Analysis of heavy pollution weather in Shenyang City and numerical simulation of main pollutants

Yunfeng Ma; Huijie Zhao; Lei Feng; Siyu Jin; Di Zhao; Shuai Wang; Qiyao Liu; Kunyu Gao; Zhengqing Xu

doi:10.1515/geo-2022-0415

Article Open Access

Analysis of heavy pollution weather in Shenyang City and numerical simulation of main pollutants

Yunfeng Ma , Huijie Zhao , Lei Feng , Siyu Jin , Di Zhao , Shuai Wang , Qiyao Liu , Kunyu Gao and Zhengqing Xu

Published/Copyright: November 13, 2023

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Open Geosciences Volume 15 Issue 1

Abstract

In the present research, a statistical analysis of all pollution incidents occurring from 2015 to 2019 in the cities of the urban agglomeration centered on Shenyang was performed. The results indicated that heavy pollution mainly occurred during the heating season, and the main pollutants were (particulate matter) PM_2.5 and PM₁₀. It was also determined that the heavy pollution that occurred during the heating season in Shenyang was of the soot type. The weather research forecast-chemistry (WRF-Chem) was used to simulate the meteorological elements and particle concentration during the two heavy pollution periods in 2019 and compared the simulation data with the monitoring data to verify the simulation performance of the model. Results demonstrated that the model had a better simulation effect on temperature and pressure than on wind speed and wind direction. By comparing the hourly particle concentration data, it was found that the simulation results for pollutants obtained with the WRF-Chem model were lower than the measured values. The simulation effect on PM_2.5 was better than that on PM₁₀, and the simulation results were basically consistent in the high- and low-value areas, and the time of peak and valley was basically synchronous. It was proven that the selected parameterization scheme properly simulated the weather situation and changes in pollutants during heavy pollution events in the Shenyang area. These results verified the application value of the WRF-Chem model during the investigation of heavy pollution events.

Keywords: WRF-Chem; PM2.5 ; PM10 ; simulation

1 Introduction

In recent years, along with economic development, air pollution in China has severely increased. Atmospheric pollution dominated by high concentrations of particulate matter has increased over the years. The environmental problems caused by air particulate pollutants have aroused widespread concern [1,2]. The pollution caused by atmospheric particles affects atmospheric visibility as well as environmental and human health [3,4].

As a national heavy industry base, Liaoning province also faces the problem of air pollution. The occurrence of air pollution is not only related to local emission sources. Changes in meteorological conditions are also a key factor in the occurrence of heavily polluted weather [5]. After analyzing air pollution characteristics in Liaoning province, it was found that the high pressure of the Changbai Mountain and the northeast topographic features helped in the formation of a stable atmospheric structure, which led to the accumulation of pollutants and the occurrence of heavy pollution weather [6]. Hong et al. [7] studied the effects of terrain and weather conditions in Northeast and North China on the haze pollution processes in the urban agglomerations in central Liaoning and the Beijing–Tianjin–Hebei region. Their results indicated cross-regional transportation of haze pollution. The main body of the central Liaoning urban agglomeration includes Shenyang, Anshan, Fushun, and Benxi. This area, where heavy industry is settled, has a long history of poor urban air quality, and pollution caused by particulate matter has occurred frequently in recent years [8]. According to previous research, the urban agglomerations in central Liaoning must adopt different measures and strategies aimed at controlling the regional pollution that is present nowadays.

Different atmospheric transport models have been applied to investigate air pollution processes. From the earliest Gaussian diffusion model to the two-dimensional diffusion model with relatively simple physical processes, to the three-dimensional diffusion model that has emerged in recent years, the research on air pollution is becoming more and more refined and accurate. In 2010, Xue et al. [9] used the Community Multiscale Air Quality Model to simulate the (particulate matter) PM_2.5 chemical composition of 333 cities across the country and applied the simulation results to the atmospheric environmental capacity accounting. In a heavily polluted weather study, Cao et al. [10] applied a diffusion model to analyze the chemical composition of PM_2.5 in the area of “2 +26” cities in Beijing–Tianjin–Hebei. Yu et al. [11] used a model to determine the effect of the regional transport of air pollutants in the central and eastern regions of China on pollution processes occurring in the middle reaches of the Yangtze River during the winter season. Wang et al. [12] used the Hybrid Single Particle Lagrangian Integrated Trajectories model to simulate the air mass trajectory during a pollution process in the urban agglomeration in central Liaoning. They identified a pollutant transport channel from the Beijing–Tianjin–Hebei region, which provided a good option for the prevention and effective control of regional pollution. Zou et al. [13] used the model to simulate the diffusion of air pollutants in Shenyang on 5 January 2006. Their results showed that the California Puff model could intuitively and truly simulate the distribution of pollutants in the atmosphere and found that the distribution of daily average concentrations was affected by the average wind. Also, they identified the dual influence of field and topography. Finally, these researchers determined that the wind field affected the distribution of real-time concentrations of pollutants. Zou et al. [14] simulated the daily average concentration distribution of PM₁₀ in Shenyang in the winter of 2010, and the simulation results showed that PM₁₀ in Shenyang was distributed eastward, southeast, and northwest, which were consistent with the surface wind direction.

