Congener Patterns of Persistent Organic Pollutants ... - ACS Publications

Subscriber access provided by University of Sussex Library

Article

Congener patterns of persistent organic pollutants establish the extent of contaminant biotransport by Pacific salmon in the Great Lakes Brandon Gerig, Dominic Chaloner, David J Janetski, Richard R. Rediske, James P. O'Keefe, Ashley Moerke, and Gary Lamberti Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b05091 • Publication Date (Web): 07 Dec 2015 Downloaded from http://pubs.acs.org on December 10, 2015

Just Accepted

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Subscriber access provided by University of Sussex Library

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication online prior to technical editing, formatting for publication and author proofing. The A Society provides “Just Accepted” as a free service to the research community dissemination of scientific material as soon as possible after acceptance. “Just Acce appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manu fully peer reviewed, but should not be considered the official version of record. They a readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optio to authors. Therefore, the “Just Accepted” Web site may not include all articles tha in the journal. After a manuscript is technically edited and formatted, it will be remo Accepted” Web site and published as an ASAP article. Note that technical editing m changes to the manuscript text and/or graphics which could affect content, and al and ethical guidelines that apply to the journal pertain. ACS cannot be held resp or consequences arising from the use of information contained in these “Just Accep

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

Environmental Science & Technology

ACS Paragon Plus Environment


1 2

Congener patterns of persistent organic pollutants establish the extent of contaminant biotransport by Pacific salmon in the Great Lakes

3 4 5 6 7 8 9 10 11 12 13 14 15

Brandon S. Gerig1*, Dominic T. Chaloner1, David J. Janetski2, Richard R. Rediske3, James P. O’Keefe3, Ashley H. Moerke4, and Gary A. Lamberti1 1

Department of Biological Sciences, University of Notre Dame, Notre Dame, IN, 46556 2 Department of Biology, Indiana University of Pennsylvania, Indiana, PA, 15705 3 Annis Water Resources Institute, Grand Valley State University, Muskegon, MI, 49441 4 School of Biological Sciences, Lake Superior State University, Sault Ste. Marie, MI, 49783

*Corresponding Author *Email: [email protected]

1


Page 2 of 33

Page 3 of 33


16

ABSTRACT: In the Great Lakes, introduced Pacific salmon (Oncorhynchus spp.) can transport

17

persistent organic pollutants (POPs), such as polychlorinated biphenyls (PCBs) and

18

polybrominated diphenyl ethers (PBDEs), to new environments during their spawning

19

migrations. To explore the nature and extent of POP biotransport by salmon, we compared 58

20

PCB and 6 PBDE congeners found in spawning salmon directly to those in resident stream fish.

21

We hypothesized that stream fish exposed to salmon spawners would have congener patterns

22

similar to those of salmon, the presumed contaminant source. Using permutational multivariate

23

analysis of variance (PERMANOVA) and non-metric multidimensional scaling (NMDS), we

24

found that POP congener patterns of Pacific salmon varied among Great Lakes basin (i.e., lakes

25

Huron, Michigan, or Superior), tissue type (whole fish or eggs), and contaminant type (PCB or

26

PBDE). For stream-resident fish, POP congener pattern was influenced by the presence of

27

salmon, location (i.e. Great Lakes Basin), and species identity (i.e., brook trout [Salvelinus

28

fontinalis] or mottled sculpin [Cottus bairdii]). Similarity in congener patterns indicated that

29

salmon are a source of POPs to brook trout in stream reaches receiving salmon spawners from

30

lakes Michigan and Huron but not from Lake Superior. Congener patterns of mottled sculpin

31

differed from those of brook trout and salmon, suggesting that brook trout and mottled sculpin

32

either use salmon tissue to differing degrees, acquire POPs from different dietary sources, or

33

bioaccumulate or metabolize POPs differently. Overall, our analyses identified the important role

34

of salmon in contaminant biotransport but also demonstrated that the extent of salmon-mediated

35

POP transfer and uptake in Great Lakes tributaries is location- and species-specific.

36

Keywords: brook trout, contaminant biotransport, ecosystem linkage, Laurentian Great Lakes,

37

mottled sculpin, PCB congener, PBDE congener, Pacific salmon

2



38

39

INTRODUCTION

Persistent organic pollutants (POPs) have been a major contaminant in aquatic

40

ecosystems since the 1950s1 and include both legacy pollutants such as polychlorinated

41

biphenyls (PCBs) and emerging pollutants such as polybrominated diphenyl ethers (PBDEs).2

42

Despite extensive regulation, POPs continue to be of concern due to their ability to

43

bioaccumulate in food webs.3,4 Recently, attention has focused on understanding the role of

44

organisms as biological vectors for contaminants to recipient ecosystems that do not have

45

sources of these contaminants.5,6 Migratory organisms deliver resource subsidies and provide

46

predictable and important pulses of nutrients and energy across ecosystem boundaries.7,8

47

However, migratory organisms, such as Pacific salmon (Oncorhynchus spp.), also transport

48

contaminants.3,9 The contaminant burden and composition in an organism represents an

49

integration of the specific suite of POPs to which they were exposed during their life cycle.

50

Hence, the chemical ‘signature’ of an organism’s contaminant burden can be used to identify

51

sources of contamination, track the movement of material, and infer interactions between

52

resident organisms and resource subsidies.10,11

53

Pacific salmon are ideal for studying contaminant biotransport and transfer. Salmon are

54

anadromous and semelparous, grow to large sizes, occupy a high trophic position, and have a

55

high lipid content.7,12 As a result, salmon can deposit contaminated tissue to tributaries and

56

influence POP levels in the aquatic ecosystems where they spawn.9,10 For example, contaminated

57

salmon material, especially eggs, can be directly ingested by fish and invertebrates13,14 or

58

become incorporated and indirectly assimilated by primary producers and primary consumers

59

through decomposition pathways.15,16

3


Page 4 of 33

Page 5 of 33

60


The Laurentian Great Lakes provide an ideal setting to study contaminant biotransport by

61

salmon. First, the Great Lakes have been extensively altered by introduced species and

62

environmental contaminants that can magnify the role of salmon as a vector for pollutants.17,18

63

Second, the Great Lakes support large populations of naturally reproducing salmon that

64

bioaccumulate POPs during their open-water phase and deposit body burdens to tributaries

65

during their spawning run.18 Previous research has shown a positive relationship between POP

66

burden and egg consumption in rainbow trout (Oncorhynchus mykiss)13 and higher total POP

67

concentrations in resident fish exposed to migratory fish in Great Lakes tributaries.19 Recently,

68

Janetski et al.14 demonstrated a direct correlation between POP concentrations in stream-resident

69

fish and the quantity of POPs delivered by salmon. Although associations between salmon and

70

resident fish POPs have been documented in the Great Lakes, no study to date has directly linked

71

the chemical composition of POPs in salmon to those in other organisms consuming salmon

72

material.

73

The source of contaminants concentrated in biota can be identified using their specific

74

chemical composition.20 PCBs and PBDEs consist of mixtures of congeners21 that differ in the

75

position and number of chlorine or bromine additions, respectively, within their molecular

76

structure. For instance, a total of 209 PCB congeners has been synthesized, thereby representing

77

a diverse array of compounds in the environment.21 The variable structure of congeners,

78

combined with their relative abundance, can be used as a fingerprint to identify sources of

79

contamination22 and suggest the extent and pathway of ecosystem linkages.23 Such an approach

80

has been used to identify sources of contaminants in tree swallows (Tachycineta bicolor),24

81

American dippers (Cinclus mexicanus),11 striped bass (Morone saxatilis),25 and sockeye salmon

82

(Oncorhynchus nerka).10 To date, POPs have not been used similarly to trace the movement of

4



83

material from migratory fish, including Pacific salmon (Oncorhynchus tshawytscha, O. kisutch)

84

to stream-resident fish (brook trout [Salvelinus fontinalis], mottled sculpin [Cottus bairdii]) in

85

the Great Lakes region.

