Intra-Annual Deposition of Atmospheric 210 Pb , 210 Po and the Residence Times of Aerosol in Xiamen , China

Atmospheric deposition of Pb and Po from Nov. 2010 to Jan. 2012 were analyzed to reveal their temporal variations, as well as applications in constraining the residence times of aerosol. The monthly depositional fluxes varied from 2.42 to 29.31 Bq/m/mon and from 0.160 to 3.388 Bq/m/mon for Pb and Po, respectively. High fluxes of Pb and Po were observed in the southwest monsoon prevailing months, while low fluxes corresponded to the northeast monsoon seasons, revealing the monsoon control over Pb and Po deposition on seasonal timescales. There were significant positive linear correlations between the daily depositional fluxes and precipitation, supporting the predominant removal passage of Pb and Po through rainfall. The inverse relations between specific activities and precipitation indicated that the removal efficiencies were much higher at the beginning of rainfall. In contrast, the dry deposition only accounted for 22% and 29% of the bulk depositional fluxes of Pb and Po, respectively. The depositional flux ratios of Po to Pb varied between 0.015 and 0.223 with a mean of 0.091 ± 0.012, corresponding to the residence times of 3.1–57.5 d with an average of 20.8 ± 3.0 d. Investigations on size-fractionated Po and Pb could provide better understanding of the resident timescales of aerosol.


INTRODUCTION
Rn in the atmosphere is mainly from the continents.The global exhalation of 222 Rn from the continents varies from 1300 to 1800 Bq/m 2 /d, while its input from the oceans is only about 17 Bq/m 2 /d (Samuelsson et al., 1986;Nazaroff, 1992).Accordingly, the concentrations of atmospheric 222 Rn are controlled by the land-sea distribution pattern (Balkanski et al., 1992).This pattern, to a large degree, decides the longitudinal distributions of its daughters, such as 210 Pb (T 1/2 = 22.3 yr) and 210 Po (T 1/2 = 138.4d) in the atmosphere and finally their spatial deposition patterns (Turekian et al., 1977;Lee et al., 2004).Apparently, the depositional fluxes of 210  Pb and 210 Po are latitude-dependent on a global scale (Turekian et al., 1977;Kim et al., 1998).Because 210 Po in surface air is from 210 Pb, the disequilibrium between 210 Po and 210 Pb (i.e., deficit of 210 Po to 210 Pb) actually reflected the residence time of 210 Pb in the atmosphere (Baskaran, 2011).Owing to their strong particle reactivity, 210  Pb and  210 Po are readily adsorbed onto aerosols after production, and then removed from the atmosphere through deposition (Marley et al., 2000).The depositional 210 Po/ 210 Pb ratios can be used to quantify the residence times of aerosol and diffusion of atmospheric contaminants (Moore et al., 1973;Nho et al., 1996;Marley et al., 2000;Papastefanou, 2009;Baskaran, 2011).
Even though 210 Pb as an aerosol proxy has been examined especially in European and North American countries (Baskaran et al., 1993;Baskaran and Shaw, 2001;McNeary and Baskaran, 2007), 210 Po has been less employed.To date, several models have been established to simulate the atmospheric 210 Pb deposition on global-scales (Turekian et al., 1977;Lee et al., 2004).However, recent researches revealed the significant intra-and inter-annual variability of 210 Pb deposition at a specific station (Su et al., 2003;Huh et al., 2006;Wan et al., 2008).Some episodic events such as volcano activity, typhoon and dust storm are responsible for the unexpected deposition of 210 Pb (Su and Huh, 2002;Du et al., 2008).For example, dust storm and anthropogenic activities in 2006 result in the abnormally enhanced 210 Pb deposition in Shanghai (Du et al., 2008).Similar local effects might lead the deposition of 210 Pb to deviate significantly from the global simulation models.Consequently, these model values may introduce unreal estimation when they are adopted in quantifying some local processes such as resident timescale of aerosol constrained by the 210 Po- 210 Pb technique.Although 210 Po has been less studied on a global-scale, limited investigations indicated its region-specific depositional patterns.Therefore, to better use the 210 Po/ 210 Pb proxy to trace atmospheric processes time-series observation is required at the specific location.
Research on the deposition of 210 Pb and 210 Po in Chinese mainland is still absent although they were examnied in several sites such as Guizhou (Wan et al., 2008), Shanghai (Du et al., 2008) and Qingdao (Yi et al., 2005).Since China accounts for 6.7% of the global continental area and the desert area makes up 30% of Chinese land, it is expected that the emanation of 222 Rn from Chinese continent is of great importance in the global atmospheric 222 Rn budget, as well as its daughters of 210 Pb and 210 Po.In Jan-Feb, 2013, many Chinese cities including Beijing experienced the most serious haze, indicating that the air pollution in China is now in a fearful situation. 210 Pb and 210 Po are effective tools to study the transport and residence of atmospheric pollutants in surface air (Lamborg et al., 2000;Kristiansen et al., 2012;Yang et al., 2012).Thus, extensive investigations on the atmospheric 210  Pb and 210 Po in China could provide important insights into the diffusion and movement of atmospheric contaminants in the background of more and more serious air pollution events.
Xiamen, a southeast coast city of China, almost represents the least polluted city in China; however, 21 days of fog and haze have also been recorded during Jan-Feb, 2013.Examining the deposition of atmospheric 210 Pb and 210 Po could promote our understanding of the residence, transport and removal processes of contaminants in the atmosphere across the Taiwan Strait.In this study, the atmospheric depositional fluxes of 210  Pb and 210 Po from November 2010 to January 2012 were investigated to examine its temporal variations and controlling factors.Further, their applications in constraining the residence times of aerosol were also investigated.

