Environ. Sci. Technol. 2001, 35, 2171-2177
The Uptake of Iron-55 by Marine Sediment, Macroalgae, and Biota Following Discharge from a Nuclear Power Station P . E . W A R W I C K , * ,† A . B . C U N D Y , ‡ I. W. CROUDACE,† M. E. D. BAINS,§ AND A. A. DALE§ Southampton Oceanography Centre, Southampton, SO14 3ZH, U.K., School of Chemistry, Physics and Environmental Science, University of Sussex, Brighton, BN1 9QJ, U.K., and Formerly UKAEA, Winfrith, Dorchester, Dorset, DT2 8DH, U.K.
Significant quantities of 55Fe, an activation product of stable iron, have been released into the environment following the atmospheric testing of nuclear weapons (mainly in the 1950s and 1960s) as well as through authorized discharges of radioactivity from nuclear power and reprocessing sites. Although some studies have been performed on the behavior of weapons’ fallout-derived 55Fe in the environment and subsequent impact on humans, little has been published on the behavior of 55Fe released as a point source discharge from nuclear sites. This study presents data on the concentration and temporal variation of 55Fe in fucoid seaweeds, shellfish, crab, and lobster collected from Weymouth Bay and adjacent coastal areas, southern England. These areas have received authorized discharges of radionuclides originating from the operation of a nowdecommissioned steam-generating, heavy water-type reactor at AEE Winfrith. The highest activities of 55Fe are found associated with marine sediments collected near the discharge pipeline and a rapid decline occurs away from the pipeline. This is consistent with rapid sorption of 55Fe by the sediment, and the data show there is only limited reworking and remobilization. Activities of 55Fe in biota generally decreased over time, due to a reduction in the amount of 55Fe discharged. The variation of 55Fe activity, revealed from the monthly sampling of seaweed, does not reflect the short-term fluctuations seen in the patterns of discharged 55Fe activity. Although discharges of 55Fe from AEE Winfrith exceeded other radionuclides, the radiological impact on local seafood consumers is considerably less than for other key radionuclides such as 60Co and 65Zn but of comparable magnitude to the global average population dose arising from fallout-derived 55Fe.
Introduction Radioactive contamination in the marine environment has predominantly originated from atmospheric fallout associated with nuclear weapons testing programs and from point Corresponding author phone: +44 (0) 2380596600; fax: +44 (0)23 80596450; e-mail:
[email protected]. † Southampton Oceanography Centre. ‡ University of Sussex. § Formerly UKAEA. 10.1021/es001493a CCC: $20.00 Published on Web 05/04/2001
2001 American Chemical Society
source discharges from nuclear facilities. Radionuclides derived from global atmospheric fallout are deposited at low activities across wide areas of the open ocean, whereas those from nuclear facilities are frequently discharged in concentrated form into the turbid, shallow coastal zone. Hence the radionuclides released from these sources may show markedly different behavior in the marine environment. The source of discharge to the ocean (global fallout or point source release) and the characteristics of the receiving environment (surface open ocean or coastal shelf sea) may influence radionuclide uptake by marine organisms and may be important when evaluating transfer through the foodchain and the resultant dose to humans. Iron-55 (t1/2 ) 2.7 years) has been released into the environment globally through atmospheric nuclear weapons testing mainly during the 1950s and 1960s. During these tests, 55Fe was produced following neutron capture by stable Fe present in weapons’ construction materials and soil in the vicinity of the test. It has been estimated that 50 MCi (1.85 EBq) of 55Fe was released into the environment as a result of tests in 1961 and 1962 (1). Iron-55 decays via electron capture to 55Mn with the emission of Auger electrons and low-energy X-rays (5.89 keV, 16.2%). Dosimetric studies of weapons’ fallout-derived 55Fe were performed in the late 1960s and 1970s by which time the 55Fe activities had decayed to undetectable levels. Of particular interest was the deposition of 55Fe on lichen and its subsequent transfer to reindeer and caribou and ultimately to Finnish Lapps and Eskimos (2). Globally, however, the average dose from 55Fe was typically much lower than for other fallout-derived radionuclides e.g. the average dose from 55Fe was 2 orders of magnitude lower than for 90Sr (1). A number of studies also measured the activity of 55Fe in rainwater, seawater, and in certain marine biota, particularly tuna where activities of 82 000 dpm/g Fe (1370 Bq/g Fe) were recorded (3). However, activity levels of 55Fe in ocean sediments were found to be small as a result of particle settling rates that were low in comparison with the physical half-life of 55Fe. More recently, 55Fe has been routinely released as part of the authorized discharges from nuclear power stations and reprocessing plants. In contrast to weapons fallout-derived 55Fe, however, there are no published data on the marine behavior of nuclear