Optical Properties of Aerosols and Chemical Composition Apportionment under Different Pollution Levels in Wuhan during January 2018

Abstract

To clarify the aerosol optical properties under different pollution levels and their impacting factors, hourly organic carbon (OC), elemental carbon (EC), and water-soluble ion (WSI) concentrations in PM_2.5 were observed by using monitoring for aerosols and gases (MARGA) and a semicontinuous OC/EC analyzer (Model RT-4) in Wuhan from 9 to 26 January 2018. The aerosol extinction coefficient (b_ext) was reconstructed using the original Interagency Monitoring of Protected Visual Environment (IMPROVE) formula with a modification to include sea salt aerosols. A good correlation was obtained between the reconstructed b_ext and measured b_ext converted from visibility. b_ext presented a unimodal distribution on polluted days (PM_2.5 mass concentrations > 75 μg⋅m⁻³), peaking at 19:00. b_ext on clean days (PM_2.5 mass concentrations < 75 μg⋅m⁻³) did not change much during the day, while on polluted days, it increased rapidly starting at 12:00 due to the decrease of wind speed and increase of relative humidity (RH). PM_2.5 mass concentrations, the aerosol scattering coefficient (b_scat), and the aerosol extinction coefficient increased with pollution levels. The value of b_ext was 854.72 Mm⁻¹ on bad days, which was 4.86, 3.1, 2.29, and 1.28 times of that obtained on excellent, good, acceptable, and poor days, respectively. When RH < 95%, b_ext exhibited an increasing trend with RH under all pollution levels, and the higher the pollution level, the bigger the growth rate was. However, when RH > 95%, b_ext on acceptable, poor and bad days decreased, while b_ext on excellent and good days still increased. The overall b_ext in Wuhan in January was mainly contributed by NH₄NO₃ (25.2%) and organic matter (20.1%). The contributions of NH₄NO₃ and (NH₄)₂SO₄ to b_ext increased significantly with pollution levels. On bad days, NH₄NO₃ and (NH₄)₂SO₄ contributed the most to b_ext, accounting for 38.2% and 27.0%, respectively.

1. Introduction

In recent years, air quality deterioration and visibility impairment have occurred in many urban areas of China [1,2,3,4]. One of the key pollutants is PM_2.5 (aerosol particle with aerodynamic diameter less than 2.5 μm), which has a great effect on visibility impairment, urban air quality, and human health [5,6,7]. PM_2.5 is mainly composed of organic and inorganic components and is typically hygroscopic. The size of PM_2.5 can increase under high relative humidity (RH) conditions, therefore influencing the aerosol optical properties and leading to an increase in the aerosol scattering coefficient (b_scat) and aerosol extinction coefficient (b_ext) [8,9].

PM_2.5 in the atmosphere can reduce visibility through light extinction (including scattering and absorption of light) and also plays an important role in the formation of haze in urban areas [10,11]. The aerosol extinction coefficient (b_ext) is an important parameter of atmospheric extinction and is particularly significant in aerosol optical research. b_ext is mainly contributed by (NH₄)₂SO₄, NH₄NO₃, organic matter (OM), elemental carbon (EC), sea salt, coarse mass, and so on. Their relative contributions to b_ext vary with time and location. The Interagency Monitoring of Protected Visual Environment (IMPROVE) formula is a method for calculating the aerosol extinction coefficient (b_ext) based on the chemical composition of a substance [12,13]. Many scholars have used it to calculate the contribution of each particle component to the extinction coefficient. Existing studies have found that in urban areas, (NH₄)₂SO₄ was the largest contributor to b_ext, followed by NH₄NO₃ and OM [14,15,16,17]. Combustion-related particles rather than wind-blown dust dominated the light extinction budget in Beijing [18]. The contributions of sulfate, ammonium, and nitrate to the PM_2.5 mass concentration were 15%, 5%, and 8%, respectively. Mineral aerosol contributed 16% to the PM_2.5 aerosol mass, showing that combustion-related particles rather than wind-blown dust dominated the light extinction budget [19]. The seasonally reconstructed b_ext was in the order of autumn (319.4 ± 207.2 Mm^–1) > winter (269.6 ± 175.5 Mm^–1) > summer (219.0 ± 129.3 Mm^–1) > spring (193.3 ± 94.9 Mm^–1) annually in Guangzhou [20]. RH is an important meteorological factor in the atmosphere and has a notable effect on the formation and optical properties of PM_2.5. RH is a key factor in visibility impairment that affects light extinction through the hygroscopic growth of particles [21,22]. b_ext values were higher in higher humidity conditions [23,24]. High RH played an important role in the formation of PM_2.5. However, when RH > 80%, PM_2.5 concentrations began to decrease due to the high frequency of precipitation [25,26].

