Sensitivity Analyses for the Atmospheric Dry Deposition of Total PCDD / Fs-TEQ for Handan and Kaifeng Cities , China

During the period 2016–2017, the atmospheric wet, dry, and total deposition fluxes and scavenging ratios of the total PCDD/Fs-WHO2005-TEQ in Handan and Kaifeng were investigated. In addition, a sensitivity analysis for the dry deposition fluxes of the total PCDD/Fs-WHO2005-TEQ was conducted. The annual wet deposition fluxes of total PCDD/Fs-WHO2005-TEQ in Handan ranged between 51.1 and 83.5 and averaged 72.3 pg WHO2005-TEQ m year, which was approximately 1.04 times of magnitude higher than that in Kaifeng (69.3 pg WHO2005-TEQ m year). From 2016– 2017, the contribution fraction of dry deposition to the total PCDD/Fs-WHO2005-TEQ deposition flux ranged between 60.8% and 100% and averaged 80.4%. Dry deposition fluxes were more dominant than wet deposition fluxes. In terms of the seasonal variations in total PCDD/Fs-WHO2005-TEQ dry deposition fluxes (the mean values for 2016 and 2017) in Handan, those in spring, summer, fall, and winter were 1084, 563, 964, and 1325 pg WHO2005-TEQ m month, respectively, while in Kaifeng, they were 963, 428, 715, and 1016 pg WHO2005-TEQ m month, respectively. The total PCDD/FsWHO2005-TEQ deposition fluxes in winter was approximately 2.0 times of magnitude higher than that in summer. The sensitivity analysis of total PCDD/FsWHO2005-TEQ dry deposition fluxes in Handan and Kaifeng showed that the PM10 concentration was the most positively correlated sensitive factor. When ΔP/P was changed from 0% to +50%, ΔS/S responded from 0% to +46.1% and +46.3%, respectively. The second positively correlated sensitive factor was the PM2.5 concentration, where when ΔP/P was changed from 0% to +50%, ΔS/S responded from 0% to +47.8% and +40.8%, respectively. For PCDD/Fs mass concentration, when ΔP/P was changed from 0% to +50%, ΔS/S responded from 0% to +32.2% and +28.1%, respectively. This was followed by the atmospheric temperature, and its effect was negatively correlated. When ΔP/P was changed from –50% to +50%, ΔS/S responded from +46.4% to –26.9% and +57.0% to –30.5%, respectively. The results of this study provide useful information that can be used to achieve more insights into both atmospheric deposition of total PCDD/Fs-WHO2005-TEQ and the sensitive factors for dry deposition fluxes.