MATERIALS AND METHODS
Time-integrated atmospheric depositions were sampled from November 2010 to January 2012 on the roof of the SHIJING Building (24°26′12′′N, 118°5′54″E) at Xiamen University (Fig. 1).The SHIJING Building is situated at the top of a hill; hence the samples represent the true atmospheric deposition extensively.In general, time resolution of sampling, depending on precipitation, is less than 15 d except four samples (Table 1).Samples were collected using a clean polyethylene container with an area of 1.09 m 2 , which was sequentially pre-cleaned by 0.3 mol/L HCl and Milli-Q water.After collection, the bulk materials, including both wet and dry deposition, were transferred into a cleaned polyethylene bottle.Then, 0.5 mol/L HNO 3 and Milli-Q water were used to wash particle reactive radionuclides of 210 Po and 210 Pb which may adsorbed onto the container walls.All solution was combined and added 6 mol/L HCl until the pH value was less than 2.0.Precise amount of 209 Po and Fe 3+ carrier (FeCl 3 ) were then added under stirring to quantify the yield recovery of 210 Po and enrich Po and Pb isotopes.The acidified solution was settled for 24 h, and its pH value was adjusted to 8.5 with ammonia solution.Po and Pb were co-precipitated with iron hydroxides/oxides (Yang et al., 2012).The precipitate was collected through centrifugation and dried, and finally it was placed in a counting vial for gamma spectroscopy analysis.Activities of 210 Pb were determined by non-destructive gamma counting using an Ortec ultra-high purity germanium detector.The counting efficiency of 210 Pb was calibrated by standard material.The net counts for 210 Pb were higher than 600, resulting in the counting errors of less than ±4%.
After measurements of 210 Pb, the precipitate was digested with mixed HClO 4 -HNO 3 -HF acids, and the residual was re-dissolved in 6 mol/L HCl solution.Ascorbic acid was added to complex iron (Fe 3+ ) until the solution showed no color.Then, 2 mL of 20% hydroxylamine hydrochloride and 2 mL of 25% sodium citrate were used to reduce other elements with oxidative valances.Finally, the pH of the solution containing polonium isotopes (i.e., 209 Po and 210 Po) was adjusted to 1.5.Po isotopes deposited onto a silver plate at 90C for 4 h (Yang et al., 2012(Yang et al., , 2013)).The silver plate was rinsed with Milli-Q water and dried.The activities of 209 Po and 210 Po were counted by an alpha spectrometer (EG&G ORTEC).The net counts of both 209 Po and 210 Po were more than 400, leading to a statistical counting error of less than ±5%.The specific activities of 210 Po were calculated based on the recovery of 209 Po.Decay correction was conducted to the mid-point of sampling for all data.