86

We used a hierarchical framework to determine whether introduced salmon were a source

87

of POP contamination to stream-resident fish in Great Lakes tributaries. First, we examined the

88

PCB and PBDE congener patterns of salmon to determine if POP pattern varied by Great Lakes

89

basin, contaminant type, and tissue type. Second, within each basin, we compared PCB and

90

PBDE congener patterns of brook trout and mottled sculpin to determine if consumer species

91

identity influenced POP congener pattern. Lastly, we determined whether resident fish POP

92

pattern was directly related to salmon spawner biomass. We predicted that (1) salmon PCB and

93

PBDE congener patterns would differ among basins, reflecting differing history, sources, and

94

extent of pollution; (2) within each lake basin, resident brook trout and mottled sculpin in stream

95

reaches with salmon would have congener patterns more similar to salmon than conspecifics in

96

reaches lacking salmon; and (3) as the amount of salmon spawner biomass increased, the

97

congener patterns of stream-resident fish would become more similar to salmon.

98 99

METHODS

100 101

Sampling Protocol. We sampled Pacific salmon and resident fish during fall 2008 and fall 2009

102

from tributaries to Lake Michigan, Lake Huron, and Lake Superior (Figure 1). Each tributary

103

received a fall spawning run of Chinook and/or coho salmon, and supported a resident fish

104

community that included brook trout and mottled sculpin (Table 1). We selected brook trout and

105

mottled sculpin because of differences in life history related to foraging mode and orientation,

5


Page 6 of 33

Page 7 of 33


106

trophic position, and diet. In general, Lake Superior streams had small runs of coho salmon while

107

Lake Michigan and Huron streams had small to large runs of Chinook salmon with occasional

108

coho salmon present. We limited our study to streams with runs of semelparous Pacific salmon

109

because salmon deposit all POPs accumulated in their bodies into streams as gametes and

110

carcasses. Salmon were collected during early October and resident fish were collected about one

111

month after the spawning run peaked. When salmon were collected, salmon spawner density was

112

estimated by counting all live and dead salmon in the wetted stream channel over a 300-m

113

reach.cf. 26 Stream-resident fish were captured using backpack electrofishing in stream reaches

114

with salmon and in reaches upstream of a barrier to salmon (e.g., dam or waterfall). Persistent

115

organic pollutant patterns in fish above barriers were assumed to reflect an alternate source of

116

contamination such as atmospheric deposition while those below the barrier reflected salmon and

117

non-point contaminant sources.

118 119

Contaminant Analyses. Procedures for tissue homogenization, extraction and clean-up, gas

120

chromatography/negative chemical ionization mass spectrometry analysis, lipid determination,

121

and quality control criteria used for contaminant analyses are described in detail elsewhere.14

122

The relationship between total POP concentrations in salmon and resident fish was described in a

123

previous publication.14 For salmon samples, we analyzed whole fish and eggs separately for PCB

124

and PBDE congener pattern while for resident species, we only considered whole fish POP

125

pattern. Mottled sculpin samples were composites of two or more individuals to obtain sufficient

126

mass (>20 g) for POP analysis. Fifty-eight PCB congeners representing penta-deca PCB

127

homolog groups and six PBDE congeners representing tetra-hexa PBDE homolog groups were

128

quantified. The PCB and PBDE congeners assessed reflect congener mixtures present in the

6



129

environment and commonly used in monitoring PCB and PBDE levels in the Great Lakes

130

(Figure 2, Figure S1).

131 132

Statistical Analyses. Prior to statistical analysis, congener concentration was converted to

133

congener relative abundance because our primary interest was in differences in congener patterns

134

rather than absolute concentrations.cf. 10,27 To address our first two predictions, we used

135

permutational multivariate analysis of variance (PERMANOVA)28 and non-metric

136

multidimensional scaling (NMDS)29 to compare (1) POP congener pattern of salmon among

137

Great Lakes basins and tissue types and (2) POP congener pattern of brook trout and mottled

138

sculpin between stream reaches within each individual lake basin. To address our third prediction

139

of a biomass-congener relationship, we used a non-linear Michaelis–Menten model to explore if

140

POP congener pattern was a function of salmon spawner biomass and a likelihood ratio test

141

(LRT) to assess species-specific differences.30 We added one to each NMDS axis value as an

142

offset prior to model fitting to ensure no data points were negative. Michaelis–Menten models

143

are commonly used to assess asymptotic relationships, and have been applied widely in

144

physiology (e.g., enzyme kinetics) and fisheries ecology (e.g., spawner-recruitment

145

relationships).30

146

Permutational multivariate analysis of variance is a multivariate, non-parametric

147

statistical method that partitions variance in a distance matrix by calculating a distance-based F-

148

statistic.28 All PERMANOVA models were analyzed using a matrix of Bray-Curtis distances

149

calculated from the congener pattern data. Similar to a Monte Carlo simulation, one thousand

150

permutations of the original data were conducted to generate p-values to assess factor

151

significance. Bray-Curtis distances are considered a robust metric for assessing dissimilarity in

7


Page 8 of 33

Page 9 of 33


152

ecological data.29 When significant differences were identified for factors of interest, pairwise

153

multiple PERMANOVA procedures were conducted to identify differences among factor

154

levels.28 We controlled the procedure-wise error rate using a Bonferroni correction where our

155

desired alpha level (0.05) was divided by the number of pairwise comparisons considered with a

156

particular model.28 This resulted in an alpha level of 0.005 for salmon and 0.006 for stream-

157

resident fish POP congener pairwise comparisons (non-significant indicated by NS in text). For

158

the statistical analyses, non-detectable concentrations were treated as zero. Although the

159

concentration of contaminants in non-detectable samples is seldom zero, this artifact did not

160

influence our results because we used non-parametric, rank-based analyses.29

161

We used NMDS ordinations, in conjunction with pairwise PERMANOVA, to further

162

explore how POP congener pattern differed among species, stream reaches, and lake basins.

163

Non-metric multidimensional scaling was used to identify clusters among factors of interest by

164

capturing variation in congener pattern along the two axes that explained the highest proportion

165

of variance. All NMDS scores were calculated from a matrix of Bray-Curtis distances, and

166

solutions were solved iteratively from 1,000 random starts to avoid local minima and ensure that

167

a convergent solution was reached that minimized model stress.29 Data points with close

168

proximity in ordination space indicate similarity in congener patterns, whereas data points far

169

away in ordination space indicate dissimilarity in congener pattern composition. Confidence

170

limits were calculated for ellipses by multiplying the standard error of the mean centroid by the

171

corresponding degrees of freedom from the Chi-squared distribution.29 The NMDS goodness-of-

172

fit was assessed using the stress statistic where values from 0 to 10 indicate a good fit, 10 to 20 a

173

moderate fit, and those over 20 a poor fit.29 All data were analyzed using R version 3.0.3;

174

PERMANOVA and NMDS were calculated using the R package VEGAN.

8



175

RESULTS

176

POP Patterns of Salmon. Salmon PCB patterns differed among lake basins (PERMANOVA,

177

P=0.001) and between egg and whole-fish samples (PERMANOVA, P=0.001; Figure 2).