INTRODUCTION
Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) are persistent organic pollutants (POPs) because they are toxic, persistent, and bio-accumulative (Micheletti et al., 2007;White and Birnbaum, 2009).They are extremely hazardous chemicals that form both naturally and anthropogenically (Hashimoto et al., 1990;Brzuzy and Hites, 1996;Kim et al., 2003).Combustion processes in nature, such as forest fires and volcanoes as well as anthropogenic activities are the main sources of the PCDD/Fs released into the environment (Chi et al., 2011).Of these two sources, anthropogenic activities are the most dominant sources contributing to the presence of PCDD/Fs in the environment, which include many manufacturing processes related to products for humans.The main sources of PCDD/Fs have been found to mostly come from the emissions of waste combustion, chemical plants, thermal sources, metal smelting process, and vehicles (Schuhmacher et al., 2000;Wang et al., 2003;Lin et al., 2007;Hsieh et al., 2009;Chuang et al., 2010Chuang et al., , 2011)).PCDD/Fs can enter the human body via ingestion, inhalation, and dermal contact (Shih et al., 2009;Chen et al., 2010).In the human body, PCDD/Fs pose health risks to the human immune system, interfere with developmental, reproductive, and regulatory hormones, and even create the risk of cancer (Lin et al., 2010;Chi et al., 2011).They are chemically stable, have low solubility in the water, and have been shown to accumulate in the food chain (Shih et al., 2009).PCDD/Fs are emitted into the atmosphere, where they are transformed, degraded, and transported from the source to receptor sites (Chi et al., 2009;Xu et al., 2009;Fang et al., 2011).In the atmosphere, PCDD/Fs are partitioned between gas and particle phases through a process that is dependent on their vapor pressures, ambient temperatures, and other parameters (Wu et al., 2009;Wang et al., 2010;Cheruiyot et al., 2015).PCDD/Fs can be degraded by chemical reactions controlled by OH radicals as well as by photochemical reactions (Chi et al., 2009).Removal of PCDD/Fs from the atmosphere occurs via both dry and wet deposition (Giorgi, 1988;Chi et al., 2009;Wu et al., 2009;Huang et al., 2011;Mi et al., 2012).Since they were first found in the fly ash of municipal solid waste incinerators (MSWIs), researchers have paid more attention to various emissions sources (Wang et al., 2003;Wang et al., 2010;Chi et al., 2015;Cheruiyot et al., 2015Cheruiyot et al., , 2016;;Wei et al., 2016;Li et al., 2017).
Wet deposition is the process by which atmospheric pollutants are removed via rainfall, cloud droplets, or snow (Lohmann and Jones, 1998) and is responsible for much of the higher chlorinated homologues in environmental sinks (Shih et al., 2006;Wang et al., 2010).The wet deposition flux of SVOCs is a combination of both vapor dissolution into rain and removal of suspended particulates by precipitation (Bidleman, 1988;Koester and Hltes, 1992).The dry deposition of PCDD/Fs is a combination of both gas-and particle-phase fluxes.The dry deposition fluxes of PCDD/Fs are usually higher than the wet deposition fluxes, demonstrating that dry deposition is the major PCDD/Fs removal mechanism in the atmosphere (Wang et al., 2010;Tseng et al., 2014;Lee et al., 2016;Zhu et al., 2017a, b).The ambient temperature, rainfall, vapor pressure, and particle size will also affect the atmospheric deposition process (Wu et al., 2009;Wang et al., 2010;Chang et al., 2004).This is usually evaluated using the total scavenging ratio (Stot), the ratio of the concentration in the precipitation to the concentration in the atmosphere.
Accumulation due to heavy industry in northern China, the winter cold in the north, and the use of coal for energy have led to significant coal combustion emissions and serious environmental pollution.In addition, the north temperate monsoon climate, characterized by a cold winter and dry, poor air flow, leads to accumulation of atmospheric pollution that is not easily disseminated, which exacerbates air pollution.Zhu et al. (2017) studied the PCDD/Fs, wet/dry deposition of total PCDD/Fs-WHO 2005 -TEQ, the total deposition of total PCDD/Fs-WHO 2005 -TEQ, and the scavenging ratio in the air of northern China's cities (Shijiazhuang and Harbin) over a period of time in 2014, and there have been few studies in recent years.Handan is an industrial city with pillar industries including steel, coal, and cement.All of these industries are characterized by high energy consumption and high air pollutant emission.Handan is also the hardest hit areas in terms of particulate matter (Ren et al., 2004;Wang et al., 2012;Zhao et al., 2012).Kaifeng City is a key city designated by the state environmental protection administration for the prevention and control of national air pollution.With the acceleration of the industrialization process and the rapid development of tertiary industries, in recent years, there have been 222 newly built, reconstructed and expanded coal-fired boilers in urban areas, together with 386 original ones, totaling 608.These boilers burn 2 million tons of coal each year, generating 430,000 tons of slag and releasing 18,000 tons of soot and 15,000 tons of sulfur dioxide into the atmosphere.Therefore, two northern cities, Handan and Kaifeng, were selected for this study to investigate their wet/dry deposition of total PCDD/Fs-WHO 2005 -TEQ, the total deposition of total PCDD/Fs-WHO 2005 -TEQ, and the scavenging ratios in both 2016 and 2017, thus providing valuable current data for the air pollution in northern cities.

METHODS
Two cities, Handan and Kaifeng in Hebei and Henan province, China, respectively, were selected and evaluated in this study.The monthly average of PM 2.5 and PM 10 concentrations, and related meteorological information including monthly average temperature and precipitation from 2016 and 2017 in Handan and Kaifeng were obtained from local air quality monitoring stations and the Weather Underground website.

Atmospheric Dry Deposition of PCDD/Fs
The atmospheric dry deposition flux of PCDD/Fs is a combination of both gas-and particle-phase fluxes, which are given by: (1) F d,T : the total PCDD/F deposition flux contributed by adding both gas-and particle-phase deposition fluxes, where F d,g : the PCDD/F deposition flux contributed by the gas phase; F d,p : the PCDD/F deposition flux contributed by the particle phase; C T : the measured concentration of total PCDD/Fs in the ambient air; V d.T : the dry deposition velocity of total PCDD/Fs; C g : the calculated concentration of PCDD/Fs in the gas phase; V d,g : the dry deposition velocity of gas-phase PCDD/Fs; C p : the calculated concentration of PCDD/Fs in the particle phase; V d,p : the dry deposition velocity of particle-phase PCDD/Fs.In this study, the mean dry deposition velocity of total PCDD/Fs (V d.T = 0.42 cm s -1 ) was as proposed by Shih et al. (2006).Dry deposition of gas-phase PCDD/Fs occurs mainly by diffusion, and due to the lack of measured data for PCDD/Fs, a selected value (0.010 cm s -1 ) for the gasphase PAH dry deposition velocity, V d,g , as proposed by Sheu et al. (1996) and used by Lee et al. (1996), was used in the current work to calculate the PCDD/F dry deposition flux contributed by its gas phase.Dry deposition of particlephase PCDD/Fs is mainly achieved by gravitational settling, and the dry deposition velocity of particle-phase PCDD/Fs, V d,g , can be calculated using Eq.(1).