The weather research forecast-chemistry (WRF-Chem) model was developed by the National Oceanic and Atmospheric Administration Forecast Laboratory in the United States. It represents a new generation of regional air quality models that couple online the meteorological model and the chemical model. The WRF-Chem model contains detailed atmospheric physical and chemical processes and is widely used in the study of air pollution incidents [15,16]. Grell et al. [17] used WRF-Chem to develop a 27-km horizontal resolution forecast and compared the results with those of other models. In this case, data showed that WRF-Chem has high accuracy in forecasting O₃ and PM_2.5. Saide et al. [18] used the WRF-Chem model to cooperate with the local government to prevent and control the haze pollution caused by PM₁₀ and PM_2.5, and the results showed that the WRF-Chem model was accurate in determining the concentration of pollutants. Similar to the central city of Liaoning province, Shenyang is an old industrial base in Northeast China, which is characterized by displaying frequent soot-type pollution. When heavy pollution occurs, PM_2.5 and PM₁₀ are the main pollutants [19]. With the implementation of a series of pollution control programs and supply-side reforms, the number of industrial enterprises that surpass the allowed size in Shenyang has been reduced from 5,252 to 1,416 since 2010 [20]. However, air pollution in this city is still higher as compared to that in other parts of the same province.

The model operation parameters suitable for the Shenyang local weather were selected to run the pollution processes. Two heavy pollution processes were: (1) from January 11th to January 15th; and (2) from February 27th to March 5th. On the one hand, the distribution characteristics of meteorological elements and weather system characteristics on the ground and at high altitude in Shenyang during heavy pollution were studied, which provides a theoretical basis for formulating scientific and effective heavy pollution control measures. On the other hand, through statistical analysis, correlation analysis, and time series comparison of the simulation results and online monitoring data, the simulation ability of WRF-Chem for the particulate pollution process in Shenyang was verified.

2 Data and methodology

2.1 Data

The initial and boundary data of WRF-Chem are the final operational global analysis reanalysis data of the National Centers for Environmental Prediction and the National Center for Atmospheric Research (http://rda.ucar.edu/) with a spatial resolution of 1° × 1° and a temporal resolution of 6 h (00:00, 06:00, 12:00, and 18:00).

Pollutant monitoring data (PM_2.5 and PM₁₀) and weather data (wind speed, wind direction, air pressure, humidity, etc.) were provided by the Shenyang Environmental Monitoring Center Station. Figure 1 shows the latitude and longitude of the 11 state-controlled environmental air automatic monitoring points in Shenyang. The monitoring data covers the cities and rural areas in Shenyang [21]. The time resolution of the collected data was in hours. PM_2.5 and PM₁₀ concentrations were monitored by Thermo 5030i Sharp Particulate Monitor, and the method used for concentration analysis was the β-ray absorption method-light scattering [22]. The missing and invalid data in the original data were eliminated, and then the linear interpolation method was used to supplement the time series of pollutants. Calculations were performed using the following equation:

(1) X i = X n + ( i − n ) ( X m − X n ) / ( m − n ) , n < i < m ,

Figure 1

Study area and monitoring point.

where X _i is the missing and invalid data; X _m is the concentration value before X _i; and X _n is the concentration value after X _i. If there are too many missing and invalid data for each hour of the day, it is replaced with the hourly data of the adjacent date.

2.2 Methodology

2.2.1 Model setting and scheme design

The simulation area of the WRF-Chem model was centered on the Hunnan East Road (41.75°N, 123.53°E). This simulation used the EDGAR-HTAP (Global Atmospheric Research Emissions Database-Hybrid Trading and Analytical Processing) emission inventory with a spatial resolution of 0.1° × 0.1° [23]. As shown in Figure 2, two layers of grid nesting were adopted. The resolution of grid d01 was 27 km, and the resolution of grid d02 was 9 km. The simulation output result d02 of the inner area was selected as the research object. The grid included the entire Liaoning province and the other surrounding areas. The specific parameterization scheme settings of the specific physical and chemical processes in the WRF-Chem mode are shown in Table 1.

Figure 2

WRF-Chem nested grid.

