Beryllium-7 Deposition and Its Relation to Sulfate ... - Springer Link

Journal of Atmospheric Chemistry 29: 217–231, 1998. c 1998 Kluwer Academic Publishers. Printed in the Netherlands.

217

Beryllium-7 Deposition and Its Relation to Sulfate Deposition YASUHITO IGARASHI, KATSUMI HIROSE and MAKIKO OTSUJI-HATORI Geochemical Research Department, Meteorological Research Institute, 1-1 Nagamine, Tsukuba, Ibaraki, 305 Japan (Received: 24 June 1996; in final form: 2 December 1997) Abstract. Deposition of 7 Be, a cosmogenic radionuclide, was observed at the Meteorological Research Institute in Tsukuba, Japan from 1986 to 1993 and compared with those of several chemical species observed in Tsukuba over the same period. We found a correlation between the monthly depositions of 7 Be and SO24, , a major acidic species. The correlation was especially strong for late spring and fall, when both species had high depositional fluxes. This correlation was also observed in precipitation samples collected daily in 1992 at the same site. The cause of this correlation is discussed in connection with the fact that the stratospheric aerosol is composed largely of SO24, . 7 Be is produced in the upper atmosphere, and detection of 7 Be, especially in spring and fall in Japan, can be regarded as detection of stratospheric aerosol. However, we conclude that the bulk of the SO24, observed did not have a stratospheric or an upper tropospheric origin. The correlation, therefore, may present a new question regarding acidic deposition: Why does the deposition of stratospheric aerosol in Japan coincide with that of nss-SO24, originally from anthropogenic sources on the Earth’s surface? Key words: 7 Be deposition, SO24, deposition, correlation, stratospheric aerosol.

1. Introduction One of the naturally occurring cosmogenic radionuclides, 7 Be, which is produced by nuclear spallation reactions between high energy cosmic-rays and atmospheric nuclei, has been used for various purposes in the field of geoscience (see for example: Brost et al., 1991; Dibb et al., 1994; Doi et al., 1993; Dutkiewicz and Husain, 1979; Harvey and Matthews, 1989; Ishikawa et al., 1995; Kritz et al., 1991; Raisbeck et al., 1981; Rangarajan and Eapen, 1990; Turekian and Tanaka, 1992; Young and Silker, 1980). The production rate of cosmogenic radionuclides increases with altitude and reaches a maximum at 12–15 km in the mid latitudes, and about 2/3 of the production of cosmogenic radionuclides occurs in the stratosphere (Lal, 1963; Lal and Suess, 1968). 7 Be (T1=2 54d) is easier to measure than other cosmogenic radionuclides such as 14 C, 32 P, 35 S, etc., because it emits -rays with -decay (electron capture) and is produced at a higher rate. Due to its electric charge, 7 Be probably attaches to the ambient aerosol soon after its production. The 7 Be-bearing aerosol is then transported downward through meteorological dynamics (stratosphere-troposphere air mass exchange) and deposit-

218

YASUHITO IGARASHI, KATSUMI HIROSE AND MAKIKO OTSUJI-HATORI

ed on the ground, mainly by wet precipitation. Thus, 7 Be is a useful tool by which to learn about atmospheric mixing and aerosol scavenging processes. The source, budget and distribution of 7 Be has been given in detail by Brost et al. (1991), in which 7 Be was used to estimate wet scavenging parameter in their global climate simulation model. We have been measuring monthly depositions of 7 Be, along with several artificial radionuclides, in our radioactivity monitoring project (Igarashi et al., 1996). As well as radioactivity, we are interested in the issue of acidic deposition and have been measuring the ionic species in precipitation. Many investigations, including a nationwide survey by the Japan Environment Agency (e.g. Hara et al., 1990), have shown that acidic deposition is a problem in Japan. It has been suggested that acidic pollutants emitted from the Asian continent are transported long distances to Japan, especially to the coastal areas along the Sea of Japan (Fujita, 1996; Kitamura et al., 1993; Ohizumi et al., 1997). The rapid increase in human activity on the Asian continent seems a potential menace to the global environment in terms of ecological impact as well as climate change. For instance, the Asian continent is going to become the largest source area of SO2 (over 10 Tg S in 1987) in the Northern Hemisphere (Akimoto and Narita, 1994). This figure is not small, because the annual global emission of gaseous sulfur to the atmosphere, including both natural and anthropogenic sources, is thought to be around 100 Tg S (Langner and Rohde, 1991). For this reason, several atmospheric research projects, such as the NASA Pacific Exploratory Mission (Hoell et al., 1996), have been conducted. We compared temporal variations in 7 Be deposition with variations in the deposition of acidic pollutants. We noticed a correlation between 7 Be and SO24, depositions which was especially high in late spring and in fall, when there were relatively high deposition rates for both species. In this report, we describe the correlation and the implications of this finding. 2. Experimental 2.1.