Scavenging Ratios
In the case of slightly soluble trace organic compounds, such as PCDD/Fs and other semi-volatile organic compounds, it is commonly believed that equilibrium partitioning occurs between the compound in the gas phase and that in a falling rain drop (Ligocki et al., 1985a;Ligocki et al., 1985b;Cheruiyot et al., 2015;Cheruiyot et al., 2016;Redfern et al., 2017).The scavenging ratio is defined as the concentration of the pollutant in the raindrop divided by the concentration of the same pollutant in the surrounding air during precipitation.The gas scavenging ratio, S g , can be estimated by: where S g : the gas scavenging ratio of PCDD/Fs (dimensionless); R: the universal gas constant (82.06 × 10 -6 m 3 atm mol -1 K -1 ); T: ambient temperature (K); H: the Henry constant (m 3 atm mol -1 ).
On the other hand, particle scavenging largely depends on meteorological factors and particle characteristics.The gas scavenging ratio is the ratio of the dissolved phase concentration in the raindrop divided by the gas phase concentration in the air, S g , and can be calculated by: where S g : the gas scavenging ratio of PCDD/Fs (dimensionless); C rain,dis : the dissolved-phase concentration of PCDD/Fs in the raindrop; C g : the concentration of PCDD/Fs in the gas phase.
The particle scavenging ratio is the ratio of the particle phase concentration in a raindrop divided by the particle phase concentration in the air, S p , which can be calculated by: where S p : the particle scavenging ratio of PCDD/Fs (dimensionless); C rain,particle : the particle-phase concentration of PCDD/Fs in the raindrop; C p : the concentration of PCDD/Fs in the particle phase.
The total scavenging of precipitation is the sum of gas and particle scavenging, S tot , which can be calculated by: where S tot : the total scavenging ratio of PCDD/Fs (dimensionless); ϕ: the fraction of the total air concentration bound to particles.
Because of a lack of measured data for the particle scavenging ratios of PCDD/Fs, the S p (S p is 42,000) and the values of OCDD and OCDF as measured by Eitzer and Hites (1989) were averaged and used here.

Wet Deposition
Wet deposition is the removal of particles in the atmosphere through precipitation (rainfall and cloud droplets).Precipitation scavenging accounts for the majority of removal of PCDD/Fs from the atmosphere by wet deposition (Huang et al., 2011b).The wet deposition flux of PCDD/Fs is a combination of both vapor dissolution into rain and the removal of suspended particulates through precipitation (Bidleman et al., 1988; Koester and Hites et al., 1992).
The wet deposition fluxes of PCDD/Fs can be evaluated by: where F w,T : the wet deposition flux of PCDD/Fs from both vapor dissolution into rain and removal of suspended particulates by precipitation; F w,dis : the wet deposition flux contributed by vapor dissolution into rain; F w,p : the wet deposition flux contributed by removal of suspended particulates by precipitation; Rainfall: monthly rainfall (m).

