Estimate of European - ACS Publications - American Chemical Society

Sellafield (Great Britain), Marcoule, and La Hague (both. France). While reliable data on 129I releases from La Hague exist for the whole period of op...
0 downloads 0 Views 160KB Size
Environ. Sci. Technol. 2006, 40, 5891-5896

Estimate of European 129I Releases Supported by 129I Analysis in an Alpine Ice Core H E R B E R T R E I T H M E I E R , * ,† VITALI LAZAREV,† WERNER RU ¨ H M , ‡,| M A R G I T S C H W I K O W S K I , §,⊥ H E I N Z W . G A¨ G G E L E R , § , ⊥ A N D ECKEHART NOLTE† Physics Department E15, TU Mu ¨ nchen, D-85748 Garching, Germany, and Institute of Radiobiology, LMU Mu ¨ nchen, D-80336 Munich, Germany, and Laboratory for Radiochemistry and Environmental Chemistry, PSI, CH-5232 Villigen, Switzerland, and Department for Chemistry and Biochemistry, University of Bern, CH-3012 Bern, Switzerland

129I in the European environment originates predominantly from the industrial nuclear fuel reprocessing plants Sellafield (Great Britain), Marcoule, and La Hague (both France). While reliable data on 129I releases from La Hague exist for the whole period of operation, less is known about the contributions from Sellafield and Marcoule. For those periods where no data are available, i.e., for the first 16 years of the Sellafield operation and for the first 3 decades of the Marcoule operation, we estimated releases into the atmosphere of 118 GBq and 825 GBq, respectively. Hence, Marcoule was the major European source of airborne 129I, contributing about 45% to the total airborne 129I releases from all the European reprocessing facilities, until it was decommissioned in 1997. The estimated total emissions were compared with 129I deposition fluxes for the time period 1970-2002, obtained from the analysis of an ice core from Fiescherhorn glacier, Swiss Alps (46°33’N, 08°04’E; 3900 m asl). The temporal evolution of the 129I deposition agrees well with the total 129I releases into the atmosphere from the European reprocessing facilities and from atmospheric nuclear weapons tests, supporting our estimated release rates. However, the 129I concentrations and deposition fluxes at Fiescherhorn glacier were a factor of 6 lower than values obtained from the analysis of rainwater collected near Zurich (408 m asl) in Switzerland in the years 1994-97. This suggests a strong vertical concentration gradient of 129I, typical for water-soluble atmospheric trace species which are removed from the atmosphere in the course of days by precipitation scavenging, and must be taken into account if glaciers are used as an archive for a retrospective quantification of 129I deposition fluxes. In addition, the temporal evolution of the contribution of 129I re-emitted from the ocean’s surface for the 129I inventory in the atmosphere was quantified for the first time. Although the annual amount of 129I released this way was very low until the early 1990s, it is similar to the

* Corresponding author e-mail: [email protected]. † TU Mu ¨ nchen. ‡ LMU Mu ¨ nchen. § Paul Scherrer Institute. ⊥ University of Bern. | Current address: Institute of Radiation Protection, GSF, D-85764 Neuherberg, Germany. 10.1021/es0605725 CCC: $33.50 Published on Web 08/23/2006

 2006 American Chemical Society

airborne 129I releases from Sellafield and La Hague in the present time.

1. Introduction Before anthropogenic input began, the radioactive isotope 129 I (T1/2 ) 15.7 My) was in natural equilibrium. Its inventory was about 390 GBq in ocean water, corresponding to an 129I concentration of 2 × 105 at/l (1). Because the major part of this iodine was dissolved in the deep water layers of the oceans, only about 2.4% of the free global 129I inventory was subject to short-scaled exchange processes (2), corresponding to an 129I activity of about 9 GBq within a period of 100 y. Since the use of nuclear fission for military and civil purposes, 129I has accumulated in the environment. Reprocessing of spent nuclear fuel was identified as a source of 129I observed in soil samples (3) and in ocean water (4). Apart from samples from the immediate surroundings of the small test facility WAK (Germany) (5), the increase of 129I in European environmental samples was ascribed to 129I releases from Sellafield (Great Britain) and La Hague (France) (4, 6-8). The importance of the reprocessing plant at Marcoule (France) for the interpretation of 129I concentrations in water from the Rhoˆne river (9, 10) and from the Mediterranean (9), and in rainwater from Bavaria (1) was also emphasized. Recently, 129I was used as a tracer for radioiodine dispersion caused by nuclear accidents or by failures in the operation of nuclear facilities (8, 11, 12). However, only sparse information on background concentrations were available for those regions. For a quantitative estimate of this background, transport models have to be used. Because such models are always based on the 129I releases from fuel reprocessing facilities, a careful determination of these 129I releases is required. In this study we reconstructed past 129I releases from atomic weapons tests and from the European reprocessing plants at Windscale/Sellafield, Marcoule and La Hague. The estimated emissions were compared with the 129I deposition history over the entire time period with anthropogenic 129I releases, obtained from the analysis of an ice core from the Swiss Alps.