7 BE IN MONTHLY DEPOSITION SAMPLES

Depositions of nuclides by both wet and dry processes were collected monthly by using plastic samplers (two different sizes: 2 or 4 m2 ) at the observation field of the Meteorological Research Institute (MRI) (lat. 36 030 N, long. 140 080 E). Samples were evaporated to dryness, and the residues obtained subjected to ray measurement using a Ge semiconductor detector coupled to a 4096 channel multi-channel analyzer. The 478 keV photo-peak was used to determine 7 Be. Since this -ray measurement is also used to determine 137 Cs at extremely low level, the measurement time is more than one week, which allows us to obtain sufficient counts to determine 7 Be. The counting error was less than 1% RSD. The radioactivity was decay-corrected to the last day of the month. The systematic

BERYLLIUM-7 DEPOSITION AND ITS RELATION TO SULFATE DEPOSITION

219

errors involved are estimated to be around 15% for 1986–1990 and less than 10% for 1991–1993. 2.2.

7 BE IN DAILY WET DEPOSITION SAMPLES

Rainwater samples were collected daily at MRI in 1992 using a wet only sampler. Insoluble matter was filtered off using a membrane filter (Millipore HA type, pore size 0.45 m). A small aliquot of the filtrate was taken to determine ionic species and the rest used for 7 Be determination. Since less than 5% of the 7 Be remains on the filter, only the filtrate was used. Samples, with an addition of HCl, were evaporated to dryness in a porcelain dish. The residue was then dissolved in a small amount of dilute HCl and transferred to a plastic vessel. The solution was dried over moderate heat. The resulting sample was then subjected to -ray spectroscopy in the same manner as described above except that the measurement time was shorter. The accompanying counting errors for these samples were around 10%. 2.3. IONIC SPECIES IN THE DAILY WET DEPOSITION SAMPLE Details of the measurement for ionic components in the daily collected samples + 2+ 2+ , are given elsewhere. Ions measured include Na+ , NH+ 4 , K , Ca , Mg , Cl , , 2, + NO3 and SO4 . pH was measured to obtain H concentrations. The data set for statistical analysis was chosen by the ion balance test. When the analysis involves little problem, the sum of anions over the sum of cations (the ion balance) will lie around unity. The data set having the ion balance between 0.7 and 1.3 were used. These data were also used to determine the monthly deposition of ions in 1991, 1992 and 1993. The precipitation records of the Aerological Observatory (Tateno), on the same campus, were used to construct the monthly deposition rates. 3. Results and Discussion 3.1. FEATURES AND CONTROLLING FACTORS OF SEASONAL CHANGES IN MONTHLY DEPOSITION AND ATMOSPHERIC CONCENTRATION OF 7 BE IN JAPAN

Monthly deposition of 7 Be during 1986–1993 are given in Table I. The main point to be addressed is the seasonal cycle in 7 Be deposition (see also Figure 1). The general seasonal pattern shows depositional peaks in spring and fall. Since 7 Be is in aerosol form, as will be discussed later, the concentration of 7 Be in rainwater is determined by its atmospheric concentration and the aerosol scavenging efficiency of rain. Since the cosmic-ray flux does not vary seasonally (Brost et al., 1991; Abe et al., 1993), the seasonal variation of 7 Be air concentration seems to arise from differences in air mass. In the mid-latitudes (lat. 30 –40 ) of the Northern Hemisphere, the concentration of 7 Be in air at ground level is maximal in spring and fall. This bi-modal seasonal trend in Japan (Abe et al., 1993; Doi et al., 1993;

220


Table I. Monthly deposition of 7 Be observed at MRI, Tsukuba during 1986–1993 (decay corrected to the last day of the month) Year

Month

7

1986

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.

33 114 151 214 318 224 131 56 189 119 86 145 83 86 152 85 204 132 81 82 171 147 134 48 66 39 182 104 223 292 157 107 254 216 94 7 47 140 68 193 228 90 67 59 76 160 53 35

1987

1988

1989

Be depo (Bq/m2 )

Year

Month

7

1990

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.

20 46 96 108 95 55 50 33 112 99 114 48 36 27 153 85 104 75 63 61 116 167 62 36 63 10 151 212 210 177 67 12 68 225 95 34 98 67 104 90 158 187 164 66 65 121 66 65

1991

1992

1993

Be depo (Bq/m2 )


221

Figure 1. Seasonal variation in 7 Be and SO24, depositions observed in Tsukuba during 1986– 1993 ( SO24, data are from Okamoto et al., 1992).

Matsunami and Megumi, 1994; Sato et al., 1994) seems to be more pronounced than those observed at U.S. sites located in 30 –40 N (Feely et al., 1989). 7 Be transport from the stratosphere has been discussed in many studies (Doi et al., 1993; Dutkiewicz and Husain, 1979, 1985; Graustein and Turekian, 1996; Kritz et al., 1991; Reiter, 1978; Vizee and Singh, 1980, etc.). Based on the concentration change of ozone and artificial radionuclides and the gradient of the potential vorticity, Danielsen et al. (1970) argued that tropopause folding associated with the jet stream plays an important role in the stratosphere-troposphere air mass exchange. The descending stratospheric airflow caused by tropopause folding is schematically shown in Danielsen (1980). The intruded stratospheric air, traceable by 7 Be, seems to fan out behind the high-pressure system at ground level (Wolff et al., 1979). Warneck (1988) has summarized the individual processes of air exchange between the stratosphere and the troposphere, seasonal relocation of the tropopause, largescale meridional circulation, tropopause folding, and small-scale eddy diffusion. Among these, tropopause relocation and folding events seems to be the major cause of the seasonal variations in the atmospheric 7 Be concentrations in Japan. In summer, the tropopause gap (the position of the jet stream and the region where the folding events occur) is shifted northward by the overwhelming activity of the Pacific high-pressure system, while in winter the gap is relocated southwards. The tropopause gap passes over Japan (lat. 30 –40 N) in spring and fall. Thus, the active air mass exchange between the stratosphere and the troposphere probably causes the concentration maxima of 7 Be in these seasons. It is important to know whether it is stratospheric or upper tropospheric budget that mainly supports the depositional flux of 7 Be at ground level (Baskaran, 1995; Dutkiewicz and Husain, 1979, 1985; Turekian et al., 1983; Reiter, 1978). Although

