Total Mercury and Methylmercury Response in Water, Sediment, and

Subscriber access provided by University of Newcastle, Australia

Article

TOTAL MERCURY AND METHYLMERCURY RESPONSE IN WATER, SEDIMENT, AND BIOTA TO DESTRATIFICATION OF THE GREAT SALT LAKE, UTAH, USA Carla Valdes, Frank J Black, Blair Stringham, Jeffrey N. Collins, James R Goodman, Heidi J. Saxton, Christopher R. Mansfield, Joshua N. Schmidt, Shu Yang, and William P. Johnson Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b05790 • Publication Date (Web): 12 Apr 2017 Downloaded from http://pubs.acs.org on April 13, 2017

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

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

Photo credit: NASA ISS040-E-043230 (6 July 2014) 82x49mm (300 x 300 DPI)

ACS Paragon Plus Environment


1

Draft 2016-03-25

2 3 4 5

TOTAL MERCURY AND METHYLMERCURY RESPONSE IN WATER, SEDIMENT, AND

6

BIOTA TO DESTRATIFICATION OF THE GREAT SALT LAKE, UTAH, USA

7 8 9 10 11

Carla Valdes‡, Frank J Black §, Blair Stringham †, Jeffrey N Collins §, James R Goodman §,

12

Heidi J Saxton §, Christopher R Mansfield §, Joshua N Schmidt §,

13

Shu Yang‡, and William P Johnson *,‡

14 15 ‡

16

§

17 18

Department of Geology & Geophysics, University of Utah, Salt Lake City, UT 84112

†

Department of Chemistry, Westminster College, Salt Lake City UT 84105

Division of Wildlife Resources, Utah Department of Natural Resources, Salt Lake City, UT 84114

19 20 21 22 23

*

Corresponding author, [email protected] (801)-664-8289

ACS Paragon Plus Environment

Page 2 of 33

Page 3 of 33


24

Abstract:

25

Measurements of chemical and physical parameters made before and after sealing of culverts in

26

the railroad causeway spanning Great Salt Lake in late 2013 documented dramatic alterations in

27

the system in response to the elimination of flow between the Great Salt Lake’s north and south

28

arms. The flow of denser, more saline water through the culverts from the north arm (Gunnison

29

Bay) to the south arm (Gilbert Bay) previously drove the perennial stratification of the south arm

30

and the existence of oxic shallow brine and anoxic deep brine layers. Closure of the causeway

31

culverts occurred concurrently with a multiyear drought that resulted in a decrease in the lake

32

elevation and a concomitant increase in top-down erosion of the upper surface of the deep brine

33

layer by wind-forced mixing. The combination of these events resulted in replacement of the

34

formerly stratified water column in the south arm with one that is vertically homogeneous and

35

oxic. Total mercury concentrations in the deep waters of the south arm decreased by

36

approximately 81%, and methylmercury concentrations in deep waters decreased by roughly

37

86%, due to destratification. Methylmercury concentrations decreased by 77% in underlying

38

surficial sediment whereas there was no change was observed in total mercury. The dramatic

39

mercury loss from deep waters and methylmercury loss from underlying sediment in response to

40

causeway sealing provides new understanding of the potential role of the deep brine layer in the

41

accumulation and persistence of methylmercury in the Great Salt Lake. Additional mercury

42

measurements in biota appear to contradict the previously implied connection between elevated

43

methylmercury concentrations in the deep brine layer and elevated mercury in avian species

44

reported prior to causeway sealing.