2. Materials and Methods 2.1. Estimation of 129I Emissions from Atomic Bomb Tests Between the first atomic bomb test on July 16, 1945 (New Mexico) by the U.S. and the last above-ground test on October 16, 1980 (Lop Nor) by China, 541 atmospheric atomic explosions were performed, mainly by the U.S. and by the former Soviet Union. These atmospheric atomic explosions produced together a fission energy of about 190 Mt TNT (13). In this work, the 129I releases to the atmosphere were calculated as a function of time, based on the above-ground tests, their fraction of energy produced in fission as listed in ref 13, and a fast fission yield of 1.6% for 129I. The latter was interpolated from the fission yields of the neighboring 125Sn and 131I taken from ref 13. The annual deposition of 90Sr due to the weapons tests was previously estimated for the northern hemisphere using a fission yield of 3.5% for 90Sr (13). From this, a 90Sr deposition fluence of 4.2 × 1012 at/m2 results for a latitude of 50 °N. For a rough estimate, the 129I dispersion and deposition was assumed to be comparable to that of 90Sr. The 129I deposition in the northern hemisphere as a function of time and the 129I deposition fluence at a latitude of 50 °N was obtained by correcting for the ratio of the fast fission yields of 129I and 90Sr. VOL. 40, NO. 19, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5891

2.2 Estimation of 129I Emissions from Fuel Reprocessing at Windscale/Sellafield. The first European reprocessing plant (B204) was put into operation at Windscale (later Sellafield) in 1951. In this plant and in its successor B205 (from 1964 on), the irradiated fuel from the British Magnox reactors was reprocessed. In 1994, an additional plant for reprocessing oxide fuels was commissioned (15). Unfortunately, there are no published data on airborne 129 I releases from Windscale/Sellafield before 1976. Therefore, the unknown 129I releases were estimated in this work. For the years 1951-1965, the 129I release was estimated from the amount of 129I produced in the British Magnox reactors (Table 1, Supporting Information), of which the irradiated fuel was reprocessed at Windscale. These estimates are based on the following: • the thermal power of the operating reactors (14, 15), • a reactor capacity factor of 0.9 per year, i.e., it was assumed that the reactors operated all the time except for about 10% maintenance time, • a hold time of 6 months for the early years and 18 months for the 1960s, i.e., it was assumed that the spent fuel was stored about 6 months and 18 months before reprocessing started, respectively, and • the thermal fission yield of 129I of 0.78% for fission of 235U. It was further assumed that the entire 129I present in the spent fuel at the time of its reprocessing was released to the atmosphere. Though the assumed scenario appears to be realistic, the 129I emissions estimates represent an upper limit for the actual gaseous 129I release caused predominantly by the uncertainties of the reactor capacity factor and the fraction of 129I released into the atmosphere, which might be overestimated by about 10% and by a factor of 2, respectively. In contrast, the uncertainty of the hold time yields only a minor time shift and does not affect the total 129I released. For the following years of plant B205 operation (19661975) it was assumed that 20% of the total 129I was released into the atmosphere. This assumption is based on the mean fraction of airborne release of 17% for the time period 19761985, calculated from the 129I releases reported in (13). The remainder was released in liquid form. The liquid 129I releases from 1966-1975 were previously reconstructed via analysis of seaweed collected in the English Channel (16). For this period, we estimated the airborne releases of 129I from the liquid releases assuming a fraction of airborne release of 20%. 2.3. Estimation of 129I Emissions from Fuel Reprocessing at Marcoule. The first French reprocessing plant was plant UP-1 at Marcoule (Languedoc-Roussillon) which operated between 1959 and 1997. Until the early 1990s, it was used predominantly for production of weapons-grade plutonium (17, 18). The irradiated fuel reprocessed at UP-1 originated from the French natural uranium reactors, and, to a lesser extent, from the Spanish natural uranium reactor Vandellos1. In addition, the smaller-scaled pilot plant APM was commissioned in 1962 for reprocessing of uranium alloys and, since 1973, of uranium oxide fuels (18). Because of its military purpose, there is almost nothing known about any radio-nuclide releases from the Marcoule facilities: Gaseous and liquid 129I releases were not published before 1988 and 1995, respectively (17, 19). Therefore, the early gaseous 129I releases from the Marcoule plants UP-1 and APM had to be estimated. This was done from the amount of 129I produced in the French natural uranium reactors. Times of operation and estimated production of 129I are listed in Table 2, Supporting Information, for the considered natural uranium reactors, of which the spent fuel was reprocessed in the Marcoule plants (17, 18). Similar to the estimates of the Sellafield releases, the thermal power of the reactors (14, 18, 19) was used and a reactor 5892