222


we cannot describe explicitly how much portion of the 7 Be observed at MRI originated from the stratosphere, we think, in Japan, that contribution of stratospheric 7 Be to the maximal 7 Be deposition, in spring and fall, is fairly large. Work by Abe et al. (1993) may provide information in this regard. They investigated the correlation between the change in the annual average of atmospheric 7 Be concentrations at Chiba, Japan (lat. 35 380 N, long. 140 060 E) and that of cosmic-ray intensity at 3 stations in the Northern Hemisphere. Note that the annual average 7 Be concentrations at Chiba was relatively depending on 7 Be concentration levels observed in spring and fall. The cosmic-ray data were obtained at Norikura (lat. 36 110 N, long. 137 550 E) and Tokyo (lat. 35 750 N, long. 139 720 E), Japan and Kiel, Germany (lat. 54 340 N, long. 10 120 E). The best correlation was found between the cosmic-ray flux in Kiel and the atmospheric concentration of 7 Be in Chiba. As the concentration of 7 Be in air observed is determined by its production, transport and removal, these processes should be taken into consideration. If the majority of 7 Be observed on the surface at Chiba were produced in the troposphere above Chiba, its surface concentration would correlate with the cosmic-ray intensities at Norikura and Tokyo. Therefore, 7 Be produced at high latitude and altitude seems to be transported through large scale meridional motion of the atmosphere to Japan. In short, the air mass that contains the 7 Be-bearing aerosol subsides from the upper atmosphere in spring and fall over Japan. Rain events occur in spring and fall in Japan due to a low-pressure-front system, associated with tropopause folding in the upper atmosphere (Browning et al., 1995; Ogawa, 1991; Ogura, 1984; Young et al., 1987). We think the combination of 7 Be transport from the stratosphere and precipitation (an effective removal process for the aerosol) causes the greater deposition of 7 Be in the months of spring and fall.

3.2. CORRELATION BETWEEN 7 BE AND IONIC SPECIES DEPOSITION We compared the monthly depositions of 7 Be and other ionic species to investigate aerosol scavenging mechanisms. The seasonal deposition patterns of 7 Be and SO24, were well correlated especially in 1992 and 1993. The summer of 1993, as anomalously cold and wet, so that the spring SO24, deposition peak shifted towards summer. Interestingly, this shift coincided with that of 7 Be deposition. To confirm the correlation between 7 Be and SO24, , the comparison was extended back to 1986 with the National Institute of Agro-Environmental Sciences (Tsukuba) –SO24, deposition data (for 1986 to 1990) collected by the bulk sampling method (Okamoto et al., 1992). In Figure 1, deposition fluxes of both species are normalized to each year’s annual average in order to focus on the seasonal variations in deposition. It is clear that, with few exceptions, peaks of 7 Be in spring and fall coincide with those of total SO24, . Note that the about 90% of the annual SO24, deposition is comprised of non sea salt fraction (nss-SO24, ), indicating, therefore, that it is the depositions of 7 Be and nss-SO24, that are correlated.


223

Table II. Linear correlation coefficients and slopes of the regression line between normalized monthly 7 Be and SO24, deposition Month

Correlation coefficient

Slope

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.

0.565 0.887 0.616 0.794 0.666 0.921 0.752 0.443 0.882 0.394 0.874 0.490

0.829 1.072 0.422 0.647 0.492 0.931 0.657 0.757 1.003 0.457 0.647 0.449

Table II shows that 7 Be and nss-SO24, were highly correlated in February, June, September and November. In June and September, relatively high depositions of the both species were recorded. Note that for June and September, not only were the correlation slopes steep, but also the correlation coefficients were high. This correlation between 7 Be and SO24, was further established in the rainwater samples collected daily in 1992. Table III gives the correlation coefficients between each of the major ionic components and 7 Be. Because their frequency distributions are lognormal, we calculated Spearman correlation coefficients. The correlation coefficients of 7 Be and SO24, was the highest among that of 7 Be and ionic species, and was far greater than the coefficient at 99% significance level (0.267 at 90 degrees of freedom). However, what produces this correlation between the depositions of 7 Be and SO24, ? The coefficient was not as high as that between SO24, and NH+ 4 , which is 2, attributable to the chemistry of these species – a portion of SO4 in the aerosol is neutralized by NH+ 4 . Since beryllium is an alkaline earth element, a similar analogy is possible. However, the number of 7 Be atoms in the air is extremely small compared to that of the SO24, . If a chemical process (neutralization, etc.) were the cause of high correlation between 7 Be and SO24, , the coefficient should be as high as that of SO24, and NH+ 4 . This not being so, a physical process seems more likely and should be considered. 3.3.