1 ACS Paragon Plus Environment


45 46

Introduction: Mercury is a global pollutant that migrates through natural systems by complex

47

transformation and transport processes that are in large part governed by redox conditions in the

48

environment.1 Mercury exists in nature as elemental mercury Hg(0), divalent mercury Hg(II),

49

and organomercury compounds such as monomethylmercury (MeHg), the bioaccumulative form

50

of mercury. Biomagnification of MeHg can result in potentially toxic levels of MeHg exposure

51

to humans and wildlife through consumption of prey, causing detrimental neurological,

52

behavioral, and reproductive effects.2

53

The methylation of inorganic mercury in the environment is predominantly biologically

54

mediated, occurs in anoxic environments, and is carried out largely by sulfate-reducing

55

bacteria.3-5 Methylation by iron-reducing bacteria and methanogens can also be important, and

56

many other microbes have been shown to contain the hgcAB gene cluster responsible for

57

mercury methylation.6-9 Abiotic parameters, such as dissolved organic matter (DOM), have also

58

been shown to increase mercury methylation under sulfidic conditions in lab experiments,5,10 and

59

MeHg concentrations in aquatic sediments are often positively correlated with concentrations of

60

dissolved organic carbon (DOC), although this is not always the case due to the complex effects

61

of DOM on mercury’s biogeochemical cycling via its role as a substrate for microbial respiration

62

and its role in inorganic mercury complexation and chemical speciation.11-13

63

The Great Salt Lake (GSL) is the largest terminal lake in the Western Hemisphere, and

64

the Western Hemisphere Shorebird Reserve Network recognizes it as a habitat of hemispheric

65

importance for millions of migratory birds (Figure 1). Over 1.4 million shorebirds use the GSL

66

for breeding and staging areas, and over 7 million waterfowl utilize the GSL and its adjacent

67

1,900 km2 of freshwater and brackish wetlands during some portion of their biannual migration.


Page 4 of 33

Page 5 of 33


68

The GSL avian community feeds on aquatic organisms adapted to highly saline conditions,

69

which includes brine shrimp (Artemia fransicana), brine fly (Ephydra spp.), water boatman

70

(corixid) and numerous algal species.14 In 2007, human consumption advisories were placed on

71

three duck species (Cinnamon Teal, Northern Shoveler, and Common Goldeneye) at the GSL

72

based on elevated mercury levels in breast muscle tissue exceeding the EPA screening value (0.3

73

µg/kg, ww).15-17

74

The construction of a railroad causeway in 1959 restricted water flow between the north

75

arm (Gunnision Bay), and the south arm (Gilbert Bay).15 Because the south arm receives nearly

76

all surface water inflows of freshwater to the GSL, the north arm is evaporatively concentrated to

77

a greater extent than the south arm, yielding brine in the north arm (250-280 g/L) that is 1.4 to

78

1.6 times more saline than the south arm (110-180 g/L), the latter being 3–5 times more saline

79

than seawater (33-36 g/L).18-19 Culverts in the causeway allowed bidirectional flow, with

80

shallow water flowing from the south arm to the north arm, and deeper denser water flowing

81

from the north to south arm. This flow of dense brine from the north arm, along with limited

82

wind-driven mixing, resulted in saline based density-driven stratification and the formation of a

83

monimolimnion, herein referred to as the deep brine layer (DBL), in the south arm that persisted

84

6-7 m below the surface of the lake, and that was not subject to annual turnover.15,18,19,20 The

85

DBL was anoxic with elevated DOC (59 – 88 mg/L) and sulfide (7 –29 mg/L)18,22,23, high

86

activities of sulfate-reducing bacteria, 24 and elevated MeHg (20 to 32 ng/L) and total mercury

87

(HgT) (38-80 ng/L) with a high percentage of HgT existing as MeHg.22 Several studies20,25-27

88

have implicated the elevated MeHg in the DBL as a potential source of elevated mercury in

89

biota, possibly via mixing into the oxic upper brine layer where uptake into the food chain could

90

subsequently occur. However, a previous laboratory study23 reported no difference in the HgT



91

concentration of brine shrimp grown for 14 days in a water column where a DBL was absent

92

compared to a stratified water column where an anoxic DBL was present. And they reported that

93

brine shrimp grown in water created by mixing different ratios of the upper brine layer and DBL

94

accumulated Hg in proportion to the Hg:POC (particulate organic carbon) ratio, such that at least

95

the short term effect of mixing in water from the DBL was a decrease in the HgT concentration

96

in the brine shrimp.

