Attributions of emission-reduction and meteorological conditions to typical heavy pollution episodes in a cold metropolis, northeast China

Heavy pollution episodes frequently occurred in winter in northeast China due to the multiple anthropogenic emissions coupled with adverse meteorological conditions, which increased the difficulty of environmental pollution control. To better enact strategies for mitigating air pollution in the post-pandemic era, daily pollutant concentration monitoring and meteorological data were used to evaluate the changes and meteorological factors of air pollutants before (2019) and during (2020) the lockdown in Harbin City, northeast China. Moreover, typical pollution episodes under COVID-19 lockdown were identified, and their emission sources, meteorology conditions, and regional pollution transportation were analyzed. The results showed significant decreases in NO₂, PM₁₀ and CO, while O₃ increased, and no differences in PM_2.5 and SO₂ during the lockdown compared with non-lockdown periods. It indicated that reduced activities of transportation resulted in reductions of NO₂ concentrations by 16%, and stationary emission sources were less affected. Correlation between PM_2.5 and O₃ tended to change from positive to negative as the threshold of PM_2.5 = 90 μg m⁻³, with the main controlling factor changed from their common gaseous precursors to meteorological conditions (temperature <0°C and wind speed <2 m s⁻¹). Pollution days were concentrated in the COVID-19 lockdown period with PM_2.5 as the primary pollutant. SO₂ dominant pollution and PM_2.5 dominant pollution were distinguished from six sustained heavy pollution events. PM_2.5 and SO₂ played essential roles in SO₂ dominant pollution, which derived from local emissions of coal combustion and firework discharge. PM_2.5 dominant pollution might be chemical transformed from coal burning, vehicle exhaust, and other secondary precursors, which was affected and aggravated by CO, NO₂, high relative humidity and low wind speed affected by local emission and long-distance transport.

1 Introduction

Harbin is a cold metropolis, which is a provincial capital city that is located at the highest latitude and experiences the lowest temperature in China. As a heavily industrialized and agricultural city, it has significant atmospheric pollution due to its large population and history as a former industrial base in northeast China (Sun et al., 2023). In response to the severe air pollution, China has been actively implementing the Atmospheric Pollution Prevention and Control Action Plan (APPCAP) since 2013, significantly reducing the concentration of particulate matter (PM) pollution and effectively alleviating long-term haze conditions (Wang et al., 2017). But even so, PM_2.5-dominated atmospheric pollution frequently occurred during winter mouths against the backdrop of continuous improvement in annual air quality in China (Sun et al., 2020). There are many causes leading to Harbin’s severe PM pollution, the most important reason is the release of coal during the long heating season (Liu et al., 2020a). The heating period can last up to 6 months due to cold temperature, requiring extensive coal burning. Furthermore, adverse meteorological conditions controlled by the Siberian high pressure and Arctic Vortex may intensify the formation and enrichment of airborne PM and hinder pollutant dispersion, such as low temperature, high humidity, low wind speed, frequent temperature inversions (Cui et al., 2019; Luo et al., 2021). Human activities, such as straw burning and fireworks, also exacerbate its pollution, making it a highly polluted city (Zhang et al., 2020a; Chen et al., 2021a). In addition, frequent strong winds in the spring often carry dust particles to Harbin from other regions (Xie et al., 2006; Wang et al., 2021). Research on Harbin’s air pollution has focused on the spatial and temporal distribution and source apportionment of air pollutants, as well as the causes and evolution of pollution by individual pollutants (Liu et al., 2020b; Wang et al., 2020). However, a lack of research exists on the annual variation of atmospheric PM, leading to uncertainties in the implementation of pollution control measures.

Coronavirus disease 2019 (COVID-19) pandemic was first identified in December 2019 in Wuhan City, Hubei Province, which spread rapidly to other countries across the world. Efforts to combat the spread of COVID-19, the government began initiating their first-level response to a major public health emergency, e.g., entailed self-isolation, reducing personal travel and outdoor activities, and business closures across all sectors. The strict control measures and social lockdowns have consequently involved switching off most pollutant emission sources, which have had a notable impact on air pollutant concentrations (Chu et al., 2021). The social lockdown initiated during COVID-19 pandemic presented an opportunity to investigate the impacts of reduced anthropogenic activities on the air quality. However, Harbin City still experienced severe pollution events when almost all avoidable outdoor anthropogenic activities stopped under COVID-19 in 2020. In particular, a rare pollution event with air quality index (AQI) reached the maximum value of 500 was occurred, indicating the cause of air pollution could be more complex due to the adverse synoptic systems, complex emission systems, secondary transformation, and regional transmission (Sun et al., 2020), and significant challenges related to improving air quality is urgently required here.

The current study utilizes pollution concentration and meteorological data collected in Harbin in 2020 to analyze continuous heavy pollution periods and identify typical pollution events and their meteorological and anthropogenic causes using radar charts and statistical methods under COVID-19 pandemic. The TrajStat model is used to cluster the backward trajectories, and weighted potential source contribution factor (WPSCF) and concentration-weighted trajectories (WCWT) analyses are conducted to quantify the potential source areas of atmospheric PM and reveal the causes, influencing factors, and the transport of typical continuous heavy pollution processes in Harbin. The specific objectives of the current study were to 1) comparatively investigate the changes and meteorological factors of air pollutants before (2019) and during (2020) the lockdown, and discuss the impacts of COVID-19 lockdown on air pollutants; 2) determine the temporal variations of AQI values and primary pollutant throughout the year 2020, identify pollution events, distinguish typical heavy pollution types, and compare the significant differences between heavy pollution types; 3) explore the impacts of meteorology conditions on the typical heavy pollution types; and 4) investigate the transmission paths and potential pollution source areas of the typical heavy pollution types. The findings of this study will provide theoretical support for the prevention, control, and prediction of atmospheric pollution in urban areas in the post-pandemic era.