178

Specifically, salmon in Lake Superior were distinct from salmon in lakes Michigan and Huron

179

(Pairwise PERMANOVA, P=0.001; Figure S2A, Table S1). Salmon PCB patterns were

180

dominated by congeners CB132, CB138, and CB118 across all basins. Higher proportions of

181

congeners CB85, CB87, CB97, CB99, CB110, CB118, CB146, and CB149 in lakes Michigan

182

and Huron contributed to the observed dissimilarity in congener patterns from Lake Superior

183

(Figure 2). In lakes Michigan and Huron, salmon eggs differed from homogenized whole-fish

184

samples (Pairwise PERMANOVA, P=0.001; Figure S2A, Table S1). In general, egg samples

185

exhibited higher proportions of CB110 and CB118, and lower proportions of CB180, CB187,

186

and CB199 when compared with whole-fish samples (Figure 2).

187

PBDE patterns in spawning salmon differed among lake basins (PERMANOVA,

188

P=0.008) and between tissue types (PERMANOVA, P=0.003; Figure S1). However, the

189

difference in salmon PBDE patterns was driven solely by egg samples from Lake Superior,

190

which had a higher proportion of congener BDE47 than Lake Michigan (pairwise

191

PERMANOVA, P=0.004, Table S1). Overall, PBDE patterns were similar between tissue types

192

and among lake basins (Figure S2B).

193 194

Lake Michigan Stream-Resident Fish POP Pattern. PCB patterns in resident fish from Lake

195

Michigan tributaries differed between stream reaches (PERMANOVA, P=0.001) and fish species

196

(PERMANOVA, P=0.001). Brook trout and mottled sculpin in stream reaches with salmon

197

spawners were distinct from conspecifics in streams lacking salmon (pairwise PERMANOVA,

9


Page 10 of 33

Page 11 of 33


198

P=0.001; Figure 3A, Table S2). Furthermore, brook trout PCB patterns in streams with salmon

199

were similar to salmon spawners, and NMDS scores in ordination space overlapped considerably

200

[pairwise PERMANOVA, P=0.020 (NS with Bonferroni correction); Figure 3A, Table S2]. In

201

brook trout, this pattern was driven by elevated proportions of PCB congeners CB82, CB83,

202

CB87, CB97, CB99, CB110, CB118, and CB138 and lower proportions of congeners CB206,

203

CB207, CB208, and CB209. In contrast, mottled sculpin in reaches with salmon spawners

204

exhibited a pattern that was distinct from salmon or brook trout (pairwise PERMANOVA,

205

P=0.001; Figure 3A, Table S2).

206

PBDE patterns in resident fish differed between reaches with and without salmon

207

spawners (PERMANOVA, P=0.001) and between fish species (PERMANOVA, P=0.001).

208

Brook trout exposed to salmon had a PBDE congener pattern similar to salmon [pairwise

209

PERMANOVA, P=0.170 (NS with Bonferroni correction); Figure 3D, Table S3] and differed

210

from conspecifics in areas without salmon spawners, which had lower proportions of BDE99

211

(pairwise PERMANOVA, P=0.001; Figure 3D, Table S3). As with PCBs, PBDEs in mottled

212

sculpin in reaches with salmon spawners exhibited a pattern that was distinct from salmon or

213

brook trout (pairwise PERMANOVA, P=0.001; Figure 3D, Table S3). Overall, PBDEs varied

214

less among species and locations than did PCBs.

215 216

Lake Huron Stream-Resident Fish POP Pattern. Resident fish PCB patterns in Lake Huron

217

tributaries differed between stream reaches (PERMANOVA, P=0.001) and between fish species

218

(PERMANOVA, P=0.001). Brook trout in stream reaches with salmon present exhibited a

219

congener pattern similar to salmon spawners [pairwise PERMANOVA, P=0.040 (NS with

220

Bonferroni correction); Figure 3B, Table S2] and differed from conspecifics in streams lacking

10



221

salmon spawners (pairwise PERMANOVA, P=0.001; Figure 3B, Table S2). Brook trout in

222

reaches with spawning salmon had elevated proportions of PCB congeners CB82, CB83, CB87,

223

CB97, CB99, CB118, and CB138 and reduced proportions of congeners CB206-209. Similar to

224

Lake Michigan, mottled sculpin PCB patterns were distinct from brook trout and salmon

225

(pairwise PERMANOVA, P=0.004; Figure 3B, Table S2). Low sample sizes of mottled sculpin

226

precluded detailed interpretation of their PCB patterns between reaches with and without salmon.

227

PBDE patterns in stream-resident fish also varied between reaches with and without

228

salmon spawners (PERMANOVA, P=0.002) and by fish species (PERMANOVA, P=0.004).

229

Brook trout PBDE pattern in reaches with salmon spawners were similar to salmon [pairwise

230

PERMANOVA, P=0.169 (NS with Bonferroni correction); Figure 3E, Table S3] and different

231

from conspecifics in reaches without salmon (pairwise PERMANOVA, P=0.003; Figure 3E,

232

Table S3). Low sample sizes of mottled sculpin precluded detailed interpretation of their PBDE

233

patterns between reaches. Overall, PBDE patterns were less distinct compared with PCBs.

234 235

Lake Superior Stream-Resident Fish POP Pattern. Brook trout and mottled sculpin in Lake

236

Superior tributaries exhibited species-specific PCB patterns (PERMANOVA, P=0.001) that did

237

not vary between stream reaches (PERMANOVA, P=0.080). Brook trout and mottled sculpin in

238

stream reaches with salmon were similar to conspecifics in streams lacking salmon (pairwise

239

PERMANOVA, P>0.05; Figure 3C, Table S2). Overall, brook trout and mottled sculpin PCB

240

congener patterns were more variable than salmon (Figure 4A).

241

PBDE patterns in stream fish did not differ between reaches (PERMANOVA, P=0.100)

242

but did differ among fish species (PERMANOVA, P=0.001). Brook trout and mottled sculpin in

243

stream reaches with salmon were similar to conspecifics in streams lacking salmon (pairwise

11


Page 12 of 33

Page 13 of 33


244

PERMANOVA, P>0.05; Figure 3F, Table S3). Overall, the patterns for PBDEs were less distinct

245

among species and locations when compared with PCBs.

246 247

Relationship between Salmon Biomass and Resident Fish POP Pattern. Brook trout and

248

mottled sculpin PCB NMDS axis-1 scores were positively related to salmon biomass, suggesting

249

that with increased salmon inputs, resident fish congener patterns became more similar to salmon

250

(Figure 4). The Michaelis-Menton model indicated that brook trout and mottled sculpin PCB

251

patterns exhibited a positive relationship that saturated at high salmon spawner biomass.

252

Including species (i.e., brook trout, mottled sculpin) in the PCB model improved model fit and

253

estimated species-specific asymptotes (LRT, F=14.31, P=0.001). This result suggests that brook

254

trout and mottled sculpin exhibit different NMDS scores but respond similarly to salmon inputs.

255

Similar to PCBs, PBDE NMDS axis-1 scores were positively related to salmon biomass

256

and saturated at high salmon biomass (Figure 4). Unlike for PCBs, including species as a model

257

variable did not improve model fit (LRT, F=1.46, P=0.24) suggesting that brook trout and

258

mottled sculpin exhibit a similar range of PBDE scores and respond similarly to increased

259

salmon inputs.