In recent years, many studies have been performed to investigate aerosol optical properties in many developed cities in China, such as Nanjing [22,27], Beijing [18,28,29], Guangzhou [20,25], and Chengdu [14,30]. In addition, many studies have been done to determine the relationship between aerosol optical properties and pollution level. The atmospheric extinction coefficient (b_ext) and the absorption coefficient of aerosols (b_abs) increased, and the single scattering albedo (SSA) decreased from excellent to polluted levels [31]. Higher aerosol optical depth (AOD) and SSAs were observed during polluted periods than during non-polluted periods [32]. However, studies in Wuhan about aerosol optical properties under different pollution levels are very limited.

As the capital of Hubei province, Wuhan covers an area of 8500 square kilometers. The permanent resident population was 10.90 million in 2017. In recent years, with the rapid development of economy and the acceleration of urbanization, haze pollution in Wuhan is increasingly serious. Therefore, many studies have been done in Wuhan. PM_2.5 was the primary pollutant, and O₃ was the secondary pollutant, in Wuhan as determined by calculating the value of the Individual Air Quality Index (IAQI) [33]. The major chemical compositions of PM_2.5 were investigated in Wuhan, and it was found that OM was the most abundant component in PM_2.5, the lowest concentrations of which were observed in summer [34]. Regional chemical transport played an important role in both short-term haze episodes and persistent haze episodes that occurred in Wuhan [35]. The physical and chemical characteristics, including chemical composition, hygroscopicity, and particle size distribution of aerosols are quite different under different pollution levels. In addition, the meteorological conditions (RH, wind speed, wind direction, and stability) are also different under different pollution levels. Therefore, the contribution of aerosol chemical components to the light extinction is different under different pollution levels. To gain a thorough knowledge of aerosol optical properties under different pollution conditions, the values of hourly concentrations of PM_2.5, water-soluble ions (WSIs), organic carbon (OC), elemental carbon (EC), and meteorological parameters were observed in Wuhan during January 2018. b_ext was reconstructed based on the IMPROVE formula, and the reconstructed b_ext was used to assess the contributions of individual components of PM_2.5 to the total light extinction. In this study, aerosol optical parameters under different pollution levels were compared and analyzed. The influence of meteorological factors on optical parameters was also discussed.

2. Materials and Methods

2.1. Data Origins

As the central city of central China, Wuhan is located in the eastern part of Hubei province, at the intersection of the Yangtze River and Hanshui. Wuhan is covered by rivers and lakes, and water accounts for about a quarter of the total area of the city. According to the statistical bulletin of Wuhan national economic and social development in 2017, the annual GDP of Wuhan had reached 134.10 billion yuan by the end of 2017, which was an increase of 8.0% over the previous year. The annual average concentrations of ambient fine particulate matter (PM_2.5) and inhalable particulate matter (PM₁₀) were 52 μg⋅m⁻³ and 87 μg⋅m⁻³, respectively, which were 8.8% and 5.4% lower than those of the previous year. Wuhan has a north subtropical monsoon climate and experiences abundant rainfall and sufficient heat annually. The city is relatively hot in the summer and cold in the winter. The observation time was from January 9 to January 26, 2018. All time mentioned in this passage were local time, and the time of sunset was 17:30 in January in Wuhan.

2.2. Instruments

Water-soluble ions (NO₃⁻, SO₄²⁻, Cl⁻, Ca²⁺, and Mg²⁺) were measured using a model ADI 2080 monitor in ambient air (MARGA, Applikon Analytical B.V., the Netherlands) with a Teflon-coated PM2.5 sampling inlet at an hourly temporal resolution. For detailed principles of the tool, please refer to related articles [36,37,38,39,40,41].

