Global and Local Sources of Mercury Deposition in Coastal New

Subscriber access provided by University of Winnipeg Library

Characterization of Natural and Affected Environments

Global and local sources of mercury deposition in coastal New England reconstructed from a multi-proxy, high-resolution, estuarine sediment record William F. Fitzgerald, Daniel R Engstrom, Chad R. Hammerschmidt, Carl H. Lamborg, Prentiss H. Balcom, Ana L. Lima-Braun, Michael H Bothner, and Christopher M. Reddy Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b06122 • Publication Date (Web): 13 Jun 2018 Downloaded from http://pubs.acs.org on June 15, 2018

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 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 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.

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 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Environmental Science & Technology

Global and local sources of mercury deposition in coastal New England reconstructed from a multi-proxy, high-resolution, estuarine sediment record William F. Fitzgerald*,1, Daniel R. Engstrom2, Chad R. Hammerschmidt3, Carl H. Lamborg4, Prentiss H. Balcom1,†, Ana L. Lima-Braun5‡, Michael H. Bothner6, and Christopher M. Reddy5 1

Department of Marine Sciences, University of Connecticut, Groton, Connecticut 06340, USA

2

St Croix Watershed Research Station, Science Museum of Minnesota, Marine on St. Croix, Minnesota 55047, USA 3

Department of Earth & Environmental Sciences, Wright State University, Dayton, Ohio 45435 USA 4

Department of Ocean Sciences, University of California Santa Cruz, Santa Cruz, California 95064, USA 5

Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, USA 6

United States Geological Survey, Woods Hole Science Center, Woods Hole, Massachusetts 02543, USA

18 19

ABSTRACT

20

Historical reconstruction of mercury (Hg) accumulation in natural archives, especially lake

21

sediments, has been essential to understanding human perturbation of the global Hg cycle. Here

22

we present a high-resolution chronology of Hg accumulation between 1727 and 1996 in a varved

23

sediment core from the Pettaquamscutt River Estuary (PRE), Rhode Island (USA). Mercury

24

accumulation is examined relative to (1) historic deposition of polycyclic aromatic hydrocarbons

25

(PAHs) and lead (Pb) and its isotopes (206Pb/207Pb) in the same core, and (2) other

26

reconstructions of Hg deposition in urban and remote settings. Mercury deposition in PRE

27

parallels the temporal patterns of PAHs, and both track industrialization and regional coal use

28

between 1850 and 1950 as well as rising petroleum use after 1950. There is little indication of

29

increased Hg deposition from late 19th-century silver and gold mining in the western USA. A

30

broad maximum of Hg deposition during 1930–1980, and not found in remote sites, is consistent

31

with the predicted influence of additional industrial sources and commercial products. Our

32

results imply that a significant portion of global anthropogenic Hg emissions during the 20th

33

century was deposited locally, near urban and industrial centers of Hg use and release.

1

ACS Paragon Plus Environment


34

INTRODUCTION

35

The pivotal role of the atmosphere as the primary medium for mobilizing mercury (Hg) from

36

natural and anthropogenic sources is well established 1-3. On a global scale, elemental Hg (Hg0)

37

is the principal species transported, and knowledge of its behavior and fate and that of other

38

airborne species such as reactive gaseous Hg is increasing 4, as is the variety and sophistication

39

of regional and global simulations of Hg cycling in nature 1, 2, 5. Broadly viewed, modeling the

40

pathways of Hg in the environment and the impact of human activities on the Hg cycle uses an

41

“industrial ecological” approach 6, 7, which analyzes historical and modern data on sources,

42

emission factors, fluxes, and depositional chronologies, which provide both model inputs and

43

validation. One of the ongoing challenges to these efforts is the attribution of emission sources

44

contributing to changing Hg deposition in different regions and over time 8.

45

Given the relatively long atmospheric residence time of Hg0 and its predominance in

46

contemporary emissions, Hg deposition today tends toward regional uniformity – with limited

47

local differentiation or “hot spots” – implying that continental to global-scale sources are

48

dominant8. However, that is unlikely to have been the case in the past, as Hg emission sources

49

have changed markedly over time, in type, location, and strength, with reactive Hg species

50

(particulate and reactive gaseous Hg) dominating certain historical sources 9. Such changes have

51

been effectively documented through reconstruction of Hg deposition trends from natural

52

archives, especially lake sediments, which show in several regional syntheses, contrasting

53

deposition trends in lakes located either remote from or near urban and industrial centers 10-12.

