Deposition of artificial radionuclides in sediments of Loch Etive, Scotland

Hamza Al-Qasmia, Gareth T.W. Lawa, L. Keith Fifieldb, John A. Howec, Tim Brandc,

Gregory L. Cowied, Kathleen A. Lawa, Francis R. Livensa

a Centre for Radiochemistry Research, School of Chemistry, The University of Manchester,

Oxford Road, Manchester, M13 9PL, UK

b Department of Nuclear Physics, Research School of Physics and Engineering, The

Australian National University, Canberra, ACT 0200, Australia

c Scottish Association for Marine Science, Oban, Argyll, PA37 1QA, UK

d School of Geosciences, The University of Edinburgh, King's Buildings, Edinburgh, EH9

3FE

1

1

2

3

4

5

6

7

8

9

10

AbstractThe nuclear fuel reprocessing plants on the Sellafield site (UK) have released low-level

effluents into the Irish Sea under authorisation since 1952. This has led to the labelling of

nearby offshore sediments with a range of artificial radionuclides. In turn, these sediments act

as a long-term secondary source of both soluble and particle-associated radionuclides to

coastal areas. These radionuclides are of interest both in assessing possible environmental

impacts and as tracers for marine processes. Here we present results from a study of the

geochemistry of natural (234, 238U) and artificial (137Cs, 241Am, 238Pu, 239+240Pu, and 236U)

radionuclides and their accumulation in sediments from Loch Etive, Scotland. The data are

interpreted in the context of the historical radioactive discharges to the Irish Sea and

biogeochemical processes in marine sediments. Loch Etive is divided into two basins; a

lower, seaward basin where the sedimentation rate (~ 0.6 cm/yr) is about twice that of the

more isolated upper basin (~ 0.3 cm/yr). These accumulation rates are consistent with the

broad distribution of 137Cs in the sediment profiles which can be related to the maximum

Sellafield discharges of 137Cs in the mid-1970s and suggest that 137Cs was mainly transported

in solution to Loch Etive during that period. Enrichments of Mn, Fe and Mo in sediment and

porewater from both Loch Etive basins result from contemporary biogeochemical redox

processes. Enrichments of 238U and 234U in the lower basin may be a result of the cycling of

natural U. By contrast, the Sellafield-derived artificial isotope 236U does not seem to be

affected by the redox-driven reactions in the lower basin. The 238Pu/239,240Pu ratios suggest

contributions from both historical Sellafield discharges and global fallout Pu. The uniform

sediment distributions of Pu and Am, which do not reflect Sellafield historical discharges,

suggest the existence of a homogenous secondary source. This could be the offshore ‘mud

patch’ in the vicinity of Sellafield from which the supply of radionuclides reflects time-

2

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

integrated Sellafield discharges. This source could also account for the continuing supply of

Cs to Loch Etive, even after substantial reductions in discharge from the Sellafield site.

Keywords: Americium; uranium; plutonium; cesium; Sellafield

1 IntroductionThe behaviour of artificial radionuclides in marine sediments and their interaction with

biogeochemical processes is of fundamental importance in developing an understanding of

their long-term fate in the environment. Such understanding can be used to underpin both

radionuclide tracing of environmental processes and assessment of possible environmental

impacts of radionuclide releases to coastal waters, such as occurred in 2011 as a result of the

accident at the Fukushima Daiichi Nuclear Power Plant.

The absence of oxygen in sediments drives successive reduction of other terminal electron

acceptors (e.g. nitrate, Fe and Mn oxyhydroxides, and sulfate) by anaerobic microorganisms.

These processes could impact the fate of natural and artificial radionuclides in sediments

either by direct cycling of redox sensitive radionuclides by anaerobic microorganisms or by

interaction of radionuclides with biogeochemical redox cycles of other elements (Brookshaw

et al., 2012; Campbell et al., 2015; Kimber et al., 2012; Lovley et al., 1991; Malcolm et al.,

1990). Because of the biogeochemical gradients present, sea-lochs provide an ideal natural

laboratory to explore these interactions (Sholkovitz, 1983; Williams et al., 1988).

Loch Etive is a sea-loch on the west coast of Scotland with well documented geochemistry

and hydrography (Edwards and Edelsten, 1977; Overnell, 2002; Ridgway and Price, 1987)

which provides perfect conditions (high sedimentation rates, anaerobic conditions, restricted

exchange, and reduced erosion) to investigate the geochemical behaviour of natural and

artificial radionuclides. Sellafield, a nuclear site located on the NW coast of England, is the

dominant source of artificial radionuclides in the NE Irish Sea. Authorised radioactive

discharges of low level liquid effluents from Sellafield to the Irish Sea started in 1951 with a 3

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

maximum in the mid 1970’s (Figure 1) (Gray et al., 1995). Other sources, such as weapons

testing which had a peak input in 1963 (MacKenzie, 2000), and accidental releases,

particularly the Chernobyl accident in 1986 (Camplin et al., 1986; Mitchell et al., 1986) have

made smaller contributions to the radioactive inventory of the Irish Sea. Further, accidental

release of 238Pu from the burn-up of a US satellite in the atmosphere in 1964 increased the

activity ratio of 238Pu/239+240Pu from global fallout in the northern hemisphere from 0.024 to

0.036 (Hardy et al., 1973).

Figure 1. Total annual quantities of (a) 137Cs, (b) 241Am, and (c) 239,240Pu released from

Sellafield between 1952 - 1992 (data from Gray et al., 1995).