260 261

DISCUSSION

262 263

Salmon Contaminant Biotransport to Great Lakes Tributaries. By conducting a detailed

264

analysis of POP congeners, we found direct evidence that Pacific salmon are a source of POPs to

265

resident fish in Great Lakes tributaries. Our results are consistent with studies in the native range

266

of salmon9,10 but represent the first study in the introduced range to demonstrate directly that

12



267

congener patterns of stream-resident fish reflect the influx of salmon-derived contaminants. Our

268

two focal resident species of interest, brook trout and mottled sculpin, exhibited contrasting POP

269

congener patterns between stream reaches with and without spawning salmon. However, these

270

two species also differed from each other in POP congener pattern, suggesting that brook trout

271

and mottled sculpin assimilate POPs differentially. Past studies have demonstrated that congener

272

pattern can vary between species due to trophic position and individual physiology.31,32,33 In

273

addition, our study clearly demonstrated that salmon in the Great Lakes have basin-specific PCB

274

but not PBDE patterns. This result suggests that source and mode of transport as integrated by

275

salmon differs between these two POP contaminant types. Overall, our data suggest that the

276

specificity and diversity of POP congeners in the environment may make them a more sensitive

277

tracer. Our research highlights the need to assess factors related to both location and species

278

when assessing the extent of contaminant biotransport and transfer by Pacific salmon in the

279

Laurentian Great Lakes.cf. 34

280 281

Influence of Environmental Context on Contaminant Biotransport by Pacific Salmon.

282

Environmental context shapes the ecological role of salmon in stream ecosystems in their

283

native34 and introduced35 ranges, and influences the pattern of contaminant dispersal in the Great

284

Lakes. In particular, the muted response of stream-resident fish congener pattern to coho salmon

285

inputs in Lake Superior tributaries reflects the interaction among lower historical inputs of

286

pollutants,2,5 lower contaminant concentrations in salmon,14 and smaller salmon runs. In contrast,

287

contaminant transport by Chinook salmon in Lake Michigan is magnified by their larger body

288

size, higher contaminant burden, higher lipid content, and larger run sizes.14,36 In Lake Huron,

289

we found evidence of contaminant transfer by salmon to stream fish, but large reductions in

13


Page 14 of 33

Page 15 of 33


290

salmon populations following an ecosystem-level change in food web structure18 appear to have

291

reduced the risk of future contaminant biotransport by Pacific salmon. Thus, basin-specific

292

differences among lakes Michigan, Huron, and Superior have the potential to influence the

293

extent of contaminant biotransport to stream tributaries and transfer to stream-resident fish.

294

In the Lake Michigan basin, brook trout PCB and PBDE congener patterns closely

295

reflected those of spawning salmon. Therefore, brook trout appear to be an important sentinel

296

organism for assessing the extent of salmon-mediated contaminant transport. Brook trout are

297

opportunistic stream predators that forage on aquatic and terrestrial insects, crustaceans, and

298

fish.37 Similar to resident char (Salvelinus alpinus, S. malma) and grayling (Thymallus arcticus)

299

in the native range of salmon,10,38,39 brook trout readily consume salmon eggs when they are

300

available.14 This dietary plasticity can be directly linked to increased body POP concentrations

301

via increased salmon egg consumption,13 trophic position,10 and salmon-mediated contaminant

302

inputs.14 Within their native range, salmon spawner density and redd superimposition have been

303

identified as key factors regulating the availability of resource subsidies (e.g., salmon eggs) to

304

resident trout.34,38 In Lake Michigan tributaries, where stream sediments are relatively small and

305

prone to disturbance,35 higher densities of spawning salmon increase egg availability for

306

consumption by resident fish, and subsequent bioaccumulation of salmon-derived pollutants.

307

This hypothesis of egg-mediated, pollutant transfer is supported by the increased similarity in

308

brook trout congener patterns compared with salmon as salmon spawner biomass increases then

309

saturates at higher spawner abundances. Similarly, resident fish stable isotope signatures reached

310

an asymptote at high salmon spawner densities in their native range.16 Taken together, these

311

results indicate that an upper limit may exist which restricts the ability of stream-resident fish to

312

bioaccumulate contaminants transported by salmon.

14



313

We found that mottled sculpin exhibited a PCB pattern that was distinct between stream

314

reaches yet different from Pacific salmon or brook trout PCB patterns in the Lake Michigan

315

basin. Mottled sculpin are small benthic fish with a relatively large gape that forage on salmon

316

material, including eggs, when available.40,41 The dissimilarity in congener patterns between

317

mottled sculpin and salmon coupled with differences in sculpin patterns between stream reaches

318

with and without salmon spawners suggests that PCBs are expressed or accumulated differently

319

in mottled sculpin when compared with brook trout.40 In addition, we found that mottled sculpin

320

congener patterns exhibited less variation at high salmon spawner abundance, likely reflecting

321

the incorporation of salmon into their diet. Previous research has shown that mottled sculpin

322

have elevated POP concentrations in stream reaches receiving salmon runs, but a lower

323

contaminant burden overall when compared to brook trout.14 These reach-specific differences

324

suggest that salmon, in part, mediate this relationship. Many fish species assimilate POPs

325

differently due to diet, trophic position, and individual physiology, which can lead to differential

326

expression of POP patterns.31,32,33 For example, deepwater sculpin (Myoxocephalus thompsonii)

327

possess the unique ability to form methyl sulfone metabolites capable of reducing PCB body

328

burdens by 10%.33

329 330

Pacific Salmon as Integrators of Basin-Specific Contamination. We found a strong contrast

331

in PCB congener patterns of salmon between lakes Superior and Michigan. These differences

332

reflect the extent of historical pollution, differences in salmon species composition, and variation

333

in salmon abundance among lake basins. Historically, Lake Michigan received 15 times the

334

direct input of PCBs compared with Lake Superior,42 and continues to receive larger atmospheric

335

inputs of POPs, especially at the southern terminus of Lake Michigan, along with higher rates of

15


Page 16 of 33

Page 17 of 33


336

sediment re-suspension.43 Together, these factors contribute to greater availability of PCBs for

337

uptake and bioaccumulation by Pacific salmon in Lake Michigan, resulting in the difference in

338

PCB congener pattern among basins.43

339

Similar to PCBs, PBDE concentrations are highest in Lake Michigan and lowest in Lake

340

Superior.14 However, PBDE congener patterns were nearly identical among basins, suggesting

341

that both basins are contaminated by atmospheric sources.14,44 Overall, our results are consistent

342

with previous research that has identified BDE-47 as the primary PBDE congener associated

343

with Pacific salmon from Lake Michigan.45 However, our samples had a much higher proportion

344

of BDE-100 when compared with other published values.45 This difference may reflect changes

345

in PBDE deposition patterns following legislative restrictions and bans that have reduced use.46

346

Overall, the congener pattern of salmon indicated that PCBs and PBDEs enter the environment

347

by different mechanisms and from different sources that contribute to the contrasts observed

348

among lake basins.

349

The similarity in PCB patterns of salmon in lakes Michigan and Huron suggests

350

substantial inter-basin movement of fish during their open water life stage. Chinook salmon

351

populations in Lake Huron collapsed in 2004 following a whole-ecosystem shift in food web

352

structure due to changes in both bottom-up and top-down processes.18,47 This food web collapse

353

led to the near extirpation of alewife (Alosa pseudoharengus) in Lake Huron, the primary prey

354

species for Chinook salmon in the upper Great Lakes.18 Such changes were important because

355

levels of PCBs in upper trophic level fish are strongly affected by diet and the PCB congener

356

pattern can be considered an integration of contamination from primary diet sources.4 Thus, the

357

PCB congener patterns suggest that Lake Huron salmon move to Lake Michigan to forage where

358

alewife are currently more abundant.18 Such behavior is consistent with large-scale mark-

16



359

recapture studies that have found significant unidirectional movement of Chinook salmon from

360

Lake Huron to Michigan in response to changing prey densities.18,48

361 362

POPs as Ecological Tracers. Researchers increasingly are using chemical tracers, such as stable

363

isotopes, to document flow of nutrients, energy, and contaminants through food webs.49 In

364

ecology and ecotoxicology, different tracers can be useful in tracking the protein, lipid, and

365

carbohydrate fractions in organisms to offer more complete profiles of consumer resource use.49

366

For instance, POPs generally track the lipid fraction of tissue,3 while heavy metals such as