Table 1

Selection of a parameterization scheme

Parameterized scheme	Parameterization scheme selection
Microphysical process scheme	Lin plan
Near ground floor plan	MM5 Monin–Obukhov scheme
Land model	Noah Land Surface Model
Boundary layer scheme	Yonsei university plan
Cumulus solution	Kain–Fritsch scheme
Meteorological chemistry mechanism	RADM2
Aerosol mechanism	MADE/SORGAM
Photolysis solution	Madronich photolysis (TUV) scheme

2.2.2 Verification of results

In order to quantitatively describe the accuracy of the model, the mean bias (MB) and the root mean square error (RMSE) were selected to verify the simulation results of meteorological elements. In addition, the normalized MB (NMB), the normalized mean error (NME), the mean fractional bias (MFB), and the mean fractional error (MFE) were used to verify the simulation results of particulate matter. The calculation method is shown in equations (2)–(6). MFE shows the real situation of the simulation error; RMSE indicates the degree of deviation of the simulated data from the real data. In this case, the smaller the value, the higher the accuracy of the simulation. The ideal simulation effects of the model were MFB ≤ ±30%, MFE ≤ ±50%, and MFB ≤ ±60%. In addition, MFE ≤ ±75% was used as the standard in order to verify the simulation results of particulate matter [24,25].

The observed value and the simulated value were used for fitting analysis, and the Pearson correlation coefficient (R) was used to reflect the relationship between the simulated value and the detected value. The calculation formula was as follows in equation (7).

(2) MB = 1 N ∑ i = 1 N ( m − b ) ,

(3) NMB = ∑ i = 1 N ( m − b ) ∑ i = 1 N b × 100 %,

(4) NME = ∑ i = 1 N | m − b | ∑ i = 1 N b × 100 %,

(5) MFB = 2 N ∑ i = 1 N m − b m + b × 100 % ,

(6) MFE = 2 N ∑ i = 1 N ∣ m − b ∣ ∣ m + b ∣ × 100 % ,

(7) R = ∑ i = 1 N ( m − m ¯ ) ( b − b ¯ ) ∑ i = 1 N ( m − m ¯ ) 2 ∑ i = 1 N ( b − b ¯ ) 2 ,

where m is simulated data; m ¯ is the average value of the simulated data; b is observed data; b ¯ is the average value of the observed data.

3 Statistical analysis of pollution incidents

3.1 Definition of the pollution incident

The urban agglomeration in central Liaoning, centered on Shenyang, includes Anshan, Tieling, Liaoyang, Fushun, Benxi, and Yingkou. A statistical analysis of all pollution processes occurring from 2015 to 2019 was performed, in addition to the analysis of the time distribution characteristics of heavy pollution events and the primary pollutants.

We referred to the ⟪Ambient Air Quality Index (AQI) Technical Regulations (Trial)⟫ (HJ 633-2012) [26] to classify the range of regional AQI values. The classification basis is shown in Table 2. In the event that the regional AQI value reached a certain pollution level for at least six consecutive hours, this was defined as a pollution process. If the interval between the two pollution processes was less than 6 h, the two pollution processes were combined into one.

Table 2

Classification of the air quality index

AQI	Air quality level	Air quality category
0–50	1	Excellent
51–100	2	Good
101–150	3	Slight pollution
151–200	4	Moderate pollution
201–300	5	Heavy pollution
＞300	6	Severe pollution

According to climate season (QX/T 152-2012), seasons in Shenyang were divided into spring (April and May), summer (June, July, and August), autumn (September and October), and winter (November and December and January, February, and March of the following year). Shenyang is heated throughout the winter, and pollution occurs frequently in winter. In the manuscript, winter was defined as the heating season.

3.2 Temporal variation of air pollution

As shown in Figure 3, heavy and above pollution events mainly occurred during the heating season. Heavy and above pollution processes with a shorter duration appeared in April. Heavy pollution events occurred in May and October. However, during this same period, no severe pollution processes appeared. From June to September, no heavy and above pollution processes appeared.

Figure 3

Monthly pollution duration from 2015 to 2019.

As shown in Figure 4, the heavy and above pollution events were relatively high between 14:00, 17:00, and 21:00. A total of 23 heavy and above pollution processes occurred during this period, accounting for 50.0% of the total. With regard to the nine severe pollution processes, the starting time for the events occurred twice at 19:00 and once at 00:00, 2:00, 3:00, 8:00, 11:00, 13:00, and 15:00. In addition, heavy and above pollution processes’ ending times were from 0:00 to 2:00, 5:00, 10:00, and 11:00. A total of 28 heavy and above pollution processes occurred during this period, accounting for 60.9% of the total. The nine severe pollution processes ended twice at 00:00 and once at 8:00, 9:00, 10:00, 12:00, 16:00, 17:00, and 23:00.

Figure 4

Distribution of the start and end times of the pollution process from 2015 to 2019.

3.3 Analysis of the primary air pollutants

According to the statistical analysis of the heavy and above pollution processes from 2015 to 2019, it was found that the main pollutants were PM_2.5 and PM₁₀. Among these, PM_2.5 was the main pollutant. In the heavy and above pollution event, the cumulative time where PM_2.5 was the primary pollutant was 872 h, which accounted for 92.0% of the total pollution process time. In addition, in the severe pollution process, the cumulative time where PM_2.5 was the primary pollutant was 103 h, which represented 66.4% of the total process time.