A semicontinuous thermal-optical transmittance (TOT) carbon analyzer (Model RT-4, Sunset Laboratory Inc., Tigard, OR, USA) was used to measure the hourly concentrations of OC and EC. The affirmation of the Sunset field carbon analyzer can be found in related reference [35,36]. Each measurement cycle is composed of 45 min of sampling and 15 min of OC/EC analysis. The detailed principle of the instrument can be found in other literature [37].

Hourly concentrations of PM_2.5, PM₁₀ and NO₂ were downloaded from the Ministry of Ecology and Environment (MEE) of China website (http://106.37.208.233:20035/), Zhuankou district station (114°150 E, 30°470 N). The meteorological data (temperature, wind speed, wind direction, relative humidity) were measured from the Wuhan meteorological station (114.03° E, 30.36° N).

2.3. Data Analysis

The black carbon (BC) mass concentrations were used to calculate the b_abs as follows:

$$b_{abs}=\alpha \times \left[BC\right],$$

(1)

where α is a conversion factor. The value of α was 8.28 m²⋅g⁻¹ based on a comparison test between the Aethalometer and the photoacoustic spectrometer in southern China [42]. BC mass concentrations were replaced by EC mass concentrations. b_abs is absorption coefficient of particles (unit: Mm⁻¹).

In this study, two methods were used to calculate b_ext. One method used the visibility data, which were converted to b_ext using the Koschmeider equation [43]:

$$b_{ext}=3912/VR\left(2\right),$$

(2)

where VR represents visual range (unit: km), and b_ext represents the aerosol extinction coefficient (unit: Mm⁻¹). Another method was to calculate b_ext using the IMPROVE formula:

$$\begin{array}{cc}{b}_{ext}=3\times \mathrm{f}\left(\mathrm{RH}\right)& \times \left[{(N{H}_4)}_{2}S{O}_4\right]+3\times \mathrm{f}\left(\mathrm{RH}\right)\times \left[(N{H}_{4}N{O}_3\right]+4\times \left[OM\right]+10\times \left[EC\right]\hfill \\ & +1\times \left[Soil\right]+1.7\times \mathrm{f}\left(\mathrm{RH}\right)\times \left[\mathrm{SS}\right]+0.6\times \left[CoarseMass\right]\hfill \\ & +0.33\times \left[N{O}_2\right]+\mathrm{Rayleigh}\mathrm{Scattering}\hfill \end{array}$$

(3)

where [x] represents the concentration of constituent x (unit: μg⋅m⁻³). [(NH₄)₂SO₄] was replaced by 1.375 times [SO₄²⁻], [NH₄NO₃] was replaced by 1.29 times [NO₃⁻], [OM] was replaced by 1.6 times [OC], [Soil] was replaced by the sum of 1.63 times [Ca⁺] and 1.67 times [Mg²⁺], and [SS] was replaced by 1.8-times [Cl⁻]. [CoarseMass] could be replaced by [PM₁₀], the unit of [NO₂] is 10⁻⁹, and Rayleigh Scattering was replaced by 10. f(RH) is the hygroscopic growth factor [8].

Therefore, b_scat can be obtained from a difference between b_ext and b_abs:

$$b_{scat}=b_{ext}-b_{abs}$$

(4)

Figure 1 shows the comparison of b_ext between the two methods, and a fair correlation was observed between the two sets of b_ext, with a correlation coefficient (R) of 0.6; the number of samples (N) was 211. This finding suggests that the reconstructed b_ext using the IMPROVE formula can be used for later analysis.

3. Results and Discussion

3.1. Summary of a Haze Episode

According to newly released ambient air quality standards in 2012, in China, air quality conditions can be classified into five categories. According to PM_2.5 mass concentrations from the best to worst were: excellent (PM_2.5 mass concentrations < 35 μg⋅m⁻³), good (35 μg⋅m⁻³ < PM_2.5 mass concentrations < 75 μg⋅m⁻³), acceptable (75 μg⋅m⁻³ < PM_2.5 mass concentrations < 115 μg⋅m⁻³), poor (115 μg⋅m⁻³ < PM_2.5 mass concentrations < 150 μg⋅m⁻³), and bad (PM_2.5 mass concentrations > 150 μg⋅m⁻³). Wuhan experienced a severe pollution event from 18 January to 21 January, during which the hourly average mass concentration of PM_2.5 was 170 μg⋅m⁻³, which was 1.9 times the overall average in January.