Atmospheric Depositional Fluxes of 210 Pb and 210 Po
The daily atmospheric depositional fluxes of 210 Pb varied from 0.039 to 2.253 Bq/m 2 /d with an average of 0.508 ± 0.049 Bq/m 2 /d (Table 1, Fig. 2), which was comparable to the average flux of 0.51 Bq/m 2 /d between 2001 and 2002 (Jia et al., 2003).In contrast, the depositional fluxes of 210 Po, ranging from 2.6 to 150.1 mBq/m 2 /d with the mean of 38.5 ± 2.1 mBq/m 2 /d, was an order of magnitude lower than 210 Pb.The flux ratios of 210 Po to 210 Pb varied from 0.015 to 0.223 (Table 1).The integrated monthly depositional fluxes of 210 Pb showed a range of 2.42-29.31Bq/m 2 /mon with an average of 10.45 ± 0.68 Bq/m 2 /mon (Table 2).The monthly fluxes of 210 Po varied from 0.160 to 3.388 Bq/m 2 /mon with the mean of 0.89 ± 0.04 Bq/m 2 /mon.The monthly vitiations of 210 Pb and 210 Po showed similar annual pattern to 2004 (Fig. 3, Yi et al., 2007).In general, higher fluxes were observed during the southwest monsoon prevailing period, while low values corresponded to the northeast monsoon months.
Based on the monthly depositional fluxes, the annual fluxes of 210 Pb and 210 Po were calculated to be 140.5 Bq/m 2 /yr and 11.8 Bq/m 2 /yr, respectively.By comparison, the annual depositonal flux of 210 Pb was lower than the global mean value of 195 ± 11 Bq/m 2 /yr between 20°N and 30°N (Baskaran, 2011).

Specific Activities of 210 Pb and 210 Po
The specific activities of 210 Pb in rainwater varied from 0.031 to 2.292 Bq/L, with a mean of 0.375 ± 0.053 Bq/L, which was close to the result of 0.31 Bq/L in Hakodate (Tokieda et al., 1996), but higher than those in Huelva (0.23 Bq/L, Lozano et al., 2011), Islamabad (0.123 Bq/L, Ali et al., 2011) and Murree (0.035 Bq/L, Ali et al., 2011).For 210 Po, its specific activity had a range of 1.4-410.2mBq/L, with an average of 34.1 ± 1.5 mBq/L which was twice as high as the result in Detroit (15.2 mBq/L, McNeary and Baskaran, 2007).

Local Effects versus Latitudinal Effects on the Atmospheric Deposition of 210 Pb and 210 Po
On a global-scale, it is generally accepted that the midlatitude regions have higher atmospheric deposition fluxes of 210 Pb compared to the Equatorial and Polar Regions (Schell, 1977) due to the coverage of water and snow in low and high latitudes, which results in less land exposure to surface air and successive less 222 Rn exhalation. 210Po, a granddaughter of 210 Pb, is also speculated to have a similar spatial pattern (Yang, 2005).Our sampling site is located in the low to mid-Northern Hemisphere, where usually is characterized by relatively low atmospheric deposition of 210 Pb (Schell, 1977;Turekian et al., 1977).However, the fluxes of 210 Pb in Xiamen were two-to threefold higher than some mid-latitude areas, such as Odawa Bay, Izmir and Huelva (Tateda et al., 2008;Lozano et al., 2011;Uğur et al., 2011).In addition, they were close to some mid-Northern regions, i.e., Galveston, Tsuyazaki, Norfolk and  Stillpond (Todd et al., 1989;Baskaran et al., 1993;Kim et al., 2000a;Tateda et al., 2008) although higher values were also observed at many mid-latitudes.Similar scenario also happened to 210 Po.Hence, the depositional fluxes of 210 Pb and 210 Po in a specific region do not strictly follow the latitudinal deposition pattern.Local effects appear to regulate the specific 210 Pb depositional flux to a large degree.
Local effects might include several dominating factors.First, land usually benefits the exhalation of 222 Rn into surface air and successive high 210 Pb compared to oceans (Lee et al., 2004), consequently resulting in high 210 Pb deposition on land.For instance, due to the surrounding oceanic settings, Peng-Chia Yu, an island close to Taiwan, has much lower atmospheric 210 Pb deposition fluxes than the almost same latitude site of Nankang (a suburb district of Taipei, Taiwan) (Su et al., 2003).Second, local meteorological conditions often generate specific deposition patterns of 210 Pb.For example, high temperature increasing exhalation of 222 Rn from soil and successive 210 Pb in surface air, together with plenty of precipitation from May to September, result in high 210 Pb deposition during summer in Xiamen.While high-pressure system and vertical mixing of air masses establish a high winter-spring deposition pattern in Chinese inland city of Guiyang (Wan et al., 2008).Last, anthropogenic activities sometimes significantly change local 210 Pb deposition.As an example, the abnormal high flux of 210 Pb in Shanghai was ascribed to storm dust, population increase and coal ash (Du et al., 2008).Therefore, the traditional model results actually represent the latitudinal depositional fluxes of 210 Pb over a global-scale.To investigate some local atmospheric and oceanographic processes, specific local fluxes of 210  Pb and 210 Po could be much more suitable than the global model simulation.