4 Numerical simulation results and analysis

In the first quarter of 2019, Shenyang launched two emergency plans for heavily polluted weather. For two heavy pollution incidents occurring from January 11th to January 15th (process 1), and from February 27th to March 5th (process 2), the primary pollutants were PM_2.5 and PM₁₀. For these events, the meteorological data and particulate matter were simulated, and the results were compared with the real-time monitoring results in order to verify the simulation capabilities of WRF-Chem.

4.1 Analysis of heavy pollution processes

Process 1: From the 11th to the 12th (Figures 5 and 7 and and 8), large-scale pollution occurred in the Beijing–Tianjin–Hebei region. The surface weather chart shows that at about 15:00 on the 11th, Shenyang was dominated by a southwest wind, and the wind direction determines the distribution of atmospheric pollutant concentrations [27]. Because of airflow, pollutants in the Beijing–Tianjin–Hebei region reached Shenyang and surrounding cities. The ground in the Shenyang area was controlled by weak low pressure, the atmosphere was in a neutral or unstable state, the bottom air convergence increased, and the pollutants near the ground were sent to a high altitude. The 700 and 850 hPa altitude maps indicated that the high-altitude humidity in Shenyang was about 60%, the high-altitude wind speed was about 2 m/s, and the airflow direction changed from southwest to west. High humidity promoted the growth of hygroscopic particles in pollutants, reduced visibility, and promoted the formation of secondary particles, resulting in the aggravation of pollution [28,29,30].

Figure 5

Ground weather map (a–c), 700 hPa (d–f), and 850 hPa (g–i) at 15:00 on January 11th and at 0:00 and 21:00 on January 12th.

On the 13th (Figures 6–8), the ground isobar was relatively sparse, the pressure gradient was small, the ground wind speed and the temperature were low, and the stratification of the lower atmosphere was stable. These situations were not favorable for the horizontal and vertical diffusion of pollutants, which continued to accumulate [31,32,33]. In the early morning of the 14th, pollution aggravated to the severe pollution stage; during the afternoon of the 14th, the wind direction near the ground turned northwest, the ground temperature increased, and the wind speed at high altitudes increased. The ground atmosphere converged, and the atmosphere moved upward. The divergence of the airflow and the increase in high-altitude wind speed improved the diffusion conditions, reducing the concentration of pollutants, improving the weather conditions, and ending the pollution event.

Figure 6

Ground weather map (a–c), 700 hPa (d–f), and 850 hPa (g–i) at 21:00 on January 13th; 6:00 on January 14th; and 00:00 on January 15th.

Figure 7

Concentrations of PM_2.5 on: January 11th at (a) 15:00; January 12th at (b) 12:00 and (c) 21:00; January 13th at (d) 21:00; January 14th at (e) 6:00; and January 15th at (f) 00:00.

Figure 8

Concentrations of PM₁₀ on: January 11th at (a)15:00; January 12th at (b) 12:00 and (c) 21:00; January 13th at (d) 21:00; January 14th at (e) 6:00; and January 15th at (f) 00:00.

Process 2: From the evening of the 26th to the early morning of the 27th (Figures 9 and 11 and and 12), a strong temperature inversion occurred in Shenyang. The ground temperature reached 8°C at 3:00 on the 27th. The ground was controlled by high pressure, and the small pressure gradient caused the low wind speed, forming a relatively stable and horizontally transmitted weather background field [34]. The high-altitude humidity was between 60 and 70%, and the air humidity was relatively high. This high relative humidity favored the adhesion of atmospheric particulates to the water vapor, increasing the concentration of particulates. In weather with high-humidity conditions, the phenomenon of temperature inversion is often present, and the particulate matter in the air is more difficult to diffuse [35]. Thus, pollution in Shenyang continued to increase.

Figure 9

Ground weather map (a–c), 700 hPa (d–f), and 850 hPa (g–i) at 12:00 on February 26th; 12:00 on March 1st; and 12:00 on March 3rd.

From March 1st to 2nd (Figures 10–12), large-scale severe pollution occurred in the Beijing–Tianjin–Hebei region. During the day on March 2nd, the surface temperature in Shenyang was 10°C, the southwest wind speed on the ground was 4 m/s, and the upper-air wind speed was as high as 6 m/s. The polluted air masses over the Tianjin–Hebei area and Bohai Bay were transported to Shenyang with the southwest airflow. In addition, on the evening of March 2nd, a strong temperature inversion occurred. The humidity in Shenyang was relatively high, and the vertical and horizontal diffusion conditions were poor. The static and high-humidity meteorological conditions caused heterogeneous chemical reactions of pollutants [36,37]. Thus, the pollution process continued to increase.

Figure 10

Ground weather map (a–c), 700 hPa (d–f), and 850 hPa (g–i) at 00:00 on March 3rd; 6:00 on March 4th; and 12:00 on March 5th.