Due to the seawater circulation in the NE Irish Sea, the discharged radionuclides disperse

northwards (Jefferies et al., 1973; Mitchell et al., 1999). Soluble radionuclides such as Cs

mostly remain in seawater and are transported northwards along the Scottish coast out of the

Irish Sea (Dahlgaard, 1995; Jefferies et al., 1973; Kershaw and Baxter, 1995), whereas Pu

and Am (particle-reactive species) mainly associate with suspended particles and are then

focussed into areas of muddy sediment (the ‘mud patch’) close to Sellafield. This mud patch

then acts as a long-term secondary source of soluble and particle-associated artificial

radionuclides, the composition of which are time-integrated due to sediment mixing

processes (Brown et al., 1999; Kershaw et al., 1984; MacKenzie et al., 1998). The ingrowth

of 241Am from decay of 241Pu (t1/2 = 14.3 years) has also dominated the direct discharges of

241Am from Sellafield since the late 1970s (Hunt et al., 2013).

4

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

Uranium occurs naturally almost entirely as a mixture of three isotopes: 238U, 234U and 235U

but natural 236U also exists in the environment in ultra-trace quantities with a 236U/238U atom

ratio of ~ 10-14 (Zhao et al., 1997). Uranium is generally soluble in oxic seawater and present

as dissolved U(VI) complexed with carbonate (Cochran et al., 1986), which may also inhibit

reduction to U(IV). Diffusion of seawater to the underlying sediments may result in the

reduction of U(VI) to insoluble U(IV) at about the depth of Fe(III) reduction, which can lead

to accumulation of U at depth in the sediment (Barnes and Cochran, 1990; Cochran et al.,

1986). The Sellafield discharges to sea contain irradiated U which would be masked by the

natural U in seawater (Zhao et al., 1997) and historical, non-nuclear discharges to the Irish

Sea from the Marchon phosphate plant on the Cumbrian coast (Kershaw et al., 1990).

However, irradiated U is unique in its enhanced level of 236U (t1/2 = 2.3 x 107 years) (Zhao et

al., 1997) which provides a clear signature for irradiated U and hence, in this setting,

Sellafield-derived U, and can be used as a tracer (Al-Qasmi et al., 2016; Marsden et al.,

2006). However, 236U has not been widely used as a tracer of environmental processes due to

the difficulties of detecting its low mass or activity concentrations with traditional methods.

Accelerator mass spectrometry can overcome these difficulties (Eigl et al., 2013; Zhao et al.,

1997).

The inventory of naturally occurring 210Pb (t1/2 = 22.3 years) in marine sediment has two

components: supported and unsupported. The supported Pb is generated in situ by decay of

226Ra contained in the sediment minerals themselves. The term ‘supported’ reflects the fact

that this component is in radioactive equilibrium with its parent isotopes and its activity

concentration will therefore not change over the timescales of interest. The unsupported Pb

originates from decay of 222Rn and this accumulates in surface sediments. Since it is not in

radioactive equilibrium with its parent isotopes, the rate at which unsupported 210Pb decays

allows estimation of the rate at which the sediment accumulates (Swan et al., 1982). The 210Pb

5

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

dating method has been widely used to date cores from sea-lochs from the west coast of

Scotland (Krom et al., 2009; Teasdale et al., 2011).

The main aim of this paper is to investigate relationships between artificial radionuclides

released from Sellafield over a well-defined time period and the biogeochemical cycles which

are well documented in Loch Etive, the study site. This will provide deeper understanding of

the behaviour of natural and artificial radionuclides in the marine environment, underpinning

both assessments of possible environmental impacts and radionuclide tracing of

environmental processes.

2 Methods

2.1 Study Site and Sampling

Loch Etive is a sea-loch on the west coast of Scotland which can essentially be divided into

two distinct parts, a lower and upper basin (Figure 2). The seaward, lower basin has

a maximum depth of ~ 70 m and experiences tidally-induced mixing, leading to

an oxygenated water column (Overnell et al., 2002). The upper basin is separated from the

lower by a submerged sill of 13 m depth and has a maximum depth of 145 m (Howe et al.,

2002). In contrast to the lower basin, which is tidally mixed, only the surface water of the

upper basin is tidally exchanged. As a result, the bottom water of the upper basin is only

exchanged irregularly depending on freshwater input, with a mean repetition time of

16 months (Edwards and Edelsten, 1977) and as a result it is often hypoxic. These

environmental conditions are also reflected in the underlying sediments, which generally

show oxic conditions in the top 5 cm of surface sediments in the lower basin, and suboxic

conditions in top 1 cm in the upper basin (Ridgway and Price, 1987). The upper basin surface

sediment is high in Mn beneath the generally hypoxic waters, contrasting with lower surficial

Mn in the lower basin beneath water which is usually well mixed (Overnell et al., 2002).

Malcolm (1985) studied the geochemistry of Mo in both basins of the loch and established

6

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

an association of Mo geochemistry with the cycling of organic matter, Mn and sulfur.

Previous studies of radioactivity in Loch Etive have used 210Pb and 137Cs as dating tools and

showed sedimentation rates between 0.3 and 0.9 cm/yr (Krom et al., 2009; Ridgway and

Price, 1987; Shimmield, 1993). A study of 25 surficial sediment samples from the length of

the loch showed that Am and Cs had recorded activity concentrations between 1.9 and 8.0

Bq/kg of 241Am and between 88 and 353 Bq/kg of 137Cs (Williams et al., 1988).

Figure 2. Right: Locations of Sellafield and the study site, Loch Etive. The Irish Sea “Mud

Patch” is located just offshore from Sellafield. Left: Map of Loch Etive and its bathymetric

depth profile (Howe et al., 2002) showing the sampling stations in the lower and upper

basins.

For this study, two sediment cores were collected from the near deepest point in the lower

basin (~ 67 m) (Figure 2) in November 2013 using a megacorer. The first core (30 cm) was

7

131

132

133

134

135

136

137

138

139

140

141

142

143

144

sliced into 25 sections in a nitrogen-filled atmosbag (20 sections from the top 20 cm and 5

sections from the next 10 cm). Porewaters (> 3 ml per sample) were then extracted via

centrifugation (~ 1600 g), filtered through 0.22 µm Acrodisc syringe filters, and preserved

via addition of 100 µl HNO3 (ARISTAR® Grade). The sediment samples were freeze-dried

and then disaggregated with a mortar and pestle to obtain homogeneous fine samples which

were used to conduct all further analysis except gamma spectroscopy. The second core (28

cm) was sliced into 24 sections (20 sections from the top 20 cm and 4 sections from the next

8 cm) which were oven-dried at 40 °C, then disaggregated by mortar and pestle and used for

gamma spectroscopy. Another sediment core was collected from the deepest point (142 m) in

the upper basin (Figure 2) in June 2014 using a Sholkovitz corer. The core and sediment /

porewater samples were processed similarly to the first core from the lower basin.