Factors Controlling the Atmospheric Fluxes of 210 Pb and 210 Pb in Xiamen
Although the sources of 210 Pb and 210 Po in surface air are fairly different, significant linear correlations between the daily depositional flux of two nuclides were observed (Fig. 4), revealing their similar removal mechanisms from the atmosphere.Overall, their daily fluxes follow the fluctuation of precipitation (Fig. 2).Thus, it is possible that precipitation is a crucial factor in controlling the daily deposition of 210 Pb and 210 Po.Indeed, statistical analysis indicated that there were significant positive relations between the fluxes and precipitation for 210 Po (y = 17.1 + 4.7x, R 2 = 0.59, P < 0.0001) and 210 Pb (y = 0.22 + 0.57x, R 2 = 0.63, P < 0.0001) respectively, supporting a precipitation-dominating mechanism.In addition, the nine complete dry deposition samples showed that the average dry depositional fluxes of 210 Pb and 210 Po were 0.11 ± 0.01 Bq/m 2 /d and 11.12 ± 0.53 mBq/m 2 /d respectively, which accounted for 22 ± 3% and 29 ± 2% of their corresponding bulk deposition (Table 1).Little difference in the contributions of dry deposition also supported that 210 Po and 210 Pb have similar removal efficiencies via dry deposition in surface air.Overall, precipitation is the main removal passage of 210 Po and 210 Pb from the atmosphere in Xiamen in spite of the fact that the contribution of dry deposition is innegligible.Similar conclusions have also been reached in northcentral Wisconsin (Lamborg et al., 2000) and southern Mississippi (Yang et al., 2012).However, the percentage of dry deposition in the bulk fluxes of 210 Pb and 210 Po in Xiamen was much higher than the global mean of 13-14% calculated from a three-dimensional aerosol model (Feichter et al., 1991;Balkanski et al., 1993).This deviation was attributed to the elevated suspended matter in surface air of Xiamen caused by the increased motor vehicles and rapid development of construction industry in the past years (Hong et al., 2007;Fan et al., 2009).
Although the 210 Pb deposition was predominantly controlled by the rainfall, its dependence on the precipitation showed seasonal variability (Fig. 5).Monsoon is the most important factor in adjusting the meteorological conditions in Xiamen.In the southwest monsoon prevailing seasons, the slope of the correlation between the depositional flux of 210   0.78, P < 0.0001), which is much larger than that in the northeast monsoon seasons (y = 0.23 + 0.03x, R 2 = 0.36, P < 0.02).In other words, the same amount of rainfall resulted in more 210 Pb deposition in southwest monsoon seasons than that in the northeast monsoon seasons.Because similar amount of rainwater has almost the same removal efficiency of 210 Pb (Fig. 6), the evident discrepancy between different monsoon seasons actually reflected the significant difference in atmospheric 210 Pb contents.The southwest monsoon usually corresponds to hot seasons.On the one hand, high temperature increases the emanation velocity of 222 Rn from soil and thus high atmospheric 222 Rn concentrations (Iskandar et al., 2004).On the other hand, southwest monsoon carries hot air masses which stay over the low latitudinal Asia continent and accumulate amount of 222 Rn though the seasonal variation of 222 Rn flux in not evident (Szegvary et al., 2009).Consequently, southwest monsoon leads to high 210 Pb contents in the atmosphere.In contrast, the 210 Pb concentrations in the northeast monsoon prevailing seasons are low.Air masses from the Pacific Ocean, usually containing low 222 Rn, may responsible for the low 210 Pb during the northeast monsoon.However, such mechanisms have little influence on the depositional fluxes of 210 Po (Fig. 5).The slopes of the correlations between the 210 Po flux and precipitation are almost the same if considering the errors.The different source term of 210 Po is responsible for this phenomenon.Because 210 Po is from the disintegration of 210 Pb, which is independent of temperature but dependent upon the residence time of 210 Pb in the atmosphere, as a result, the 210 Po concentrations and finally depositional fluxes are insensitive to temperature and monsoons.
Notably, all the specific activities of 210 Pb and 210 Po showed significant inverse relations with the amount of precipitation (Fig. 6).At the beginning stage of rain, corresponding to less than 10 mm of precipitation, these nuclides usually have much higher activities in per liter rainwater.However, their specific activities decreased sharply with the increasing of precipitation and finally varied little when the rainwater was more than 20 mm.Similar "dilution effect" was also observed in Detroit (McNeary andBaskaran, 2003, 2007).Therefore, the removal efficiencies of 210 Pb and 210 Po at the beginning of rainfall were much higher, probably owing to their initial higher contents in surface air.According to the above discussion, it is obvious that depositional fluxes of 210  Pb and 210 Po in a specific location, over seasonal-to annual-timescales, are controlled by its local meteorological conditions such as monsoon and precipitation at all.However, on an intra-seasonal timescale, precipitation might be the deciding factor.