54

Sediments in remote lakes typically reveal a steady increase in Hg deposition from preindustrial

55

times to the present day with an overall rise of 3.6 ± 1.2 × the preindustrial rate 13, while lakes

56

near urban centers frequently exhibit much larger increases with marked declines in recent

57

decades associated with emission reduction efforts 10.

58

Here we add a critical new record of near-urban Hg deposition that more fully illustrates the

59

contrasting trends of Hg inputs to aquatic environs resulting from the changing strengths of local

60

versus global-scale emission sources. The record at hand is a high-resolution chronology of Hg

61

deposition between 1727 and 1996, as determined in a well-studied sediment core from the

62

Pettaquamscutt River Estuary (PRE) in Rhode Island, northeastern U.S. 14-17. The PRE core is

63

unique among lacustrine archives of near-source deposition in that Hg accumulation rates are

2


Page 2 of 21

Page 3 of 21


64

relatively unaffected by local land-use disturbance and thus primarily record changes in

65

atmospheric inputs. These new results for Hg accumulation in this scrupulously dated, varved

66

repository (1–3 year resolution) provide a quantitative means for assessing the magnitude,

67

timing, and relative source strengths of human-related Hg emissions and deposition over nearly

68

three centuries. This information should be especially useful in improving models of the global

69

and regional atmospheric cycling of Hg, and assessing the biogeochemical impacts associated

70

with changing emission sources (e.g., coal combustion; metal mining and smelting; commercial

71

uses and products) in modern and historic settings 18, 19.

72

The significance of these new findings is examined relative to (1) the historical depositional

73

patterns of pyrogenic polycyclic aromatic hydrocarbons (PAHs) and lead (Pb) and its isotopes

74

(206Pb/207Pb) as previously determined in the same core, and (2) other temporal reconstructions

75

of Hg deposition from dated sediment cores. The question of whether Hg accumulating in the

76

estuary is derived primarily from either local or more regional sources is explored based on

77

known processes of atmospheric deposition and environmental cycling of PAHs, Hg, and Pb.

78

EXPERIMENTAL SECTION

79

Study Area.

80

The Pettaquamscutt River Estuary (PRE), also known as the Narrow River, is oriented in a north-

81

south direction in Washington County, southern Rhode Island and flows into the western part of

82

Narragansett Bay (Fig. 1) 17. A key feature of this 9.7-km long estuary is the presence of two

83

permanently stratified kettle basins, and upper (10.5 m maximum depth) and lower (19.5 m

84

depth), which were created during the last deglaciation, about 19,000 years ago 17. The lower

85

basin was chosen for this investigation because of its greater depth and intense density

86

stratification. Its morphometry, hydrology, biology, and geochemistry have been investigated

87

and well characterized through a variety of studies beginning in the 1970s 20-24. A fjord-like

88

estuarine circulation has caused its bottom waters and sediments to remain continuously anoxic

89

for the past 1700 ± 300 years 20, 25. In the early 1990s this basin was the location of one of the

90

first multidisciplinary investigations of the biogeochemical cycling of Hg speciation (including

91

methyl- and dimethyl Hg) in a stratified estuarine system 26. The depth of the oxic/anoxic

92

transition lies between 3.5 and 6 m, below which a sulfidic zone exists 15, 26. The anoxic, sulfidic

3



93

conditions prevent sediment infaunal activity, thereby creating a stable environment for the

94

undisturbed accumulation of settling particles and the development of varved sediment

95

sequences caused by seasonal diatom blooms in surface waters. Such laminated archives are

96

ideal for a high-resolution temporal examination of the deposition of Hg and other substances 14-

97

17

98

Sampling and Dating.