7 BE-BEARING SULFATE AEROSOL

Since 7 Be originates from the upper atmosphere, the correlation between 7 Be and SO24, reminds us of the fact that the stratospheric aerosol is largely composed of

224


Table III. Matrix of Spearman correlation coefficients for the ions and 7 Be in the 1992 precipitation samples (February to December, n 97)

=

H+ Na+ NH+ 4 K+ Ca2+ Mg2+ Cl, NO, 3 SO24, 7 Be

H+

Na+

NH+ 4

K+

Ca2+

Mg2+

Cl,

NO, 3

SO24,

1 0.515 0.799 0.705 0.577 0.603 0.649 0.868 0.870 0.806

1 0.722 0.864 0.774 0.951 0.935 0.592 0.737 0.705

1 0.837 0.803 0.800 0.828 0.874 0.944 0.792

1 0.824 0.910 0.914 0.762 0.887 0.805

1 0.834 0.806 0.763 0.799 0.660

1 0.956 0.678 0.820 0.774

1 0.686 0.836 0.800

1 0.893 0.778

1 0.839

7

Be

1

sulfate or sulfuric acid (e.g. Junge et al., 1961; Junge, 1963; Lazrus and Gandrud, 1977). Turco et al. (1982) reviewed observational and theoretical knowledge of the stratospheric aerosol in detail. The source of the aerosol sulfate is thought to be organic gaseous sulfur (Crutzen, 1976), an idea that has wide acceptance (Warneck, 1988). When a large scale volcanic eruption occurs, stratospheric sulfate is known to increase (Turco et al., 1982), which is discussed later. 7 Be has an electric charge and high kinetic energy just after production by spallation. This charged species, a ‘hot atom’, will be neutralized by attachment to ambient aerosol, stratospheric sulfate. This leads to the hypothesis that sulfate is the main carrier of 7 Be in the stratosphere and this may also hold in the upper troposphere. An analogy of this ‘’radio-labeling’ process can be seen in the air surrounding high-energy accelerators. Kondo et al. (1984) collected the radioactive aerosol in the air around a beam of high-energy particles from an accelerator and investigated the size distribution of 7 Be formed by the spallation reaction. They found that the size distribution of 7 Be may depend upon that of the sulfate aerosol which is formed through radiation-induced oxidation of SO2 in the ambient air ( and -radiations produce oxidants.). Another indication that sulfate aerosol is the carrier of cosmogenic Be isotopes comes from Delmas’ (1992) work with polar ice cores. In cores from pre-industrial periods free from volcanic sulfate, Delmas found a linear relationship between nssSO24, and another cosmogenic Be isotope, 10 Be (T1=2 : 1:5 106 y). The nss-SO24, and the 10 Be contents in the ice core samples collected at the 5 different points from the Arctic and Antarctic regions surprisingly fell on the same regression curve. This means that over sufficiently long time intervals the ratio of 10 Be to sulfate remained constant in the upper atmosphere of both polar regions. The ‘labeling’ process apparently occurs in the upper atmosphere. These two facts indicate that 7 Be-bearing sulfate aerosol comes from the upper atmosphere.


225

Table IV. The result of principle component analysis for the samples from spring 1992 (April to June, n 37)

=

Species

Component loadings 1st 2nd 3rd

4th

H+ Na+ NH4 + K+ Ca2+ Mg2+ Cl, NO3 , SO4 2, 7 Be

0.620 0.914 0.964 0.954 0.803 0.927 0.958 0.829 0.965 0.823

–0.664 0.285 –0.070 0.159 0.369 0.289 0.106 –0.260 –0.030 –0.430

–0.283 –0.204 0.148 –0.061 0.249 –0.200 –0.191 0.433 0.072 0.020

–0.297 0.084 0.079 0.023 –0.348 0.008 –0.006 –0.007 0.123 0.211

Eigenvalues Variance explained (%)

7.775 77.75

1.037 10.37

0.478 4.78

0.283 2.83

Direct proof of 7 Be-bearing sulfate aerosol may be given by the statistical analysis of the daily wet deposition data of 7 Be and ionic species in the spring of 1992. Table IV gives the result of a principal component analysis for the 37 precipitation events collected daily, from April to June. The fourth component seems to be composed of just 7 Be and sulfate. In this connection, the first and second components are composed of a mixture of all species measured and species from sea salt, respectively. The third one is fairly complicated; a mixture of calcium nitrate and ammonium nitrate. Finally, although we must avoid the oversimplification, we shall argue that stratospheric aerosols are ‘labeled’ by 7 Be. In other words, detection of 7 Be can be regarded as the detection of the stratospheric aerosol. Therefore, the correlation between 7 Be and nss-SO24, depositions indicates that the deposition of nss-SO24, is associated with the stratospheric aerosol.