Residence Times of Atmospheric Aerosol in Xiamen
The residence times of atmospheric aerosol not only help us establish atmospheric cycling models (Kristiansen et al., 2012) but also quantify the diffusion and transport of contaminants in the atmosphere (Yang et al., 2012;Huh et al., 2013).For example, the residence times of aerosol constrained by 131 I and 210 Po/ 210 Pb approaches revealed that 131 I released from the Fukushima nuclear accident mainly affect the regions accessible within two weeks' transport (Yang et al., 2012).With the assumption that the atmospheric 210 Po completely comes from its grandparent of 210 Pb, the disequilibria between 210 Po and 210 Pb could record the residence time of 210  Pb and 210 Po in the atmosphere.Based on the correlation between the daily fluxes of 210 Po and 210 Pb (Fig. 4), the daily 210 Po flux was close to zero when the 210 Pb flux was zero, indicating that atmospheric 210 Po is Fig. 6.Relationships of the specific activities of 210 The residence time (τ) is the reciprocal of k.We used the flux ratios of 210 Po/ 210 Pb instead of activity ratios to calculate the residence time of atmospheric aerosol (Baskaran, 2011;Yang et al., 2012) as 210 Po and 210 Pb were removed from the atmosphere through the same passages (Fig. 2).
The residence times of aerosol varied from 3.1 d to 57.5 d, with an average of 20.8 ± 3.0 d (Table 1), indicating the possibility of inter-continent transport for the atmospheric contaminants in Xiamen.Previous reports showed the residence times of aerosol ranged between 1 d and 32.9 d for the 210 Bi/ 210 Pb approach (Fry et al., 1962;Poet et al., 1972;Lambert et al., 1983;El-Hussein et al., 1998;Ahmed et al., 2000;Papastefanou, 2009) and between 1 d and 135 d for 210 Po/ 210 Pb method (Poet et al., 1972;Gavini et al., 1974;Tokieda et al., 1996;Marley et al., 2000;Baskaran and Shaw, 2001;McNeary and Baskaran, 2007), respectively.This discrepancy was often attributed to the extra sources of atmospheric 210 Po besides 210 Pb disintegration, such as combustion (Nho et al., 1996), volcanic emissions (Su and Huh, 2002) and oceanic contribution (Bacon et al., 1980;Kim et al., 2000b).However, it should be noted that the largest residence time calculated by the 210 Bi/ 210 Pb ratios was close to the upper timescale limit of this approach, because the half-life of 210 Bi is 5.01 d the 210 Bi/ 210 Pb ratios theoretically only track processes within 35 d.Accordingly, the 210 Bi/ 210 Pb approach might only trace the residence times of partial aerosol, residing less than 35 d in the atmosphere.For aerosols with much longer residence times, the 210 Bi/ 210 Pb technique is invalid.In contrast, the range of the residence times constrained by the 210 Po/ 210 Pb ratios covered the 210 Bi/ 210 Pb determined scope.Additionally, it also included aerosol excluded by 210 Bi/ 210 Pb ratios, which was supported by the residence times of >35 d from the 210 Po/ 210 Pb