reactor-derived 55Fe in the wider scientific literature. Rapid scavenging and accumulation of 55Fe by sediments in the vicinity of a discharge point may be expected, depending on the degree and stability of complexation, followed by transfer of a significant proportion to benthic filter feeders. This study has investigated the marine biogeochemistry of 55Fe discharged from a nuclear reactor in southern England and assesses its uptake and accumulation by marine organisms in the coastal environment. The differences in behavior between weapons’ fallout-derived and nuclear power station-derived 55Fe and its contribution to critical group dose are also considered. The Study Area. Measurements of 55Fe activities were made in a wide range of biota over a 6-year period as part of an unpublished study performed by the then United Kingdom Atomic Energy Authority at Winfrith, U.K. The AEA Winfrith site in Dorset (Figure 1) began discharging radioactive effluent in 1970. Wastes were produced during the operation of the experimental steam-generating heavy water reactor, other on-site research reactors, fuel inspection facilities and other supporting operations, a significant proportion of which were subsequently collected as liquid effluent wastes. After a period of holding and dilution, the effluents were discharged under authorization into the VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2171
FIGURE 1. Location of sampling sites. marine environment via a 3 km pipeline into Weymouth Bay. This effluent mainly consisted of the major activation products of Fe, Co, Cr, Mn, Ni, and Zn released following annual cleaning of the SGHW reactor cooling circuit. Discharges peaked during the 1980s as the operations at the site increased. In 1990 the radionuclide discharges fell dramatically following the shut-down and subsequent decommissioning of the SGHW reactor.
Methodology A range of sample types was collected in the period 19891993. Seaweed samples were washed in seawater at the sampling site. Fresh growth (clearly distinguishable by its lighter color) was removed for analysis. Marine fauna were prepared as for human consumption; viz. oysters and scallops were not cooked. Crab, lobster, whelks, cockles, and winkles were cooked by boiling in water for approximately 1 h with only the edible parts of the organism being analyzed. All samples were dried at 110 °C and then ignited at 450 °C. Ignited samples were ground to produce a homogeneous sample prior to analysis. Unlike the biota samples a time-series of sediment samples was not available for this study. However, a survey of marine sediments was conducted in August 1990. Sediment samples were collected by divers while performing routine inspection and maintenance of the pipeline from predetermined locations in a radius of 300 m around the end of the discharge pipeline (Figure 2). Samples were oven-dried and ground prior to analysis of 55Fe and stable Fe. Approximately 5 g of ashed biota or dried sediment was leached for 1 h with 80 mL of 6 M HCl and 1 mL of concentrated HNO3. The sample was cooled and filtered through a Whatman No. 40 filter paper. The filtrate was diluted to a known volume, and an aliquot was removed for stable Fe determination. Fe was extracted into two 30 mL aliquots of a 3:1 mixture of ethyl acetate and butyl acetate, and the aqueous fraction was discarded. The two extracts were combined, and the Fe was back-extracted into two aliquots of 25 mL of 2 M HNO3. The nitric acid fractions were combined and neutralized with ammonia to form Fe(OH)3. The precipitate was isolated by centrifugation and washed with dilute ammonia solution. All washings were discarded. The Fe(OH)3 precipitate was dissolved in a minimum of concentrated HCl and diluted to 1 mL with 6 M HCl. The solution was transferred to an anion exchange column 2172
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
FIGURE 2. Schematic of marine sediment sampling locations relative to the discharge pipeline outlet at P12 (center) - August 1990 survey. (Dowex 1 × 8, 4 × 0.4 cm i.d. conditioned with 5 mL of 6 M HCl). The centrifuge tube was washed with 2 × 1 mL of 6 M HCl which was transferred onto the anion exchange column. The column was washed with 10 mL of 6 M HCl, and the Fe was eluted with 10 mL of 0.5 M HCl. The sample was diluted to a known volume, and an aliquot was removed for stable Fe determination. The remaining solution was evaporated to incipient dryness. Samples were counted by liquid scintillation analysis using a Wallac 1411 liquid scintillation counter. The purified Fe fraction was dissolved in 0.2 mL of 2 M H3PO4 and 0.8 mL of water. The sample was transferred to a 20 mL polythene scintillation vial with 2 × 1 mL of water, and 15 mL of Optiphase Hisafe 3 liquid scintillant cocktail was added (for a full discussion of this technique see Warwick et al. (4)). In some samples 0.2 mL of H3PO4 was insufficient to produce a colorless solution. In such cases, more 2 M H3PO4 was added while ensuring that the total aqueous volume including washings was maintained at 3 mL. Stable Fe was determined using either ICP-AES or a spectrophotometric method based on 1,10-phenanthroline (5)).