367

mercury are associated with the protein fraction of tissue.50 Non-traditional tracers, such as

368

POPs, can offer additional insight when assessing ecosystem linkages and resource use when

369

isotopic differentiation is not sufficient to identify prey items or when isotope data are not

370

available. Pollutants have been used to document the export of POPs from streams via emergent

371

insects,51 to assess the food web implications of heavy metal transport by colonial nesting birds,

372

50

373

suggest that POP congener patterns can be used to assess interactions between migratory salmon

374

and resident fish, infer basin-scale movement patterns, and reveal differential transport dynamics

375

of PCBs and PBDEs. However, much like stable isotopes, many factors influence pollutant

376

patterns including habitat-specific foraging, location-specific contamination, individual

377

physiology, and life history attributes (e.g., age, gender, body size) that can make the

378

interpretation of tracer data challenging.49,52 While not perfect, POPs offer an additional tool for

379

establishing pathways by which energy and contaminants move through and between

380

ecosystems. In addition, POP data are routinely collected by state and federal agencies to inform

381

consumption advisories across broad geographical areas and river networks. Large pollutant

and to identify differences in foraging areas used by Atlantic salmon.32 Similarly, our data

17


Page 18 of 33

Page 19 of 33


382

datasets, such as the US EPA Great Lakes Environmental Database (GLENDA,

383

https://catalog.data.gov/dataset/great-lakes-environmental-database-glenda), could be leveraged

384

in ecological and ecotoxicological studies to evaluate factors that influence contaminant

385

concentration and pattern. When possible, future studies should incorporate both traditional

386

isotope approaches and other non-traditional tracers such as POPs or heavy metals to help

387

elucidate pathways of uptake and assimilation of resource subsidies.49,52

388

ASSOCIATED CONTENT

389

Supporting Information

390 391

Bar plot of salmon PBDE congener relative abundance (Figure S1). Tables of pairwise comparisons used to identify significant differences between factor levels (Table S1-S3).

392

AUTHOR INFORMATION

393

Corresponding Author

394

*E-mail: [email protected]; phone: 574-631-0580; fax: 574-631-7413

395

Notes

396

The authors declare no competing financial interest.

397

ACKNOWLEDGEMENTS.

398 399 400 401 402 403 404 405 406

We thank M. Brueseke, K. Garvy, R. Gay, K. Harriger, J. Kosiara, E. Kratschmer, J. Ruegg, P. Shirey, T. Spear, S. Sura, K. Thompson, J. McClaren, and B. Curell for assistance with sample collection and data processing. We are also grateful to J. Leonard (Northern Michigan University), C. Helker (Batchewana First Nation), M. Holtgren and S. Ogren (Little River Band of Ottawa Indians), and R. Espinoza (Michigan Department of Natural Resources) for assistance with site selection. We also thank Dr. David Jude and four anonymous reviewers for their useful comments on previous versions of this manuscript. Funding was provided by the Great Lakes Fishery Trust (Projects 2007.857 and 2012.1244) and the Illinois−Indiana Sea Grant College Program (Grant 2014-02342-02).

407 408 409 410 411

18



412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456

REFERENCES 1. Schwarzenbach, R.P.; Egli, T.; Hofstetter, T.B.; Von Gunten, U.; Wehrly, B. Global water pollution and human health. Ann. Rev. Environ. Res. 2010, 35, 109-136. 2. Venier, M.; Dove, A.; Romanak, K.; Backus, S.; Hites, R. Flame retardants and legacy chemicals in Great Lakes water. Environ. Sci. Technol. 2014, 48, 9563-9572. 3. Mackay, D.; Fraser, A. Bioaccumulation of persistent organic chemicals: mechanisms and models. Environ. Pollut. 2000, 110, 375-391. 4. Hornbuckle, K.C.; Carlson, D.L.; Swackhamer, D.L.; Baker, J.E.; Eisenreich, S.J. Polychlorinated biphenyls in the Great Lakes, In Persistent Organic Pollutants in the Great Lakes. Springer: New York, 2006; pp. 13-70. 5. Blais, J.M.; Macdonald, R.W.; Mackay, D.; Webster, E.; Harvey, C.; Smol, J.P. Biologically mediated transport of contaminants to aquatic systems. Environ. Sci. Technol. 2007, 41, 1075-1084. 6. Flecker, A.S.; McIntyre, P.B.; Moore, J.W.; Anderson, J.T.; Taylor, B.W.; Hall Jr, R.O. Migratory fishes as material and process subsidies in riverine ecosystems, In American Fisheries Society Symposium 73; American Fisheries Society: Bethesda, 2010; 559-592. 7. Polis, G.A.; Power, M.E.; Huxel, G.R. Food webs at the landscape level. University of Chicago Press: Chicago, 2004. 8. Bauer, S.; Hoye, B.J. Migratory animals couple biodiversity and ecosystem functioning worldwide. Science 2014, 344, 54-52. 9. Ewald, G.; Larsson, P.; Linge, H.; Okla, L.; Szarzi, N. Biotransport of organic pollutants to an inland Alaska lake by migrating sockeye salmon (Oncorhynchus nerka). Arctic 1998, 51, 40-47. 10. Gregory‐Eaves, I.; Demers, M.J.; Kimpe, L.; Krümmel, E.M.; Macdonald, R.W.; Finney, B.P.; Blais, J.M. Tracing salmon‐derived nutrients and contaminants in freshwater food webs across a pronounced spawner density gradient. Environ. Toxicol. Chem. 2007, 26, 11001108. 11. Morrissey, C.A.; Pollet, I.L.; Ormerod, S.J.; Elliott, J.E. American dippers indicate contaminant biotransport by Pacific salmon. Environ. Sci. Technol. 2012, 46, 1153-1162. 12. Gende, S.M.; Edwards, R.T.; Willson, M.F.; Wipfli, M.S. Pacific salmon in aquatic and terrestrial ecosystems. Bioscience 2002, 52, 917-928. 13. Merna, J.W. Contamination of stream fishes with chlorinated hydrocarbons from eggs of Great Lakes salmon. Trans. Am. Fish. Soc. 1986, 115, 69-74. 14. Janetski, D.J.; Chaloner, D.T.; Moerke, A.H.; Rediske, R.R.; O’Keefe, J.P.; Lamberti, G.A. Resident fishes display elevated organic pollutants in salmon spawning streams of the Great Lakes. Environ. Sci. Technol. 2012, 46, 8035-8043. 15. Kline Jr, T.; Kline Jr, J.; Goering, O.; Mathisen, P.; Poe, P.; Parker, R. Recycling of elements transported upstream by runs of Pacific salmon: II. δ15N and δ13C evidence in the Kvichak River watershed, Bristol Bay, Southwestern Alaska. Can. J. Fish. Aquat. Sci. 1993, 50, 23502365. 16. Chaloner, D.T.; Martin, K.M.; Wipfli, M.S.; Ostrom, P.H.; Lamberti, G.A. Marine carbon and nitrogen in southeastern Alaska stream food webs: evidence from artificial and natural streams. Can. J. Fish. Aquat. Sci. 2002, 59, 1257-1265. 17. Ng, C.A.; Gray, K.A. Forecasting the effects of global change scenarios on bioaccumulation patterns in Great Lakes species. Global Change Biol. 2011, 17, 720-733.