ratios.At the same time, the largest value determined by 210 Po/ 210 Pb technique was much less than the timescale (2 yr) of this approach, revealing that the residence times quantified by the 210 Po/ 210 Pb ratios represented the residence times of the bulk aerosol.Hence, the combination of 210 Bi/ 210 Pb with 210 Po/ 210 Pb techniques might provide more insight into the inconsistent residence times of sizefractionated aerosols.CONCLUSIONS 210 Pb and 210 Po in atmospheric depositional samples, collected from November 2010 to January 2012, were determined to examine the temporal variability of their deposition in Xiamen.The following results and conclusions can be reached.
(1) The annual depositional fluxes of 210 Pb and 210 Po were 140.5 Bq/m 2 /yr and 11.8 Bq/m 2 /yr, respectively. 210Pb and 210 Po showed similar monthly variations of deposition, characterized by high values in the southwest monsoon prevailing seasons while low values during the northeast monsoon seasons.
(2) Precipitation plays a dominant role in the removal of 210 Pb and 210 Po from the atmosphere although the contribution of dry deposition cannot be neglected in Xiamen.The highest removal efficiencies for these radionuclides occurred at the beginning of rainfall and then descended sharply to a nearly constant.
(3) The residence times of aerosols in Xiamen varied from 3.1 d to 57.5 d with an average of 20.8 ± 3.0 d. 210 Po/ 210 Pb ratios in size-fractionated aerosols could provide more insight into the residence of aerosols in the atmosphere.
(4) For a specific site, local effects, including geographical and meteorological conditions, determine the depositional fluxes of 210  Pb and 210 Po rather than latitudinal effects.Thus, the local-specific fluxes are appropriate for site-concerned process investigations, while the global-scale model simulated results would meet the requirements of regional-to globalscale concerned researches.

Fig. 1 .
Fig. 1.Schematic map of time-series station for atmospheric deposition investigation in Xiamen.

Fig. 4 .
Fig. 4. Correlation of the daily depositional fluxes between 210 Po and 210 Pb in Xiamen in 2011.

Table 1 .
Daily depositional fluxes, specific activities of 210 Po,210Pb and the residence times of aerosol in Xiamen, from November 5, 2010 to January 19, 2012.

Table 2 .
Monthly depositional fluxes of210 Pb, 210Po and rainfall in Xiamen from November 2010 to December 2011.
(Baskaran and Shaw, 2001 of rainfall in Xiamen.Pb in surface air in Xiamen.Thus, the 210 Po/ 210 Pb disequilibria model(Baskaran and Shaw, 2001) was adopted to calculate the residence times of aerosol.Usually, the variations in the specific activities of210Po with time can be expressed as Po and A Pb are the activities of 210 Po and 210 Pb (dpm), respectively; λ is the decay constant of 210 Po (0.005 d -1 ); k denotes the removal rate constant of 210 Po from the atmosphere (d -1 ).With a steady-state, k can be estimated by