Results and Discussion General Fate of Discharges. Discharges of 55Fe from AEA Winfrith were higher than for any other radionuclide discharged to sea. In 1987 32.3 TBq of 55Fe was discharged into Weymouth Bay compared with 4.2 TBq of 60Co (the next
TABLE 1. Discharges of 55Fe from AEE Winfritha year
TBq 55Fe discharged
year
TBq 55Fe discharged
1987 1988 1989 1990 1991 1992
32.3 2 2.7 1.24 0.217 0.019
1993 1994 1995 1996 1997
0.011 0.0054 0.0042 0.0084 0.0037
a
Source: official records of UKAEA Winfrith.
FIGURE 3. Spatial distribution of 55Fe activity on marine sediments along transects - values in Bq/kg dry wt (see Figures 1 and 2 for sediment sampling locations). Samples collected in August 1990. highest discharged radionuclide). Subsequent discharges of 55Fe were significantly reduced, and in 1990 the SGHW reactor was finally shut down leading to further reductions in 55Fe discharge (Table 1). Under aerobic marine conditions iron, as Fe3+, rapidly reacts with hydroxyl groups to produce the insoluble Fe(OH)3 species (ksp ) 10-38M). Iron-55 therefore should rapidly precipitate and be deposited in the vicinity of the discharge point. If this is the case, marine biota must therefore derive any 55Fe from the very low levels in seawater or directly from the particulate phase. Measurement of 55Fe in seabed sediment, collected as part of a Winfrith marine sediment survey in August 1990 (Figure 3), showed that 55Fe activities were at least 6 times higher (P12 and F1) in the immediate vicinity and 100 m east of the pipeline compared with the rest of the area surveyed. Slightly higher 55Fe activities were observed for samples collected in the northern sector compared to those collected to the south of the pipeline, consistent with a predominant onshore transport of sediment. Uptake of Fe by Seaweeds. Fucoid seaweeds have been widely used as bioindicators for monitoring radionuclide concentrations in seawater, and concentration factors of greater than 104 are found for many radionuclides (6, 7). Fucus serratus is widespread in the Weymouth Bay area and samples were systematically collected from three locations (Figure 1). Sample details included a record of the location of the plant on the shoreline, the portion of the thallus taken, and the physiological state of the plant.
FIGURE 4. Temporal variation of stable Fe concentration in seaweeds sampled from Weymouth, Kimmeridge, and Swanage (see Figure 1 for site locations). Vertical lines mark the 1st January of each year.