2 Material and methods

2.1 Study area

Harbin City is located in the southern part of Heilongjiang Province (between 125°42′–130°13′E and 44°04′–46°40′N), with an average elevation of 151 m (Figure 1). The Songnen Plain is located in the west, and the Xiaoxing’an Mountains are located in the northeast. The city is adjacent to the Zhangguangcai Mountains, which are part of the Changbai Mountains to the south. The terrain is characterized by slightly higher elevations in the east, lower and flatter areas in the west, and a gradual slope from east to west. Harbin has a temperate continental monsoon climate, with short and cool summers and long, cold winters. The temperature in 2020 ranged from −23°C to 28°C, and the total precipitation was 792.8 mm, with summer precipitation accounting for 59.5% of the yearly total. The snow season is from November to February. According to the Harbin Statistical Yearbook 2021, the city supplied 175.97 million GJ of heating water in 2020, with the heating period ranging from October 20th to April 20th. Coal emissions are significantly higher during the long heating period, and pollutants accumulate due to relatively poor atmospheric conditions in winter, leading to significant differences in air quality between the heating and non-heating seasons. Pollution days mostly occur during the heating period.

FIGURE 1

FIGURE 1. Map of China (A) and study area (B).

2.2 Data source

The daily air quality data (AQI, PM_2.5, PM₁₀, CO, NO₂, SO₂, O₃-8 h) for Harbin from 1 January 2019 to 31 December 2020 were obtained from the real-time national urban air quality platform of the China Environmental Monitoring Station (http://106.37.208.233:20035). The AQI classes were based on the “Technical Regulations for the Environmental Air Quality Index (Trial)” (HJ633-2012) and included excellent (0–50), good (51–100), mild pollution (101–150), moderate pollution (151–200), severe pollution (201–300), and serious pollution (>300). The ground meteorological data (temperature T), wind speed (WS), and relative humidity (RH)) were obtained from the China Meteorological Data Network (http://data.cma.cn/) as hourly monitoring data. The meteorological data used in the backward trajectory model were obtained from the Global Data Assimilation System (GDAS) data for the period of 2019–2020 and were provided by the United States National Centers for Environmental Prediction (NCEP) (ftp://ftp.arl.noaa.gov/pub/archives/gdas1/), with a spatial resolution of 1°×1° and a temporal resolution of 1 h. The land-use data were sourced from Esri (https://www.esri.com/) and were produced using Sentinel-2 satellite remote sensing images combined with artificial intelligence land classification models.

2.3 Radar chart

Radar charts are typically used to depict the normalized components of atmospheric pollutants (PM_2.5, PM_2.5-10, SO₂, NO₂, CO) at specific times or locations to eliminate the influence of the pollution concentrations of different pollutants. The upper and lower limits of pollutants can be compared with eigenvalue during the same period to determine the pollution type in the region (Duan et al., 2018).

The normalized component at specific times is defined as:

$Z_{ij}={c_i}_j/{\displaystyle \sum _{j=1}^n}{c}_{ij}(1)$

where Z_ij represents the normalized components of the jth type of pollutant at time i, and c_ij represents the original mass concentration (μg/m³) of the jth type of pollutant at time i.

The characteristic value of the pollutant at a specific time is defined as:

$C{V}_{ij}=Z_{ij}/{\overline{Z}}_j(2)$

where CV_ij represents the char.

acteristic value of the jth pollutant at time i; ${\overline{Z}}_j$ denotes the average normalized value of the jth pollutant during a certain period; Z_ij represents the normalized value of the jth pollutant at time i.

The characteristic the upper and lower limit values of the pollutants during a certain period are defined as:

$\text{The\hspace{0.17em}upper\hspace{0.17em}limit}:Ma{x}_j=\left({\overline{Z}}_j+S_j\right)/{\overline{Z}}_j(3)$

$\text{The\hspace{0.17em}lower\hspace{0.17em}limit}:Mi{n}_j=\left({\overline{Z}}_j-S_j\right)/{\overline{Z}}_j(4)$

where Max_j represents the upper limit of the jth pollutant, Min_j represents the lower limit of the jth pollutant, and S_j represents the standard deviation of the jth pollutant.

2.4 Backward trajectory model and potential source analysis

This study used backward trajectory clustering, WPSCF, and WCWT analysis of the atmospheric data of Harbin city using the TrajStat model (Draxler and Hess, 1998) in the Meteoinfo software developed by Wang et al. (2009). We used the daily PM_2.5 concentration data and GDAS meteorological data from 2020 to conduct a backward trajectory simulation of sustained heavy pollution events in Harbin. The city center (126.61^oE, 45.74^oN) was selected as the starting point for the computations. The previous period was 72 h, and the calculation interval was 2 h. The altitude was 500 m to incorporate the transport characteristics of near-surface air masses and reduce the influence of surface friction. The air mass trajectories in the study area were clustered using the Euclidean distance based on a significant increase in the total spatial variation (TSV), and the proportion of each trajectory category to the total trajectories was calculated.

The PSCF model has been widely used to identify potential pollution sources for high-concentration pollution events (Zong et al., 2018; Sun et al., 2020). It was used in this study to analyze the potential pollutant sources. Harbin city and its surrounding areas were divided into horizontal grids with a size of 0.25°×0.25°. The extent of the grid’s contamination was evaluated by determining the pollutant amount and time of the pollution trajectory in each grid as follows:

$PSC{F}_{ij}=m_{ij}/n_{ij}(5)$

where PSCF_ij represents the potential source contribution of a grid cell ij; m_ij represents the number of pollution trajectory endpoints with pollutant values exceeding a threshold in grid cell ij; n_ij represents the total number of trajectory endpoints passing through grid cell ij.

An analysis of the weighted potential source contribution factors (WPSCF) was conducted by using the weighting factor W_ij to eliminate uncertainties in grid cells with small n_ij and reduce the error values of PSCF. The weight functions are typically defined as follows (Zhang et al., 2019):

$W_{ij}=\left\{\begin{array}{l}1.0\left(n_{ij}>80\right)\\ 0.7\left(20<n_{ij}\le 80\right)\\ 0.42\left(10<n_{ij}\le 20\right)\\ 0.05\left(n_{ij}\le 10\right)\end{array}\right.(6)$

The PSCF analysis only provides the number of pollution trajectories within a grid cell. When the WPSCF values of grid cells are the same, it is impossible to differentiate the impact of the trajectory endpoints with different pollutant concentrations on the grid cells. However, the CWT method provides a more precise description of the influence of the trajectory endpoints with different pollutant concentrations on the grid cells. The CWT is calculated as follows:

$C_{ij}=\frac{1}{{\displaystyle \sum _{i=1}^n}{\tau }_{ij}}{C}_i{\tau }_{ij}(7)$

where C_ij represents the weighted average concentration of the (i,j) grid cell, i is the index of the trajectory, n is the total number of trajectories, C_l represents the concentration of pollutants corresponding to trajectory l passing through the network unit (i,j), and τ_ijl represents the duration of trajectory l remaining in grid cell (i,j). Similar to the PSCF method, the CWT method also cannot distinguish the number of trajectory endpoints within each grid cell. Therefore, this study employed the same weight factor W_ij. A higher WCWT value indicates a higher pollutant concentration in the grid cell and a greater contribution to pollution.