2.2 Radionuclide Measurements

Samples (2 g dry) were spiked with a known activity of internal standards and digested using

aqua regia. Uranium and Pu isotopes were chemically separated using UTEVA and AG1-X8

resins. The U separation procedure used was based on extraction chromatography (Eichrom

Technologies, 2005). The Pu separation procedure used was based on ion exchange

chromatography (Keith-Roach, 1998). Certified 232U and 242Pu standards (AEA Technology,

Harwell and NPL Teddington, respectively) were used as internal standards in α-

spectrometry. Uranium and Pu were prepared as sources for alpha spectroscopy by

electroplating onto stainless steel planchettes. The alpha sources were counted using passivated

implanted planar silicon (PIPS) detectors (Canberra, Belgium, model A450), with counting times

of up to two weeks, as required, to obtain suitable counting statistics. Uranium (234 and 238)

and Pu (238, 239+240) activities were obtained from alpha spectra. The 235U/238U atom ratio

was measured by ICP-MS, and the 236U/234U atom ratio by accelerator mass spectrometry

(AMS; see 2.2.1 below for full details). Gamma spectrometry was performed using an Ortec

LO-AX 51370/20-S hyper-pure germanium (HPGe) detector housed in a 5 cm thick lead shield

8

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

(15 -1200 keV). The activities of 210Pb, 241Am, and 137Cs were evaluated from the gamma ray

peaks at 46.5, 59.5 and 661.6 keV respectively by employing γ-spectrometry standards. Here,

matrix-matched standards were prepared by spiking one of the Loch Etive samples with

known activities of certified 210Pb, 241Am, and 137Cs standards (AEA Technology, Harwell).

The standards were then counted in the same geometry as the samples. The 210Pb data was

used to date the cores using the following equation and assuming that the initial activity of

210Pb and the rate of sediment accumulation are constants (Faure, 1977):

t=2.303λ log( Pb210 0

Pb210 )A

where: 210PbA = activity of 210Pb per unit weight of sample at depth h, 210PbA0 = activity of

210Pb at the surface (h = 0), λ = decay constant of 210Pb (3.11 X 10-2 y-1) and t = age of the

sample.

2.2.1 236U/234U ratio Measurement

The AMS samples were prepared using the procedure described by Marsden et al. (2001)

with slight modification (Al-Qasmi et al., 2016). The U was separated as described above

with no tracer added to the sample. The purified U was eluted from the separation column by

15 ml 1 M HCl which was reduced in volume to ~ 3 ml and then 1 ml of iron (III) nitrate

nonahydrate in 0.1 M HCl (5 mg Fe per ml) was added. This mixture was evaporated to

dryness on a hotplate and 15 ml 1 M HNO3 was added to remove any residual chlorides. This

was then taken to dryness and baked on a hotplate for 1 h at 250 °C to convert the iron nitrate

into iron oxide. The oxide residue was placed in a furnace at 450 °C for 8 h to ensure

complete conversion to oxide. The oxide was mixed with 1 mg of aluminium powder and the

mixture packed into an aluminium sample holder. Samples were then loaded into a 32-sample

wheel for insertion into the ion source of the AMS system. The measurements were

performed using the 14UD pelletron accelerator at the Australian National University. The

9

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

detection system is based upon combined high-resolution velocity measurement by time-of-

flight and energy measurements (Fifield, 2008; Fifield et al., 2013). The counting times per

cycle were 1 minute at 234U followed by 5 min at 236U. This cycle was performed twice for

each sample, ending with a 234U count.

2.3 Stable Metal Measurements

Homogenised, freeze-dried sediment samples were totally digested in acid matrices (cHCl,

cHNO3 and cHF in a ratio 6:3:2) assisted by a microwave digestion system (CEM Mars)

based on that of Law et al. (2009). After digestion, the supernatant was evaporated to

incipient dryness on a hotplate and the residue was dissolved in 30 ml 2 % HNO3. Inductively

coupled plasma mass spectrometry (ICP-MS, Agilent 7500cx) was used to analyse U and Mo

while inductively coupled plasma atomic emission spectroscopy (ICP-AES, Perkin-Elmer

Optima 5300) was used to analyse Fe (238.2 nm) and Mn (257.6 nm) in the porewaters and

the digestion supernatants.

3 Results and Discussion

3.1 Stable Metal Profiles

The metal concentrations in porewater and solid phases are plotted against the core depth in

the upper and lower basin cores (Figure 3). Iron and Mn data are similar to those previously

reported for Loch Etive (Malcolm, 1985; Overnell, 2002). Solid phase Fe data show

approximately uniform profiles in both the lower basin (4.2-5.3 wt%) and the upper basin

(3.9-4.6 wt%) whereas porewater Fe shows concentration changes with depth at both sites.