3.4. IS MOST SULFATE DERIVED FROM THE UPPER ATMOSPHERE? The simplest interpretation of the correlation between monthly 7 Be and SO24, depositions is that both are from the same source; that most of the SO24, originates from the stratosphere and upper troposphere. However, this hypothesis is not rational, because the global stratospheric sulfate budget is estimated to be only 0.16 Tg S (Turco et al., 1982). Furthermore, Delmas, in the above-mentioned work (1992), estimated that the global sulfate budget in the upper troposphere was 3 Tg S during volcanically quiescent pre-industrial periods. In order to confirm that most sulfate does not come from the upper atmosphere, the following calculations were

226


conducted. According to Brown et al. (1989), the average ratio of cosmogenic Be isotopes in rainwater is

[10 Be]=[7 Be] = 1:5 104 atom cm,3=1:2 104 atom cm,3:

(1)

On the other hand, the relation between concentrations of 10 Be (in atom/g) and nss-SO24, (in g/g) in the polar ice is as follows (Delmas, 1992);

[10 Be] = 0:079 1013 [nss , SO24,] , 0:807 104:

(2)

The annual deposition (simply, the sum of monthly depositions) of 7 Be at MRI in 1992 was 1324 Bq/m2 ; 8:7 109 atoms. Incorporation of the number into the equations gives deposition of the nss-SO24, at MRI from the upper atmosphere, 0.014 g/m2 . This is less than 1% of the nss-SO24, annual deposition observed at MRI (about 2 g/m2 ). Golombek et al. (1993) modeled the stratospheric sulfate layer using a global three-dimensional model. In their model, 12.3 Tg of sulfur is deposited annually on the ground from the free troposphere. This estimate is only 4 times larger than that of Delmas (1992), thus confirming that most SO24, is not derived from the stratosphere or upper troposphere. 3.5. DOES VOLCANIC ACTIVITY AFFECT SULFATE DEPOSITION IN TSUKUBA? The eruption of Mt. Pinatubo recently caused an increase of SO24,in the stratosphere. The total mass of SO2 injected into the stratosphere was estimated to be 20 Tg, 10 Tg S (Bluth et al., 1993; McCormick et al., 1995). This flux is comparable to the 1987 annual emission of SO2 from anthropogenic sources in the Far East region (China, North Korea, South Korea and Japan), 11.3 Tg S (Akimoto and Narita, 1994). Although this amount is not small, Pinatubo sulfate increased the budget in the upper atmosphere by only three-fold that which it would be in volcanically quiescent pre-industrial periods. This increase might have affected the SO24, deposition in Japan in 1991 and 1992, but not in 1993 as discussed below. Sato et al. (1994) noted an increase of 210 Pb deposition at Tsukuba during the winter months of 1991–1992 and, based on the variation in 210 Pb/7 Be activity ratio, concluded that this excess deposition was derived from the Mt. Pinatubo eruption. Presumably, the deposition of Pinatubo SO24, coincided with this excess 210 Pb deposition. Assuming that the Pinatubo sulfate (10 Tg S) was deposited evenly over the globe (510 1012 m2 ), this would have resulted in a small annual flux of approximately 0.02 g S/m2 (0.06 g SO24, /m2 , equivalent to less than 5% of the nss-SO24, annually deposited at MRI). Considering the latitudinal distribution of Pinatubo aerosol, even if the stratospheric fallout from the eruption affected chemicals in the surface air in 1991 and 1992, the extent should not have been so large as to explain most of the nss-SO24, deposition found at Tsukuba. The Pinatubo sulfate in the stratosphere decreased with a halfresidence time of approximately 1 y (Lambert et al., 1997), which is the same figure as previously reported for stratospheric artificial radioactivity (e.g. Igarashi


227

Figure 2. Annual trend in depositions of SO24, and nss-SO24, observed in Tsukuba from 1984 to 1993 (1984–1985 data are from Dokiya et al., 1987 and the 1986–1990 data are from Okamoto et al., 1992).

et al., 1996). Therefore, half of the Pinatubo sulfate would be removed from the stratosphere in 1991, meaning that the deposition of Pinatubo sulfate on the ground would be about 0.01 g S/m2 over the globe in 1991. In 1992 the deposition of the Pinatubo sulfate would be half of the initial deposition in 1991, and so on in the following years. In the 1980s, although there were some other volcanic eruptions which may have affected SO24, budget in the stratosphere, all of SO2 emissions were smaller than that of Pinatubo (Bluth et al., 1993). There was no large increase in annual SO24, deposition during 1984 to 1993 in Tsukuba city (Figure 2), suggesting little contribution of volcanic SO24, from the stratosphere to total SO24, deposition in Japan. 3.6. MIXING OF AEROSOL FROM THE STRATOSPHERE AND THE BOUNDARY LAYER Important controlling factors of the deposition of a given chemical, which we must consider, are source (including chemical transformation), transport and removal. As discussed above, it is clear that 7 Be and nss-SO24, have definitely different source regions. Given the pollution situation in the Far East region, the major part of nss-SO24, deposition must originate from surface anthropogenic SO2 sources. We need, then, to examine the transport and removal processes. The 7 Be ‘labeled’ SO24, aerosol may mix well and may coagulate with the ‘unlabeled’ SO24, aerosol produced within the boundary layer. As described in Section 3.1, the transport mechanism of 7 Be aerosol from the upper atmosphere relates to the existence of migratory cyclones on the ground. The mixing process, thus, must be linked to the air mass descent process.