2.5 Statistical analysis

Statistical analysis and graphs were carried out using IBM SPSS Version 20.0 and OriginLab^@ OriginPro 2017C. Independent−samples t-test was applied to determine the statistical significant differences between lockdown and non-lockdown periods. Nonparametric Kolmogorov−Smirnov test was used to determine significance of the differences between two typical heavy pollution types. Pearson correlation coefficients were applied to test the linear correlation among atmospheric pollutants and meteorological data. The results were considered significant if a p-value was less than 0.05.

3 Results and discussion

3.1 Changes in atmospheric pollutant concentrations

Soon after the outbreak of COVID-19 pandemic, Wuhan announced lockdown on 23 January 2020 due to the dramatically increase of COVID-19 case numbers. Subsequently, Harbin government officially initiated first-level response to the major public health emergency on 26 January 2020 to reduce gatherings and travel, and this strictest control measure lasted until 2 May 2020. Nevertheless, along with the prohibition of almost all avoidable activities, severe air pollution events still occurred in Harbin (for example, PM_2.5 exceeded 900 μg/m³ on April 16), causing discussions regarding why severe air pollution was not avoided. To analyze the impacts of emission-reduction on the air quality, and also minimize the effects of meteorological conditions on temporal characteristics of air pollutants, pollutant concentrations during lockdown (January 26 to May 2) and non-lockdown (January 1 to January 25, May 3 to December 31) with respect to similar periods between 2019 and 2020 were further investigated (Figure 2).

FIGURE 2

FIGURE 2. Comparison of air pollutant concentrations during the corresponding periods of lockdown (A) and non-lockdown (B) between 2019 and 2020 (* indicates significant difference between the two periods).

During the lockdown period, statistically significant decrease in average concentrations of NO₂, PM₁₀ and CO were found under COVID-19 compared to levels in 2019 (p < 0.05), with the reductions of 16%, 25% and 13%, respectively (Figure 2A). The concentrations of PM_2.5 and SO₂ during the lockdown period in 2020 had no significant differences with those in 2019 with p values of 0.275 and 0.959, respectively. The relatively stable SO₂ concentrations indicated that the emissions from stationary sources, such as coal-fired heat plant, household heating systems, as well as firework discharge did not affected by the lockdown as much as traffic-related sources. However, significant concurrent decrease in PM_2.5 concentrations were found at global, continental, and national scales in the corresponding period (He et al., 2021). An important reason for this could be possibly due to more PM_2.5 were generated under adverse meteorological conditions since NO₂ and SO₂ are the key precursors of secondary PM_2.5 (Sun et al., 2023). It is worth noting that O₃ concentrations exhibited a significant increase by 23% during COVID-19 (p < 0.05). Lockdown measures such as restricting traffic, industry and community activities, have transformed the emission of primary pollutants into a simple pattern, which provides opportunities for the analysis of the causes of secondary pollution and the improvement of treatment plans. The more prominent cause of the increase of O₃ may due to the unbalanced reduction of NOx and VOCs emissions, which makes the rate of O₃ generation by VOCs greater than the rate of O₃ consumption by NO, resulting in a large amount of O₃ accumulation (Chu et al., 2021; Liu et al., 2023).

As shown in Figure 2B, during 2020 non-lockdown period, no statistically significant reduction was found for NO₂, O₃, PM_2.5, PM₁₀, and CO (p > 0.05), only SO₂ showed a less significant increasing (6.4%, p < 0.05) trend compared to 2019. Overall, these data further support the hypothesis that less anthropogenic emissions due to COVID-19 lockdown could improve air quality, and stationary emission sources were less affected by the lockdown. Moreover, it is important to note that the increase of O₃ and the high concentrations of PM_2.5 have become the main environmental issue in Harbin, which require further investigation.

To better understand the pollution causes during 2020 lockdown period, relationships among the main air pollutants, and their associated meteorology conditions effects were investigated (Figure 3). It is noted that PM_2.5 were correlated highest with NO₂ and SO₂, with R² of 0.63 and 0.5, respectively (p < 0.01), while the coefficient of NO₂ and SO₂ was the lowest (R² = 0.28, p < 0.01), which can be largely explained by the closely relationships between fine particles and gaseous pollutants released from coal combustion, vehicle exhaust, etc. O₃ existed no significant correlations with NO₂ and SO₂ (p > 0.05), however, it is important to note that when PM_2.5 < 90 μg m⁻³, O₃ was positive correlated with PM_2.5, and the impacts of meteorological factors on the O₃ was not obvious. Whereas O₃ was negative correlated with PM_2.5 and NO₂ when PM_2.5 > 90 μg m⁻³ under T < 0°C and WS < 2 m s⁻¹. According to Chu et al. (2020), when PM_2.5 < 50 μg m⁻³, PM_2.5 does not significantly reduce the actinic flux and HO₂ free radicals due to sufficient solar irradiation. PM_2.5 and O₃ have common precursors and they are produced simultaneously in the photochemical reaction, thus showing a positive correlation. In Harbin, the unique winter climate (cold, dry and less windy) raises the threshold to 90 μg m⁻³.

FIGURE 3

FIGURE 3. Relationships among the main air pollutants (A–F), and the relationships of T (G), RH (H), and WS (I), with O₃, PM_2.5 and NO₂.