In the lower basin, solid phase Mn decreased from ~ 0.33 wt% at the surface to ~ 0.06 wt% at

5 cm depth, which was then maintained to the bottom of the core (30 cm depth). There was

also an increased Mn concentration in the porewater profile between 2 and 9 cm. In the upper

basin, the solid phase Mn concentration decreased with depth from ~ 0.53 wt% at the surface

10

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

to ~ 0.23 wt% at 12 cm and then stayed approximately constant to the bottom of the core

(30 cm depth). There was also a gradual decrease of porewater Mn concentrations with depth,

from 9-10 mg/l at the surface to 5-6 mg/l at 30 cm. The profile distributions of Fe and Mn in

porewater in both sites indicate active microbially-mediated metal reduction (Overnell et al.,

1996). Reductive dissolution of Fe and Mn oxyhydroxides occurs a few centimetres below

the sediment/water interface when the sediment becomes suboxic, releasing some Fe and Mn

into the porewater. This proposition is supported by the surficial enrichment of the Mn solid

phase at both sites, which is consistent with dissolution and precipitation, driven by the redox

cycling of Mn. During coring, sediments collected from the lower basin also changed colour

from brown/grey to black between 10-15 cm and the core smelt of H2S. In contrast, the core

from the upper basin remained brown/grey beyond 30 cm and had no noticeable smell of

H2S.

11

219

220

221

222

223

224

225

226

227

228

229

230

Molybdenum concentrations showed minor enrichments in the surface sediments in both the

lower and upper basins (top 5 cm and top 11 cm, respectively), coincident with surface

increases in solid phase Mn. Due to the affinity of MoO42- for Mn oxyhydroxides, Mo is

scavenged onto Mn oxyhydroxides in the water-column. As the Mn oxyhydroxides are buried

in the sediment and used for microbial respiration, the Mo is lost to the porewater (Crusius et

al., 1996). In the lower basin, solid phase Mo concentrations then show increases below

11 cm, peaking at ~ 13 cm. These features have been observed in other marine sediments and

ascribed to sulfate reduction leading to reduction of Mo(VI)(aq) to particle reactive Mo(IV),

followed by Mo incorporation into sulfide minerals such as pyrite (Chaillou et al., 2002;

Morse and Luther, 1999; Zheng et al., 2000). In the upper basin solid phase Mo showed no

increase with depth which indicates that this sediment does not undergo sulfate reduction at

the depths sampled in this study (Crusius et al., 1996; Zheng et al., 2000). Similar behaviour

has been observed in Loch Etive’s lower basin before where solid phase Mo shows two zones

of elevated concentrations in the sediment, coincident with a Mn redox cycling zone near the

12

231

232

233

234

235

236

237

238

239

240

241

242

243

244

sediment surface and sulfate reduction at greater depth (Malcolm, 1985).

Figure 3. Porewater and solid phase profiles of selected redox-sensitive metals in the lower

(top) and the upper basin (bottom) sediment cores.

3.2 Radionuclide Profiles

3.2.1 Lower Basin

Activity concentrations (Bq/kg) of natural U isotopes (238 and 234), together with 236U,

239+240Pu, 241Am, 137Cs, 210Pb and atom ratios of 236U/238U are plotted against core depth for the

lower basin core (Figure 4). Activity ratios of 238Pu/239+240Pu are between 0.10 ± 0.01 and

0.17 ± 0.02 which indicate a binary mixture of Sellafield-derived Pu with an average ratio of

0.18 (MacKenzie et al., 1998) and fallout (weapons testing and SNAP-9A accident) with

a ratio of 0.036 in the northern hemisphere (Hardy et al., 1973). The activity profile of 210Pb

was used to date the sediment core and gives an estimated accumulation rate of ~ 0.6 cm/yr.

A plot of the date against the depth is also shown in Figure 4. 13

245

246

247

248

249

250

251

252

253

254

255

256

257

258

Figure 4. Radionuclide activity concentrations with depth in the lower basin core, including

atom ratios for 236U/238U. Error bars are ± 1σ, based on counting statistics. Reference lines for

the years ~ 1977 (maximum Sellafield discharge) and ~ 1986 (introduction of operations on

site to reduce discharges) are provided for reference. For more detail on the discharges see

Figure 1.

The 210Pb data provide estimated ages (uncertainty ± 1-3 years) for the core and help with

interpretation of other radionuclide profiles. The activity profile of 137Cs shows a broad peak

dated to the mid-1970s which could be related to the maximum Sellafield discharges of Cs

from ~ 1970 to the mid-1980s (Gray et al., 1995). This would suggest a direct solution input

of Sellafield-derived Cs to Loch Etive until the mid-1980s, consistent with a significant

proportion of the Cs discharges remaining in the solution phase, rather than associating with

particulate matter (Baxter et al., 1979). Although Sellafield discharges of Cs decreased

significantly after the commissioning of the SIXEP effluent treatment plant in 1985 (Kershaw

14

259

260

261

262

263

264

265

266

267

268

269

270

271

272

et al., 1992), and this would diminish the supply of Cs in solution, the sediment profile shows

that a Sellafield input of Cs to Loch Etive continues to the present day. This probably reflects

a continuing supply of 137Cs, either re-dissolved from the Irish Sea mud-patch or similar

sediment sources, or that associated with supply of fine grained, resuspended particulates

(Hunt and Kershaw, 1990; MacKenzie et al., 1998).

The activity profile of 239,240Pu is uniform (~ 25 Bq/kg) to about 17 cm and then decreases

gradually to reach the limit of detection at 30 cm depth (which was dated to be ~ the mid-

1960s). The activity concentrations of 241Am remain approximately constant (~ 11 Bq/kg) to

about 24 cm and then drop to ~ 5 Bq/kg. It is clear that the Pu and Am profiles do not show

peaks corresponding to the maximum historical discharges from Sellafield in the mid-1970s

as it would be expected for particle-reactive radionuclide profiles if the sediment simply

preserves the historical discharges of Sellafield (e.g. Al-Qasmi et al., 2016; Lindahl et al.,

2011; Marsden et al., 2006).