TABLE 2. Variation of Fe Concentration in Seaweeds Weymouth Kimmeridge Swanage
average Fe ppm
range Fe ppm
42 45 83
18-80 9.5-131 36-173
Stable iron concentrations in seaweeds collected from the three sites varied markedly throughout the year (Figure 4). Clear seasonal or annual trends were not apparent, although concentrations of stable Fe were higher in seaweed collected from Swanage compared with those collected from Weymouth and Kimmeridge (Table 2). A wide variation in Fe concentrations for fucoid seaweeds collected from different locations has been previously shown (9). This variation was also found in this study (Figure 5). The rapid rise and decline for both 55Fe and Fe in the seaweed indicate a fast turnover of Fe in the seaweed thallus. As only the thallus tips were analyzed, it is possible that the 55Fe was translocated to the older parts of the frond known to have a higher storage VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2173
FIGURE 5. Variation of 55Fe activity in seaweeds with time from three locations. FIGURE 7. Ratios of 55Fe/60Co in Winfrith discharge and seaweed samples.
FIGURE 6. 55Fe activity (normalized to stable Fe) in seaweeds with time from three locations. capacity for trace metals. Normalization of the 55Fe to stable Fe concentration was useful in revealing some trends (Figure 6). A peak in specific activity of 0.25 Bq/mg Fe was found in one seaweed sample collected from Kimmeridge in July 1990. This elevated activity was not observed at other sites. Elevated specific activities of 55Fe (between 0.20 and 0.32 Bq/mgFe) were observed at all three sites over the period MarchOctober 1991. An activity peak was also observed in Weymouth seaweed (at 0.25 Bq/mgFe) in February 1992 although this was not seen in seaweeds collected from the other sites. Although direct transfer of metal ions from sediment to seaweed (via complexation with algal polyphenols excreted from the seaweed frond onto adhering particulate matter) has been postulated for certain elements, no correlation between the concentration of Fe in sediments and that in seaweed has been shown (10). The main source of Fe in the seaweeds must therefore be from soluble species in the seawater (such as humic acid-complexed Fe) or from colloidal iron hydroxide (8). Little is known about the exact chemical form of Fe in the discharge. However, picolinic acid was used to decontaminate pipework within the SGHW reactor and a proportion of the discharged Fe may have been complexed with this acid. The Fe-picolinate complex could potentially be more soluble in seawater and be transported in the aqueous phase. Such a process has been postulated for the dispersion of 60Co from the discharge point (11), and the wider distribution of 60Co along the south coast of England is consistent with the presence of some of the 60Co in a complexed, soluble form (12). Rapid scavenging of 55Fe onto sediments, however, is indicated by the localization of 55Fe activity on sediments collected from the immediate vicinity of the outfall point (see above) and by the ratio of 55Fe to 60Co in the seaweed samples (Figure 7). The 55Fe/60Co activity ratio in seaweeds prior to 1991 is typically < 0.5 compared with a ratio in Winfrith discharges in excess of 5, indicating much more rapid removal of 55Fe than 60Co from the water column. From 1991 onward 55Fe/60Co values in seaweeds are higher than found previously. In particular high values of 55Fe/60Co were observed in all three seaweed sampling locations between March 1991 and October 1991 with isolated peaks 2174
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
occurring in individual locations during 1992. Elevated 55Fe/60Co values also occur during mid-1991 in some species which graze the seaweed and seaweed detritus (e.g. whelks and spider crabs), although the comparatively poor sampling resolution for marine fauna means that trends over monthly periods are difficult to discern (see section below). The high 55 Fe activities and 55Fe/60Co ratio observed in seaweeds between March and October 1991 were not associated with any elevated discharge. In addition, a significantly higher discharge of 55Fe which occurred in March 1990 (Figure 6) was not reflected in the seaweed 55Fe activities (Figure 5), showing that there is no clear correlation between the activity of 55Fe discharged and that found in seaweed samples. Sudden sediment reworking through storm action, anthropogenic disturbance, early diagenetic remobilization, analytical error, and input from COGEMA-La Hague were all discounted as possible explanations for the elevated 55Fe activities and 55Fe/60Co ratio between March and October 1991. Weather records do not indicate any unusual storm events occurring during this time (data from the U.K. Meteorological Office), and there is no evidence of anthropogenic disturbance around the discharge pipeline during 1991 (L. Chance, AEA Winfrith, personal communication). Early diagenetic remobilization of sediment-bound 55Fe would be unlikely to release 55Fe over such a clearly defined period (rather, this process would be expected to be more continuous), while analytical error can also be discounted as the samples showing the elevated 55Fe activities were analyzed in separate analytical runs over a period of a few months. 