3.2 Pollution types and causes of typical pollution events

Heavy air pollution has occurred frequently in the past few years in Northeast China especially during winters, which was distinct from other regions in China because of increased levels of coal and biomass combustion during the heating season, particular geographical condition, and adverse meteorological conditions (Sun et al., 2020; Jia et al., 2023; Sun et al., 2023). The range of the AQI values in Harbin throughout the 2020 was 19–500 (Figure 4), with a variation coefficient of 81.1%, indicating strong variability and a large difference in air quality at different times due to natural and human factors. Overall, the air quality was good in summer and autumn, while air pollution was severe in winter and spring in Harbin. It was strongly endured PM_2.5 pollution and haze almost one-third of that critical period, which require further air pollution control.

FIGURE 4

FIGURE 4. Temporal variations of air quality in Harbin in 2020.

According to the “Ambient Air Quality Standards” (GB3095-2012), there were 62 days throughout the 2020 AQI exceeded the Class II environmental assessment standard, with 27, 14, 15, and 6 days of mild, moderate, heavy, and severe pollution, respectively. These pollution days were concentrated from January 1 to April 19 which were under the lockdown period. The number of days when PM_2.5 was the primary pollutant was 55 days, accounting for 88.7% of all pollution days. The average concentration of PM_2.5 (ρ(PM_2.5)) was 161.07 μg/m³, which was 2.15 times higher than the Class II requirement (75 μg/m³). The ρ(PM_2.5)/ρ(PM₁₀) ratio was consistently greater than 0.7, indicating that atmospheric pollution in Harbin during the winter and spring was primarily attributed to fine PM, whereas natural dust sources had a limited impact (Robert and Brigitte, 2003; Zha et al., 2021).

In order to further analyze the typical pollution characteristics of Harbin in winter and spring, six pollution events were defined based on the AQI values. A pollution event was identified when the AQI value exceeded mild pollution levels and persisted for more than 6 consecutive days. These events were categorized as follows: January 1–9 (Event I), January 10–20 (Event II), January 25–February 1 (Event III), February 7–12 (Event IV), March 1–7 (Event V), and April 13–19 (Event VI). The historical characteristic radar chart has been extensively used for identifying the main pollution types (Li et al., 2020; Sun et al., 2023). For the current study, two typical atmospheric pollution types were identified, i.e., SO₂ dominant pollution type and PM_2.5 dominant pollution type (Figure 5), and the impacts of meteorology conditions and human activities on the pollution types were further discussed.

FIGURE 5

FIGURE 5. Temporal variations of pollution events and typical atmospheric pollution types.

The characteristic radar charts of the three sustained pollution periods (I, III, and IV) showed that the characteristic values of PM_2.5 and SO₂ significantly exceeded the maximum value, whereas the characteristic values of PM_2.5-10, NO₂, and CO were either close to or less than the standard values (Figure 5). Previous research indicated that fireworks and firecrackers had a significant impact on PM_2.5, PM₁₀, and SO₂ (Camilleri and Vella, 2010; Godri et al., 2010). Thus, the occurrence of large-scale fireworks could be determined based on whether the PM_2.5 and SO₂ concentrations significantly exceeded the upper limits. These three sustained pollution periods occurred during the traditional Chinese festivals of New Year’s Day, Spring Festival, and Lantern Festival, which are peak periods for fireworks. In addition, SO₂ is mainly released from coal combustion in winter in residential areas of north China, which is directly from the primary emissions of anthropogenic sources, makes the three sustained pollution periods are SO₂ dominant pollution type.

Numerous studies have shown that SO₂, NO₂, and CO emissions significantly affect PM_2.5 concentrations. SO₂ and NO₂ act as precursors for PM_2.5, and secondary aerosol formation also contributes to increased PM_2.5 concentrations (Zhang et al., 2015; Li et al., 2018; Bao et al., 2019; Chu et al., 2020). According to the NCBI (National Center for Biotechnology Information), in the air in a standard room (21% O₂), the half-life of CO is 320 min. CO has a relatively long atmospheric lifetime and is an inert chemical pollutant whose concentration is affected by meteorological factors. Therefore, it can be used as a tracer to analyze the secondary generation of fine PM by examining ρ(PM_2.5)/ρ(CO) (Song et al., 2017). In periods Ⅱ, Ⅴ, and Ⅵ, the value of ρ(PM_2.5)/ρ(CO) was 0.131, higher than the annual value of 0.05, indicating a significant influence by the formation of secondary PM. The radar chart showed that only the characteristic value of PM_2.5 exceeded the maximum value, indicating that these three periods of sustained pollution were PM_2.5 dominant pollution type events and highly correlated with the secondary transformation of PM_2.5.

Nonparametric Kolmogorov−Smirnov test was performed to determine significance of the differences of air pollutants between SO₂ dominant pollution and PM_2.5 dominant pollution (Figure 6). No significant differences were found for the concentrations of PM_2.5, PM₁₀, NO₂ and CO between the two types (p > 0.05), while SO₂ and O₃ were significantly higher in the SO₂ dominant pollution and PM_2.5 dominant pollution, respectively (p < 0.05). As already discussed, these pollution events occurred during the COVID-19 lockdown period in 2020. It indicated the SO₂, O₃, together with PM_2.5 were the most important contributors to heavy pollution formation when anthropogenic activities were reduced. The results reflected some typical and new characteristics of air pollution in northeast China. On one hand, coal combustion from heating supply due to the low temperature, and fireworks set off on traditional festivals were the major sources in Harbin, all of which were directly derived from primary emissions of anthropogenic sources. On the other hand, a combined air pollution dominated by PM_2.5 and O₃ was also observed, which were considered to damage human health severely. However, the formation regime of PM_2.5−O₃ are difficult to be accurately determined, due to the impacts of precursors and meteorological conditions (Liu et al., 2023; Xu et al., 2023), which require further investigation.

FIGURE 6

FIGURE 6. Comparison of air pollutant concentrations during the SO₂ dominant pollution and PM_2.5 dominant pollution episodes in 2020 (* indicates significant difference between the two pollution episodes).