Wet Deposition
Wet deposition of PCDD/Fs is defined as the removal of both particle-phase and vapor-phase PCDD/Fs from the atmosphere through rainfall or other precipitation (Lee et al., 2016) (Zhu et al., 2017).The impact of rainfall on wet deposition is significant.Under normal conditions, an increase in rainfall may cause a rise in wet deposition (Chen et al., 2017).In 2016, the maximum wet deposition in Handan occurred in July (333.3pg WHO 2005 -TEQ m -2 month -1 ), while the minimum occurred in March (almost zero), with an annual wet deposition flux of 988.7 pg WHO 2005 -TEQ m -2 year -1 , and the rainfall was 414.6,0 and 786.4 mm, respectively.In 2017, the maximum wet deposition occurred in May (122.1 pg WHO 2005 -TEQ m -2 month -1 ) was approximately 111 orders of magnitude higher than the minimum occurring in November (1.1 WHO 2005 -TEQ m -2 month -1 ), with an annual wet deposition flux of 613.7 pg WHO 2005 -TEQ m -2 year -1 and rainfall of 37.6, 0.3, and 318.4 mm, respectively.From Table 1 we can clearly see that the maximum rainfall occurring in June (71 mm)  2015).In 2016, the average wet deposition fluxes were 31.9,170.8, 84.2, and 42.7 pg WHO 2005 -TEQ m -2 month -1 in spring, summer, fall, and winter, respectively, which were associated with rainfall of 13.4, 210.8, 28.9 and 9.0 mm, respectively.In 2017, the average wet deposition fluxes were 80.5, 68.8, 27.4, and 28.0 pg WHO 2005 -TEQ m -2 month -1 in spring, summer, fall, and winter, respectively, which were associated with rainfall of 25.0, 64.6, 11.4, and 5.0 mm, respectively.
In the case of Kaifeng, in 2016, the monthly average wet deposition flux was 65.3 pg WHO 2005 -TEQ m -2 month -1 , ranging from 1.0 to 107.0 pg WHO 2005 -TEQ m -2 month -1 .The annual wet deposition flux of total PCDD/Fs-WHO 2005 -TEQ was 783.4 pg WHO 2005 -TEQ m -2 year -1 , which was a 43.0%order of magnitude higher than that of Taisi City (548 pg WHO 2005 -TEQ m -2 year -1 ) in the coastal area of central Taiwan according to Chen et al.'s investigating in the same year.In 2017, the monthly average was 67.9 pg WHO 2005 -TEQ m -2 month -1 , with wet deposition fluxes between 2.3 and 216.9 pg WHO 2005 -TEQ m -2 month -1 .The annual wet deposition flux of total PCDD/Fs-WHO 2005 -TEQ was 814.8 pg WHO 2005 -TEQ m -2 year -1 , which was a 1.2 times of magnitude increased by that of Harbin (369 pg WHO 2005 -TEQ m -2 year -1 ) in 2014 (Zhu et al., 2017).As for the impact of rainfall on wet deposition, in 2016, the maximum wet deposition occurred in July (107.0pg WHO 2005 -TEQ m -2 month -1 ) with monthly rainfall of 212.4 mm, which was far above the minimum occurring in March (1.0 pg WHO 2005 -TEQ m -2 month -1 ), with monthly rainfall of 0.3 mm and an annual wet deposition flux of 783.4 pg WHO 2005 -TEQ m -2 year -1 and annual rainfall of 672.8 mm.In 2017, the maximum wet deposition occurred in May (216.9 pg WHO 2005 -TEQ m -2 month -1 ) (with monthly rainfall of 36.8 mm), which was approximately 94.3 orders of magnitude higher than the minimum occurring in December (2.3 pg WHO 2005 -TEQ m -2 month -1 ) (0.8 mm).The annual wet deposition flux was 814.8 pg WHO 2005 -TEQ m -2 year -1 , with annual rainfall of 424.9 mm.Similarly, from Table 1, we can see that the maximum rainfall in 2017 occurred in August (140.0mm) and the minimum occurred in November (0.5 mm), which were different from May, in which the maximum value of wet deposition fluxes occurred (36.8 mm) and the minimum occurring in December (0.8 mm).This further proves that rainfall is an important factor affecting wet deposition fluxes but that it is not the only factor.In 2016, the average wet deposition fluxes were 55.9, 78.4,78.3 and 48.6 pg WHO 2005 -TEQ m -2 month -1 in spring, summer, fall, and winter, respectively, which were associated with rainfall of 24.3, 138.0, 48.7 and 13.2 mm, respectively.In 2017, the average wet deposition fluxes were 117.2, 60.7, 79.9 and 13.7 pg WHO 2005 -TEQ m -2 month -1 in spring, summer, fall, and winter, respectively, which were associated with rainfall of 21.1, 79.7, 34.4 and 6.5 mm, respectively, since the summer season has more rainfall compared to the winter season, which has fewer rainy days (Lee et al., 2016).In general, the summer wet deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ are greater than those for winter.The highest wet deposition fluxes occurred in the spring, but the maximum of rainfall occurred in the summer, which further proved that wet deposition is not only affected by rainfall, but also by factors such as particulate matter concentration, temperature, and wind speed (Wang et al., 2010;Huang et al., 2011).
Based on the above monthly rainfall as obtained from local air quality monitoring stations and monthly wet deposition fluxes, the concentrations of total PCDD/Fs-WHO 2005 -TEQ in rain can be calculated by the wet deposition fluxes divided by rainfall intensity.The monthly average total PCDD/Fs-WHO 2005 -TEQ concentrations in the rain in Handan and Kaifeng from 2016 to 2017 are presented in Figs.1(c) and Fig. 1(d).As the results show, in Handan, the monthly average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain were ranged between 0 (March) and 7.38 (December) pg WHO 2005 -TEQ L -1 , and averaged 2.87 pg WHO 2005 -TEQ L -1 in 2016.This value was 8.3 times of magnitude higher than Kaohsiung in Taiwan in 2015, which averaged 0.307 pg WHO 2005 -TEQ L -1 (Lee et al., 2016).In 2017, the monthly average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain were ranged between 0.67 (July) and 6.70 (January) pg WHO 2005 -TEQ L -1 , and averaged 2.98 pg WHO 2005 -TEQ L -1 , which was approximately 1.3 orders of magnitude higher than that of Harbin in 2014, which averaged 2.28 pg WHO 2005 -TEQ L -1 (Zhu et al., 2017).As for seasonal variations, the average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain were 2.38, 0.81, 2.91, and 4.74 pg WHO 2005 -TEQ L -1 , in spring, summer, fall, and winter, respectively, in 2016.In 2017, the average concentrations were 3.22, 1.07, 2.40, and 5.60 pg WHO 2005 -TEQ L -1 , in spring, summer, fall, and winter, respectively.It can be seen that the highest value occurred in the summer and the lowest occurred in the winter.This is related to the fact that summer is the rainy season.
In Kaifeng, the monthly average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain were ranged between 0.47 (August) and 4.49 (January) pg WHO 2005 -TEQ L -1 , with an average of 2.30 pg WHO 2005 -TEQ L -1 in 2016.This value was 13.7 times of magnitude higher than that in Meinong, Taiwan in 2015 (averaged 0.167 pg WHO 2005 -TEQ L -1 ) (Lee et al., 2016).In 2017, the monthly average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain were ranged between 0.49 (July) and 21.2 (November) pg WHO 2005 -TEQ L -1 , with an average 4.15 pg WHO 2005 -TEQ L -1 , which was 1.1 times higher than that of Shijiazhuang in 2014 (averaged 3.95 pg WHO 2005 -TEQ L -1 ) (Zhu et al., 2017).As for seasonal variations, in 2016, the average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain were 2.30, 0.57, 1.61, and 3.68 pg WHO 2005 -TEQ L -1 in spring, summer, fall, and winter, respectively, and those in 2017, were 5.37, 1.38, 7.67, and 2.18 pg WHO 2005 -TEQ L -1 , respectively.In summer, more rain and lower PCDD/F concentrations in the atmosphere resulted in a lower PCDD/F concentration in the rain.
The total scavenging ratios (Stot) in Handan City and Kaifeng City from 2016 to 2017 are shown in Figs.1(e) and 1(f).It can be clearly seen that the scavenging ratio (Stot) trend in the two cities in 2016 and 2017 was similar.In Handan, the scavenging ratio (Stot) of total PCDD/Fs-WHO 2005 -TEQ ranges from 22970 (July) to 41350 (January) and from 21040 (July) to 41400 (January), with an average of 33040 and 32980 in 2016 and 2017, respectively.In Kaifeng, the scavenging ratio (Stot) of total PCDD/Fs-WHO 2005 -TEQ ranged between 20120 (July) and 41110 (January) and ranged between 18980 (July) and 40730 (January), averaging 31750 and 31890 in 2016 and 2017, respectively.The  maximum and minimum values for both cities occurred in January (winter) and July (summer), respectively.The above results were very similar to those of previous studies, where the total scavenging ratio (Stot) increased with decreased in temperature (Chen et al., 2017).