55Fe input from the French reprocessing facility at COGEMA La Hague, approximately 100 km from the study area on the Cherbourg peninsula, is not consistent with prevailing sediment and water transport pathways in the Channel (13). The lack of characteristic La Hague products (e.g. 125Sb) in seaweed samples from the Weymouth Bay region also indicates that the influence of La Hague is comparatively minor. In the absence of a likely environmental or analytical process that might explain the elevated 55Fe/ 60Co ratio in 1991, it seems likely that a change in the discharge form of 55Fe may have been responsible. The elevated 55Fe/60Co ratio in 1991 coincides with the commencement of decommissioning of the SGHW reactor on the Winfrith site. This decommissioning operation may have altered the proportion of 55Fe in the water-soluble fraction and hence affected the environmental behavior of 55Fe. It is not possible to fully evaluate this hypothesis because there is no detailed chemical characterization of the effluent. There is therefore no clear explanation, from the available data, for the increase in 55Fe activity in seaweeds during March to October 1991. Uptake of Fe by Marine Fauna. Samples of marine fauna collected during the sampling period included brown crab, spider crab, lobster, whelks, winkles, scallops, prawns, cockles, and plaice. These samples were collected as part of the routine environmental monitoring program conducted
TABLE 3. Typical Fe Concentrations in Marine Biota Collected as Part of the Current Study sample type brown crab
plaice scallops meat scallops guts whelks lobster spider crabs cockles oysters squid winkles
location Portland Chapman’s Pool Lulworth Swanage Weymouth Lulworth Lulworth Poole Lulworth Chapman’s Pool Lulworth Chapman’s Pool Poole Poole Weymouth Poole Kimmeridge
feeding habit detrital scavengers
suspension feeder predators (barnacles, mussels, etc.) detrital scavengers detrital scavengers predators (barnacles, mussels, etc.) active predators grazers
mean Fe ppm
SD
IAEA CFe
18 20 23 22 6 70 95 21 51 12 17 38 29 96 30 14 125 156
8 3 13 6 3 54 53 7 54 11 12 18 9 45 8 10 60 61
5 × 103 a
3 × 103 b 3 × 104 c
2 × 103 d
a General value for crustaceans assuming a mean wet weight Fe concentration of 10 ppm. b General value for fish (highly dependent on species and organs included in the analysis). c General value for molluscs assuming a mean wet weight Fe concentration of 184 ppm. It was noted that there was a wide range of Fe concentrations in different species. d General value for cephalopods assuming a mean wet weight Fe concentration of 3.2 ppm. e Concentration factors from ref 14.
at AEA Winfrith and were prepared as normally required prior to human consumption with only edible parts being dissected and reserved for analysis. In addition, many of the samples were cooked, potentially resulting in a loss of Fe to the water that the samples were boiled in. Results therefore do not necessarily represent the total Fe content or 55Fe activities present in a species but do indicate the likely input of stable Fe and 55Fe into the human foodchain. Mean stable Fe concentrations were highest in winkles (156 ppm) and lowest in plaice (6 ppm) (Table 3). Stable Fe concentrations were similar in identical species collected from different locations but show considerable variability over time. 55Fe activities are also erratic, even when corrected for stable Fe concentration. The ratio of 55Fe/stable Fe was highest in whelks and in spider crab and highest at sites closest to the discharge pipeline (Lulworth and Chapman’s Pool) (Figures 8 and 9). The highest activities per kilogram dry mass of biota was also observed in whelks at Lulworth, presumably due to the limited migration of the organism increasing exposure to localized contaminated sediments in the vicinity of the pipeline. Winkles collected at Kimmeridge showed a low 55Fe activity even though they contained the highest levels of stable Fe. As this species was collected from intertidal areas it is likely that exposure to 55Fe-labeled seabed sediments is lower than for other benthic species and that the transfer of 55Fe through fauna to winkles is limited. In general, the specific activities found in biota collected in the vicinity of the pipeline were comparable with the highest activities reported in tuna resulting from weapons’ fallout (1.37 Bq/mg Fe (3)). The observed high activities of 55Fe for whelks compared with other biota is in agreement with published concentration factors (Table 3). While 55Fe (Bq/mg Fe) in marine fauna shows considerable variability over short time scales, most species show a general decline in 55Fe activity between 1988 and 1993, due to the reduction of 55Fe discharges over this period (Figure 9). The most prominent trends were observed for brown crabs, spider crabs, and whelks. The 55Fe specific activity was higher in these organisms compared with other species collected in the same area, and the dataset is more extensive. The rate of this decline can be assessed by calculating the so-called “effective half-life” using linear regression analysis of the natural logarithm of the activity versus time. Effective halflives for 55Fe for different organisms range from 0.65 to 1.20 years (Table 4). 55Fe activities are decay corrected to the date