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 19, 2006

capacity factor of 0.9 per year was assumed. The amounts of 129I arisen in reprocessing was deduced assuming a hold time of half a year for fuel irradiated for military purposes and of 1.5 years for civil fuel, and they are summarized in Table 3, Supporting Information. The first efforts to reduce the radioiodine releases were reported in 1967 (18). Thus, it was assumed that all the 129I arisen in the reprocessing process was released to the atmosphere for the period 1958-1966. For the time period 1995-1997 the mean fraction of airborne release was 82% of the total release (19). We therefore assumed that 80% of 129I was released to the atmosphere for the period 19671997 and the remaining 20% was released in liquid form to the Rhoˆne river. The very high percentage of gaseous 129I releases is supported by the following consideration: The mean amount of plutonium produced at Marcoule was 879 kg/y for the years 1994-1996 (17). For fuel irradiated in natural uranium reactors, a corresponding 129I inventory of 44 GBq/y can be deduced. The mean release of gaseous 129I in Marcoule is reported to be 39 GBq/y for this period (17), which is almost 90% of the total 129I. Though the estimate of the Marcoule releases is based on assumptions quite similar to that used for the Windscale releases, we expect the uncertainty of the total 129I released from Marcoule to be lower than about 20% caused by the high percentage of gaseous 129I released. However, we cannot exclude again a minor shift regarding the year of release due to the uncertainty of the hold time assumed. The 129I release rates have to be interpreted as a moving average for the actual 129I release rates. 2.4. 129I Emissions from Fuel Reprocessing at La Hague. Industrial fuel reprocessing in France started in 1966 at La Hague (Normandy), when plant UP-2 was put into operation. It was built for reprocessing of spent fuel from the French natural uranium reactors, mainly from the commercial power reactors. Plant UP-2 was modified in 1975 for reprocessing uranium oxide fuel from the French pressurized water reactors (23). Therefore, reprocessing of fuel from the natural uranium reactors was then transferred to plant UP-1 at Marcoule (18). A second plant, AT1, operated in the period 1969-1979 to reprocess fuel from fast breeding reactors (23). Fuel from German, Japanese, Swiss, Dutch, and Belgian reactors is reprocessed in a third plant (UP-3) commissioned in 1989. In contrast to Sellafield and Marcoule, reliable data on the 129I releases of the La Hague facility exist for the whole period of operation. Thus, the gaseous 129I releases of La Hague presented here are based on data from the literature (13, 19, 20, 23). 2.5. Estimation of 129I Re-emission from the Ocean. Although nowadays approximately 99% of the total 129I in Europe is released in liquid form to the English Channel and to the Irish Sea, gaseous releases are expected to still be the main source of 129I observable in European rainwater (1, 24). However, liquid releases can also contribute to the 129I concentration in the atmosphere by re-emission from the ocean surface before deep water formation occurs in the Northern Atlantic. To quantify this process, we used the emission rate of stable iodine from the oceans, and the time required for the iodine to descend to deeper water layers (see Supporting Information). 2.6 Ice Core Samples and 129I Analysis. An ice core reaching bedrock a depth of 150 m was retrieved in December 2002 from the Fiescherhorn glacier (46° 33’ N, 08° 04’ E; 3900 m asl, Bernese Alps, Switzerland) (25). Samples of 4-5 cm length were prepared for analysis of major ion concentrations and stable isotope ratios. The portion of the samples remaining after analysis was stored at a temperature of -20 °C. The upper 108 m of this core was dated using the seasonal

FIGURE 1. Estimated 129I input to the atmosphere as a result of the above-ground atomic tests (O) and expected total activity of 129I deposited in the northern hemisphere (dashed line). variation of the NH+ 4 concentration and the stable isotope ratio δ 2H. The dating uncertainty is 1-2 years (26, 27). In order to reconstruct historic 129I depositions, the 129I concentrations in the upper 64.2 m water equivalent of this ice core were analyzed by means of accelerator mass spectrometry (AMS) at the Maier-Leibnitz laboratory in Garching (Germany), corresponding to the period 19702002. The melted ice samples were combined to obtain annual or bi-annual samples, depending on their estimated 129I contents. 0.5 mg of Woodward Iodine with an 129I/127I ratio of (2.3 ( 0.7) × 10-14 (1) was added to the samples as carrier. Total inorganic iodine and an unknown fraction of organic iodine was extracted from the samples and purified using a carbon tetrachlorid method and was finally precipitated as AgI (1). To estimate the 129I input by the chemical treatment of the sample material, blank samples were prepared from the carrier material. Analysis of these blank samples indicated an 129I input between 2 × 105 and 6 × 105 atoms per sample during the chemical treatment and the concentration in ice samples was corrected for this value. A detailed description of sample preparation and of AMS measurement can be found in ref 1.