Figure 11

Concentration distribution of PM_2.5 on February 27th at (a) 00:00; February 28th at (b) 21:00; March 1st at (c) 12:00; March 2nd at (d) 12:00; March 4th at (e) 6:00; and March 5th at (f) 12:00.

Figure 12

Concentration distribution of PM₁₀ on February 27th at (a) 00:00; February 28th at (b) 21:00; March 1st at (c) 12:00; March 2nd at (d) 12:00; March 4th at (e) 6:00; and March 5th at (f) 12:00.

From March 2nd to 5th, the Shenyang area was controlled by a weak high pressure, which was accompanied by a weak downdraft. This favored the transmission of low-temperature air from the upper floors, inhibiting the development of a mixed layer and helping in the formation of temperature inversions. This process combined with the low wind speed on the ground was not favorable for the diffusion of pollutants, leading to a cumulative increase in their concentrations [38]. During the evening of March 5th, the surface weather system was dominated by weak low pressure, the lower atmosphere converged, the air current increased, and the upper atmosphere diverged. In the upper air map, the relative humidity was about 45%, and the wind speed was relatively high. During the cold advection control, the vertical convection of the atmosphere was strong, and the disturbance was good. These conditions favored the entry of strong cold air, the diffusion conditions were improved, and the pollutants were gradually dissipated.

4.2 Validation of simulation results of meteorological elements

As shown in Table 3, WRF-Chem provided good simulation effects for temperature and air pressure in pollution processes 1 and 2, but the simulation effect on wind direction and wind speed was relatively poor.

Table 3

Statistical analysis of observed and simulated values of meteorological elements

Process	Meteorological	Observed	Simulated	MB	RMSE
1	Temperature (°C)	−6.53	−6.4	0.13	2.81
	Air pressure (hPa)	1025.68	1020.64	−5.04	5.11
	Wind direction (°)	183.33	222.34	39.01	101.11
	Wind speed (m/s)	1.94	3.57	1.62	2.07
2	Temperature (°C)	1.99	3.68	1.69	3.71
	Air pressure (hPa)	1019.64	1015.18	−4.46	4.61
	Wind direction (°)	156.33	205.52	49.19	104.08
	Wind speed (m/s)	1.85	3.75	1.9	2.37

In the time series diagram displayed in Figures 13 and 14, the observed and simulated values of temperature and air pressure were consistent during the heavy pollution. The simulated peak and valley values were consistent with the observed values, and the occurrence time was basically synchronized with the observed values. The fitting diagram was shown in Figure 15, the correlation coefficients of temperature (R ₁ = 0.8500, R ₂ = 0.9252) and air pressure (R ₁ = 0.9597, R ₂ = 0.9693) were relatively higher than wind speed (R ₁ = 0.7044, R ₂ = 0.7991) and wind direction (R ₁ = 0.5674, R ₂ = 0.6055) in process 1 and process 2. It indicated that the model had a good simulation effect on temperature and pressure, but the simulated values were smaller than the observed values on the whole. The simulation results of wind speed were consistent in the high- and the low-value areas. The simulated value was larger than the observed value as a whole, but it was lower than the observed value in the high-value area, which may be due to the influence of the earth’s underlying surface. The roughness of different underlying surfaces, radiation balance, and soil vegetation may have a great impact on airflow and climate. The simulation results of wind direction were worse, the instantaneous fluctuation with time was also relatively large, and the correlation coefficient was the lowest.

Figure 13

Comparison of (a) temperature, (b) air pressure, (c) wind direction and (d) wind speed between observed and simulated values of process 1.

Figure 14

Comparison of (a) temperature, (b) air pressure, (c) wind direction and (d) wind speed between observed and simulated values of process 2.

Figure 15

Correlation of the observed and simulated values for meteorological elements during period 1 (a–d) and period 2 (e–h).

4.3 Verification of particle simulation results

MFB and MFE results shown in Table 4 indicated that PM_2.5 and PM₁₀ during pollution processes 1 and 2 were within the acceptable range. In addition, the simulation results of PM_2.5 in pollution period 1 reach the ideal effect of model simulation.

Table 4

Statistical analysis of observed and simulated values of particulate matter

Process	Pollution	Observed (μg/m³)	Simulated (μg/m³)	MB	NMB (%)	NME (%)	MFB (%)	MFE (%)
1	PM_2.5	111.83	88.64	−23.19	−20.74	42.77	−6.96	52.42
1	PM₁₀	157.98	100.90	57.08	56.57	79.96	30.71	61.71
2	PM_2.5	167.92	95.60	−72.32	−43.07	48.07	−50.39	65.73
2	PM₁₀	227.05	115.34	−111.71	−49.20	53.80	−56.24	68.74

The time series diagram (Figures 16 and 17) proved that in processes 1 and 2, the simulation results were consistent in the high-value area and low-value area, and the peak and valley times were also more consistent. In addition, the fitting diagram (Figure 18) showed that the simulation effect of WRF-Chem mode for PM_2.5 (R ₁ = 0.6286, R ₂ = 0.7639) was better than that for PM₁₀ (R ₁ = 0.6243, R ₂ = 0.5464).