According to Figure 2, the value of PM_2.5 mass concentration was moderately low, with an average of 58.9 μg⋅m⁻³ before 8:00 on 14 January. Excellent and good days were observed during this period. Both the visibility and RH were relatively high, and average values of 7.4 km and 68.9% were observed, respectively. The temperature was as low as 2.8 °C. The mean wind speed was 1.3 m⋅s⁻¹ on average, and the maximum was 4.1 m⋅s⁻¹. The mean values of light scattering components (OC, NO₂, and water-soluble ions) and light absorbent components (EC) were 91.5 μg⋅m⁻³ and 2.7 μg⋅m⁻³ respectively.

From 8:00 on 14 January to 16:00 on 18 January, both acceptable and poor days were observed, and the average concentration value of PM_2.5 was 109.0 μg⋅m⁻³. During this period, the temperature was low, and there was a high RH, presenting mean values of 5.3 °C and 84.9%, respectively. The mean value of wind speed remained unchanged at 1.3 m⋅s⁻¹, while the maximum reached 5.1 m⋅s⁻¹. The mean concentration values of light scattering components and light absorbent components increased, with mean values of 108.5 μg⋅m⁻³ and 4.3 μg⋅m⁻³, respectively, resulting in a decrease in visibility to 2.5 km.

The mean value of PM_2.5 mass concentration increased rapidly to 170.0 μg⋅m⁻³ from 16:00 on 18 January to 15:00 on 21 January. Wuhan experienced a severe pollution event composed of bad days during this period, which was related to unfavorable climate conditions such as low wind speed and high RH. On one hand, the wind speed reduced to 0.8 m⋅s⁻¹, which is not conducive to the horizontal and vertical diffusion of PM_2.5. On the other hand, the average RH reached as high as 91.7%, which was approximately 1.3 times higher than the first period. Increasing RH facilitates the aerosol hygroscopic growth and further enhances the aerosol liquid water, facilitating the secondary aerosols formation by serving as an important medium for liquid-phase and heterogeneous reactions [26]. Therefore, high RH will increase PM_2.5 concentrations and worsen pollution. The temperature increased to 6.1 °C. Finally, the sum of the concentrations of light scattering components and light absorbent components reached as high as 124.3 μg⋅m⁻³, leading to a quick reduction of visibility, with an average of 1 km.

After this severe pollution event, due to the impact of precipitation and high wind speed (3.3 m⋅s⁻¹), the air quality improved rapidly with an average PM_2.5 mass concentration of 66.0 μg⋅m⁻³. The average value of temperature decreased to 1.1 °C as a result of evaporative cooling by precipitation. RH decreased slightly to 89.4%. The mean values of light scattering components and light absorbent components reduced to the same range as seen in the first period. The visibility increased to 3.0 km.

3.2. Optical Properties under Different Pollution Levels

3.2.1. Diurnal Variations of Optical Parameters on Clean and Polluted Days

Clean days (PM_2.5 mass concentrations < 75 μg⋅m⁻³) and polluted days (PM_2.5 mass concentrations > 75 μg⋅m⁻³) were defined in order to analyze the influence of meteorological elements on optical properties under different pollution conditions. According to Figure 3, b_abs peaked at 9:00 on clean days, while it did not vary much on polluted days. b_ext had an identical diurnal variation to that of b_scat on both clean and polluted days. b_ext presented a unimodal distribution that peaked at 19:00 on polluted days and it was pretty flat on clean days. Clear unimodal diurnal patterns of b_ext might be related to the evolution of the planetary boundary layer (PBL) and daily pollution emission trends. Since from the sunset, the height of the PBL decreased, and the turbulent mixing weakened due to the decrease of temperature. Therefore, PM_2.5 accumulated in the lower atmosphere. In addition, emission from vehicles and factories increased in the evening rush hour, which led to high concentrations of PM_2.5. High PM_2.5 concentrations led to the increase of b_ext, which peaked at 19:00. As time passed and temperature increased, the turbulent mixing was enhanced, and the height of the PBL increased, leading to the decrease of PM_2.5 concentration and b_ext values during the daytime. Besides, b_scat was pretty low than b_abs on polluted days in the early morning hours, indicating that more absorbent particles appeared in the morning.