A similar Pu profile has been observed in the shelf sediments off the west coast of Scotland

in which 239,240Pu activities range between 20 and 25 Bq/kg in the top 10 cm and then

gradually decrease to less than 10 Bq/kg at 20 cm depth (MacKenzie et al., 2006). Since Pu

and Am are strongly particle-associated, this was interpreted as demonstrating a continual

supply of Pu and Am from a homogeneous secondary source, specifically the offshore

sediment described in MacKenzie et al. (1998), which has accumulated Pu and Am from the

Sellafield discharges over time and thus supplies material reflecting time-integrated Sellafield

discharges to Loch Etive. This indirect transport pathway for Pu and Am to Loch Etive means

that the sediment functions as a reservoir or buffer. This interpretation is certainly consistent

with the 210Pb data, which show an exponential decrease and hence do not suggest extensive

post-depositional mixing of the Loch Etive sediments. Ridgway and Price (1987) similarly

15

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

concluded that 137Cs and 210Pb chronologies do not suggest extensive bio-mixing in Loch

Etive.

Some differences in the distribution of Pu and Am, similar to those seen here, have been seen

before in the Irish Sea basin, and were attributed to a combination of the greater affinity of

Am than Pu for particulate matter, and the in situ generation of 241Am from 241Pu decay

(Marsden et al., 2006). The differences in Pu and Am in Loch Etive sediments probably

reflect similar effects. Alternatively, remobilisation of Pu may be occurring in the deeper

sediment sections. Under oxic to suboxic conditions particle-reactive Pu(IV) is thought to

dominate speciation. However, under more reducing conditions in in the presence of organic

ligands, soluble Pu(III)-species can form (e.g. Boukhalfa et al., 2007; Plymale et al., 2012).

As such, sulfate reduction at depths > 10-15cm in the lower basin may serve to slowly

remobilise Pu relative to Am, as is unresponsive to changes in redox conditions.

The 238U/235U atom ratios (136-146; data not plotted) in the lower basin core are typical of the

natural baseline. Uranium-238 activities are approximately uniform in the top 10 cm

(~ 30 Bq/kg) followed by a rise to 44 Bq/kg and then fluctuate in the range between 39 and

50 Bq/kg. Uranium-234 shows a similar trend to 238U, with a generally higher activity than

238U which might suggest that it is predominantly marine-derived (Cochran et al., 1986). The

increase of 238U and 234U activities below 10 cm coincides with the depletion of porewater

Fe(II), and suggests a potential response to the redox chemistry in this core as the redox

cycling of U in marine sediments is closely coupled with the redox cycling of Fe and the

onset of sulfate reduction (Zheng et al., 2002). Here, either biotic reduction of soluble U(VI)

to insoluble U(IV) by Fe(III)-reducing bacteria, or abiotic reduction of U(VI) with reduced

by-products of microbially metabolism can lead to U(IV) enrichment in then sediment

(Campbell et al., 2015; Lovley et al., 1991). Attributing any of the 238U and 234U enrichment

in this core to Sellafield discharges difficult given that U released from the nuclear site

16

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

rapidly equilibrates with seawater natural U and U released from the Marchon phosphate

plant (Kershaw et al., 1990). However, atom ratios of 236U/238U measured in Loch Etive

sediments are around 10-8. This is well above natural baseline levels (10-14) (Zhao et al., 1997)

suggesting that Sellafield-derived U is present in Loch Etive sediments.

As anthropogenic 236U cannot be incorporated in rock-forming minerals, it is expected that it

would respond to redox changes in marine systems (Al-Qasmi et al., 2016). However, the

236U profile in the lower basin (Figure 4) shows no systematic variations with depth that can

be attributed to redox processes in this core (i.e. 236U is not enriched in or below the area of

microbially mediated Fe(III) reduction). In fact, the 236U profile in the lower basin is

approximately uniform to 26 cm depth and then significantly decreases at 30 cm, and as such

is qualitatively similar to Pu and Am. Reflecting this, it appears that Sellafield derived 236U is

transported to Loch Etive in a form that appears unresponsive to redox change (e.g.

particulate bound U rather than dissolved or readily exchangeable U).

3.2.2 Upper Basin

Activity concentrations (Bq/kg) of 239+240Pu, 241Am, 137Cs, 238U, 234U, 210Pb and activity ratios of

238Pu/239+240Pu are plotted against core depth for the upper basin core (Figure 5). Both 238U and

234U profiles are relatively uniform and do not show any clear evidence of redox-driven

redistribution in this core. This may reflect that this core is not sufficiently reducing for

aqueous U(VI)-carbonate reduction. Indeed, there was no evidence for sulfate reduction in

this core. Activity ratios of 238Pu/239+240Pu are between 0.055 ± 0.005 and 0.147 ± 0.010 which

indicate a binary mixture of Sellafield-derived Pu with an average ratio of 0.18 (MacKenzie

et al., 1998) and fallout Pu with a ratio of 0.036 in the northern hemisphere (Hardy et al.,

1973). The activity ratios of 238Pu/239+240Pu start to decrease gradually at 12 cm (~ 1972),

which is consistent with a higher contribution from fallout Pu, the dominant source of Pu in

the mid-1960s, at greater depth.

17

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

Dating using the 210Pb data shows an estimated accumulation rate in the upper basin of

~ 0.3 cm/yr, slower than in the lower basin (~ 0.6 cm/yr) since, as described earlier, the upper

basin is isolated from the open sea source by a sill of 13 m depth (Figure 1). The activity

profile of 137Cs shows a broad peak dated to the mid-1970s which would be consistent with

the Sellafield maximum discharges. The broad peak of 137Cs is followed by a distribution of

Cs to the core bottom which could be attributed to a redissolution and redistribution of 137Cs

within the core (Hunt and Kershaw, 1990; MacKenzie et al., 1998). In contrast, Pu and Am

show uniform profiles to about 11-12 cm then a gradual decrease to less than 5 Bq/kg for Pu

and to below detection limits for Am. The deep distribution of Pu to the bottom of the core

could reflect a partial redissolution of Pu as has been observed in the Irish Sea (Cook et al.,

1997; Hunt and Kershaw, 1990; MacKenzie et al., 1998; McCartney et al., 1994) or input of

fallout plutonium to the Etive sediments before Sellafield-derived Am was transported into

the Sea Loch. Interestingly, and in contrast to the lower basin, Pu was also not lost from the

solids relative to Am in the deep basin core. This may reflect that the deep basin sediments do

not transition to sulfate reduction by 30 cm. Otherwise, the Pu and Am trends are similar in

both cores, whereas Cs behaves in a different way in both. That would suggest a predominant

solution input of Cs to Etive and a particulate input of Pu and Am from a homogeneous

secondary source such as the Irish Sea Mud Patch.