3. Results and Discussion 3.1. 129I Emissions from Atomic Bomb Tests. The amount of 129I produced during the atmospheric atomic explosions as a function of time is shown in Figure 1. According to our calculations, in total 600 GBq of 129I were produced. A significant fraction of the fission products was introduced into the stratosphere and thus dispersed on a global scale. The expected annual 129I fallout in the northern hemisphere is also shown in Figure 1. Based on data for the 90Sr deposition fluxes (13), we estimated a total 129I deposition fluence from the weapons fallout of 1.9 × 1012 at/m2 for a latitude of 50 °N. This is lower by almost 2 orders of magnitude compared to the mean 129I deposition fluence of 1.2 × 1014 at/m2, determined in seven soil samples from Lower Saxony (28), and only slightly exceeds the recent annual 129I deposition of 1.2 × 1012 at/m2, deduced from analysis of rainwater from Upper Bavaria (1). Thus the predominant fraction of total 129I deposited in Europe originates from reprocessing activities. 3.2. 129I Emissions from Windscale/Sellafield. The estimated amount of 129I produced in the British Magnox reactors for the years 1951-1965 is given in Table 1, Supporting Information. We conclude the accumulated airborne 129I releases to be 84 GBq for Windscale from 1951 to 1964. Since our estimate represents an upper limit for the

airborne 129I releases, based on the total 129I produced in the British reactors, the number of 308 GBq given in (29) seems to overestimate the actual releases of the years 1951-1964 by almost a factor of 4. It should be emphasized that a constant release rate of 22 GBq was assumed in ref 29 for the considered period, which is not realistic, as can be seen from Table 1, Supporting Information. The estimated 129I releases from the Windscale/Sellafield facilitiy are shown in Figure 2a. Although the estimations for 1951-1965 and for 1966-1975 are based on quite different assumptions, they fit well to each other in the 1960s. The drop in airborne 129I releases by about a factor of 5 in 19641965 is due to the decommissioning of B204 in 1964 and the beginning of operation at plant B205 in 1965. Only 20% of the 129I arisen at B205 was assumed to be released to the atmosphere. The obtained liquid release is in good agreement with the value reconstructed by ref 16 for 1966. This is also the case for the estimated 129I releases from 1966 to 1975 and the published data for the 1970s (13). Both data sets from the literature are also shown in Figure 2a. It is evident that the liquid releases dominated, by far, the total releases in Sellafield. At present, gaseous releases amount only a few percent of the total releases. Including the year 2004, the total release of airborne and liquid 129I from Sellafield were estimated here to be 1 TBq and 9 TBq, respectively. 3.3. 129I Emissions from Marcoule. Estimated 129I releases from the Marcoule plants UP-1 and APM are shown in Figure 2b, along with published data of gaseous and liquid releases for the years 1988-1997 and 1995-1997, respectively (17, 19). Our estimates indicate that an 129I activity of 470 GBq has arisen in the Marcoule plants for the years 1988-1997, and that about 375 GBq was released to the atmosphere. For the same period of time, an airborne 129I release of 448 GBq was published (17), which is only slightly higher (Figure 2b). In total, our estimates suggest that 1200 GBq of 129I were released to the atmosphere by the Marcoule facility. In addition to that, an estimated 129I activity of 300 GBq was released in liquid form to the Mediterranean via the Rhoˆne river. Thus it is evident that the Marcoule plants must be expected to have a major influence on the 129I inventory of the European environment, comparable in magnitude to the better known reprocessing facilities in Sellafield and La Hague. The importance of the 129I releases from the Marcoule plants for the 129I concentration in the Mediterranean (9) and in rainwater from Southern Germany (1) was already emphasized. 3.4. 129I Emissions from La Hague. The 129I releases of the La Hague facility are shown in Figure 2c. Since commissioning in 1966 about 0.44 TBq of gaseous 129I was released until 2004. For comparison, the overall liquid release of 129I to the English Channel was much larger and amounts to 20.7 TBq. For the La Hague facility, a reconstruction of the amount of 129I arisen in reprocessing of irradiated fuel from the French natural uranium reactors was performed for the years 19661975. The corresponding total 129I activity of 141 GBq is in good agreement with the total 129I release of 157 GBq reported in (20). 3.5. 129I Re-emission from the Ocean. According to our estimates, the emission rate of stable iodine from the ocean’s surface is 2 × 109 kg/y. This corresponds to a yearly evaporation of the total iodine inventory in the upper 12 cm of the oceans. Former estimates delivered values of 1.3 × 109 kg/y and 8 × 108 kg/y for the emission rate of methyl iodide from the ocean’s surface (21, 22), which are only somewhat lower than our estimate. The resulting 129I re-emission rates from ocean water are compared to the total annual releases from the European reprocessing activities and the atomic bomb fallout in the VOL. 40, NO. 19, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5893