97

Mixing between the DBL and upper brine layer in the south arm of the GSL is believed

98

to be relatively limited, with previous research28 indicating that the pools of dissolved inorganic

99

nitrogen (DIN) and total dissolved phosphorous (TDP) in the shallow brine layer are only weakly

100

coupled to those in the deep brine layer. But despite persistent stratification in the GSL, Beisner

101

et al.29 reported evidence that limited mixing between the DBL and shallow brine layer occurred

102

during conveyance of the DBL from the northern to southern areas of the south arm, and Jones

103

and Wurtsbaugh23 estimated that 40% of the DBL was entrained into the shallow brine layer

104

annually. Such mixing would allow for the transfer of MeHg from the DBL to the overlying

105

shallow brine layer, and by extension, into the base of the food chain, and eventually to brine fly

106

larvae and brine shrimp that serve as diet for many avian species.23,27,28,30,31 Such an indirect

107

connection was inferred by Johnson et al.22 based on elevated Hg concentrations previously

108

reported in birds (eared grebes)32 and brine flies26 on the southwestern side of Gilbert Bay

109

(where the DBL existed at depth in the most proximal water body), as well as due to the

110

temporal correspondence of reduced Hg(II)-methylation rates in the DBL and reduced Hg

111

burdens among aquatic invertebrates (brine shrimp30,33 and brine flies26).

112 113

In November 2012 and December 2013, two railway causeway culverts that allowed limited water flow between the north and south arms of the GSL were closed due to deteriorating


Page 6 of 33

Page 7 of 33


114

structural integrity. These closures provided the opportunity to determine the geochemical

115

response of the system to elimination of flow between the north and south arms and

116

destratification and disappearance of the DBL in the south arm (described below). This event

117

also allowed us to quantify how HgT and MeHg concentrations in the water column, surficial

118

sediment, and biota responded to the disappearance of the DBL, thus providing insight into the

119

role of the DBL in the biogeochemical cycling of mercury in the GSL.

120 121

Methods

122

Water and Sediment Sampling and Analysis

123

Periodically from October 2015 to December 2016, water column chemical and physical

124

conditions were characterized at seven locations across the south arm of GSL (Figure 1). Water

125

samples were collected at five stations (2267, 2565, 2767, N1018, 2820, 3510, and 4069) at 0.2

126

m below lake surface and 0.5 m above lake bottom, referred to as shallow and deep samples,

127

respectively. The data from these samples collected after the disappearance of the DBL were

128

compared to previous data collected from the shallow and deep brine layers from May 2007 to

129

December 2012, prior to causeway closure when the DBL was present22,34 (Figure 1). Water

130

column temperature, specific conductance, pH, and dissolved oxygen (DO) were measured in the

131

field using a YSI Professional Series Quatro probe that was calibrated within 12 hours prior to

132

sampling.22 The YSI probe corrects for temperature but not salinity. Thus, the DO values

133

presented have a positive bias and are qualitative, but are still useful for accurately denoting oxic

134

versus anoxic conditions. Sulfide was measured in filtered water samples in the field

135

immediately after collection using a photometric method (V-2000 Multi-analyte LED

136

Photometer and Vacu-vials®, CHEMetrics).22



137

Clean hands - dirty hands protocol was followed during sample collection and analysis.35

138

Unfiltered Hg water samples were collected by peristaltic pump using acid-washed PTFE tubing

139

into pre-cleaned FLPE bottles. Bottles used to collect anoxic water samples were filled to

140

overflowing in order to minimize headspace. Filtered water samples were passed through a 0.45-

141

micron pore size, pre-acid rinsed capsule filters in the field (Geotech Environmental). After

142

collection, water samples for HgT and MeHg analyses (unfiltered) were stored on ice in the field

143

and acidified to 0.5% using trace-metal grade sulfuric acid for preservation the same day as

144

sampling. All samples were then refrigerated and analyzed within two weeks of collection.

145

HgT samples were oxidized by amendment to 5% BrCl at least 24 hours prior to analysis.