99

Mercury was determined in sediment aliquots from the same core samples used by Lima and co-

100

workers 14-17 in their investigations of the depositional history of PAHs, Pb, and 206Pb/207Pb. This

101

sediment core and six others were collected in 1999 with a rectangular freeze corer by Hubeny

102

and co-workers 17. Their study provides full details on the coring techniques, sample processing,

103

and dating methodologies. The dating of the sediment layers is well constrained by excellent

104

agreement between the 210Pb chronology and that determined by counts of annually resolved

105

varved layers in the core 14-16. Additional dating verification was provided by both the 1963 137Cs

106

peak associated with pre-test ban treaty nuclear detonations and that associated with the 1986

107

Chernobyl reactor explosion. The presence of well-defined laminated deposits from the 1938 and

108

1954 hurricanes provided an additional check on the varve chronology.

109

Dry mass sediment accumulation rates (DMAR) and chemical fluxes through time were derived

110

from this chronology. Chemical fluxes were corrected for a sediment focusing

111

enhancement/factor of 1.3, which was established from a comparison of the sedimentary

112

inventory for 210Pb with that expected from atmospheric 210Pb deposition in this region 16. The

113

aforementioned papers 14-17 detail the chronology, flux determinations, and critical aspects of the

114

analysis of PAHs, Pb, and Pb isotopes. The focus here is on the Hg results and their connections

115

to the indices provided by the additional atmospherically borne tracers of anthropogenic activity

116

(PAHs, Pb) in the lower basin of PRE. It is important to note that DMAR is relatively constant

117

over the period of record in this core, indicating minimal impact of human land-use change on

118

erosion rates and associated delivery of soil-bound pollutants from the watershed. Such impacts

119

can otherwise confound interpretation of atmospheric deposition trends in sediment records from

120

near-urban lakes 10.

.

4


Page 4 of 21

Page 5 of 21


121

Hg analyses.

122

Concentrations of total Hg in freeze-dried sediment were determined using a Milestone DMA80

123

pyrolytic mercury analyzer 27-29. A certified reference material (either MESS-3 estuarine

124

sediment or TORT-2 lobster hepatopancreas; National Research Council Canada) was analyzed

125

approximately every ten samples, and two or three duplicate samples were analyzed during each

126

analytical day. The average relative percent difference of duplicate analyses of sediment was 1.9

127

(range 0.1 to 3.9), and mean (± standard deviation) recoveries were 97 ± 4% for MESS-3 and

128

103 ± 3% for the high organic TORT-2.

129

Mercury fluxes were calculated as the product of the focusing-corrected DMAR and Hg

130

concentration for each interval. Mercury flux ratios were determined from the flux of Hg at key

131

time periods normalized to the mean (background) flux in preindustrial times.

132

RESULTS AND DISCUSSION

133

Hg concentrations and fluxes.

134

The concentration (ng g-1) and sedimentary flux (µg m-2 y-1) of Hg between 1727 and 1996 are

135

presented in Figure 1. The resolution is unusually fine with 139 determinations over the 269-year

136

interval. As a consequence, not only is the broad historic pattern of variation evident, but short-

137

term changes and fine features are also apparent. Background values are approached between

138

1832 and 1822, when the concentration and depositional flux show little change. A background

139

flux of 13.6 ± 2.0 µg m-2 y-1 was determined by averaging the measurements prior to 1832 (n =

140

41). There is a near linear increase of ~2.6% yr-1 between 1832 (17.3 µg m-2 y-1) and 1901 (49.2

141

µg m-2 y-1), interrupted by a small peak between 1842 and 1863. The Hg accumulation rate

142

changes little between 1901 and 1926 (48.4 ± 2.2 µg m-2 y-1) after which it increases abruptly to

143

a broad maximum (average = 164 ± 21 µg m-2 y-1 ; n = 23) between 1949 and 1975, and then

144

declines to the most recent period (1989–1996) where it varies between 53.2 and 80.5 µg m-2 y-1

145

(average = 65.0 ± 9.9 µg m-2 y-1 ; n = 8), yielding a coefficient of variation of 15%, which is

146

comparable to the degree of uncertainty observed for the background flux. Both preindustrial and

147

recent Hg fluxes in the PRE core are substantially higher than present-day measurements of wet

5



148

deposition (6-8 µg m-2 y-1) 30, indicating that a significant portion of Hg inputs to the estuary is

149

ultimately derived from watershed runoff.