19


Page 20 of 33

Page 21 of 33

457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502


18. Dettmers, J.M.; Goddard, C.I.; Smith, K.D. Management of alewife using Pacific salmon in the Great Lakes: Whether to manage for economics or the ecosystem? Fisheries 2012, 37, 495-501. 19. Giesy, J.; Verbrugge, D.; Othout, R.; Bowerman, W.; Mora, M.; Jones, P.; Newsted, J.; Vandervoort, C.; Heaton, S.; Aulerich, R. Contaminants in fishes from Great Lakes influenced sections and above dams of three Michigan rivers. I: concentrations of organochlorine insecticides, polychlorinated biphenyls, dioxin equivalents, and mercury. Arch. Environ. Contam. Toxicol. 1994, 27, 202-212. 20. Yunker, M.B.; Ikonomou, M.G.; Sather, P.J.; Friesen, E.N.; Higgs, D.A.; Dubetz, C. Development and validation of protocols to differentiate PCB patterns between farmed and wild salmon. Environ. Sci. Technol. 2011, 45, 2107-2115. 21. Erickson, M.D. Analytical chemistry of PCBs, 2nd ed. CRC Press: Boca Raton, 1997. 22. Van den Brink, Paul J; Van den Brink, Nico W; Ter Braak, C.J. Multivariate analysis of ecotoxicological data using ordination: demonstrations of utility on the basis of various examples. Australas. J. Ecotoxicol. 2003, 9, 141-156. 23. Lamberti, G.A.; Chaloner, D.T.; Hershey, A.E. Linkages among aquatic ecosystems. J. N. Am. Benthol. Soc. 2010, 29, 245-263. 24. Echols, K.R.; Tillitt, D.E.; Nichols, J.W.; Secord, A.L.; McCarty, J.P. Accumulation of PCB congeners in nestling tree swallows (Tachycineta bicolor) on the Hudson River, New York. Environ. Sci. Technol. 2004, 38, 6240-6246. 25. Ashley, J.T.; Horwitz, R.; Steinbacher, J.C.; Ruppel, B. A comparison of congeneric PCB patterns in American eels and striped bass from the Hudson and Delaware River estuaries. Mar. Pollut. Bull. 2003, 46, 1294-1308. 26. Chaloner, D.; Lamberti, G.; Merritt, R.; Mitchell, N.; Ostrom, P.; Wipfli, M. Variation in responses to spawning Pacific salmon among three south‐eastern Alaska streams. Freshwat. Biol. 2004, 49, 587-599. 27. Morrissey, C.A.; Elliott, J.E.; Ormerod, S.J. Local to continental influences on nutrient and contaminant sources to river birds. Environ. Sci. Technol. 2010, 44, 1860-1867. 28. Anderson, M.J. A new method for non-parametric multivariate analysis of variance. Aust. Eco. 2001, 26, 32-46. 29. Oksanen, J.; Blanchet, F.G.; Kindt, R.; Legendre, P.; Minchin, P.R.; O'Hara, R.B.; Simpson, G.L.; Solymos, P., Stevens, M.H.; Wagner, H. Vegan: Community Ecology Package. R package version 2.0-10. 2013. http://CRAN.R-project.org/package=vegan. 30. Bolker, B.M., Ecological models and data in R. Princeton University Press: New Jersey, 2008. 31. Kiljunen, M.; Peltonen, H.; Jones, R.I.; Kiviranta, H.; Vourinen, P.J.; Verta, M.; Karhalainen. Coupling stable isotopes with bioenergetics to evaluate sources of variation in organochlorine concentrations in Baltic salmon (Salmo salar). Can. J. Fish. Aquat. Sci. 2008, 65, 2114-2126. 32. Svendsen, T.C.; Vorkamp, K.; Svendsen, J.C.; Aarestrup, K.; Frier, J. Organochlorine fingerprinting to determine foraging areas of sea-ranched Atlantic salmon: a case study from Denmark. N. Am. J. Fish. Manage. 2009, 29, 598-603. 33. Stapleton. M.; Letcher, R. J.; Baker, J. E. Metabolism of PCBs by the deepwater sculpin (Myoxocephalus thompsoni) Environ. Sci. Technol. 2001, 35, 4747– 4752. 34. Janetski, D.J.; Chaloner, D.T.; Tiegs, S.D.; Lamberti, G.A. Pacific salmon effects on stream ecosystems: a quantitative synthesis. Oecologia 2009, 159, 583-595.

20



503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548

35. Janetski, D.J.; Chaloner, D.T.; Moerke, A.H.; Levi, P.S.; Lamberti, G.A. Novel environmental conditions alter subsidy and engineering effects by introduced Pacific salmon. Can. J. Fish. Aquat. Sci. 2014, 71, 502-513. 36. Johnson, S.P.; Schindler, D.E. Trophic ecology of Pacific salmon (Oncorhynchus spp.) in the ocean: a synthesis of stable isotope research. Ecol. Res. 2009, 24, 855-863. 37. Scott, W.B.; E.J. Crossman. Freshwater fishes of Canada. Bull. Fish. Res. Board Can. 1973, 184, 1-966. 38. Moore, J.W.; Schindler, D.E.; Ruff, C.P. Habitat saturation drives thresholds in stream subsidies. Ecology 2008, 89, 306-312. 39. Denton, K.P.; Rich, H.B.; Moore, J.W.; Quinn, T.P. The utilization of a Pacific salmon Oncorhynchus nerka subsidy by three populations of charr Salvelinus spp. J. Fish Biol. 2010, 77, 1006-1023. 40. Swain, N.; Hocking, M.; Harding, J.; Reynolds, J.; Richardson, J. Effects of salmon on the diet and condition of stream-resident sculpins. Can. J. Fish. Aquat. Sci. 2014, 71, 521-532. 41. Schindler, D.E.; Scheuerell, M.D.; Moore, J.W.; Gende, S.M.; Francis, T.B.; Palen, W.J. Pacific salmon and the ecology of coastal ecosystems. Front. Ecol. Environ. 2003, 1, 31-37. 42. Golden, K.; Wong, C.; Jeremiason, J.; Eisenreich, S.; Sanders, G.; Hallgren, J.; Swackhamer, D.; Engstrom, D.; Long, D. Accumulation and preliminary inventory of organochlorines in Great Lakes sediments. Water Sci. Technol. 1993, 28, 19-31. 43. Li, A.; Rockne, K.; Sturchio, N.; Song, W.; Ford, J.; Wei, H. PCBs in sediments of the Great Lakes: Distribution and trends, homolog and chlorine patterns, and in situ degradation. Environ. Pollut. 2009, 157, 141-147. 44. Streets, S.S.; Henderson, S.A.; Stoner, A.D.; Carlson, D.L.; Simcik, M.F.; Swackhamer, D.L. Partitioning and bioaccumulation of PBDEs and PCBs in Lake Michigan. Environ. Sci. Technol. 2006, 40, 7263-7269. 45. Manchester Neesvig, J.; Valters, K.; Sonzogni, W. Comparison of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) in Lake Michigan salmonids. Environ. Sci. Technol. 2001, 35, 1072-1077. 46. Crimmins, B.S.; Pagano, J.J.; Xia, X.; Hopke, P.K.; Milligan, M.S.; Holsen, T.M. Polybrominated diphenyl ethers (PBDEs): turning the corner in Great Lakes trout 1980– 2009. Environ. Sci. Technol. 2012, 46, 9890-9897. 47. Bunnell, D.B.; Barbiero, R.P.; Ludsin, S.A.; Madenjian, C.P.; Warren, G.J.; Dolan, D.M.; Brenden, T.O.; Briland, R.; Gorman, O.T.; He, J.X. Changing ecosystem dynamics in the Laurentian Great Lakes: bottom-up and top-down regulation. Bioscience 2014, 64, 26-39. 48. Adlerstein, S.A.; Rutherford, E.S.; Claramunt, R.M.; Clapp, D.F.; Clevenger, J.A. Seasonal movements of Chinook salmon in Lake Michigan based on tag recoveries from recreational fisheries and catch rates in gill-net assessments. Trans. Am. Fish. Soc. 2008, 137, 736-750. 49. Ramos, R.; González-Solís, J. 2012 Trace me if you can: the use of intrinsic biogeochemical markers in marine top predators. Front. Ecol. Environ. 2012, 10, 258–266. 50. Walters, D.M.; Fritz, K.M.; Otter, R.R. The dark side of subsidies: adult stream insects export organic contaminants to riparian predators. Ecol. Appl. 2008, 18, 1835-1841. 51. Mallory, M. L.; Mahon, L.; Tomlik, M.D.; White, C.; Milton, G.R.; Spooner, I. Colonial marine birds influence island soil chemistry through biotransport of trace elements. Water Air Soil Poll. 2015, 226, 31-39 52. Layman, C.A.; Araujo, M.S.; Boucek, R.; Hammerschlag-Peyer, C.M.; Harrison, E.; Jud, Z.R.; Matich, P.; Rosenblatt, A.E; Vaudo, J.J.; Yeager, L.A., Post, D.M.; Bearhop, S.