18

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

Figure 5. Radionuclide profiles for the upper basin core, including activity ratios for 238Pu/239,240Pu. Error bars are ± 1σ, based on counting statistics.

4 ConclusionsDespite the significant decrease in Sellafield discharges, Sellafield-derived radionuclides

continue to arrive to Loch Etive suggesting that the offshore sediments in the vicinity of

Sellafield continue to play an important role as a source of radionuclides over time.

Plutonium and Am distribution profiles in Loch Etive do not preserve the Sellafield historical

discharges, while 137Cs profile showed a broad peak corresponding to the Sellafield maximum

discharges. Although U is expected to behave as 137Cs, it actually behaves similarly as Pu and

Am with a pseudo-uniform profile. The Loch Etive settings develop anoxic sediments which

results in the redox cycling of several metals (Mn, Fe, Mo and natural U). However, redox-

active artificial radionuclides did not seem to be affected by the redox chemistry in this

system.

19

365

366

367

368

369

370

371

372

373

374

375

376

377

378

Acknowledgements

This work has been supported by the Natural Environmental Research Council grant

‘LO-RISE’ – Long-lived Radionuclides in the Surface Environment (NE/L000547/1; part of

the RATE programme (Radioactivity and the Environment) co-funded by the Environment

Agency and Radioactive Waste Management Ltd.). Al-Qasmi would like to acknowledge the

financial support from Al-Baath University, Syria and the British Council. Law and Livens

would also like to acknowledge support from NERC grant NE/M014088/1.

References

Al-Qasmi, H., Law, G.T.W., Fifield, L.K., Livens, F.R., 2016. Origins of Artificial Radionuclides in Soil and Sediment in North Wales. J. Environ. Radioact. 151, 244–249. doi:10.1016/j.jenvrad.2015.10.013

Barnes, C.E., Cochran, J.K., 1990. Uranium removal in oceanic sediments and the oceanic U balance. Earth Planet. Sci. Lett. 97, 94–101. doi:10.1016/0012-821X(90)90101-3

Baxter, M.S., McKinley, I.G., Mackenzie, A.B., Jack, W., 1979. Windscale radiocaesium in the Clyde sea area. Mar. Pollut. Bull. 10, 116–120. doi:10.1016/0025-326X(79)90104-8

Boukhalfa, H., Icopini, G.A., Reilly, S.D., Neu, M.P., 2007. Plutonium(IV) reduction by the metal-reducing bacteria Geobacter metallireducens GS15 and Shewanella oneidensis MR1. Appl. Environ. Microbiol. 73, 5897–903. doi:10.1128/AEM.00747-07

Brookshaw, D.R., Pattrick, R.A.D., Lloyd, J.R., Vaughan, D.J., 2012. Microbial effects on mineral-radionuclide interactions and radionuclide solid-phase capture processes. Mineral. Mag. 76, 777–806. doi:10.1180/minmag.2012.076.3.25

Brown, J.E., McDonald, P., Parker, A., Rae, J.E., 1999. The vertical distribution of radionuclides in a Ribble Estuary saltmarsh: transport and deposition of radionuclides. J. Environ. Radioact. 43, 259–275. doi:10.1016/S0265-931X(98)00041-1

Campbell, K.M., Gallegos, T.J., Landa, E.R., 2015. Biogeochemical aspects of uranium mineralization, mining, milling, and remediation. Appl. Geochemistry 57, 206–235. doi:10.1016/j.apgeochem.2014.07.022

Camplin, W.C., Mitchell, N.T., Leonard, R.D.P., Jefferies, D.F., 1986. Radioactivity in Surface and Coastal Waters of the British Isles. Monitoring of fallout from the Chernobyl reactor accident. Ministry of Agriculture, Fisheries and Food, Directorate of Fisheries Research, Lowestoft.

Chaillou, G., Anschutz, P., Lavaux, G., Schäfer, J., Blanc, G., 2002. The distribution of Mo, U, and Cd in relation to major redox species in muddy sediments of the Bay of Biscay. Mar. Chem. 80, 41–59. doi:10.1016/S0304-4203(02)00097-X

Cochran, J.K., Carey, A.E., Sholkovitz, E.R., Surprenant, L.D., 1986. The geochemistry of uranium and thorium in coastal marine sediments and sediment pore waters. Geochim.

20

379

380

381

382

383

384

385

386

387388389

390391

392393

394395396

397398399

400401402

403404405

406407408409

410411412

413414

Cosmochim. Acta 50, 663–680. doi:10.1016/0016-7037(86)90344-3

Cook, G.T., MacKenzie, A.B., McDonald, P., Jones, S.R., 1997. Remobilization of Sellafield-derived radionuclides and transport from the north-east Irish Sea. J. Environ. Radioact. 35, 227–241. doi:10.1016/S0265-931X(96)00070-7

Crusius, J., Calvert, S., Pedersen, T., Sage, D., 1996. Rhenium and molybdenum enrichments in sediments as indicators of oxic, suboxic and sulfidic conditions of deposition. Earth Planet. Sci. Lett. 145, 65–78.

Dahlgaard, H., 1995. Transfer of European coastal pollution to the arctic: Radioactive tracers. Mar. Pollut. Bull. 31, 3–7. doi:10.1016/0025-326X(95)00003-6

Edwards, A., Edelsten, D.J., 1977. Deep water renewal of Loch Etive: A three basin Scottish fjord. Estuar. Coast. Mar. Sci. 5, 575–595. doi:10.1016/0302-3524(77)90085-8

Eichrom Technologies, I., 2005. Uranium in soil (2 gram sample). Analytical Procedures, ACS07, Rev. 1.5.