Dry Deposition
The dry deposition fluxes of PCDD/Fs are contributed by both the gas and particle phases.The monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ in the ambient air of Handan and Kaifeng from 2016 to 2017 are shown in Figs.2(a) and 2(b), respectively.For Handan, in 2016, the monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged between 467 and 1959 pg WHO 2005 -TEQ m -2 month -1 , with an annual dry deposition fluxes of 11590 pg WHO 2005 -TEQ m -2 year -1 , which was approximately 2.9 times higher than those in Lunbei, Taiwan (3970 pg WHO 2005 -TEQ m -2 year -1 ) in the same year (Chen et al., 2017).The maximum monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ (1959 pg WHO 2005 -TEQ m -2 month -1 ) occurred in December at a level that was approximately 2.8 times higher than the minimum level (467 pg WHO 2005 -TEQ m -2 month -1 ), which occurred in August.In 2017, the monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged between 489 and 1765 pg WHO 2005 -TEQ m -2 month -1 , with an annual  dry deposition flux of 12020 pg WHO 2005 -TEQ m -2 year -1 , which was a 22.0% of magnitude decreased by that of Shijiazhuang (15400 pg WHO 2005 -TEQ m -2 year -1 ) in 2014 (Zhu et al., 2017).The maximum monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ (1765 pg WHO 2005 -TEQ m -2 month -1 ) occurred in January at a level that was approximately 3.6 times higher than the minimum level (489 pg WHO 2005 -TEQ m -2 month -1 ), which occurred in July.In terms of seasonal variations, the average values of the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ were in the following order: winter (1324 pg WHO 2005 -TEQ m -2 month -1 ) > spring (1084 pg WHO 2005 -TEQ m -2 month -1 ) > fall (964 pg WHO 2005 -TEQ m -2 month -1 ) > summer (562 pg WHO 2005 -TEQ m -2 month -1 ).
The value in summer (562 pg WHO 2005 -TEQ m -2 month -1 ) was 57.6% lower than the value in winter (1324 WHO 2005 -TEQ m -2 month -1 ).The maximum for both occurred in winter, and the minimum for both occurred in the summer.This is due to the fact that a higher temperature will cause a greater fraction of PCDD/Fs in the gas phase in summer and the gas phase scavenging ratio will be less than that of the particle phase (Chen et al., 2017).which was approximately 2.7 times higher than that in Taisi, Taiwan (3580 pg WHO 2005 -TEQ m -2 year -1 ) in the same year (Chen et al., 2017).The maximum monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005-TEQ (1213 pg WHO 2005 -TEQ m -2 month -1 ) occurred in January, with a level that was approximately 3.4 times higher than the minimum level (354 pg WHO 2005 -TEQ m -2 month -1 ), which occurred in August.In 2017, the monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged between 377 and 1140 pg WHO 2005 -TEQ m -2 month -1 , with an annual dry deposition flux of 9094 pg WHO 2005 -TEQ m -2 year -1 , which was a 1.1 times of magnitude higher than Harbin (8240 pg WHO 2005 -TEQ m -2 year -1 ) in 2014 (Zhu et al., 2017).The maximum monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ (1140 pg WHO 2005 -TEQ m -2 month -1 ) occurred in January, at a level that was approximately 3.0 times higher than the minimum level (377 pg WHO 2005 -TEQ m -2 month -1 ), which occurred in July.Similar to the seasonal variations in Handan City, the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ were in the following order: (1016 pg WHO 2005 -TEQ m -2 month -1 ) > spring (963 pg WHO 2005 -TEQ m -2 month -1 ) > fall (714 pg WHO 2005 -TEQ m -2 month -1 ) > summer (427 pg WHO 2005 -TEQ m -2 month -1 ).The value in summer (427 pg WHO 2005 -TEQ m -2 month -1 ) was a 60.0% of magnitude lower than the value in winter (1016 pg WHO 2005 -TEQ m -2 month -1 ).
It was also verified that as the temperature increased, the dry deposition fluxes were reduced.Similar to previous studies, in Taiwan, the seasonal variations in dry deposition fluxes were high in winter and low in summer (Wu et al., 2009;Wang et al., 2010;Mi et al., 2012;Zhu et al., 2017).

