Ding, A., Mei, Z., Zhao, B., Wong, D. C., Zhou, W., Zheng, G.,
531 Wang, L., 2014. Impact of aerosol–meteorology interactions on fine particle pollution during
532 China’s severe haze episode in January 2013. Environ. Res. Lett. 9, 094002.
533 doi:10.1088/1748-9326/9/9/094002.
534 Yuskiewicz, B. A., Orsini, D., Stratmann, F., Wendisch, M., Wiedensohler, A., Heintzenberg, J.,
535 Martinsson, B. G., Frank, G., Wobrock, W., Schell, D., 1998. Changes in submicrometer
536 particle distributions and light scattering during haze and fog events in a highly polluted
538 Zhang, R, Li, Q., Zhang, R, 2014. Meteorological conditions for the persistent severe fog and haze
539 event over eastern China in January 2013. Sci. China-Earth Sci. 57, 26-35.
540 doi:10.1007/s11430-013-4774-3.
541 Zou, J., Sun, J., Ding, A., Wang, M., Guo, W., Fu, C., 2017. Observation-based estimation of
542 aerosol-induced reduction of planetary boundary layer height. Adv. Atmos. Sci. 34,