The relationships among air pollutants and meteorological conditions during SO₂ dominant pollution and PM_2.5 dominant pollution were evaluated to investigate the attributions of emission-reduction and meteorological conditions (Figure 7). Positive correlations between PM_2.5 and key gaseous pollutants (SO₂, NO₂ and CO) (p < 0.01), and negative correlation between PM_2.5 and T (p < 0.05) were observed in SO₂ dominant pollution, clearly indicating that combustion source release is important contributor in northeast China due to the low temperature and low wind speed. As for the PM_2.5 dominant pollution, the influence of meteorological factors on pollutants is different from that of SO₂ dominant pollution, with higher correlations between RH and PM_2.5 and gaseous pollutants (p < 0.01), and less or no significant correlations were found between T and PM_2.5 and gaseous pollutants. These phenomena may result from a significant influence by the formation of secondary particulate matter, that the hygroscopic growth of particles may be an essential factor, as it could promote homogeneous or heterogeneous reactions by providing sufficient reaction sites by increasing the surface area (Li et al., 2021). High humidity conditions accelerate the conversion of gaseous precursors to particles, and secondary fine particles are generated under stable meteorological conditions that hinder pollutant diffusion (Li et al., 2017).

FIGURE 7

FIGURE 7. Correlation between atmospheric pollutants and meteorological factors during SO₂ dominant pollution (A) and PM_2.5 dominant pollution (B) pollution episodes (* indicates p < 0.05, ** indicates p < 0.01).

PM_2.5 was the main pollutant exceeding air quality standards during pollution episodes, which comes from both primary pollution release and secondary transformation. The distributions of PM_2.5 concentrations under different levels of temperature, relative humidity, and wind speed during SO₂ dominant pollution and PM_2.5 dominant pollution episodes were further investigated (Figure 8). The results showed that concentrations of PM_2.5 increased with the decrease of temperature, due to coal-fired heating in cold winter, and increased with the increase of relative humidity, indicating that the effect of hygroscopic growth increased gradually. High concentrations of PM_2.5 occurred more frequently under lower wind speed, which was unfavorable for the dispersion of atmospheric pollutants.

FIGURE 8

FIGURE 8. Distributions of PM_2.5 concentrations under different levels of temperature (A), relative humidity (B), and wind speed (C) during SO₂ dominant pollution and PM_2.5 dominant pollution episodes.

3.3 Pollutant transport pathways and potential source analysis of different types of pollution episodes

3.3.1 SO₂ dominant pollution

Regional transportation was found to contribute considerably to the rapid accumulation of PM_2.5. Therefore, the trajectories of the air masses and potential geographical origins were analyzed to explore its influence. Five air mass trajectories were identified in the SO₂ dominant pollution (I, III, and IV) pollution episode (Figure 9A). Trajectories 1, 3, and 5 were similar, originating from high-altitude regions in Russia and entering China from the southeast. As they passed through the mountains in the Greater Khingan Range, their altitude and speed decreased, accumulating atmospheric pollutants. Finally, they were transported to Harbin in the east via Daqing City. Trajectory 4 originated in high-altitude regions in southern Russia and entered China from the southeast, passing through Inner Mongolia at the border of Jilin Province. The trajectory turned northeastward, and the altitude decreased, ultimately reaching Harbin City at low speed. Trajectory 2 involved a short-distance transfer from neighboring cities, with a slow speed and small altitude changes. Nonetheless, it contributed the most to Harbin’s pollution (29.63%). It originated from central Hulunbuir City in the south, moved southeastward to Changchun, and then turned northward. The ups and downs became gentle, and the pollutants accumulated and were slowly transported to Harbin. In conjunction with the wind-rose diagram, due to the strong wind from the southwest contributed to the transmission of atmospheric particles, several surrounding cities west and southwest of Harbin had the greatest impact on Harbin’s pollution during the SO₂ dominant pollution. In addition, atmospheric particles transmitted from Russia might have contributed to the severe pollution event in Harbin.

FIGURE 9

FIGURE 9. Backward trajectory (A,D), weighted potential source contribution factors (WPSCF) (B,E), and weighted concentration-weighted trajectories (WCWT) (C,F) of PM_2.5 during SO₂ dominant pollution and PM_2.5 dominant pollution episodes.

The potential source area of PM_2.5 during the SO₂ dominant pollution is shown in Figures 9B,C. A WPSCF value exceeding 0.6 indicates that the region contributes significantly to PM_2.5 pollution in Harbin and is a significant potential source area (Zhang et al., 2020b; Ren et al., 2021; Chen et al., 2022). The source area of PM_2.5 was centered in the western part of Harbin and formed a central zone that extended to the southern parts of Daqing and Suihua cities and the northern parts of Changchun and Songyuan cities. This area mainly contributed to the short-distance transport of PM_2.5. The potential source areas of PM_2.5 in Tongliao, Tieling, and Siping were highly consistent with trajectory 4, which mainly contributed to the long-distance transport of PM_2.5. The source area of mild pollution expanded outward from the source area for severe pollution and was located in the southwestern part of Heilongjiang Province, the central-western part of Jilin Province, and at the border area of Inner Mongolia. The results of the WCWT and WPSCF methods were similar. The pollution source area west of Harbin contributed 80–120 μg/m³ to Harbin’s WCWT concentration. Meanwhile, the source areas for severe pollution in Tongliao, Tieling, and Siping contributed 80–100 μg/m³. The area at the junction of Baicheng City, Songyuan City, and Daqing City also significantly contributed to the combustion source pollution in Harbin, with a WCWT value of 60–80 μg/m³. The results of the potential source areas combined with PM_2.5 carried by strong winds from the southwest, are highly consistent with the backward trajectories, indicating that Harbin and the surrounding cities suffered from pollution emissions far beyond the environmental capacity due to fireworks and firecracker discharges, as well as coal combustion in cold season.