FIGURE 2. Estimated airborne (O) and liquid (×) 129I releases from Windscale/Sellafield (a.), from Marcoule (b.), and from La Hague (c.). For comparison, the known airborne (b) and liquid (*) releases from Windscale/Sellafield (13, 16, 19, 34), from Marcoule (17, 19), and from La Hague (13, 19, 20, 23) are also shown, respectively. Comparison of annual 129I releases is shown (d.): total activity of 129I produced during the atomic bomb explosions and deposited in the northern hemisphere (dotted line), gaseous (O) and liquid (b) emissions from reprocessing in Windscale/Sellafield, in Marcoule and in La Hague, and 129I re-emitted from ocean water (*). northern hemisphere in Figure 2d. The contribution to the atmospheric 129I inventory in Europe by 129I re-emitted from the oceans was small before 1997, when plant UP-1 at Marcoule was decommissioned and an improved iodine filter technique was installed at La Hague. As a result, nowadays re-emission from the ocean’s surface is almost as important as direct gaseous releases from reprocessing for the 129I concentration in the atmosphere. We estimate the total 129I activity re-emitted from the ocean’s surface to be 180 GBq until the year 2004. An upper limit for the 129I re-emission rate from the English Channel and coastal regions of the North Sea was formerly estimated to be 8.9 GBq/y for the early 1990s (24). This value agrees well with our estimate of 3-4 GBq/y for the period 1990-1993. 3.6. 129I Measurements at Fiescherhorn Glacier. The resulting 129I concentrations in ice samples are given in Table 4, Supporting Information, and visualized in Figure 3. They were combined with 129I concentration data for the period 1950-1986, previously analyzed (2) from another ice core drilled at the same glacier in 1988 (30). The combined ice core records provide unique information on the 129I deposition from the very beginning of the anthropogenic 129I releases up to the year 2002. In the period of overlap (1970-1986) they generally agree well. Deviations might be caused by slight differences in dating of the two ice cores. In addition, a relocation of 129I deposited in 1 year to the layer of the previous year(s) due to meltwater percolation and refreezing cannot be excluded. This effect does not result in a net loss of 129I, since the meltwater refroze totally. This is corroborated by the fact 5894

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 19, 2006

FIGURE 3. 129I concentrations in samples from the Fiescherhorn ice core drilled in 2002 (*) compared to values published in (2) (O). that the mean 129I concentrations in the period of overlap (1975-1986) for the new core is slightly higher than for the old core with values of (3.1 ( 0.2) × 108 at/L and (2.4 ( 0.1) × 108 at/L, respectively. The net accumulation rate of the new core, investigated in this work, is 19.7 m water equivalent (m we) for the period 1975-1986, while it is only 16.8 m we for the old one. This difference is caused by local topographic features at the drilling sites and has to be taken into account, if the 129I deposition fluxes are compared. The resulting 129I deposition fluxes (now corrected for the different net accumulation rates)