Eigl, R., Srncik, M., Steier, P., Wallner, G., 2013. U-236/U-238 and Pu-240/Pu-239 isotopic ratios in small (2 L) sea and river water samples. J. Environ. Radioact. 116, 54–8. doi:10.1016/j.jenvrad.2012.09.013

Faure, G., 1977. The U-Series Disequilibrium Methods of Dating, in: Principles of Isotope Geology. John Wiley & Sons, pp. 238–304.

Fifield, L.K., 2008. Accelerator mass spectrometry of the actinides. Quat. Geochronol. 3, 276–290. doi:10.1016/j.quageo.2007.10.003

Fifield, L.K., Tims, S.G., Stone, J.O., Argento, D.C., De Cesare, M., 2013. Ultra-sensitive measurements of Cl-36 and U-236 at the Australian National University. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 294, 126–131. doi:10.1016/j.nimb.2012.04.028

Gray, J., Jones, S.R., Smith, A.D., 1995. Discharges to the environment from the Sellafield site, 1951-1992. J. Radiol. Prot. 15, 99–131. doi:10.1088/0952-4746/15/2/001

Hardy, E.P., Krey, P.W., Volchok, H.L., 1973. Global Inventory and Distribution of Fallout Plutonium. Nature 241, 444–445.

Howe, J.A., Shimmield, T.M., Austin, W.E.N., Longva, O., 2002. Post-glacial depositional environments in a mid-high latitude glacially-overdeepened sea loch, inner Loch Etive, western Scotland. Mar. Geol. 185, 417–433. doi:10.1016/S0025-3227(01)00299-7

Hunt, G.J., Kershaw, P.J., 1990. Remobilisation of artificial radionuclides from the sediment of the Irish Sea. J. Radiol. Prot. 10, 147–151. doi:10.1088/0952-4746/10/2/009

Hunt, G.J., Leonard, K., Hughes, L., 2013. Artificial radionuclides in the Irish Sea from Sellafield: remobilisation revisited. J. Radiol. Prot. 33, 261. doi:10.1088/0952-4746/33/2/261

Jefferies, D.F., Preston, A., Steele, A.K., 1973. Distribution of caesium-137 in British coastal waters. Mar. Pollut. Bull. 4, 118–122. doi:10.1016/0025-326X(73)90185-9

Keith-Roach, M.J., 1998. The Biogeochemistry of Transuranium Elements. Unpublished thesis (PhD), University of Manchester.

Kershaw, P.J., Baxter, A., 1995. The transfer of reprocessing wastes from north-west Europe to the Arctic. Deep Sea Res. Part II Top. Stud. Oceanogr. 42, 1413–1448. doi:10.1016/0967-0645(95)00048-8

21

415

416417418

419420421

422423

424425

426427

428429430

431432

433434

435436437438

439440

441442

443444445

446447

448449450

451452

453454

455456457

Kershaw, P.J., Pentreath, R.J., Woodhead, D.S., Hunt, G.J., 1992. A review of radioactivity in the Irish Sea: a report prepared for the Marine Pollution Monitoring Management Group. MAFF Dir. Fish. Res. Aquat. Environ. Monit. Rep. No., 65.

Kershaw, P.J., Swift, D.J., Pentreath, R.J., Lovett, M.B., 1984. The incorporation of plutonium, americium and curium into the Irish Sea seabed by biological activity. Sci. Total Environ. 40, 61–81. doi:10.1016/0048-9697(84)90342-5

Kershaw, P.J., Woodhead, D.S., Malcolm, S.J., Allington, D.J., Lovett, M.B., 1990. A Sediment History of Sellafield Discharges. J. Environ. Radioact. 12, 201–241. doi:10.1016/0265-931X(90)90024-P

Kimber, R.L., Boothman, C., Purdie, P., Livens, F.R., Lloyd, J.R., 2012. Biogeochemical behaviour of plutonium during anoxic biostimulation of contaminated sediments. Mineral. Mag. 76, 567–578. doi:10.1180/minmag.2012.076.3.08

Krom, M.D., Carbo, P., Clerici, S., Cundy, A.B., Davies, I.M., 2009. Sources and timing of trace metal contamination to sediments in remote sealochs, N.W. Scotland. Estuar. Coast. Shelf Sci. 83, 239–251. doi:10.1016/j.ecss.2009.03.028

Lindahl, P., Worsfold, P.J., Keith-Roach, M.J., Andersen, M.B., Kershaw, P.J., Leonard, K., Choi, M.-S., Boust, D., Lesueur, P., 2011. Environment Temporal record of Pu isotopes in inter-tidal sediments from the northeastern Irish Sea. Sci. Total Environ. 409, 5020–5. doi:10.1016/j.scitotenv.2011.08.019

Lovley, D.R., Phillips, E.J.P., Gorby, Y.A., Landa, E.R., 1991. Microbial reduction of uranium. Nature 350, 413–416.

MacKenzie, A.B., 2000. Environmental radioactivity: experience from the 20th century-trends and issues for the 21st century. Sci. Total Environ. 249, 313–29.

MacKenzie, A.B., Cook, G.T., McDonald, P., Jones, S.R., 1998. The influence of mixing timescales and re-dissolution processes on the distribution of radionuclides in Northeast Irish Sea sediments. J. Environ. Radioact. 39, 35–53. doi:10.1016/S0265-931X(97)00044-1

MacKenzie, A.B., Stewart, A., Cook, G.T., Mitchell, L., Ellet, D.J., Griffiths, C.R., 2006. Manmade and natural radionuclides in north east Atlantic shelf and slope sediments: Implications for rates of sedimentary processes and for contaminant dispersion. Sci. Total Environ. 369, 256–272. doi:10.1016/j.scitotenv.2006.04.012

Malcolm, S.J., 1985. Early diagenesis of molybdenum in estuarine sediments. Mar. Chem. 16, 213–225.