3.3.2 PM_2.5 dominant pollution

Five air mass trajectories were obtained from backward trajectory simulation (Figure 9D). The altitudes of these five pollution trajectories were significantly lower than those of the SO₂ dominant pollution episode, resulting in a smoother transport and the conversion of secondary PM precursors. Trajectory 1 exhibited short-distance transportation from the surrounding cities. It had the shortest airflow trajectory, slowest velocity, and the highest proportion (27.67%), which originated south of Heihe City, slowly descending in altitude as it moved south, passing through Yichun City and Suihua City, and transporting pollutants to Harbin City, contributing significantly to PM pollution in Harbin City. Trajectories 2 and 3 originated from the northwest airflow of Russia. When it crossed the Greater Khingan Range to the southeast, the altitude suddenly dropped, and the speed slowed down, and it moved eastward from Daqing City to Harbin. Trajectory 4 started from the Xing’an League and was transported to the border between Jilin Province and Liaoning Province in a southeastern direction with a continuous altitude decrease. Subsequently, the airflow turned northeast, and the altitude remained unchanged. A large number of pollutants accumulated during a relatively long stay in Jilin Province. Eventually, the trajectory slowly moved toward Harbin. Trajectory 5 was the only northeasterly airflow. It had a relatively high altitude and was moving fast before arriving in Harbin. Its height and speed suddenly dropped after entering Harbin, transporting pollutants from the eastern region to the central urban area. The wind frequency rose diagram shows that the wind speed was low, although the frequency of westward winds that dominated the PM_2.5 dominant pollution episode was high. Therefore, locally generated PM from Harbin City’s pollution sources accounted for the main portion. In contrast, on days with high east and north winds, short-distance transport of pollutants, such as coal burning and vehicle exhaust from surrounding cities, occurred. The pollutants were transformed into secondary PM pollutants under favorable conditions, such as low temperatures, high humidity, and stagnant weather, accumulating pollutants and resulting in heavy pollution in Harbin (Li et al., 2021).

The range of the potential source areas of PM_2.5 dominant pollution with WPSCF values >0.6 was larger than that of the SO₂ dominant pollution, and the source areas extended beyond the central area. The potential source areas were centered on Harbin and covered most parts of Suihua City, southern Daqing city, western Harbin, and the north of Songyuan City. Another potential source area of heavy PM_2.5 pollution formed at the junction of Changchun, Jilin, and Harbin, and there were other sporadic potential source areas in the surrounding cities. The WCWT results showed extremely high pollution areas east of Harbin, north of Suihua City, and at the junction of Changchun, Jilin, and Harbin, with WCWT values above 180 μg/m³. Due to the extreme pollution in mid-April, the highest WCWT value was 410 μg/m³. The PM_2.5 contribution range of other potential source areas of PM_2.5 dominant pollution in Harbin city was 90–180 μg/m³, whereas that of potential source areas of slight pollution was 60–90 μg/m³.

In summary, local source pollutants generated by combustion emission at specific times, coupled with pollutants transmitted by southwestern winds, caused SO₂ dominant pollution in Harbin City. Many secondary pollutant precursors were transferred to Harbin from the east and north due to strong winds, causing a sharp deterioration of the atmospheric environment in Harbin City and form PM_2.5 dominant pollution. The potential sources of PM_2.5 in such a large area are mainly due to precursors of secondary pollution from coal-fired heating, biomass burning and vehicle exhaust (Cheng et al., 2021b; Li et al., 2023).

4 Conclusion

To better understand the impacts of reduced anthropogenic activities on the air quality, and develop more effective measures to control air pollution in the post-pandemic era, daily pollutant concentration monitoring and meteorological data in 2019 and 2020 were used to investigate the changes of air pollutants during lockdown and non-lockdown between the 2 years, respectively. And also to discriminate the dominant air pollution processes and pollution types, analyze the influence of meteorological factors and human activities on the pollution processes, and clarify the potential source areas and their contributions in Harbin City, a cold metropolis in northeast China under COVID-19 in 2020.

Compared to the same time spans in 2019, significant concurrent decrease in NO₂, CO and PM₁₀ concentrations, while no significant differences in SO₂ and PM_2.5 were found during the lockdown in 2020, indicating that reduced activities of transportation did result in reductions of NO₂, and stationary emission sources were less affected. When PM_2.5 < 90 μg m⁻³, O₃ was positive correlated with PM_2.5, and the impacts of meteorological factors on the O₃ was not obvious. Whereas O₃ was negative correlated with PM_2.5 and NO₂ when PM_2.5 > 90 μg m⁻³ under temperature <0°C and wind speed <2 m s⁻¹.

SO₂ dominant pollution and PM_2.5 dominant pollution were distinguished from six sustained heavy pollution periods by radar charts. PM_2.5 and SO₂ played driving role in SO₂ dominant pollution derived from local pollutants emission of coal combustion and firework discharge. PM_2.5 dominant pollution might be chemical transformed from coal burning, vehicle exhaust, and other secondary precursors, which was affected and aggravated by CO, NO₂, high relative humidity and low wind speed affected by local emission and long-distance transport.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Author contributions

JL: Writing–original draft, Data curation, Methodology, Resources, Software. LS: Conceptualization, Funding acquisition, Project administration, Writing–original draft, Writing–review and editing. XY: Methodology, Software, Writing–original draft. SZ: Data curation, Project administration, Writing–review and editing. JW: Data curation, Writing–original draft. DM: Resources, Writing–review and editing.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This research was funded by National Natural Science Foundation of China (NSFC) (42371138 and 42171127), Key Joint Program of National Natural Science Foundation of China (NSFC) and Heilongjiang Province for Regional Development (U20A2082), Science and Technology Fundamental Resources Investigation Program (2022FY100701), and Fundamental Research Funds for the Central Universities (2022-KYYWF-0155).