150

The temporal pattern for Hg accumulation in PRE sediments follows the historical source

151

function for coal consumption nationally in the U.S., especially during the period when coal was

152

the primary energy source 31. Coal usage began about 1850, and this date coincides reasonably

153

with the increase in the PRE Hg flux that commenced, as noted, around 1830. An earlier increase

154

in wood and biomass burning is a likely explanation for the nearly 20-year difference between

155

the initial rise in Hg and that of coal usage. A general correspondence with U.S. coal

156

consumption continues to about 1930 when Hg flux in the PRE core increases abruptly, likely

157

owing to large additional emissions from other industrial sectors 19.

158

PAHs, Pb, and Pb-isotopes.

159

Fluxes of Hg from this work and those for Pb and the pyrogenic PAHs – phenanthrene, pyrene,

160

and benzo(a)pyrene – follow strikingly similar temporal patterns (Fig. 2). The Pb flux is near

161

constant before about 1820, increases – steadily at first and then more sharply after 1930 – to

162

peak values in the 1970s; the peak is followed by an equally precipitous decline to the present

163

day. PAHs are relatively constant prior to 1840, increase gradually from about 1840 to 1930, and

164

then rise sharply to maximum values in the mid-1950s. Thereafter, fluxes decline – at first

165

sharply and then irregularly – into the 1990s.

166

The initial increase of Pb flux to PRE around 1820 (Fig. 2b) has been attributed to increasing

167

lead ore smelting in the upper Mississippi Valley of the U.S. and is manifest in a marked peak in

168

the 206Pb/207Pb ratio about 1840 (Fig. 2a) 14. Thereafter, coal combustion overtook lead-ore

169

smelting as the dominant source of atmospheric Pb, lowering 206Pb/207Pb ratios and increasing

170

both Pb and Hg fluxes to PRE. A more substantial increase, by a factor of 4, in the Pb flux began

171

about 1920 and peaked in 1975. This enhancement is coincident with the combustion of

172

tetraethyl lead in gasoline. Use of tetraethyl Pb in the U.S. peaked during 1972–1973, declined

173

25% by 1979, and decreased 65% by 1983 14. This correspondence between the temporal history

174

of Pb use in U.S. gasoline and that of Pb accumulating in the PRE archive is striking. It provides

175

additional validation of the integrity and fidelity of this sedimentary record in capturing the

176

atmospheric depositional history of Pb, Hg, and PAHs. Further, a significant difference between

6


Page 6 of 21

Page 7 of 21


177

the Hg and Pb depositional histories is the dramatic increase in concentration and sedimentary

178

flux of Hg between 1930 and 1980 with a maximum occurring during the 1950–1975 period

179

(Figs. 1, 2). This difference points to additional sources of Hg which, as suggested, are probably

180

related to the increasing releases of Hg from commercial use 19.

181

Parallel trends in the sedimentary accumulation of PAHs and Hg also point to pyrogenic

182

emissions and atmospheric deposition as a primary input of these substances to the estuary 15.

183

Thus, the increases between 1820 and 1930 reflect the transformation of energy sources from

184

wood burning to coal combustion, which became increasingly significant in the 1880s as

185

regional industrialization began to slowly surpass agriculturally based economies (Fig. 2) 15, 32.

186

The marked rise in Hg fluxes after 1930 coincides with further expansion of industrial activity

187

and use of Hg in commercial products that greatly increase emission sources and releases to the

188

environment 19. According to their assessment of global Hg releases from commercial activities

189

between 1850 and 2010, Horowitz and co-workers 19 posit 1) a maximum in 1890 in accord with

190

the Streets et al. 33 projections, 2) a comparable peak in 1930, and 3) a larger peak in 1970

191

followed by a decline (Fig. 3a). Their hypothesized Hg emissions peak in 1890 – attributed to Hg

192

use in silver and gold extraction – is not present in the sedimentary record of PRE, but the broad

193

maximum in sedimentary Hg accumulation during the 1930–1980 time period is consistent with

194

the influence of additional and variable sources. Not only do such sources, and those from fossil

195

fuel combustion, vary temporally and spatially, but additional variability should be anticipated

196

from evolving regulatory actions at the state and federal level (e.g., U.S. Clean Air and Water

197

Acts of 1970 and 1972).

198

Comparisons with other studies.