228


We have found that there may be two modes in the annual acidic deposition on the Pacific side of Japan, and that these modes are relevant to the seasonal change of the air mass over the Japanese islands. In the 1st mode each ionic species, Na + , NH+ 4, , 2, + 2 + 2 + , K , Ca , Mg , Cl , NO3 and SO4 , are deposited. While in the second mode, , 2, three major species, NH+ 4 , NO3 and SO4 , are deposited. The 1st mode prevails in spring and fall and the second mode prevails only in summer. Therefore, the good correlations between 7 Be and SO24, deposition for spring and fall suggest that the 7 Be (one can regard it as the stratospheric aerosol) is deposited by the 1st mode of acidic deposition. In order to check this suggestion, the component analysis of the daily precipitation sample from February to December, 1992 ( n = 97) was further enlarged to include 7 Be. Loading factors for 7 Be in the 1st and 2nd component were 0.729 and 0.111, respectively. The first component showed high loading factors not only for 7 Be and SO24, (0.895) but also for other species including Ca2+ of soil origin (0.751). It seems that, in spring and fall, convection as well as horizontal advection is active due to the seasonal change in meteorology discussed here. nss-Ca2+ ions in the precipitation seem to originate from the Asian continent. Presumably, the deposition of 7 Be and nss-SO24, in spring and fall is controlled by coincidence of three important factors, stratosphere-troposphere air mass exchange, advection of polluted air from the continent, and precipitation by the cyclone-frontal system. 3.7. CLOUD PROCESSES The interpretation described in Section 3.6 is not completely clear, because it seems to suggest that, when it rains, it rains 7 Be (stratospheric aerosols) and nssSO24, . A global three-dimensional model study by Langner and Rodhe (1991) depicts well the tropospheric cycle of sulfur and the annual average deposition of a few sulfur species on the ground. Wet deposition of SO24, is thought to be the most efficient process for removal of sulfur from the troposphere. SO24, formation from gaseous sulfur is, therefore, the major pathway to deposition. They described the importance of the aqueous-phase SO2 oxidation, that occurs in liquid water clouds. Hobbs (1993) summarized important processes that occur in clouds, to produce the acidic wet precipitation. These include nucleation scavenging of aerosols by clouds, scavenging of aerosols by precipitation, and chemical reactions in the aqueous-phase. Thus, the correlation between cloud coverage (monthly mean cloudiness by noninstrumental observation) at Tateno and 7 Be and SO24, deposition was examined for the years of 1991, 1992 and 1993. In 1993, patterns of seasonal change in mean cloudiness and normalized 7 Be and SO24, depositions were similar, but in the other 2 years, a clear correlation was not found. Showing a similar correlation as described here, but between atmospheric concentrations of 7 Be and NO, 3 observed at Pacific islands sites, Uematsu et al. (1994) pointed out the importance of processes in the free troposphere. Probably, similar mechanisms are involved in the association of acidic species with stratospheric


229

aerosols. Further research is needed to better understand the processes in the free troposphere that may control the cycles of chemical species in the atmosphere. 4. Conclusions 1. Monthly depositions of 7 Be and SO24, in Tsukuba, Japan were correlated. The correlation was especially strong in spring and fall, when high depositions of 7 Be and SO24, were recorded. The correlation was also significant for daily precipitation samples taken in 1992. 2. Although 7 Be is derived from the upper atmosphere, most of the SO24, (nss-SO24, ) originates from sources at lower altitudes (more than 99%). 3. The correlation between 7 Be and SO24, implies a link between the meteorological dynamics (air mass descent from the upper atmosphere) and the formation of acidic deposition. Observations of 7 Be in the air seem important in this regard. Acknowledgements The authors express their sincere thanks to Professor Yukiko Dokiya and Mr. Kazuhiro Tsuboi, Meteorological College, Kashiwa, Japan, Dr. Michiko Abe, National Institute of Radiological Sciences, Chiba, Japan, Mr. Michio Aoyama, Mr. Takashi Miyao and Dr. Kikuo Okada, MRI for discussion and encouragement. The authors acknowledge Dr. Yuichi Oki, National Laboratory for High Energy Physics (KEK), Tsukuba, Japan for his kind discussion on radioactive aerosols in the air around high energy accelerators. The first author benefited from attending the seminar on the acidic deposition held by Dr. H. Hara, National Institute of Public Health, Tokyo, Japan. Thanks are also due to Dr. Richard Weisburd, ELLS, Tsukuba, Dr. Ian J. Graham, Institute of Geological and Nuclear Sciences Limited, New Zealand, and to Ms. Kim Todd and Mr. Ben Thomas for their help in english correction of this manuscript. References Abe, M., Kurotaki, K., Shibata, S., Takesita, H., and Abe, S., 1993: Trend analysis of ten years of changes in radioactive substances in the atmosphere at Chiba, Japan, Isotope Techniques in the Study of Past and Current Environmental Changes in the Hydrosphere and the Atmosphere, IAEA-SM-329/8, International Atomic Energy Agency, Vienna, pp. 35–42. Akimoto, H. and Narita, H., 1994: Distribution of SO 2 , NOx and CO2 emissions from fuel combustion and industrial activities in Asia with 1 1 resolution, Atmos. Environ. 28, 213–225. Baskaran M., 1995: A search for the seasonal variability on the deposition fluxes of 7 Be and 210 Pb, J. Geophys. Res. 100, 2833–2840. Bluth, G. J., Schnetzler, C. C., Krueger, A. J., and Walter, L. S., 1993: The contribution of explosive volcanism to global atmospheric sulfur dioxide concentrations, Nature 366, 327–329. Brost, R. A., Feichter, J., and Heimann, M., 1991: Three-dimensional simulation of 7 Be in a global climate model, J. Geophys. Res. 96, 22423–22445. Brown, L., Stensland, G. J., Klein, J., and Middleton, R., 1989: Atmospheric deposition of 7 Be and 10 Be, Geochim. Cosmochim. Acta 53, 135–142.