Total (Dry +Wet) Deposition
The total (dry + wet) deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ were calculated by adding the dry and wet deposition fluxes, and shown in Figs.3(a) and 3(b), respectively.Figs.3(c) and 3(d) show the contribution fractions for the wet deposition and dry deposition fluxes, respectively, to the total deposition fluxes in Handan and Kaifeng for 2016-2017.For Handan, the monthly total (dry + wet) deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged from 545 (August) to 2046 (December) pg WHO 2005 -TEQ m -2 month -1 as well as the annual 12580 pg WHO 2005 -TEQ m -2 year -1 in 2016, which were approximately 2.9 times higher than those in Lunbei, Taiwan (4360 pg WHO 2005 -TEQ m -2 year -1 ) in the same year (Chen et al., 2017).The contribution fractions of dry deposition to the total deposition fluxes ranged from 60.8% to 100%, where the maximum occurred in March (100%), and the minimum occurred in July (60.8%).This is because the maximum and minimum rainfall for 2016 occurred in July (414.6 mm) and March (0 mm) respectively, resulting in a variation in the dry deposition contribution fraction to the total (dry + wet) deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ.In 2017, the monthly total (dry + wet) deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged from 534 (July) to 1813 (January) pg WHO 2005 -TEQ m -2 month -1 , with an annual value of 12630 pg WHO 2005 -TEQ m -2 year -1 , which was a 21.5% of magnitude decreased by that in Shijiazhuang (16100 pg WHO 2005 -TEQ m -2 year -1 ) in 2014 (Zhu et al., 2017).The contribution fraction of dry deposition to total deposition ranged from 88.0% to 99.9%; the maximum occurred in November (99.9%).and the minimum occurred in June (88.0%).In these two months, the rainfall was 71 mm and 0.3 mm, respectively.As to the seasonal variations in Handan, the averages for these two years were 1141, 682, 1020, and 1360 pg WHO 2005 -TEQ m -2 month -1 , and that of winter was approximately 2.0 times higher than that of summer.Previous studies have mentioned that the total dry deposition flux increases as the ambient air temperature decreases (Shih et al., 2006;Huang et al., 2011).When ambient air temperature decreases in the winter season, the PM 10 , PM 2.5 and total PCDD/F concentration will increase and thus results in higher total deposition fluxes in January than that occur in August (Lee et al., 2016).
For Kaifeng, the monthly total (dry + wet) deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged from 391 (August) to 1227 (January) pg WHO 2005 -TEQ m -2 month -1 , with an annual value of 10410 pg WHO 2005 -TEQ m -2 year -1 in 2016, which was approximately 2.5 times higher than that in Taisi, Taiwan (4130 pg WHO 2005 -TEQ m -2 year -1 ) in the same year (Chen et al., 2017).The contribution fraction of dry deposition to the total deposition ranged from 78.5% to 99.9%, and averaged 90.9%; the maximum occurred in March (99.9%), and the minimum occurred in July (78.5%), and the maximum and minimum rainfall for 2016 occurred in July (212.4mm) and March (0.25 mm) respectively.In 2017, the monthly total (dry + wet) deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged from 412 (July) to 1395 (May) pg WHO 2005 -TEQ m -2 month -1 , with an annual value of 9909 pg WHO 2005 -TEQ m -2 year -1 , which was a 15% order magnitude higher than that in Harbin (8610 pg WHO 2005 -TEQ m -2 year -1 ) in 2014 (Zhu et al., 2017).The contribution fraction of dry deposition to the total deposition ranged from 84.5% to 99.7% and averaged 91.8%; the maximum occurred in December (99.7%), and the minimum occurred in May (84.5%), and the rainfall for these two months was 36.8 mm and 0.8 mm, respectively.As to the seasonal variations, the average values for 2016 and 2017 of total PCDD/Fs-WHO 2005 -TEQ deposition fluxes in Handan, were 1049, 497, 794, and 1047 pg WHO 2005 -TEQ m -2 month -1 in spring, summer, fall, and winter, respectively, for which the values for spring and winter were approximately 2.1 orders of magnitude higher than that for summer.