146

This higher BrCl concentration relative to established protocols that call for 1% BrCl was

147

necessary36 to fully oxidize the high levels of dissolved organic matter (75.4 ± 11.1 mg/L) in

148

these water samples (as described below). HgT concentrations in water were determined via

149

reduction with SnCl2, purge and trap onto gold traps, thermal desorption, with quantification by

150

cold vapor atomic fluorescence spectroscopy (CVAFS) using a MERX-T automated system

151

(Brooks Rand) using established techniques.36 Mean laboratory blank concentrations were 0.07 ±

152

0.008 ng/L HgT. HgT matrix spike recovery averaged 98.3 ± 3.2% (n=3). MeHg concentration

153

in water was measured after distillation with ammonium pyrrolidine dithiocarbamate (APDC),

154

followed by aqueous phase ethylation, purge and trap onto tenax traps, thermal desorption,

155

pyrolitic decomposition, and CVAFS detection using a MERX-M automated system (Brooks

156

Rand) using established techniques.37 MeHg blanks averaged 0.15 ± 0.10 ng/L. Because no

157

certified reference material exists for MeHg in water, MeHg matrix spikes were included in each

158

distillation; MeHg spike recoveries (n=2) averaged 93 ± 29%. Water samples for dissolved

159

organic carbon (DOC) analysis were placed on ice in the field then placed in a refrigerator upon


Page 8 of 33

Page 9 of 33


160

return tio the laboratory. DOC was measured in water samples (TOC-5000a, Shimadzu) within

161

one week of sample collection using EPA Method 1684.38

162

Surficial sediment was sampled at eleven sites (A1, A2, B1, B2, C1, C2, CB, 2565, 3510,

163

4069, and N1018) in the GSL (Figure 1), and were collected by peristaltic pump using acid-

164

washed PTFE tubing into pre-cleaned 500 mL FLPE bottles filled to overflowing to minimize

165

headspace, and stored on ice in the field. The preferred method for examining vertical variations

166

involves collecting a sediment core that is subsequently sectioned. However, because our focus

167

was geographic variations in surficial sediment, we employed a faster method to collect

168

sediment. The method composites surficial sediment across depths of a few cm. Naftz et al.39

169

reported 40% variation of HgT concentration within the top 2 cm of GSL sediment. Previous

170

studies22 using this method showed relatively low variability in HgT and MeHg in surficial

171

sediment, indicating that collected sediment was representative.

172

Subsamples were oven dried at 105°C for 12 hours and re-weighed to determine water

173

content and allow conversion between wet and dry weight concentrations (without salt

174

correction). HgT was extracted from sediment by digestion in a 7:3 mixture of HNO3:H2SO4

175

(trace metal grade) at 80°C for 6 hours, followed by amendment to 5% BrCl.36 MeHg was

176

leached from sediment with a mixture of potassium bromide, sulfuric acid, and copper sulfate,

177

extracted into methylene chloride, back extracted into water, followed by aqueous phase

178

ethylation, purge and trap, thermal desorption, pyrolitic decomposition, and CVAFS detection

179

according to established techniques.36,37,41 Certified reference materials (CRMs) for HgT

180

(MESS-3) and MeHg (CC-580) in sediment were also analyzed, with recoveries averaging 99%

181

and 119%, respectively.

182



183 184

Biota Collection and Analyses Adult brine flies (Ephydra spp.) were collected from Lady Finger Point on Antelope

185

Island of the GSL (Figure 1) from spring to fall of 2012 to 2015. Flies were collected with nets,

186

transferred into polypropelene tubes, placed on ice in the field, and frozen back in the lab the

187

same day. Seventy-six waterfowl were harvested in mid-September of 2014 and ninety-nine

188

were harvested in late August and early September of 2015 from the GSL and surrounding

189

wetlands by the Utah Division of Wildlife Resources. The waterfowl were harvested from the

190

Farmington Bay, Ogden Bay, Howard Slough, and Harold Crane Waterfowl Management Areas

191

(Figure 1). Ducks were harvested with shotguns using non-lead shot then frozen. The age of each

192

bird was determined by physical characteristics; examining rectrices, wing and body plumage,

193

and cloacal characters. The species sampled were Northern Shovelers (Anas clypeata) (DBL