Figure 16

Comparison of the observed and simulated values for (a) PM_2.5 and (b) PM₁₀ concentration during period 1.

Figure 17

Comparison of the observed and simulated values for (a) PM_2.5 and (b) PM₁₀ concentration during period 2.

Figure 18

Correlation of the observed and simulated values for PM_2.5 and PM₁₀ concentrations during period 1 (a and b) and period 2 (c and d).

The simulated values of PM_2.5 and PM₁₀ were lower than the observed values. There are many reasons for simulation errors. In addition to the error of the WRF-Chem model, the uncertainty of the spatial positioning of the emission inventory is also one of the reasons. The traditional inventory often allocates the space for emission documents according to the parameters such as population and gross domestic product, while the heavily polluted industries gradually move to sparsely populated places such as suburbs, which will make the pollutant concentration of urban stations high. At the same time, human activities also have great uncertainty, and the model may not accurately predict local real-time pollution. On the whole, the changing trend of pollutant (PM_2.5 and PM₁₀) concentration was consistent with the actual situation, which had certain reference values for studying the formation and dissipation processes of heavily polluted weather in Shenyang and analyzing its evolution process.

5 Conclusion

The severe and heavy pollution processes mainly occurred during the heating season (January–March, November, and December), and the pollutants were PM_2.5 and PM₁₀, of which PM_2.5 was the main pollutant.
Combining the surface weather map and the high-altitude weather map during the heavy pollution period, it was proven that under the weather conditions of high pressure, high humidity, low wind speed, and temperature inversion, heavy pollution was prone to occur. Also, warm advection was present at high altitudes. It is likely that warm advection occurred in front of and behind the ridge, causing severely polluted weather. Moreover, when the ground was controlled by high pressure, the high-pressure central airflow descended and diverged, further aggravating the pollution process.
The WRF-Chem model displayed the best simulation effect on temperature and air pressure, compared to that on wind speed and wind direction. The simulation effect on wind direction was lower than the observation value because the instantaneous fluctuation was high with the change of time. The simulation effect of wind speed was the worst of all.
The simulation results for pollutants were lower than the observed values. Specifically, the simulation effect for PM_2.5 was better than that for PM₁₀. The results also showed that the simulation results were consistent in the high- and low-value areas. In addition, the peak and valley times were also synchronized.

Acknowledgments

Thanks are due to the authors Yunfeng Ma, Lei Feng, Siyu Jin and Qiyao Liu for assistance with the experiment, and to Shuai Wang, Di Zhao, Kunyu Gao and Zhengqing Xu for the valuable discussion.

Funding information: The research was financially funded by the following projects: “New Infrastructure + University Informatization” Research Project: Research on the Evaluation of Course Teaching and Training Platform Based on the Environmental Impact of Virtual Reality (Grant No. XJJ202205008); 2021 Liaoning Provincial Natural Science Foundation-funded Project: Research on Integrated Air-ground Early Warning Technology for Toxic and Harmful Gases in Chemical Industry Parks (Grant No. 2021-MS-079); Liaoning Provincial Department of Ecology and Environment’s 2021 Eco-environmental Scientific Research Plan: Research on The Fine Simulation Technology of Block-level Air Pollution Based on The Three-Dimensional Model of the Urban Underlying Surface (Grant No. 2021-036); 2021 Shenyang Science and Technology Talent Project (Young and Middle-aged Science and Technology Innovation Talent Support Program): Chemical Park Toxic and Harmful Gas Early Warning Technology Research Project (Grant No. RC210349).
Author contributions: Y.M. and H.Z. designed the experiments. H.Z., L.F., S.J., D.Z., S.W., Q.L., K.G., and Z.X. performed the experiments, analyzed the results, and wrote the article.
Conflict of interest: 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 article.
Data availability statement: The raw data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

References

[1] Su WK, Bao XL, Ni SY, Zhao WF. Typical characteristics of haze pollution in Shijiazhuang City in autumn and winter in 2018. Env Sci. 2019;40(11):4755–63.Search in Google Scholar

[2] Gui K, Che HZ, Wang YQ, Wang H, Zhang L, Zhao HJ, et al. Satellite-derived PM2.5 concentration trends over Eastern China from 1998 to 2016: Relationships to emissions and meteorological parameters. Env Pollut. 2019;247:1125–33.10.1016/j.envpol.2019.01.056Search in Google Scholar PubMed

[3] Dockery DW, Pope CA. Acute respiratory effects of particulate air pollution. Annu Rev Publ Health. 1994;15:107–32.10.1146/annurev.pu.15.050194.000543Search in Google Scholar PubMed