230


Browning, K. A. and Golding, B. W., 1995: Mesoscale aspects of a dry intrusion within a vigorous cyclone, Q. J. R. Meteorol. Soc. 121, 463–493. Crutzen, P. J., 1976: The possible importance of CSO for the sulfate layer of the stratosphere, 1976, Geophys. Res. Lett. 3, 73–76. Danielsen, E., Bleck, R., Shedlovsky, J., Wartburg, A., Haagenson, P., and Pollock, W., 1970: Observed distribution of radioactivity, ozone, and potential vorticity associated with tropopause folding, J. Geophys. Res. 75, 2352–2361. Danielsen, E., 1980: Stratospheric source for unexpectedly large values of ozone over the Pacific Ocean during Gametag, August 1977, J. Geophys. Res. 85, 401–412. Delmas, R. J., 1992: Free tropospheric reservoir of natural sulfate, J. Atmos. Chem. 14, 261–271. Dibb, J. E., Meeker, L. D., Finkel, R. C., Southon, J. R., Caffee, M. W., and Barrie, L. A., 1994: Estimation of stratospheric input to the Arctic troposphere: 7 Be and 10 Be in aerosols at Alert, Canada, J. Geophys. Res. 99, 12855–12864. Doi, T., Hosomi, M., Mizoguchi, T., and Sato, J., 1993: Variation of atmospheric ozone and 7 Be concentrations in Tsukuba district, Tenki 40, 827–834 (in Japanese). Dokiya, Y., Aoyama, M., Katsuragi, Y., Yoshimura, E., and Toda, S., 1987: Deposition of chemical components in Japan, The Chemistry of Acid Rain Sources and Atmospheric Processes, ACS symposium series 349, American Chemical Society, Washington, D.C., pp. 258–272. Dutkiewicz, V. A. and Husain, L., 1979: Determination of stratospheric ozone at ground level using 7 Be/ozone ratios, Geophys. Res. Lett. 6, 171–174. Dutkiewicz, V. A. and Husain, L., 1985: Stratospheric and tropospheric components of 7 Be in surface air, J. Geophys. Res. 90, 5783–5788. Feely, H. W., Larsen, R. J., and Sanderson, C. G., 1989: Factors that cause seasonal variations in beryllium-7 concentrations in surface air, J. Environ. Radioactivity 9, 223–249. Fujita, S., 1996: An estimation for atmospheric sulfur budget over the Japanese Archipelago, Environmental Science 9, 185–199 (in Japanese). Graustein, W. C. and Turekian, K. K., 1996: 7 Be and 210 Pb indicate an upper troposphere source for elevated ozone in the summertime subtropical free troposphere of the eastern North Atlantic, Geophys. Res. Lett. 23, 539–542. Harvey, M. J. and Matthews, K. M., 1989: 7 Be deposition in a high-rainfall area of New Zealand, J. Atmos. Chem. 8, 299–306. Hara, H., Ito, E., Katou, T., Kitamura, Y., Komeiji, T., Oohara, M., Okita, T., Sekiguchi, K., Taguchi, K., Tamaki, M., Yamanaka, Y., and Yoshimura, K., 1990: Analysis of two-year results of acid precipitation survey within Japan, Bull. Chem. Soc. Jpn. 63, 2691–2697. Hoell, J. M., Davis, D. D., Liu, S. C., Newell, R., Shipham, M., Akimoto, H., McNeal, R. J., Bendura, R. J., and Drewry, J. W., 1996: Pacific Exploratory Mission-West A (PEM-West A): September–October 1991, J. Geophys. Res. 101, 1641–1653. Igarashi, Y., Otsuji-Hatori, M., Hirose, K., 1996: Recent deposition of 90 Sr and 137 Cs observed in Tsukuba, J. Environ. Radioactivity 31, 157–169. Ishikawa, Y., Murakami, H., Sekine, T., and Yoshihara, K., 1995: Precipitation scavenging studies of radionuclides in air using cosmogenic 7 Be, J. Environ. Radioactivity 26, 19–36. Junge, C. E., 1963: Air Chemistry and Radioactivity, International Geophysics Series Vol. 4, Academic Press Inc., New York, U.S.A. Junge, C. E., Chagnon, E. W., and Manson, J. E., 1961: Stratospheric aerosol, J.Meteorol. 18, 81–108. Kitamura, M., Sugiyama, M., Ohhashi, T., and Nakai, N., 1993: An estimation of the origin of sulfate ion in rain water in view of sulfur isotopic variations, Chikyukagaku 27, 109–118 (in Japanese). Kondo, K., Muramatsu, H., Kanda, Y., and Takahara, S., 1984: Particle size distribution of 7 Beaerosols formed in high energy accelerator tunnels, Int. J. Appl. Radiat. Isot. 35, 939–944. Kritz, M. A., Rosner, S. W., Danielsen, E. F., and Selkirk, H. B., 1991: Air mass origins and troposphere-to-stratosphere exchange associated with mid-latitude cyclogenesis and tropopause folding inferred from 7 Be measurements, J. Geophys. Res. 96, 17405–17414. Lal, D., 1963: On the investigations of geophysical processes using cosmic ray produced radioactivity, Earth Science Meteoritics, North-Holland Publishing Company, Amsterdam, pp. 115–142. Lal, D. and Suess, H. E., 1968: The radioactivity of the atmosphere and hydrosphere, Annual Rev. Nucl. Sci. 18, 407–434.