21


Page 22 of 33

Page 23 of 33

549 550


Applying stable isotopes to examine food-web structure: an overview of analytical tools. Biol. Rev. 2012, 87, 545-562

22



551 552 553 554 555 556

Table 1. Site characteristics and mean (standard error) and sample size (N) of persistent organic contaminants levels for salmon and stream-resident fish collected in this study. Live salmon were not collected from the Boyne River, but Chinook salmon carcasses provided evidence of a salmon run. CHS=Chinook salmon, COS=coho salmon, BKT=brook trout; MTS=mottled sculpin. Order indicates which salmon species was dominant during the run. NA=not available due to absence of live salmon (e.g., Boyne River), lack of an upstream barrier (e.g., Elliot Creek), or species not captured at the site. Biomass estimate was based on a survey of 300 m of stream length. For more detailed information on salmon and resident fish contaminant burden see Janetski et al.14

Lake Basin Site* Stream Name Michigan 1 Pine Creek, MI

Superior

Huron

Resident Fish POP Levels (µg kg -1 wet mass) Salmon Present Salmon Absent

Salmon POP Levels (µg kg -1 wet mass)

Salmon Spawners Biomass Species (kg m -2 ) CHS 0.305

PCB 390.97 (34.57)

PBDE 34.38 (3.99)

2

Kids Creek, MI

CHS

0.199

456.39 (73.88) 50.35 (12.03)

3

Deer Creek, MI

CHS

0.160

197.63 (32.47)

21.59 (3.71)

4

Boyne River, MI

CHS, COS

NA

NA

5

Thompson Creek, MI CHS, COS

0.785

325.47 (34.66)

30.26 (2.83)

6

Mosquito River, MI

COS

0.002

36.55 (13.83)

7.72 (2.40)

7

Pendills Creek, MI

COS, CHS

0.006

45.74 (12.13)

11.83 (2.57)

8

Garden River, ON

CHS

0.042

495.67 (16.25)

50.18 (0.16)

9

Crystal Creek, ON

COS

0.003

51.47 (2.55)

9.76 (0.04)

10

Elliot Creek, MI

CHS, COS

0.080

164.62 (75.05)

17.02 (6.31)

NA

557 558

Page 24 of 33

N Resident Fish 8 BKT MTS 2 BKT MTS 4 BKT MTS 0 BKT MTS 14 BKT MTS 5 BKT MTS 7 BKT MTS 2 BKT MTS 2 BKT MTS 5 BKT MTS

*See Figure 1 for geographic location of study sites.

23


PCB PBDE N 289.41 (39.22) 22.99 (3.32) 9 80.11 (19.49) 5.65 (1.03) 4 44.63 (15.17) 5.48 (1.39) 4 17.50 (1.56) 2.17 (0.48) 4 3.08 1.51 1 9.75 (1.34) 2.35 (0.73) 4 NA NA 0 159.58 (128.25) 13.17 (12.79) 2 134.29 (98.37) 8.23 (4.88) 3 74.90 (16.33) 5.94 (1.61) 5 5.73 (1.91) 2.29 (0.01) 2 2.51 (0.68) 0.56 (0.09) 3 4.29 (1.32) 0.74 (0.33) 3 1.80 (0.24) 0.43 (0.08) 5 NA NA 0 18.81 5.41 1 NA NA 0 20.86 (12.19) 4.20 (1.62) 4 55.19 (16.31) 5.63 (1.47) 4 NA NA 0

PCB 4.83 (1.18) 2.08 (0.34) 4.48 (0.70) 3.25 (0.43) 2.35 (0.66) 3.01 (1.95) 2.40 2.61 (0.53) 5.25 (2.99) 6.15 (1.33) 3.32 (0.30) NA 2.10 (0.31) 1.77 (0.58) 2.70 (0.48) 1.91 3.05 (0.23) NA NA NA

PBDE 0.75 (0.18) 0.70 (0.20) 1.23 (0.29) 0.70 (0.12) 0.97 (0.08) 2.30 (1.83) 0.32 1.37 (0.83) 0.87 (0.31) 1.15 (0.23) 1.33 (0.51) NA 0.47 (0.09) 0.33 (0.06) 0.58 (0.20) 0.79 0.92 (0.18) NA NA NA

N 5 2 4 4 2 3 1 2 2 5 2 0 4 3 2 1 3 0 0 0

Page 25 of 33


559

Lake Superior

Lake Huron

Lake Michigan

560 561 562

Figure 1. Location of study streams within the upper Laurentian Great Lakes. Site characteristics are provided in Table 1.

563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 24



580 581 582 583 584 585 586 587 588 589

Page 26 of 33

Figure 2. PCB congener patterns (mean percentage ± standard error) of salmon eggs and whole fish taken from lakes Michigan (eggs N=16, tissue N=28), Huron (eggs N=4, tissue N=9), and Superior (eggs N=7, tissue N=12). PCB congeners quantified were: Penta-(CB82, CB83, CB85, CB87, CB97, CB99, CB105, CB107/123, CB110, CB118, CB124), Hexa-(CB130, CB132/153, CB134, CB137, CB138, CB141, CB144, CB146, CB149, CB151, CB156, CB157, CB158, CB163, CB164, CB167), Hepta-(CB170, CB171, CB172, CB174, CB175, CB176, CB177, CB178, CB179, CB180, CB183, CB185, CB187, CB189, CB190, CB191, CB193), Octa-(CB194, CB195, CB196, CB197, CB199, CB200, CB201, CB202, CB203, CB205), Nona-(CB206, CB207, CB208), and Deca-(CB209).

590 591

25


Page 27 of 33


92

93 94 95 96 97 98

Figure 3. Non-metric multidimensional scaling (NMDS) plots of PCB (A-C) and PBDE (D-F) pattern for salmon spawners and resident fish in stream reaches with and without salmon from lakes Michigan (A, D), Huron (B, E), and Superior (C, F). Brook trout with salmon present are gold circles. Brook trout with salmon absent are red circles. Mottled sculpin with salmon present are blue squares. Mottled sculpin with salmon absent are green squares. Pacific salmon are pink triangles. Ellipses represent 95% confidence limits of the mean, ellipses are the same color as corresponding data points, and the arrow in each sub-plot points to the salmon ellipse. Low sample sizes of mottled sculpin precluded detailed interpretation of their PCB patterns between reaches with and without salmon for Lake Huron. 26



599

600 601 602 603 604 605 606

Figure 4. Relationship between salmon spawner biomass and PCB and PBDE patterns for brook trout (A, B) and mottled sculpin (C, D) pooled over all three Great Lakes (n=58). Likelihood ratio tests comparing Michaelis-Menten models demonstrated that species was a significant factor for PCBs (P=0.001) but not for PBDEs (P=0.24). Note that with increasing spawner biomass PCB and PBDE congener patterns of brook trout and mottled sculpin become increasingly similar to salmon.