The diurnal variation of optical parameters on polluted days was quite different from that on clean days. As shown in Figure 3, the values of b_abs, b_scat, and b_ext on polluted days were higher than those on clean days because the concentration of PM_2.5 was higher on polluted days, which means there will be more particles in the atmosphere absorbing and scattering light. Furthermore, according to Figure 3c, b_ext increased rapidly starting at 12:00 and peaked at 19:00, with a value of 818.1 Mm⁻¹ on polluted days. The value of b_ext did not change much during the day on clean days. The peak of b_ext on polluted days was related to meteorological conditions. According to Figure 3, the wind speed decreased significantly starting at 6:00 and continuously decreased from 2.2 m⋅s⁻¹ to 1.0 m⋅s⁻¹ over the next 6 h. Therefore, PM_2.5 began to accumulate due to the low wind speeds, leading to a significant increase in PM_2.5 concentration at 12:00. RH began increasing at 8:00 and reached 95% at 17:00, after which it remained almost unchanged. With high RH, aerosol liquid water not only augments particle sizes, which enhances aerosol scattering, but also substantially enhances secondary aerosol formation. Therefore, these two factors together led to the rapid increase of PM_2.5 concentrations at 12:00.

3.2.2. Analysis of Optical Parameters under Different Pollution Levels

According to Figure 4, PM_2.5 mass concentrations increased with pollution levels. The average mass concentrations of PM_2.5 were 25.95 μg⋅m⁻³, 56.53 μg⋅m⁻³, 93.83 μg⋅m⁻³, 130.9 3μg⋅m⁻³, and 168.43 μg⋅m⁻³ on excellent, good, acceptable, poor, and bad days, respectively. The high levels of PM_2.5 led to a rapid increase of b_ext, the value of which was 854.72 Mm⁻¹ on bad days, which was 4.86, 3.1, 2.29, and 1.28 times that on excellent, good, acceptable, and poor days, respectively. The value of b_scat was 816.36 Mm⁻¹ on bad days and was 4.98, 3.19, 2.19, and 1.29 times that on excellent, good, acceptable, and poor days, respectively. The value of b_abs was 31.72 Mm⁻¹ on acceptable days, which was 1.6 times that on good days. However, as pollution levels continued to increase, the value of b_abs did not change substantially and was 38.74 Mm⁻¹ and 38.36 Mm⁻¹ on poor and bad days, respectively.

The single scattering albedo (SSA) is the ratio between the aerosol scattering coefficient and the aerosol extinction coefficient, reflecting the relative importance of aerosol scattering and absorption, and it is an important factor affecting global climate change. As shown in Figure 4, the value of SSA varied from 91.5% to 95.5%. The minimum value appeared on acceptable days, indicating an enhancement of absorption ability. SSA value increased with pollution levels on polluted days (including acceptable, poor, and bad days) and reached a maximum of 95.5% on bad days, showing that the contribution of scattering to light extinction was greater for higher pollution levels on polluted days.

RH is an important factor affecting the optical properties of PM_2.5. RH was divided into five sections: 20–60%, 60–80%, 80–90%, 90–95% and 95–100%, in order to analyze the effect of RH on optical properties. According to Figure 5, b_scat showed a similar trend as b_ext. The values of b_scat and b_ext at the same RH range value increased with pollution levels. The growth rate of b_scat and b_ext grew with increasing pollution levels. In addition, b_scat and b_ext under all pollution levels increased with the increase of RH when RH was less than 95%. Many components of atmospheric aerosols are hygroscopic, meaning that they take up water as RH increases, and the sizes and effective radius of particles will be augmented, therefore, aerosol light scattering and extinction will be enhanced [26]. However, b_scat and b_ext began to decrease when RH exceeded 95%.