FIGURE 8. Variation of 55Fe activity per mg Fe [A] and 55Fe activity concentration [B] for a range of species collected from Lulworth Cove in 1990-1991. of sample collection, hence the effective half-life includes the decline in activity due to physical decay of 55Fe (t1/2 ) 2.7 years - Jef PC, 1996). Only datasets where the linear regression fit was significant at the 95% confidence level are included. The effective half-life calculation assumes complete cessation of discharge at the beginning of 1988 and so may overestimate the true effective half-life of 55Fe due to continued, albeit considerably reduced, discharges after January 1988. The general decline in 55Fe activity between 1988 and 1993 in marine fauna contrasts with the seaweed data, which are characterized by rapid fluctuations in 55Fe activity, and which shows no clear trend in activity over the same period (see above). The likely uptake route of 55Fe in marine fauna is a combination of adsorption from seawater and ingestion of particulate matter. The nature of uptake will differ between species due to differences in physiology and feeding habits, although the effective half-lives calculated for brown crabs, spider crabs, and whelks show little evidence for speciesdependence. A consequence of uptake both from seawater and from particulate material is likely to be the continued absorption of 55Fe even after cessation of discharge, due to VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2175
FIGURE 9. Variation in 55Fe activities with time in various processed species collected from different locations.
TABLE 4. Effective Half-Lives of 55Fe for Selected Species and Locationsa species
location
effective half-life (years)
range (2 SD) years
brown crab brown crab brown crab spider crab whelks
Chapman’s Pool Lulworth Cove Swanage Lulworth Cove Lulworth Cove
0.82 0.92 1.20 0.65 0.70
0.55-1.67 0.68-1.42 0.66-6.81 0.39-1.98 0.55-0.97
a
See text for details.
ingestion of 55Fe-labeled particles. Hence 55Fe activity variations with time in marine fauna will show a partially timeintegrated record of 55Fe discharges from Winfrith and consequently a slower decline of 55Fe than in seaweed, where 55Fe uptake is dominantly from solution. Cundy et al. (12) found a similar reduction in AEA Winfrithderived 60Co and 65Zn over the period of 1988-1998 and 2176
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
used the same approach to calculate effective half-lives for these radionuclides in sediments, seaweed and marine fauna. While effective half-lives in marine fauna were similar to those reported in the current study, 60Co and 65Zn showed much longer effective half-lives (1-4 years) than 55Fe in seaweed. This difference may be caused by the more rapid and permanent removal of 55Fe to oxic coastal sediments. Comparison of Nuclear Reactor-Derived 55Fe and Weapons Fallout 55Fe. Although the activities of 55Fe in the Winfrith discharge were higher than for 54Mn, 60Co, and 65Zn, the CEDE (committed effective dose equivalent) for 55Fe is low (1.6 × 10-10 Sv/Bq by ingestion for an adult (15)). The contribution of this isotope to the critical group dose is therefore comparatively low. For Winfrith discharges to sea, the critical group is defined as local inhabitants who consume 210 g of fish, 110 g of whelks, and 70 g of crabs each day (16). Taking an average of the 55Fe activities measured for crab and whelks collected near Lulworth and plaice collected at Weymouth for 1990, the 55Fe dose to critical group is 0.32 µSv/year. This compares with an estimated total critical group dose from
all radionuclides in the Winfrith discharges of 14 µSv/year (16). The average dose from 55Fe from weapons fallout over the period 1954-1962 was 0.7 µSv (0.6 mRad to bone marrow (1)), assuming a radiation weighting factor, WT, of 1 and a tissue weighting factor, WR, of 0.12 (17)), although this would have been significantly higher for certain critical groups. Whereas weapons’ fallout-derived 55Fe was deposited globally and was subsequently diluted and dispersed, reactorderived 55Fe was discharged at a discrete point source near to the seabed. The present study shows that the Winfrith discharged 55Fe sorbed rapidly on to seabed sediments in the vicinity of the pipeline. Uptake by biota was predominantly observed in benthic scavengers and filter feeders and was less evident in intertidal biota such as seaweeds that are more likely to adsorb radionuclides from seawater rather than from particulate matter. Potential doses from power station derived 55Fe are therefore more likely to be concentrated in a limited number of local inhabitants having a diet rich in molluscs. As the 55Fe is dispersed over a limited area, doses per unit discharge are likely to be higher than for an equivalent activity released during weapons detonations. In both instances the doses to the general public, and to the respective critical groups, are extremely low in comparison with legal limits and doses received from other discharged radionuclides.