194

present: n = 16; DBL absent: n= 16), Mallard (Anas platyrhynchos) (DBL present: n= 16; DBL

195

absent: n= 21), Gadwall (Anas strepera) (DBL present: n = 15; DBL absent: n =22), and

196

Cinnamon Teal (Anas cyanoptera) (DBL present: n = 23; DBL absent: n = 31). Tissue samples

197

were thawed, then a scalpel was used to remove the skin, and breast muscle tissue was harvested

198

and stored in a Whirl-Pak (NASCO) polyethylene bag and frozen. Breast muscle was analyzed

199

because it is the most likely tissue to be consumed by duck hunters.

200

Both brine fly and waterfowl samples were freeze-dried and homogenized prior to

201

analysis, and thus all HgT concentrations in biota are reported on a dry weight basis. Brine flies

202

were digested in Teflon vials with 10 mL of a 2:1 mixture of trace metal grade nitric and sulfuric

203

acids. Samples were allowed to predigest at room temperature for 1 hour, then heated to 100 °C

204

for 4 hours. Following the digestion, samples were amended to 1% BrCl. All digestions included

205

at least 2 digestion blanks and 2 certified reference materials (TORT-2 and DORM-3, National


Page 10 of 33

Page 11 of 33


206

Research Council Canada), each digested in duplicate. Aliquots of the digested samples were

207

measured by oxidation with BrCl, reduction with SnCl2, purge and trap using dual-stage gold

208

trap amalgamation and quantification by CVAFS.36 The average daily HgT detection limit,

209

based on 3 times the standard deviation of digestion blanks, was 2.3 ng g-1, assuming a 100-mg

210

sample. Recoveries for HgT in the biota certified reference materials averaged 101.4% ± 7.8%, n

211

= 88. The duck muscle tissue was analyzed using thermal decomposition, amalgamation, and

212

atomic absorption spectrophotometry using a DMA-80 (Milestone) following established

213

protocols.42 Muscle tissue samples were only analyzed for HgT because previous studies have

214

shown that >95% of Hg in most bird muscle is MeHg43. Each analysis run included each of the

215

two certified reference materials (TORT-2 and DORM-3) analyzed in duplicate, with recoveries

216

of HgT ranging from 88% to 123%. Analytical precision of the HgT measurements was

217

assessed by analyzing eight samples in triplicate, and the average percent relative standard

218

deviation was 5.2%.

219

The HgT concentrations in the waterfowl were log transformed prior to statistical

220

analysis in order to meet the assumptions of parametric statistics. The data were analyzed using

221

multifactor ANOVA using a crossed design. While we did not anticipate a priori that the sex of

222

the waterfowl would have any effect on their HgT concentrations, we initially included sex in the

223

statistical model, which included the following five variables as fixed factors: duck species, age,

224

year, site, and sex. The fully crossed model could not be run due to missing cells because not all

225

sites were sampled both years. We therefore used a reduced model made by removing the five-

226

way and four-way interactions.44 In this model none of the interactions involving sex, nor sex as

227

a main factor, were significant (p>0.49 in all cases). Given this result along with our a priori

228

assumption regarding the importance of sex, this factor was removed from the model. The



229

multifactor ANOVA using a crossed design with the fixed factors duck species, age, year, and

230

site was then reduced by removing the four-way interaction and the three-way interaction Age ×

231

Site × Year due to the missing cells, as described above.

232 233

Results

234

Water Quality Parameters

235

Before 2013, prior to closure of the causeway culverts, shallow waters in the south arm

236

were oxic (DO = 9.48 ± 0.70 mg/L), whereas deep waters in areas where the DBL exists were

237

anoxic (DO = 0.05 ± 0.02 mg/L) (Figure 2). After causeway closure, mean DO was 9.1 ± 3.4

238

mg/L in shallow waters and 7.2 ± 5.2 mg/L in all deep waters. Prior to culvert closure DO

239

concentrations were significantly higher in surface waters than in the DBL (t-test, p

Total Mercury and Methylmercury Response in Water, Sediment, and

Recommend Documents