FIGURE 4. 129I deposition fluxes determined at the Fiescherhorn glacier (*), and those based on data published in (2) (O). For this comparison, the latter data were scaled with the ratio of the mean net accumulation rates of the period 1950-1974. In addition, airborne emissions of 129I from the European reprocessing facilities (thick lines) and the total 129I which was deposited in the northern hemisphere as a result of the atmospheric atomic bomb explosions (dotted line), are shown. 129I releases before 1988, which had to be estimated, are dashed. derived from the two cores are presented in Figure 4. Because the net accumulation rates of both cores differ by 13% for the period 1950-74, the 129I deposition fluxes retrieved from the old core were scaled with the ratio in net accumulation rates in Figure 4. The 129I deposition fluence obtained in this work is (1.4 ( 0.1) × 1013 at/m2 for the years 1970-2002, and (8.7 ( 0.5) × 1011 at/m2 for the years 1950-1969, deduced from the old core. Hence, the total 129I deposition fluence at the Fiescherhorn glacier is (1.5 ( 0.1) 1013 at/m2. 3.7. Comparison with European 129I Releases. The time dependence of the 129I deposition fluxes deduced from the two ice cores are compared in Figure 4 to the combined airborne 129I releases from the European reprocessing facilities Sellafield, Marcoule and La Hague. For the 1950s, the atomic bomb tests dominated the 129I input to the atmosphere. This contribution is, therefore, also shown in Figure 4. The temporal trend in 129I deposition at the Fiescherhorn glacier is in good agreement with that of the estimated airborne 129I releases. In contrast to this, the trends of the liquid releases from Sellafield and La Hague, and of the 129I re-emitted from the ocean show a clearly different pattern (see Figure 2d). This is a further indication for the previous suggestion (1, 24) that the airborne 129I releases from the European reprocessing facilities were the predominant source for atmospheric 129I in Europe. Despite the general agreement, the ice-core-derived 129I deposition fluxes show a substructure which cannot be explained by the 129I releases. An example is the period 1980-1990, with distinct maxima and minima in 129I deposition. Besides a variation in net accumulation and in the wind trajectories to the Fiescherhorn glacier which could explain part of the observed variations, postdepositional processes must also be considered: Due to an elevated air temperature or an exceptional dust deposition on the glacier surface, surface melting events can occur. The meltwater can then percolate to deeper firn layers, refreeze there and produce ice lenses. As a result, downward motion of ions dissolved in the meltwater can occur, which was observed at the Fiescherhorn glacier for sulfate (27). It is interesting to note that there also seems to be a correlation between melt horizons and low 129I deposition fluxes on the one hand, and between accumulation of ice

lenses and elevated 129I deposition fluxes on the other hand. A downward motion of deposited 129I ions could thus have occurred in the firn from the Fiescherhorn glacier. By this process, part of the substructure in 129I deposition record between 1980-1990 can possibly be explained. 3.8. Comparison with 129I Concentrations in Rainwater. 129I concentration in rainwater from Du ¨ bendorf (Zurich, Switzerland) was previously analyzed, and the resulting 129I deposition fluxes were reported in ref 24. In Table 5, Supporting Information, these values are given for the considered years (1994-1997), and compared with 129I deposition fluxes deduced from the Fiescherhorn ice core. From 1994 to 1997, a decrease in 129I deposition fluxes is observed in the rainwater as well as in the ice core. However, in terms of absolute values, a substantial difference was found for the 129I concentrations and the 129I deposition fluxes. The mean 129I deposition flux at the Fiescherhorn glacier is about a factor of 6 lower than that reported for rainwater from Zurich for the considered period. This might be due to a 129I concentration gradient in the atmosphere. Such a gradient was also observed for stable iodine with a 60% lower concentration in air at an altitude of 1000 m asl compared to that at sea level (31). This observation suggests iodine to be diluted in air with increasing altitudes, most probably because of scavenging by clouds and precipitation in the lower troposphere. A similar altitude gradient was observed for the concentration of SO4 in the atmosphere (32, 33). To investigate this issue further, we analyzed the stable iodine concentration in a combined ice sample representing the period 1963-2002 using the photometric method described in (1), and found an 127I concentration of < 0.6 µg/l. In contrast, the mean 127I concentration in rainwater from Upper Bavaria, collected between July 2003 and December 2003, was reported to be 2.2 µg/l (1). This might also be an indication that a dilution of iodine (both stable 127I and radioactive 129I) occurs with increasing altitude. Thus, 129I deposition fluxes deduced from glacier ice from high altitudes cannot be compared directly with 129I deposition fluxes at lower altitudes. Rather, the 129I deposition fluxes at high altitudes deduced from glacier ice analysis have to be corrected for the dilution by means of an altitude correction factor fa, if the 129I deposition flux at lower altitudes is to be estimated. Hence, from the total 129I deposition fluence of (1.5 ( 0.1) × 1013 at/m2 at Fiescherhorn glacier, an 129I deposition fluence of (8.9 ( 0.4) × 1013 at/m2 can be deduced for the region of Zurich, using an altitude factor of fa ) 6. This is only slightly lower than the previously reported value of 1.2 × 1014 at/m2 for the region of Lower Saxony (28).

Acknowledgments The Fiescherhorn ice core 2002 was drilled in the frame of the NCCR climate project VITA, funded by the Swiss National Science Foundation. Financial support from the Deutsche Forschungsgemeinschaft is gratefully acknowledged. We regret to announce that E. Nolte deceased before publication of this manuscript.