199

An historical record of the sedimentary accumulation of Hg, remarkably similar to that at PRE,

200

was observed in a dated core from Oyster Pond, a partially anoxic coastal estuary in Falmouth,

201

Massachusetts, about 80 km east of the lower basin of PRE (Fig. 1) 34. The historical Hg flux

202

ratios in Oyster Pond and PRE are plotted together in Figure 4a. Although dating of the Oyster

203

Pond core is not as detailed as that for PRE, the agreement of the magnitude and timing of

204

changing Hg fluxes is quite reasonable. Indeed, the maximum flux ratios in both basins occur

205

between 1940 and 1960, and the overall temporal patterns show a consistent relationship. This

7



206

regional coherence suggests that the Hg deposited in both water bodies was transported from a

207

similar suite of emission sources.

208

A key complement to the present work is also provided by the historical reconstruction of the

209

accumulation of Hg, Pb, stable Pb isotopes, and PAHs in a dated marine sediment core from the

210

southern Lebanese continental margin in the Levantine Basin (LB), eastern Mediterranean (Fig.

211

3b) 35. Contaminant inputs to the LB site were primarily atmospheric, but in contrast to PRE,

212

emission sources were longer range (e.g., Central and Eastern Europe) rather than local/regional,

213

and the temporal pattern for the sedimentary accumulation of Hg was more similar to that

214

reported for remote lakes worldwide 13. Mercury flux in the LB core increases gradually from

215

about 1850 (i.e., preindustrial) to 1970 and then more sharply to modern times (1987–2003),

216

when the flux ratio is 4.5 ± 0.3 (n = 5). This modern value is comparable to the PRE results

217

where the flux ratio is 4.7 ± 1.0 (n = 8) between 1989 and 1996. By contrast with the LB core,

218

the PRE record shows a broad region of enhanced Hg deposition between 1930 and 1980 with a

219

peak flux ratio of 14.4 in 1950 and an average of 12.1 ± 2.4 (n = 23) between 1950 and 1975

220

(Fig. 3). This period, as noted, coincides with the assessment by Horowitz et al. 19 of major

221

increases in human-related Hg releases from commercial products with maxima in 1940 and

222

1970.

223

Local versus regional/global differences in Hg depositional patterns have also been observed in

224

other investigations. For example, Engstrom et al. 10 compared historical Hg accumulation in 20

225

lake cores from the mid-continent metropolitan environs of Minneapolis/St. Paul (Twin Cities),

226

Minnesota with those obtained from 20 remote lakes located in the northeast sector of the state

227

(Fig. 4). The Twin Cities cores display historically elevated Hg accumulation that is attributed to

228

deposition from local emission sources and variable inputs from catchment erosion. The

229

temporal patterns vary among lakes, which is expected for an urban area. However, the mean

230

historical deposition pattern in the Twin Cities cores is very similar to the magnitude and timing

231

of Hg fluxes to PRE (Fig. 4b). The Twin Cities cores show a broad enhancement between about

232

1900 and 2007 with maximum deposition occurring during the 1940–1970 period (Fig. 4c). In

233

contrast, the temporal pattern of sedimentary Hg accumulation in the remote Minnesota lakes is

234

similar to reconstructions reported for other remote lakes 13, and the Levantine Basin 34, and

235

displays a gradually increasing trend starting at about 1850 (i.e., preindustrial ) and continuing to

8


Page 8 of 21

Page 9 of 21


236

modern times (Fig. 4d). The strikingly enhanced Hg depositional signals associated with

237

localized/regional Hg emissions and releases are not evident in remote, northeastern Minnesota

238

lakes, at a modest distance of 350–400 km from the Twin Cities.

239

Environmental implications.

240

Dated lake sediments have been the primary source of secular data for reconstructing Hg

241

accumulation and depositional patterns associated with preindustrial periods to the modern era.

242

An examination of these archives from remote regions reveals a gradually increasing trend from

243

about 1850 to modern times and an accumulation factor relative to preindustrial levels, or

244

anthropic global increase, that is remarkably uniform at 3.6 ± 1.2 13. In contrast, results from the

245

present study of sedimentary Hg accumulation in the lower basin of PRE from 1727 to 1996

246

display a broad maximum during the 1930–1980 time period, with greatest deposition occurring

247

between 1950 and 1975. Similar temporal changes of Hg deposition were found on a

248

local/regional scale (

Global and Local Sources of Mercury Deposition in Coastal New

Recommend Documents