Malcolm, S.J., Kershaw, P.J., Cromar, N.J., Botham, L., 1990. Iron and manganese geochemistry and the distribution of Pu-239,240 and Am-241 in the sediments of the north east Irish sea. Sci. Total Environ. 95, 69–87. doi:10.1016/0048-9697(90)90054-X

Marsden, O.J., Abrahamsen, L., Bryan, N.D., Day, J.P., Fifield, L.K., Gent, C., Goodall, P.S., Morris, K.M., Livens, F.R., 2006. Transport and accumulation of actinide elements in the near-shore environment: field and modelling studies. Sedimentology 53, 237–248. doi:10.1111/j.1365-3091.2005.00761.x

Marsden, O.J., Livens, F.R., Day, J.P., Fifield, L.K., Goodall, P.S., 2001. Determination of U-236 in sediment samples by accelerator mass spectrometry. Analyst 126, 633–636. doi:10.1039/b009764k

McCartney, M., Kershaw, P.J., Woodhead, D.S., Denoon, D.C., 1994. Artificial radionuclides in the surface sediments of the Irish Sea, 1968–1988. Sci. Total Environ. 141, 103–138.

22

458459460

461462463

464465466

467468469

470471472

473474475476

477478

479480

481482483484

485486487488

489490

491492493

494495496497

498499500

501502

doi:10.1016/0048-9697(94)90022-1

Mitchell, N.T., Camplin, W.C., Leonard, D.R.P., 1986. The Chernobyl reactor accident and the aquatic environment of the UK: a fisheries viewpoint. J. Soc. Radiol. Prot. 6, 167–172. doi:10.1088/0260-2814/6/4/002

Mitchell, P.I., Condren, O.M., Vintró, L.L., McMahon, C.A., 1999. Trends in plutonium, americium and radiocaesium accumulation and long-term bioavailability in the western Irish Sea mud basin. J. Environ. Radioact. 44, 223–251. doi:10.1016/S0265-931X(98)00135-0

Morse, J.W., Luther, G.W., 1999. Chemical influences on trace metal-sulfide interactions in anoxic sediments. Geochim. Cosmochim. Acta 63, 3373–3378. doi:10.1016/S0016-7037(99)00258-6

Overnell, J., 2002. Manganese and Iron Profiles During Early Diagenesis in Loch Etive, Scotland. Application of two Diagenetic Models. Estuar. Coast. Shelf Sci. 54, 33–44. doi:10.1006/ecss.2001.0834

Overnell, J., Brand, T., Bourgeois, W., Statham, P.J., 2002. Manganese Dynamics in the Water Column of the Upper Basin of Loch Etive, a Scottish Fjord. Estuar. Coast. Shelf Sci. 55, 481–492.

Overnell, J., Harvey, S.M., Parkes, R.J., 1996. A biogeochemical comparison of sea loch sediments. Manganese and iron contents, sulphate reduction and oxygen uptake rates. Oceanol. Acta 19, 41–55.

Plymale, A.E., Bailey, V.L., Fredrickson, J.K., Heald, S.M., Buck, E.C., Shi, L., Wang, Z., Resch, C.T., Moore, D.A., Bolton, H., 2012. Biotic and Abiotic Reduction and Solubilization of Pu(IV)O 2 • x H 2 O (am) as Affected by Anthraquinone-2,6-disulfonate (AQDS) and Ethylenediaminetetraacetate (EDTA). Environ. Sci. Technol. 46, 2132–2140. doi:10.1021/es2030752

Ridgway, I.M., Price, N.B., 1987. Geochemical associations and post-depositional mobility of heavy metals in coastal sediments: Loch Etive, Scotland. Mar. Chem. 21, 229–248. doi:10.1016/0304-4203(87)90061-2

Shimmield, T.M., 1993. Study of radionuclides, lead and lead isotope ratios in Scottish sea loch sediments. The University of Edinburgh.

Sholkovitz, E.R., 1983. The geochemistry of plutonium in fresh and marine water environments. Earth-Science Rev. 19, 95–161. doi:10.1016/0012-8252(83)90029-6

Swan, D.S., Baxter, M.S., McKinley, I.G., Jack, W., 1982. Radiocaesium and Pb-210 in Clyde sea loch sediments. Estuar. Coast. Shelf Sci. 15, 515–536. doi:10.1016/0272-7714(82)90004-X

Teasdale, P.A., Collins, P.E.F., Firth, C.R., Cundy, A.B., 2011. Recent estuarine sedimentation rates from shallow inter-tidal environments in western Scotland: implications for future sea-level trends and coastal wetland development. Quat. Sci. Rev. 30, 109–129. doi:10.1016/j.quascirev.2010.08.002

Williams, T.M., MacKenzie, A.B., Scott, R.D., Price, N.B., Ridgway, I.M., 1988. Radionuclide distributions in the surface sediments of Loch Etive. Radionuclides a tool Oceanogr.

Zhao, X.-L., Kilius, L.R., Litherland, A.E., Beasley, T., 1997. AMS measurement of environmental U-236 Preliminary results and perspectives. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 126, 297–300. doi:10.1016/S0168-

23

503

504505506

507508509510

511512513

514515516

517518519

520521522

523524525526527

528529530

531532

533534

535536537

538539540541

542543544

545546547

583X(96)01034-8

Zheng, Y., Anderson, R.F., van Geen, A., Fleisher, M.Q., 2002. Remobilization of authigenic uranium in marine sediments by bioturbation. Geochim. Cosmochim. Acta 66, 1759–1772. doi:10.1016/S0016-7037(01)00886-9

Zheng, Y., Anderson, R.F., van Geen, A., Kuwabara, J., 2000. Authigenic molybdenum formation in marine sediments: A link to pore water sulfide in the Santa Barbara Basin. Geochim. Cosmochim. Acta 64, 4165–4178. doi:10.1016/S0016-7037(00)00495-6

24

548

549550551

552553554

555