Sensitivity Analysis
The parameters that can affect gas-particle partitioning will be the parameters influencing the dry deposition fluxes.The sensitivity analysis of this study is focused on the ambient temperature, the PM 2.5 concentration, the PM 10 concentration, and the total PCDD/F concentration, which are sensitive variables for altering dry deposition fluxes.The sensitivity analyses conducted in this study in Handan and Kaifeng were dependent on the initial values of ambient air temperature = 13.5°C,PM 2.5 = 68.0µg m -3 , PM 10 =130.0 µg m -3 , total PCDD/F mass concentration = 1.76 pg m -3 , ambient air temperature = 13.7°C,PM 2.5 = 54.0 µg m -3 , PM 10 = 107.0µg m -3 , and total PCDD/F mass  The PM 2.5 and PM 10 concentrations have equivalently important sensibility to the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ.The effect of both PM 10 concentration and PCDD/Fs can be divided into two parts: For PM 10 concentration, whenΔP/P was changed from 0% to -50%, ΔS/S responded from 0% to -46.1%; when ΔP/P was changed from 0% to +50% and +80%, ΔS/S responded from 0% to +46.1% and from 0% to +73.7%, respectively.In terms of PM 2.5 concentration, when ΔP/P was changed from 0% to -50%, ΔS/S responded from 0% to -53.8%; when ΔP/P was changed from 0% to +50% and +80%, ΔS/S responded from 0% to +47.8% and from 0% to +73.6%, respectively in Handan.In the case of Kaifeng, for PM 10 , when ΔP/P was changed from 0% to -50%, ΔS/S responded from 0% to -46.3%; when ΔP/P was changed from 0% to +50% and +80%, ΔS/S responded from 0% to +46.3% and from 0% to +74.1%, respectively.In terms of  PM 2.5 concentration, when ΔP/P was changed from 0% to -50%, ΔS/S responded from 0% to -65.3%; when ΔP/P was changed from 0% to +50% and +80%, ΔS/S responded from 0% to +40.8% and from 0% to +53.6%, respectively The PM 10 can reflect the status of particle phase PCDD/Fs, where at low concentrations, the effect of PM 10 concentration on dry deposition is greater than that of PM 2.5 , but as the concentration of PM increases, the effect of PM 2.5 concentration on dry deposition gradually increases.In addition, after the PM 2.5 concentration reaches a certain level, the effect of increasing concentration on the dry deposition flux decreases.This because when the PM 2.5 concentrations is further increased, its sensitivity is weaker than that incurred when the PM 10 concentration is decreased.
The sensitivity analysis indicated that PCDD/Fs mass concentration has a significant, positive effect on the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ.In the case of Handan, the effect of PCDD/Fs mass concentration can be divided into two parts: When ΔP/P was changed from 0% to -50%, ΔS/S responded from 0% to -69.7%; when ΔP/P was changed from 0% to +50% and +100%, ΔS/S responded from 0% to +32.2% and +27.0%, respectively.In Kaifeng, when ΔP/P was changed from 0% to -50%, ΔS/S responded from 0% to -50.6%; when ΔP/P was changed from 0% to +50% and +100%, ΔS/S responded from 0% to +28.1% and +33.8%, respectively.It can be seen that at low concentrations, the mass concentration of PCDD/Fs has a significant effect on the dry deposition fluxes, but the effect is weakened at high concentrations, indicating that the mass concentration of PCDD/Fs has a saturation value in the case of the dry deposition fluxes, so this factor is weaker than other factors.
Figs. 4(a) and 4(b) both show that air temperature has a negative correlation with the dry deposition of total PCDD/Fs-WHO 2005 -TEQ.In the case of Handan, when ΔP/P was changed from -50% to +50%, ΔS/S responded from +46.4% to -26.9%.In Kaifeng, when ΔP/P was changed from -50% to +50%, ΔS/S responded from +57.0% to -30.5%.The temperature will affect the vapor pressure of PCDD/Fs and influence gas-particle partitioning because a higher temperature will cause a greater fraction of PCDD/Fs in the gas phase in the summer, and the gas phase scavenging ratio will be less than that of the particle phase (Chen et al., 2017).As the air temperature increases, this will result in a greater amount of particle-bound PCDD/Fs evaporating to the gas phase, and the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ will decrease with increases in air temperature.
The above results suggest that the most sensitive parameters for the dry deposition of total PCDD/Fs-WHO 2005 -TEQ are atmospheric PM 2.5 and PM 10 concentrations, followed by the PCDD/Fs mass concentration and air temperature,