607 608 609 610 611 612 613 27


Page 28 of 33

Page 29 of 33


614 615 616

Supporting Information

617 618 619 620 621

Figure S1. PBDE congener pattern (mean percentage ± standard error) of salmon eggs and whole fish from lakes Michigan (eggs N=16, tissue N=28), Huron (eggs N=4, tissue N=9), and Superior (eggs N=7, tissue N=12). PBDE congeners quantified were BDE28, BDE47, BDE99, BDE100, BDE153, and BDE154.

622

28



623 624 625 626 627 628

Figure S2. Non-metric, multidimensional scaling (NMDS) plots with 95% confidence ellipses of the mean for PCBs (A.) and PBDEs (B.) in salmon whole fish and eggs from lakes Michigan, Huron, and Superior. A convergent solution for PCBs was found after two iterations (stress=7.9) and for PBDEs after three iterations (stress=17.54). Tissue samples are represented by triangles and egg samples by circles. Ellipses are the same color as corresponding data points.

629 630 631 632 633 634 635 636 637 638 639 640 641 29


Page 30 of 33

Page 31 of 33

642 643 644 645 646


Table S1. Pairwise Bonferroni comparisons for PCB and PBDE PERMANOVA models in salmon from lakes Michigan, Huron, and Superior. The alpha value for each pairwise comparison was 0.006 (0.05/9 comparisons per POP class). Significant comparisons are in bold. For each contaminant we compared differences in tissue and pattern among basins. In addition, we also compared tissue and egg patterns within a given basin. POP PCB

Source Tissue

Eggs

Tissue-Eggs

PBDE

Tissue

Eggs

Tissue-Eggs 647

Comparison Michigan-Huron Michigan-Superior Superior-Huron Michigan-Huron Michigan-Superior Superior-Huron Michigan Huron Superior Michigan-Huron Michigan-Superior Superior-Huron Michigan-Huron Michigan-Superior Superior-Huron Michigan Huron Superior

648 649 650 651 652 653 654 655 656 657 658 659

30


F 4.37 42.72 24.20 3.46 79.62 18.17 15.35 6.88 3.10 2.71 0.83 1.24 0.76 6.74 3.73 3.10 0.54 4.20

p-value 0.006 0.001 0.001 0.020 0.001 0.003 0.001 0.001 0.010 0.060 0.420 0.266 0.460 0.004 0.040 0.040 0.590 0.020


660 661 662 663

Table S2. Pairwise Bonferroni comparisons for PCB models in PERMANOVA brook trout and mottled sculpin in reaches with salmon present and salmon absent from lakes Michigan, Huron, and Superior. The alpha value for each pairwise comparison was 0.00625 (0.05/8 comparisons per basin). Significant comparisons are in bold. POP PCB

664

Page 32 of 33

Basin Comparison Michigan Brook Trout-Salmon Present: Brook Trout-Salmon Absent Brook Trout-Salmon Present: Salmon Spawners Brook Trout-Salmon Absent:Salmon Spawners Mottled Sculpin-Salmon Present: Mottled Sculpin-Salmon Absent Mottled Sculpin-Salmon Present: Salmon Spawners Mottled Sculpin-Salmon Absent:Salmon Spawners Brook Trout-Salmon Present: Mottled Sculpin-Salmon Present Brook Trout-Salmon Absent: Mottled Sculpin-Salmon Absent Huron Brook Trout-Salmon Present: Brook Trout-Salmon Absent Brook Trout-Salmon Present: Salmon Spawners Brook Trout-Salmon Absent:Salmon Spawners Mottled Sculpin-Salmon Present: Mottled Sculpin-Salmon Absent Mottled Sculpin-Salmon Present: Salmon Spawners Mottled Sculpin-Salmon Absent:Salmon Spawners Brook Trout-Salmon Present: Mottled Sculpin-Salmon Present Brook Trout-Salmon Absent: Mottled Sculpin-Salmon Absent Superior Brook Trout-Salmon Present: Brook Trout-Salmon Absent Brook Trout-Salmon Present: Salmon Spawners Brook Trout-Salmon Absent:Salmon Spawners Mottled Sculpin-Salmon Present: Mottled Sculpin-Salmon Absent Mottled Sculpin-Salmon Present: Salmon Spawners Mottled Sculpin-Salmon Absent:Salmon Spawners Brook Trout-Salmon Present: Mottled Sculpin-Salmon Present Brook Trout-Salmon Absent: Mottled Sculpin-Salmon Absent

665 666 667 668 669 670 671 672 673 674

31


F 5.61 1.86 18.35 4.96 50.17 39.15 13.20 8.95 3.40 2.23 8.90 NA 12.17 NA 4.50 NA 14.52 6.66 0.922 1.14 23.70 33.38 3.50 2.14

p-value 0.001 0.020 0.001 0.001 0.001 0.001 0.001 0.001 0.003 0.039 0.001 NA 0.001 NA 0.004 NA 0.001 0.001 0.5 0.335 0.001 0.001 0.008 0.080

Page 33 of 33


675 676 677 678 679

Table S3. Pairwise Bonferroni comparisons for PBDE PERMANOVA models for brook trout and mottled sculpin in reaches with salmon present and salmon absent from lakes Michigan, Huron, and Superior. The alpha value for each pairwise comparison was 0.00625 (0.05/8 comparisons per basin). Significant comparisons are italicized.

680

POP Basin Comparison PBDE Michigan Brook Trout-Salmon Present: Brook Trout-Salmon Absent Brook Trout-Salmon Present: Salmon Spawners Brook Trout-Salmon Absent:Salmon Spawners Mottled Sculpin-Salmon Present: Mottled Sculpin-Salmon Absent Mottled Sculpin-Salmon Present: Salmon Spawners Mottled Sculpin-Salmon Absent:Salmon Spawners Brook Trout-Salmon Present: Mottled Sculpin-Salmon Present Brook Trout-Salmon Absent: Mottled Sculpin-Salmon Absent Huron Brook Trout-Salmon Present: Brook Trout-Salmon Absent Brook Trout-Salmon Present: Salmon Spawners Brook Trout-Salmon Absent:Salmon Spawners Mottled Sculpin-Salmon Present: Mottled Sculpin-Salmon Absent Mottled Sculpin-Salmon Present: Salmon Spawners Mottled Sculpin-Salmon Absent:Salmon Spawners Brook Trout-Salmon Present: Mottled Sculpin-Salmon Present Brook Trout-Salmon Absent: Mottled Sculpin-Salmon Absent Superior Brook Trout-Salmon Present: Brook Trout-Salmon Absent Brook Trout-Salmon Present: Salmon Spawners Brook Trout-Salmon Absent:Salmon Spawners Mottled Sculpin-Salmon Present: Mottled Sculpin-Salmon Absent Mottled Sculpin-Salmon Present: Salmon Spawners Mottled Sculpin-Salmon Absent:Salmon Spawners Brook Trout-Salmon Present: Mottled Sculpin-Salmon Present Brook Trout-Salmon Absent: Mottled Sculpin-Salmon Absent

32


F 4.95 1.68 20.53 4.72 92.52 10.69 23.21 2.92 4.22 1.89 13.07 NA 12.35 NA 3.28 NA 0.81 2.19 16.81 1.18 25.65 10.45 2.93 1.34

p-value 0.009 0.170 0.001 0.001 0.001 0.001 0.001 0.040 0.003 0.169 0.001 NA 0.001 NA 0.032 NA 0.470 0.130 0.001 0.350 0.001 0.001 0.095 0.282

Congener Patterns of Persistent Organic Pollutants ... - ACS Publications

Recommend Documents