231

Lambert, A., Grainger, R. G., Rodgers, C. D., Taylor, F. W., Mergenthaler, J. L., Kumer, J. B., and Massie, S. T., 1997: Global evolution of the Mt. Pinatubo volcanic aerosols observed by the infrared limb-sounding instruments CLAES and ISAMS on the upper atmosphere research satellite, J. Geophys. Res. 102, 1495–1512. Langner, J. and Rohde, H., 1991: A global three-dimensional model of the tropospheric sulfur cycle, J. Atmos. Chem. 13, 225–263. Matsunami, T. and Megumi, K., 1994: Variation of 7 Be concentrations in surface air and in deposition with the solar activity, Radioisotopes 43, 334–340 (in Japanese). McCormick, M. P., Thomason, L. W., and Trepte, C. R., 1995: Atmospheric effects of the Mt. Pinatubo eruption, Nature 373, 399–404. Ogawa, T., 1991: Taiki No Butsuri Kagaku (Physical Chemistry of the Atmosphere), Tokyo-do Shuppan, Tokyo, Japan (in Japanese). Ogura, Y., 1984: Ippan Kishogaku (General Meteorology), Tokyo Daigaku Shuppankai, Tokyo, Japan (in Japanese). Ohizumi, T., Fukuzaki, N., and Kusakabe, M., 1997: Sulfur isotopic view on the sources of sulfur in atmospheric fallout along the coast of the Sea of Japan, Atmos. Environ. 31, 1339–1348. Okamoto, R., Ohoshima, H., Yamaguchi, T., Ozaki, Y., Kawakami, K., and Fujii, K., 1992: Chemical Monitoring of Precipitation in Tsukuba Area (1985–1990), Misc. Publ. Nat. Inst. Agro-Environ. Sci. No.13 (in Japanese). Raisbeck, G. M., Yiou, F., Fruneau, M., Loiseaux, J. M., Lieuvin, M., and Ravel, J. C., 1981: Cosmogenic 10 Be/7 Be as a probe of atmospheric transport processes, Geophys. Res. Lett. 8, 1015–1018. Rangarajan, C. and Eapen, C. D., 1990: The use of natural radioactive tracers in a study of atmospheric residence times, Tellus 42B, 142–147. Reiter, E. R., 1978: Atmospheric Transport Processes Part 4: Radioactive Tracers, National Technical Information Center, U.S. Department of Energy, Springfield, Virginia. Sato, J., Doi, T., Segawa, T., and Sugawara, S., 1994: Seasonal variation of atmospheric concentrations of 210 Pb and 7 Be at Tsukuba, Japan, with a possible observation of 210 Pb originating from the 1991 eruption of Pinatubo volcano, Philippines, Geochem. J. 28, 123–129. Turco, R. P., Whitten, R. C., and Toon, O. B., 1982: Stratospheric aerosols: observation and theory, Rev. Geophys. Space Phys. 20, 233–279. Turekian, K. K. and Tanaka, N., 1992: The use of atmospheric cosmogenic 35 S and 7 Be in determining depositional fluxes of SO2 , Geophys. Res. Lett. 19, 1767–1770. Turekian, K. K., Benninger, L. K., and Dion, E. P., 1983: 7 Be and 210 Pb total deposition fluxes at New Haven, Connecticut and at Bermuda, J. Geophys. Res. 88, 5411–5415. Uematsu, M. and Duce, R. A., 1994: Atmosphere beryllium-7 concentrations over the Pacific Ocean, Geophys. Res. Lett. 21, 561–564. Vizee, W. and Singh, H. B., 1980: The distribution of beryllium-7 in the tropososphere: implications on stratospheric/tropospheric air exchange, Geophys. Res. Lett. 7, 805–808. Warneck, P., 1988: Chemistry of the Natural Atmosphere, International Geophysics Series, Vol. 41, Academic Press Inc., California, U.S.A. Wolff, G. T., Ferman, M. A., and Monson, P. R., 1979: The distribution of beryllium-7 within high pressure systems in the eastern United States, Geophys. Res. Lett. 6, 637–639. Young, J. A. and Silker, W. B., 1980: Aerosol deposition velocities on the Pacific and Atlantic oceans calculated from 7 Be measurements, Earth Planet. Sci. Lett. 50, 92–104. Young, M. V., Monk, G. A., and Browning, K. A., 1987: Interpretation of a rapidly deepening cyclone, Q. J. R. Meteorol. Soc. 113, 1089–1115.