CONCLUSION
The results of this study on the atmospheric deposition in Handan and Kaifeng can be summarized as follows: 1.In 2016, the annual wet deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ in Handan was 988.7 pg WHO 2005 -TEQ m -2 year -1 , which was 1.3 times higher than that in Kaifeng (783.4 pg WHO 2005 -TEQ m -2 year -1 ).
In 2017, the annual wet deposition flux of total PCDD/Fs-WHO 2005 -TEQ in Handan was 613.7 pg WHO 2005 -TEQ m -2 year -1 , which was approximately 24.7% lower than that in Kaifeng (814.8 pg WHO 2005 -TEQ m -2 year -1 ). 2. In Handan, the monthly average concentrations of total PCDD/Fs-WHO 2005 -TEQ in the rain ranged between null and 7.38 and between 0.67 and 6.70 pg WHO 2005 -TEQ L -1 in 2016 and 2017, respectively.In Kaifeng, the monthly average concentrations ranged between 0.47 and 4.49 and between 0.49 and 21.2 pg WHO 2005 -TEQ L -1 , respectively.3. The trend of the scavenging ratios (Stot) for the two cities in 2016 and 2017 was similar.The average total scavenging ratios of total PCDD/Fs-WHO 2005 -TEQ in Handan were 33040 and 32980, respectively.The maximum and minimum values for both cities occurred in January in winter and July in summer, respectively.4. In 2016, the annual dry deposition flux of total PCDD/Fs-WHO 2005 -TEQ in Handan was 11590 pg WHO 2005 -TEQ m -2 year -1 , which was 1.2 times higher than that in Kaifeng (9630 pg WHO 2005 -TEQ m -2 year -1 ).In 2017, the annual dry deposition flux of total PCDD/Fs-WHO 2005 -TEQ in Handan was 12020 pg WHO 2005 -TEQ m -2 year -1 , which was approximately 1.3 times higher than that in Kaifeng (9094 pg WHO 2005 -TEQ m -2 year -1 ).For the seasonal distribution, the average values of the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ were in the following order: winter > spring > fall > summer in both Handan and Kaifeng from 2016 to 2017.This is due to the fact that a higher temperature will cause a greater fraction of PCDD/Fs in the gas phase in summer, and the gas phase scavenging ratio will be less than that of the particle phase. 5.In 2016, the annual total (wet + dry) deposition flux of total PCDD/Fs-WHO 2005 -TEQ in Handan was 12580 pg WHO 2005 -TEQ m -2 year -1 , which was 1.2 times higher than that in Kaifeng (10410 pg WHO 2005 -TEQ m -2 year -1 ).
In 2017, the annual total (wet + dry) deposition flux of total PCDD/Fs-WHO 2005 -TEQ in Handan was 12630 pg WHO 2005 -TEQ m -2 year -1 , which was approximately 1.3 times higher than that in Kaifeng (9900 pg WHO 2005 -TEQ m -2 year -1 ).In addition, the average contribution fraction of dry deposition to the total deposition was 92.7% and 91.4% for Handan and Kaifeng in both 2016 and 2017, respectively.6.The sensitivity analysis of the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ in Handan and Kaifeng showed that the PM 2.5 and PM 10 concentrations were the most positively correlated sensitive factor.When ΔP/P was changed from 0% to +50%, ΔS/S responded from 0% to +46.1% and +46.3%, respectively.The second positively correlated sensitive factor was the PM 2.5 concentration.When ΔP/P was changed from 0% to +50%, ΔS/S responded from 0% to +47.8% and +40.8%.For PCDD/Fs mass concentration, when ΔP/P was changed from 0% to +50%, ΔS/S responded from 0% to +32.2% and +28.1%, respectively.This was followed by the atmospheric temperature, for which the effect was negatively correlated.When ΔP/P was changed from -50% to +50%, ΔS/S responded from +46.4% to -26.9% and +57.0% to -30.5%.

Fig. 1
Fig. 1(c).Monthly average concentration of total PCDD/Fs-WHO 2005 -TEQ in the Rain of Handan and Kaifeng in 2016.

Fig. 1
Fig. 1(d).Monthly average concentration of total PCDD/Fs-WHO 2005 -TEQ in the Rain of Handan and Kaifeng in 2017.
For Kaifeng, in 2016, the monthly average dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ ranged from 354 to 1213 pg WHO 2005 -TEQ m -2 month -1 , with an annual dry deposition flux of 9630 pg WHO 2005 -TEQ m -2 year -1 ,
Fig. 3(c).The fraction of total deposition flux in total PCDD/Fs-WHO 2005 -TEQ contributed by the dry and wet deposition, in Handan and Kaifeng in 2016.

Fig. 3
Fig. 3(d).The fraction of total deposition flux in total PCDD/Fs-WHO 2005 -TEQ contributed by the dry and wet deposition, in Handan and Kaifeng in 2017.
Fig. 4(a.)Sensitivity analysis for the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ in Handan from 2016 to 2017.

Fig. 4
Fig. 4(b) Sensitivity analysis for the dry deposition fluxes of total PCDD/Fs-WHO 2005 -TEQ in Kaifeng from 2016 to 2017.
which had a negative correlation with the dry deposition of total PCDD/Fs-WHO 2005 -TEQ.

Table 1 .
Monthly Rainfall (mm) of Handan and Kaifeng from 2016 to 2017.