Acknowledgments The authors would like to thank A. Kingsbury, L. A. Chance, and D. Wickenden (AEA Technology, Winfrith) and R. Carpenter (AEA Technology, Harwell) for their help in this study. A. Neal and colleagues at UKAEA, Winfrith are thanked for reviewing the technical content of the paper and authorizing the inclusion of previously unpublished data. The U.K. Meteorological Office is thanked for access (via the British Atmospheric Data Centre) to UKMO surface meteorological data sets. H. Cox (AEA Technology, Winfrith) performed much of the sample collection and preparation, and J. Mudge and L. Barker (formerly of AEA Technology, Winfrith) performed some of the analytical work.
Literature Cited (1) UNSCEAR. Sources and effects of ionising radiation. United Nations Scientific Committee on the Effects of Atomic Radiation; United Nations, New York, 1977. (2) Jaakkola, T. Analysis of iron-55 produced by nuclear tests and its enrichment in Finnish Lapps; Annales Academiae Scientiarum Fennicae, Series A, 150, Suomalainen Tiedeakatemia, Helsinki, Finland, 1969. (3) Rama, Koide M.; Goldberg E. D. Nature 1961, 191, 162. (4) Warwick, P. E. Croudace, I. W.; Bains, M. E. D. Radioact. Radiochem. 1998, 9(2), 19. (5) Vogel, A. Textbook of quantitative inorganic analysis; Longman: London, 1978. (6) Dahlgaard, H.; Boelskifte S. J. Environ. Radioact. 1992, 16, 4963. (7) Boelskifte, S. J. Environ. Radioact. 1985, 2, 215-227. (8) Lobban, C. S.; Harrison, P. J.; Duncan, M. J. The physiological ecology of seaweeds; Cambridge University Press: Cambridge, U.K., 1985. (9) Fuge, R.; James, K. H. Mar. Poll. Bull. 1973, 9. (10) Luoma, S. N.; Bryan, G. W.; Langston, W. J. Mar. Poll. Bull. 1982, 13(11), 394. (11) Leonard, K. S.; McCubbin, D.; Harvey, B. R. J. Environ. Radioact. 1993, 20, 1. (12) Cundy, A. B.; Croudace, I. W.; Warwick, P. E. Environ. Sci. Technol 1999, 33 (17), 2841. (13) Guegueniat, P.; Bailly du Bois, P.; Salomon, J. C.; Masson, M.; Cabioch, L. J. Mar. Sys. 1995, 6, 483. (14) IAEA. Sediment Kds and concentration factors for radionuclides in the marine environment; Technical reports series No. 247; IAEA: Vienna, 1985. (15) NRPB-GS7. Committed doses to selected organs and committed effective doses from intakes of radionuclides; National Radiological Protection Board: Oxfordshire, 1987. (16) AEAT. Annual report on radioactive discharges from Winfrith and monitoring the environment 1990; Health & Safety Division, AEA Technology, Winfrith Technology Centre: Dorchester, U.K., 1991.
Received for review July 17, 2000. Accepted February 23, 2001. ES001493A
VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2177