[4] Conner WD, Bennett RL, Weathers WS, Wilson WE. Particulate characteristics and visual effects of the atmosphere at Research Triangle Park. J Air Waste Manage Assoc. 1991;41(2):154–60.10.1080/10473289.1991.10466832Search in Google Scholar

[5] Miao YC, Guo JP, Liu SH, Zhao C, Li XL, Zheng G, et al. Impacts of synoptic condition and planetary boundary layer structure on the transboundary aerosol transport from Beijing-Tianjin-Hebei region to Northeast China. Atmos Env. 2018;181:1–11.10.1016/j.atmosenv.2018.03.005Search in Google Scholar

[6] Ma YJ, Liu NW, Wang YF. Distribution characteristics of air pollution cause of serious pollution in the mid cities of Liaoning province. Urban Env Urban Ecol. 2006;6:32–5.Search in Google Scholar

[7] Hong Y, Ma YJ, Wang XQ, Zhang YH, Zhong ZY, Wang YF, et al. The external influence of haze weather in the urban agglomeration in central Liaoning: A case study. J Env Sci. 2013;33(8):2115–22.Search in Google Scholar

[8] Ma YJ, Wang YF, Liu NW. Numerical simulation of the temporal and spatial distribution characteristics of main air pollutants in the urban agglomeration in central Liaoning. J Meteorol Env. 2006;22(2):7–10.Search in Google Scholar

[9] Xue WB, Fu F, Wang JN, He KB, Lei Y, Yang JT, et al. Modeling study on atmospheric environmental capacity of major pollutants constrained by PM2.5 compliance of Chinese cities. China Environ Sci. 2014;34(10):2490–6.Search in Google Scholar

[10] Cao YQ, Wang TJ, Han JC, Wang DY, Xie XD, Wu H, et al. Study on chemical composition and source apportionment of PM2.5 during a pollution episode in “2+26” cities. Acta Sci Circum. 2020;40(02):361–72.Search in Google Scholar

[11] Yu C, Zhao TL, Bai YQ, Zhang L, Chang JC. Heavy air pollution with a unique “non-stagnant” atmospheric boundary layer in the Yangtze River middle basin aggravated by regional transport of PM2.5 over China. Atmos Chem Phys. 2020;20(12):7217–30.10.5194/acp-20-7217-2020Search in Google Scholar

[12] Wang XQ, Yang T, Wang ZF. The impact of haze pollution across control areas: A case study of haze pollution in Beijing-Tianjin-Hebei and Northeast China. Climatic Env Res. 2011;16(06):690–6.Search in Google Scholar

[13] Zou XD, Yang HB, Liu YC. Application of CALPUFF in air pollution simulation research in Shenyang area. J Meteorol Env. 2008;24(6):24–8.Search in Google Scholar

[14] Zou XD, Yang HB, Liu YC, Yun YH. Numerical simulation of haze weather in eastern Liaoning. Enviro Prot Circ Econ. 2016;36(10):53–61.Search in Google Scholar

[15] Fei J, Wang TJ, Wang TT, Xie M, Zhao H. Numerical modeling of a continuous photochemical pollution episode in Hong Kong using WRF-chem. Atmos Env. 2008;42(38):8717–27.10.1016/j.atmosenv.2008.08.034Search in Google Scholar

[16] Wei W, Lv ZF, Li Y, Wang LT, Cheng SY, Liu H. A WRF-Chem model study of the impact of VOCs emission of a huge petro-chemical industrial zone on the summertime ozone in Beijing, China. Atmos Env. 2018;175:44–53.10.1016/j.atmosenv.2017.11.058Search in Google Scholar

[17] Grell GA, Peckham SE, Schmitz R, Mckeen SA, Frost G, Skamarock WC, et al. Fully coupled online chemistry within the WRF model. Atmos Env. 2005;39(37):6957–75.10.1016/j.atmosenv.2005.04.027Search in Google Scholar

[18] Saide PE, Carmichael GR, Spak SN, Gallardo L, Osses AE, Mean-Carrasco MA, et al. Forecasting urban PM10 and PM2.5 pollution episodes in very stable nocturnal conditions and complex terrain using WRF-Chem CO tracer model. Atmos Env. 2011;45:2769–80.10.1016/j.atmosenv.2011.02.001Search in Google Scholar

[19] Yang HB, Li YY, Zou XD. Analysis of weather types of moderately polluted and heavily polluted air in Liaoning. J Meteorol Env. 2009;25(6):15–7.Search in Google Scholar

[20] Shenyang Bureau of Statistics, Shenyang Survey Team of the National Bureau of Statistics. Shenyang Statistical Yearbook 2019. Beijing: China Statistics Press; 2019.Search in Google Scholar

[21] Li XL, Ma YJ, Wang YF, Liu NW, Hong H. Temporal and spatial analyses of particulate matter (PM10 and PM2.5) and its relationship with meteorological parameters over an urban city in Northeast China. Atmos Res. 2017;198:185–93.10.1016/j.atmosres.2017.08.023Search in Google Scholar