The b_abs showed a clear relationship with RH only on excellent and good days (Figure 5b). On excellent and good days, b_abs decreased with the increase of RH when RH was less than 90% but grew with the increase of RH when RH was more than 90%. The values of SSA under all pollution levels were low when RH was less than 80%, which means the contribution of scattering to light extinction was low (Figure 5d). The maximum of SSA on excellent and good days appeared at the RH range value of 80–90%, while the maximum on acceptable and poor days appeared at 95–100%.

3.2.3. Contribution of PM_2.5 Chemical Components to Light Extinction under Different Pollution Levels

The values of b_ext and contributions of aerosol chemical components to b_ext were calculated in order to determine the effect of chemical components on visibility impairment (Figure 6). During the entire study period, the average value of b_ext was 366.8 Mm⁻¹, and NH₄NO₃ was the largest contributor to b_ext, accounting for 25.2%, followed by OM (20.1%), coarse mass (19.4%), (NH₄)₂SO₄ (15.1%), and soil made the smallest contribution, accounting for only 1.1%.

According to Figure 6a, the value of b_ext increased with the level of pollution. The value of b_ext was 854.7 Mm⁻¹ on bad days and was 4.9 and 2.3 times that on excellent days and overall, respectively. In addition, the contribution rate of each chemical component to b_ext changed as the pollution level increased. OM was the largest contributor to b_ext on excellent days, accounting for 22.1%, and NH₄NO₃ and (NH₄)₂SO₄ accounted for 16.7% and 12.8%, respectively. The contribution of NH₄NO₃ and (NH₄)₂SO₄ to b_ext increased with pollution level. NH₄NO₃ contributed the most to b_ext on poor days, accounting for 42.1%. It was followed by (NH₄)₂SO₄, which accounted for 20.7%. The contribution of OM decreased to 12.5%. The contribution of NH₄NO₃ to b_ext decreased slightly from poor days to bad days, but it was still the largest contributor. NH₄NO₃ was the largest contributor to b_ext on bad days, accounting for 38.2%, followed by (NH₄)₂SO₄ (27.0%), and the contribution of OM to b_ext decreased to 10.9%. EC, sea salt, soil, coarse mass, NO₂, and Rayleigh made a minor contribution, together accounting for 23.9%.

NH₄NO₃ and (NH₄)₂SO₄ were the major components of light extinction in Wuhan, indicating that NH₄NO₃ and (NH₄)₂SO₄ played an important role in visibility impairment. The other chemical components, such as OM, EC, sea salt, coarse mass, and so on, were not the major factors affecting visibility on polluted days since their contribution decreased significantly on poor and bad days.

4. Conclusions

The optical properties of PM_2.5 and the influence of meteorological parameters under different pollution levels were studied in Wuhan in January 2018. Wuhan experienced a severe pollution event from 18 January to 21 January, during which time the average concentration value of PM_2.5 was 170 μg⋅m⁻³, which was 1.9 times the overall average in January. b_ext showed a unimodal distribution on polluted days, peaking at 19:00. b_ext was larger on polluted days than on clean days for the whole day. b_ext on clean days did not change much during the day. On polluted days, b_ext increased significantly starting at 12:00 and occurred along with adverse meteorological factors, including a decrease in wind speed and an increase in RH. The value of b_ext increased with pollution level and was 854.72 Mm⁻¹ on bad days, which was 4.86, 3.1, 2.29, and 1.28 times that on excellent, good, acceptable, and poor days, respectively. The value of SSA varied from 91.5% to 95.5%, increasing with pollution levels on polluted days. b_ext and b_scat under all pollution levels increased with RH when RH was less than 95%. When RH exceeded 95%, b_ext and b_scat began to decrease on acceptable, poor and bad days, but they still increased on excellent and good days. NH₄NO₃ was the largest contributor to overall b_ext, accounting for 25.2%, followed by OM (20.1%), while the contribution of soil was the lowest, accounting only for 1.1%. It is apparent that the contribution of (NH₄)₂SO₄ and NH₄NO₃ to b_ext was much higher with increasing pollution levels. NH₄NO₃ contributed the most to b_ext on bad days, accounting for 38.2%, followed by (NH₄)₂SO₄ (27.0%), and the contribution of OM to b_ext reduced to 10.9%.