Supporting Information Available Estimated annual production of 129I in the British and in the French natural uranium reactors (2 tables), estimated 129I arisen in reprocessing at Marcoule (1 table), detailed method to estimate the 129I re-emission rate from the ocean, results of the 129I measurement at Fiescherhorn glacier (1 table), and previous 129I data from Swiss rainwater (1 table). This material is available free of charge via the Internet at http:// pubs.acs.org. VOL. 40, NO. 19, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5895

Literature Cited (1) Reithmeier, H.; Lazarev, V.; Kubo, F.; Ru ¨ hm, W.; Nolte, E. 129I in precipitation using a new TOF system for AMS-Measurements. Nucl. Instrum. Methods. Phys. Res., Sect. B 2005, 239 (3), 273-280. (2) Wagner, M. J. M.; Dittrich-Hannen, B.; Synal, H.-A.; Suter, M.; Schotterer, U. Increase of 129I in the environment. Nucl. Instrum. Methods. Phys. Res., Sect. B 1996, 113, 490-494. (3) Kantelo, M. V.; Tiffany, B.; Anderson, T. J. Iodine-129 distribution in the terrestrial environment surrounding a nuclear fuel reprocessing plant after 25 years of operation. In: Environmental Migration of Long-Lived Radionuclides, Proceedings of an International Atomic Energy Agency Conference, Knoxville (USA), 27-31 July 1981; IAEA: Vienna, 1982, 495-500. (4) Yiou, F.; Raisbeck, G. M.; Zhou, Z. Q.; Kilius, L. R. 129I from nuclear fuel reprocessing; potential as an oceanographic tracer. Nucl. Instrum. Methods. Phys. Res., Sect. B 1994, 92, 436-439. (5) Robens, E.; Hauschild, J.; Aumann, D. C. Iodine-129 in the environment of a nuclear fuel reprocessing plant: IV. 129I and 127I in undisturbed surface soils. J. Environ. Radioact. 1989, 9, 17-29. (6) Paul, M.; Fink, D.; Hollos, G.; Kaufman, A.; Kutschera, W.; Magaritz, M. Measurement of 129I concentrations in the environment after the Chernobyl reactor accident. Nucl. Instrum. Methods. Phys. Res., Sect. B 1987, 29, 341-345. (7) Bachhuber, H.; Bunzl, K. Background levels of atmospheric deposition to ground and temporal variation of 129I, 127I, 137Cs and 7Be in a rural area of Germany. J. Environ. Radioact. 1992, 16, 77-89. (8) Straume, T.; Marchetti, A. A.; Anspaugh, L. R. The feasibility of using I-129 to reconstruct I-131 deposition from the Chernobyl reactor accident. Health Phys. 1996, 71, 733-740. (9) Zhou, Z. Q. Evaluation des rejets marins d’iode-129 par les usines de retraitement de La Hague (France) et Sellafield (U.K.) en vue de leur utilisation comme traceur oc’eanographique. Thesis, Universite´ Paris XI. Orsay. 1995. (10) Moran, J. E.; Oktay, S. D.; Santschi, P. H. Sources of Iodine and Iodine-129 in Rivers. Water Res. Res. 2002, 38 (8), 24-1 to 2410. (11) Mironov, V.; Kudrjashov, V.; Yiou, F.; Raisbeck, G. M. Use of 129I and 137Cs in soils for the estimation of 131I deposition in Belarus as a result of the Chernobyl accident. J. Environ. Radioact. 2002, 59, 293-307. (12) Michel, R.; Handl, J.; Ernst, T.; Botsch, W.; Szidat, S.; Schmidt, A.; Jakob, D.; Beltz, D.; Romantschuk, L. D.; Synal, H.-A.; Schnabel, C.; Lo´pez-Gutie´rrez, J. M. Iodine-129 in soils from Northern Ukraine and the retrospective dosimetry of the iodine131 exposure after the Chernobyl accident. Sci. Total Environ. 2005, 340, 35-55. (13) UNSCEAR. United Nations Scientific Committee on the Effects of Atomic Radiation. Sources and effects of ionizing radiation Vol 1: Sources. Report to the General Assembly; United Nations: New York. 2000; Vol. 1. (14) Commission of the European Communities. Radioactive effluents from nuclear power stations and nuclear fuel reprocessing plants in the Community, discharge data 1969-1974; Commission of the European Communities Report: Luxembourg. 1975. (15) Bellona Foundation. Working paper 5: Sellafield. Bellona Foundation, http://www.bellona.no, Oslo. 2001. (16) Yiou, F.; Raisbeck, G. M.; Zhou, Z. Q.; Kilius, L. R.; Kershaw, P. J. Improved Estimates of oceanic discharges of 129I from Sellafield and La Hague. In: Proceedings of the 2nd International Conference on Environmental Radioactivity in the Arctic and Antarctic, Oslo, 1995, pp 113-117.