[22] Liu M. Study on the impact of setting off fireworks and firecrackers on air quality during the Spring Festival in Shenyang. Env Prot Circ Econo. 2018;38(6):57–60.Search in Google Scholar

[23] Janssens-maenhout G, Dentener F, Van Aardenne J, Monni S, Pagliari V, Orlandini Z, et al. EDGAR-HTAP: A harmonized gridded air pollution emission dataset based on national inventories. Ispra: European Commission Publications Office; 2012.Search in Google Scholar

[24] Boylan JW, Russell AG. PM and light extinction model performance metrics, goals, and criteria for three-dimensional air quality models. Atmos Env. 2006;40(26):4946–59.10.1016/j.atmosenv.2005.09.087Search in Google Scholar

[25] Wang LT, Zhang Y, Wang K, Zheng B, Zhang Q, Wei W. Application of weather research and forecasting model with chemistry (WRF-Chem) over northern China: Sensitivity study, comparative evaluation, and policy implications. Atmos Env. 2014;124:337–50.10.1016/j.atmosenv.2014.12.052Search in Google Scholar

[26] Department of Science, Technology and Standards, Ministry of Environmental Protection. Technical regulations for ambient air quality index (AQI) (Trial): HJ633-2012. Beijing: China Environmental Science Press; 2012.Search in Google Scholar

[27] Li ZK, Pan YX, Sun RQ. Principles and application of air pollution meteorology. Beijing: Meteorological Press; 1985.Search in Google Scholar

[28] Ding YH, Liu YJ. The long-term characteristics of fog and haze in my country in the past 50 years and its relationship with atmospheric humidity. Sci China Earth Sci. 2014;44(1):37–48.10.1007/s11430-013-4792-1Search in Google Scholar

[29] Hu M, Shang DJ, Guo S, Wu ZJ. Mechanism of new particle formation and growth as well as environment effects under complex air pollution in China. Acta Chim Sin. 2016;74(5):385–91.10.6023/A16020105Search in Google Scholar

[30] Chen WC, Tang HZ, Zhao HM. Diurnal, weekly and monthly spatial variations of air pollutants and air quality of Beijing. Atmos Env. 2015;119:21–34.10.1016/j.atmosenv.2015.08.040Search in Google Scholar

[31] He XH, Ma JZ, Xu J, Ma ZQ, Xue M, Jin JL. Simulation of a heavy PM2.5 pollution event over Beijing-Tianjin-Hebei region in October 2014. Meteor Mon. 2016;42(7):827–37.Search in Google Scholar

[32] Miao YC, Guo JP, Liu SH, Liu H, Li ZQ, Zhang WC, et al. Classification of summertime synoptic patterns in Beijing and their associations with boundary layer structure affecting aerosol pollution. Atmos Chem Phys. 2017;17:3097–3110.10.5194/acp-17-3097-2017Search in Google Scholar

[33] Guo JP, Miao YC, Zhang Y, Liu H, Li ZQ, Zhang WC, et al. The climatology of planetary boundary layer height in China derived from radiosonde and reanalysis data. Atmos Chem Phys. 2016;16:13309–19.10.5194/acp-16-13309-2016Search in Google Scholar

[34] Cao WH, Liang XD, Li QC. Phase characteristics and influencing factors analysis of a continuous haze process in Beijing. Acta Meteorol Sin. 2013;71(5):940–51.Search in Google Scholar

[35] Zhao CX, Wang YQ, Wang YJ, Zhang HL, Zhao BQ. Temporal and spatial distribution of PM2.5 and PM10 pollution levels in winter and spring in Beijing and their relationship with meteorological conditions. Env Sci. 2014;35(2):418–27.Search in Google Scholar

[36] Wang YJ, Bao SW, Wang SX, Hu YT, Shi X, Wang JD, et al. Local and regional contributions to fine particulate matter in Beijing during heavy haze episodes. Sci Total Env. 2017;580:283–96.10.1016/j.scitotenv.2016.12.127Search in Google Scholar PubMed

[37] Liu PF, Zhao CS, Gobel T, Hallbauer E, Nowak A, Ran L, et al. Hygroscopic properties of aerosol particles at high relative humidity and their diurnal variations in the North China Plain. Atmos Chem Phys. 2011;11(7):3479–94.10.5194/acp-11-3479-2011Search in Google Scholar

[38] Qi H, Zhang XL, Kang P, Schaefer K, Xiang WG, Deng ZC, et al. Analysis of characteristics and meteorological causes of surface ozone pollution in Chengdu during the COVID-19 epidemic. Acta Sci Circum. 2021;41(10):1–12.Search in Google Scholar

Received: 2021-07-01

Revised: 2022-09-18

Accepted: 2022-09-20

Published Online: 2023-11-13

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

https://doi.org/10.1515/geo-2022-0415

Keywords for this article

WRF-Chem; PM2.5; PM10; simulation

Creative Commons

BY 4.0