Acknowledgments

The authors would like to thank the reviewers for their valuable and constructive comments.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Bao, Z., Chen, L., Li, K., Han, L., Wu, X., Gao, X., et al. (2019). Meteorological and chemical impacts on PM_2.5 during a haze episode in a heavily polluted basin city of eastern China. Environ. Pollut. 250, 520–529. doi:10.1016/j.envpol.2019.04.045

PubMed Abstract | CrossRef Full Text | Google Scholar

Camilleri, R., and Vella, A. J. (2010). Effect of fireworks on ambient air quality in Malta. Atmos. Environ. 44 (35), 4521–4527. doi:10.1016/j.atmosenv.2010.07.057

CrossRef Full Text | Google Scholar

Chen, L. Y., Zhang, J. K., Huang, X. J., Li, H., Dong, G. M., and Wei, S. Y. (2022). Characteristics and pollution formation mechanism of atmospheric fine particles in the megacity of Chengdu, China. Atmos. Res. 273, 106172. doi:10.1016/j.atmosres.2022.106172

CrossRef Full Text | Google Scholar

Cheng, Y., Yu, Q., Liu, J., Cao, X., Zhong, Y., Du, Z., et al. (2021a). Dramatic changes in Harbin aerosol during 2018–2020: the roles of open burning policy and secondary aerosol formation. Atmos. Chem. Phys. 21, 15199–15211. doi:10.5194/acp-21-15199-2021

CrossRef Full Text | Google Scholar

Cheng, Y., Yu, Q., Liu, J., Du, Z., Liang, L., Geng, G., et al. (2021b). Strong biomass burning contribution to ambient aerosol during heating season in a megacity in Northeast China: effectiveness of agricultural fire bans? Sci. Total Environ. 754, 142144. doi:10.1016/j.scitotenv.2020.142144

PubMed Abstract | CrossRef Full Text | Google Scholar

Chu, B., Ma, Q., Liu, J., Ma, J., Zhang, P., Chen, T., et al. (2020). Air pollutant correlations in China: secondary air pollutant responses to NO_x and SO₂ control. Environ. Sci. Technol. 7, 695–700. doi:10.1021/acs.estlett.0c00403

CrossRef Full Text | Google Scholar

Chu, B., Zhang, S., Liu, J., Ma, Q., and He, H. (2021). Significant concurrent decrease in PM_2.5 and NO₂ concentrations in China during COVID-19 epidemic. J. Environ. Sci. 99, 346–353. doi:10.1016/j.jes.2020.06.031

PubMed Abstract | CrossRef Full Text | Google Scholar

Cui, S., Song, Z. H., Zhang, L. M., Zhang, Z. L., Hough, R., Fu, Q., et al. (2019). Polycyclic aromatic hydrocarbons in fresh snow in the city of Harbin in northeast China. Atmos. Environ. 215, 1–9. doi:10.1016/j.atmosenv.2019.116915

CrossRef Full Text | Google Scholar

Draxler, R. R., and Hess, G. D. (1998). An overview of the HYSPLI_4 modelling system for trajectories, dispersion, and deposition. Aust. Meteorol. Mag. 47 (4), 295–308.

Google Scholar

Duan, J. C., Hu, J. N., Tan, J. H., and Chen, H. (2018). Design of characteristic radar chart and its application in air pollution analysis. Res. Environ. Sci. 31 (8), 1329–1336. doi:10.13198/j.issn.1001-6929.2018.06.12

CrossRef Full Text | Google Scholar

Godri, K. J., Green, D. C., Fuller, G. W., Dall'Osto, M., Beddows, D. C., Kelly, F. J., et al. (2010). Particulate oxidative burden associated with firework activity. Environ. Sci. Technol. 44 (21), 8295–8301. doi:10.1021/es1016284

PubMed Abstract | CrossRef Full Text | Google Scholar

He, C., Hong, S., Zhang, L., Mu, H., Xin, A., Zhou, Y., et al. (2021). Global, continental, and national variation in PM_2.5, O₃, and NO₂ concentrations during the early 2020 COVID-19 lockdown. Atmos. Pollut. Res. 12, 136–145. doi:10.1016/j.apr.2021.02.002

PubMed Abstract | CrossRef Full Text | Google Scholar

Jia, H. J., Zang, S. Y., Zhang, L. Q., Yakovleva, E., Sun, H. J., and Sun, L. (2023). Spatiotemporal characteristics and socioeconomic factors of PM_2.5 heterogeneity in mainland China during the COVID-19 epidemic. Chemosphere 331, 138785. doi:10.1016/j.chemosphere.2023.138785

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, B., Lu, L., Shi, X., Ma, L., Jiang, J., Yabo, S. D., et al. (2023). Drivers of 2015–2021 trends in cold winter surface PM_2.5 in the Harbin-Changchun megalopolis in China: meteorology vs. anthropogenic emission. Atmos. Res. 285, 106623. doi:10.1016/j.atmosres.2023.106623

CrossRef Full Text | Google Scholar

Li, H., Duan, F., Ma, Y., He, K., Zhu, L., Ma, T., et al. (2018). Case study of spring haze in Beijing: Characteristics, formation processes, secondary transition, and regional transportation. Environ. Pollut. 242 (Pt A), 544–554. doi:10.1016/j.envpol.2018.07.001

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, H., Ma, Y., Duan, F., He, K., Zhu, L., Huang, T., et al. (2017). Typical winter haze pollution in Zibo, an industrial city in China: characteristics, secondary formation, and regional contribution. Environ. Pollut. 229, 339–349. doi:10.1016/j.envpol.2017.05.081

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, L., Wang, K., Chen, W., Zhao, Q., Liu, L., Liu, W., et al. (2020). Atmospheric pollution of agriculture-oriented cities in Northeast China: a case in Suihua. J. Environ. Sci. 97, 85–95. doi:10.1016/j.jes.2020.04.038

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, W., Duan, F., Zhao, Q., Song, W., Cheng, Y., Wang, X., et al. (2021). Investigating the effect of sources and meteorological conditions on wintertime haze formation in Northeast China: a case study in Harbin. Sci. Total Environ. 801, 149631. doi:10.1016/j.scitotenv.2021.149631

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, C., Liang, J., Li, Y., and Shi, K. (2023). Fractal analysis of impact of PM_2.5 on surface O₃ sensitivity regime based on field observations. Sci. Total Environ. 858, 160136. doi:10.1016/j.scitotenv.2022.160136

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, J., Zheng, Y., Geng, G., Hong, C., Li, M., Li, X., et al. (2020a). Decadal changes in anthropogenic source contribution of PM_2.5 pollution and related health impacts in China, 1990–2015. Atmos. Chem. Phys. 20, 7783–7799. doi:10.5194/acp-20-7783-2020

CrossRef Full Text | Google Scholar

Liu, Y., Dong, J., Xu, X., and Jiang, Y. (2020b). PM_2.5 mass concentration variation in urban residential buildings during heating season in severe cold region of China: a case study in Harbin. Sci. Total Environ. 722, 137945. doi:10.1016/j.scitotenv.2020.137945