5896

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 19, 2006

(17) Cogema. Rapport Environnement; Cogema: Marcoule. 1997. (18) Barrillot, B.; Davis, M. Les dechets nucleaires militaires franc¸ ais; Centre de Documentation et de Recherche sur la Paix et les Conflicts (CRDS): Lyon. 1994. (19) Radiation Protection 127. Radioactive effluents from nuclear power stations and nuclear fuel reprocessing plants in the European Union, 1995-99; Stricht, S. V., Janssens, A., Eds.; European Commission: Luxembourg. 1999. (20) Groupe Radioe´cologie Nord-Cotentin. Inventaire des rejets radioactifs des installations nucleaires: Historique des rejets, Rapport Finale Detaille´; Groupe Radioe´cologie Nord-Cotentin: Fonteney-aux-Roses: 1999; Vol. 1. (21) Rasmussen, R. A.; Khalil, M. A.; Gunawardena, R.; Hoyt, S. T. Atmospheric methyl iodide (CH3I). J. Geophys. Res. 1982, 87, 3086. (22) Reifenha¨user, W.; Heumann, K. G. Determinations of methyl iodide in the Antarctic atmosphere and the south Polar Sea. Atmos. Environ. 1992, 26A, 2905. (23) Cogema. Rapport Environnement; Cogema: La Hague 2003. (24) Schnabel, C.; Lo´pez-Gutie´rrez, J. M.; Szidat, S.; Sprenger, M.; Wernli, H.; Beer, J.; Synal, H.-A. On the origin of 129I in rain water near Zu ¨ rich. Radiochim. Acta 2001, 89, 815-822. (25) Jenk, T.; Ga¨ggeler, H. W.; Schwerzmann, A. A.; Schwikowski, M. Recovery of a 150 m ice core down to bedrock from the Fiescherhorn glacier. In: PSI, Laboratory for Radiochemistry and Environmental Chemistry, Annual Report; Paul Scherrer Institut: Villigen, 2002; p 25. (26) Jenk, T.; Ga¨ggeler, H. W.; Palmer, A. S.; Schwikowski, M.; Leuenberger, M. Temperature reconstruction from the Fiescherhorn 2002 ice core. In: PSI, Laboratory for Radiochemistry and Environmental Chemistry, Annual Report; Paul Scherrer Institut: Villigen, 2004; p 32. (27) Palmer, A. S.; Schwikowski, M.; Leuenberger, M. Melt-water percolation events in the Fiescherhorn 2002 ice core. In: PSI, Laboratory for Radiochemistry and Environmental Chemistry, Annual Report; Paul Scherrer Institut: Villigen, 2004; p 28. (28) Ernst, T.; Szidat, S.; Handl, J.; Jakob, D.; Michel, R.; Schnabel, C.; Synal, H.-A., et. al. Migration of 129I and 127I in soils. Kerntechnik 2003, 68 (4), 155-167. (29) Gray, J.; Jones, S. R.; Smith, A. D. Discharges to the environment from the Sellafield Site, 1951-1992. J. Radiol. Prot. 1995, 15 (2), 99-131. (30) Schwikowski, M.; Bru ¨ tsch, S.; Ga¨ggeler, H. W.; Schotterer, U. A high-resolution air chemistry record from an Alpine ice core: Fiescherhorn glacier, Swiss Alps. J. Geophys. Res. 1999, 104 (D11), 13709-13719. (31) Fuge, R.; Johnson, C. C. The geochemistry of iodine - a review. Environ. Geochem. Health 1986, 8, 31-54. (32) Schwikowski, M. Reconstruction of European air pollution from Alpine ice cores. In: Earth Paleoenvironments: Records preserved in Mid- and Low-Latitude Glaciers; Cecil, L. D.; Green, J. R.; Thompson, L. G. Eds.; Kluwer Academic Publishers: Dordrecht. 2004, pp 95-119. (33) Preunkert, S.; Legrand, M.; Wagenbach, D. Sulfate trends in a Col du Dome (French Alps) ice core: A record of anthropogenic sulfate levels in the European midtroposphere over the twentieth century. J. Geophys. Res. 2001, 106, 31991-32004. (34) BNFL. Annual Report on Discharges and Monitoring of the Environment. BNFL: Sellafield, 2003.

Received for review March 11, 2006. Revised manuscript received June 24, 2006. Accepted July 21, 2006. ES0605725