PubMed Abstract | CrossRef Full Text | Google Scholar

Luo, Y., Liu, S., Che, L., and Yu, Y. (2021). Analysis of temporal spatial distribution characteristics of PM_2.5 pollution and the influential meteorological factors using Big Data in Harbin, China. J. Air Waste Manag. Assoc. 71 (8), 964–973. doi:10.1080/10962247.2021.1902423

PubMed Abstract | CrossRef Full Text | Google Scholar

Ren, B., Xie, P. H., Xu, J., Li, A., Tian, X., Hu, Z. K., et al. (2021). Use of the PSCF method to analyze the variations of potential sources and transports of NO2, SO2, and HCHO observed by MAX-DOAS in Nanjing, China during 2019. Sci. Total Environ. 782, 146865. doi:10.1016/j.scitotenv.2021.146865

CrossRef Full Text | Google Scholar

Robert, G., and Brigitte, B. (2003). Characterising seasonal variations and spatial distribution of ambient PM₁₀ and PM_2.5 concentrations based on long-term Swiss monitoring data. Atmos. Environ. 37 (19), 2571–2580. doi:10.1016/S1352-2310(03)00221-8

CrossRef Full Text | Google Scholar

Song, C., Wu, L., Xie, Y., He, J., Chen, X., Wang, T., et al. (2017). Air pollution in China: status and spatiotemporal variations. Environ. Pollut. 227, 334–347. doi:10.1016/j.envpol.2017.04.075

PubMed Abstract | CrossRef Full Text | Google Scholar

Sun, L., Ai, X., Yao, X., An, Q., Liu, X. M., Yakovleva, E., et al. (2023). Relationship between atmospheric pollution and polycyclic aromatic hydrocarbons in fresh snow during heavy pollution episodes in a cold city, northeast China. Ecotox. Environ. Safe. 260, 115091. doi:10.1016/j.ecoenv.2023.115091

CrossRef Full Text | Google Scholar

Sun, X., Wang, K., Li, B., Zong, Z., Shi, X., Ma, L., et al. (2020). Exploring the cause of PM_2.5 pollution episodes in a cold metropolis in China. J. Clean. Prod. 256, 120275. doi:10.1016/j.jclepro.2020.120275

CrossRef Full Text | Google Scholar

Wang, J., Zhao, B., Wang, S., Yang, F., Xing, J., Morawska, L., et al. (2017). Particulate matter pollution over China and the effects of control policies. Sci. Total Environ. 584–585, 426–447. doi:10.1016/j.scitotenv.2017.01.027

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, P., Chen, K., Zhu, S., Wang, P., and Zhang, H. (2020). Severe air pollution events not avoided by reduced anthropogenic activities during COVID-19 outbreak. Resour. Conserv. Recycl. 158, 104814. doi:10.1016/j.resconrec.2020.104814

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, Y. L., Sun, Y. W., Zhang, Z. Q., and Cheng, Y. (2021). Spatiotemporal variation and source analysis of air pollutants in the Harbin-Changchun (HC) region of China during 2014–2020. Environ. Sci. Technol. 8, 100126. doi:10.1016/j.ese.2021.100126

CrossRef Full Text | Google Scholar

Wang, Y. Q., Zhang, X. Y., and Draxler, R. R. (2009). TrajStat: GIS-based software that uses various trajectory statistical analysis methods to identify potential sources from long-term air pollution measurement data. Environ. Model. Softw. 24 (8), 938–939. doi:10.1016/j.envsoft.2009.01.004

CrossRef Full Text | Google Scholar

Xie, Y. Y., Zhang, Y., He, K., Zhou, J., and Kang, C. G. (2006). Features of sand-dust deposits in Harbin city, China. Chin. Geogr. Sci. 16, 327–333. doi:10.1007/s11769-006-0327-4

CrossRef Full Text | Google Scholar

Xu, T., Zhang, C., Liu, C., and Hu, Q. (2023). Variability of PM_2.5 and O₃ concentrations and their driving forces over Chinese megacities during 2018–2020. J. Environ. Sci. 124, 1–10. doi:10.1016/j.jes.2021.10.014

CrossRef Full Text | Google Scholar

Zha, H., Wang, R., Feng, X., An, C., and Qian, J. (2021). Spatial characteristics of the PM_2.5/PM₁₀ ratio and its indicative significance regarding air pollution in Hebei Province, China. Environ. Monit. Assess. 193 (8), 486. doi:10.1007/s10661-021-09258-w

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, J., Huang, X., Chen, Y., Luo, B., Luo, J., Zhang, W., et al. (2019). Characterization of lead-containing atmospheric particles in a typical basin city of China: seasonal variations, potential source areas, and responses to fireworks. Sci. Total Environ. 661, 354–363. doi:10.1016/j.scitotenv.2019.01.079

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, L. Y., Qiao, L. P., Lan, J., Yan, Y., and Wang, L. N. (2020a). Three-years monitoring of PM_2.5 and scattering coefficients in Shanghai, China. Chemosphere 253, 126613. doi:10.1016/j.chemosphere.2020.126613

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, M. D., Chen, W. W., Shen, X. J., Zhao, H. M., Gao, C. K., Zhang, X. L., et al. (2020b). Comprehensive and high-resolution emission inventory of atmospheric pollutants for the northernmost cities agglomeration of Harbin-Changchun, China: implications for local atmospheric environment management. J. Environ. Sci. (China) 104, 150–168. doi:10.1016/j.jes.2020.11.026

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, Q., Quan, J., Tie, X., Li, X., Liu, Q., Gao, Y., et al. (2015). Effects of meteorology and secondary particle formation on visibility during heavy haze events in Beijing, China. Sci. Total Environ. 502, 578–584. doi:10.1016/j.scitotenv.2014.09.079

PubMed Abstract | CrossRef Full Text | Google Scholar

Zong, Z., Wang, X., Tian, C., Chen, Y., Fu, S., Qu, L., et al. (2018). PMF and PSCF based source apportionment of PM_2.5 at a regional background site in North China. Atmos. Res. 203, 207–215. doi:10.1016/j.atmosres.2017.12.013

CrossRef Full Text | Google Scholar