Climate and vegetation as primary drivers for global mercury storage

Subscriber access provided by The Library | University of Bath

Environmental Processes

Climate and vegetation as primary drivers for global mercury storage in surface soil Xun Wang, Wei Yuan, Che-Jen Lin, Leiming Zhang, Hui Zhang, and Xinbin Feng Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b02386 • Publication Date (Web): 21 Aug 2019 Downloaded from pubs.acs.org on August 24, 2019

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 33

Environmental Science & Technology

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

Climate and vegetation as primary drivers for global mercury storage in surface soil Xun Wang†, Wei Yuan†,‡, Che-Jen Lin ⊥,§, Leiming Zhang∥, Hui Zhang†,Xinbin Feng†, #*, †State

Key Laboratory of Environmental Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences, Guiyang, China. ‡University

of Chinese Academy of Sciences, Beijing, China. for Advances in Water and Air Quality, Lamar University, Beaumont, TX, USA §Department of Civil and Environmental Engineering, Lamar University, Beaumont, TX, USA ∥Air Quality Research Division, Science and Technology Branch, Environment and Climate Change Canada, Toronto, ON, Canada #Center for Excellence in Quaternary Science and Global Change, Chinese Academy of Sciences, Xian 710061, China ⊥Center

15 16 17

*To

whom correspondence should be addressed. Email: [email protected]

18

ACS Paragon Plus Environment


19

Abstract: Soil is the largest Hg reservoir globally. Data of Hg concentration in surface soil are

20

fundamental to understanding environmental Hg cycling. However, present knowledge on the

21

quantity and global distribution of Hg in soil remains deficient. Using stable Hg isotopic

22

analyses and geospatial data, the concentration and global spatial distribution of Hg in surface

23

soil of 0-20 cm depth have been developed. It is estimated that 1088±379 Gg of Hg is stored in

24

surface soil globally. Thirty-two percent of the surface Hg storage resides in

25

tropical/subtropical forest regions, 23% in temperate/boreal forest regions, 28% in grassland

26

and steppe and shrubland, 7% in tundra, and 10% in desert and xeric shrubland. Evidence from

27

Hg isotopic signatures points to atmospheric Hg0 dry deposition through vegetation uptake as

28

the primary source of Hg in surface soil. Given the influence of changing climate on vegetative

29

development, global climate change can act as an important forcing factor for shaping spatial

30

distribution of Hg in surface soil. This active forcing cycle significantly dilutes the impacts

31

caused by Hg release from anthropogenic sources, and needs to be considered in assessing the

32

effectiveness of reducing Hg use and emissions as specified in Minamata Convention on

33

Mercury.

34


Page 2 of 33

Page 3 of 33


35 36



37

Page 4 of 33

1. Introduction

38

The ratification of Minamata Convention on Mercury (Hg) in 2017 is widely considered

39

a success to protect human health and the environment from the detrimental effects of Hg1.

40

However, existing Hg data and scientific knowledge are insufficient to adequately assess the

41

policy benefits in reducing human and wildlife Hg exposure 2.To date, the spatial distribution

42

of Hg concentration in soils across the globe remains largely uncertain

43

reported 235-1150 Gg of Hg storage in surface soil across the globe

44

critical for a complete understanding of the Hg cycling. Uncertainties in soil Hg data limit our

45

ability to quantify legacy Hg emission from soil to the atmosphere 2, estimating Hg runoff from

46

uplands to the aquatic environment 9, and assessing potential risks of human and wildlife Hg

47

exposure globally10.

3, 4.

3, 5-8.

Earlier studies

Such a dataset is

48

The concentration and distribution of soil Hg are best estimated with representative soil

49

samples collected in a global observational network, but such an approach is not feasible

50

because of the limitation of access and cost to perform such a global study. Statistic regression

51

modeling (e.g., Multiple Linear Regression) using the large geological database provides

52

another alternative for reliably constructing global concentration and distribution of Hg in soil.

53

Using available data of Hg concentration and distribution in China and US 11-13, coupled with

54

global organic carbon, meteorological factors, and anthropogenic Hg emissions datasets, it is

55

possible to gain insight on the contributing sources and accumulation pathways of Hg in soil at

56

a global scale.

57

The sources of Hg in natural surface soils are from atmospheric deposition of elemental

58

Hg (Hg0) (i.e., atmospheric Hg0 uptake by vegetation and input to soil with plant detritus) and

59

oxidized Hg (Hg2+)9,

60

accumulation from contaminated waters is likely an important source in areas that have

61

received air emission or water discharge of Hg from human activities

62

localized near anthropogenic sources. Since direct or standardized quantification methods for

63

measuring Hg dry deposition is not available, Hg dry deposition is often estimated by models,

64

or by surrogate measurements such as measuring Hg deposition through litterfall or on artificial

65

surfaces

66

terrestrial and aquatic ecosystems. Networks of Hg wet deposition measurement have been

15.

14, 15,

and from weathering of Hg from rocks9,

14, 16.

4, 17,

Runoff-related

which is largely

Mercury wet deposition has been the primary focus in quantifying Hg input to


Page 5 of 33


67

established globally, such as Global Mercury Observation System (GMOS)

68

Atmospheric Deposition Program’s (NADP) Mercury Deposition Network (MDN) sites in

69

North America

70

deposition to the terrestrial ecosystem. Specifically, atmospheric Hg0 uptake by vegetation is

71

responsible for substantial Hg accumulation in forest ecosystems

72

observed seasonal variation of atmospheric Hg concentrations23, 24. Climate is the main driver

73

for vegetation distribution globally 25-28.The atmospheric Hg0 dry deposition through vegetation

74

uptake whose spatial distribution is driven by vegetative development

75

changing climate. Therefore, global climate change is an important forcing factor for shaping

76

spatial distribution of Hg in surface soil. Time profiles of Hg deposition reconstructed from

77

lake sediments and peat bogs have demonstrated that the changes of climate and vegetation

78

coverage as well as the related forest fire paleo-history are the main drivers for the natural

79

changes in Hg deposition flux and subsequent concentrations in terrestrial ecosystems 31-33.

15.

18

and National

However, recent studies have highlighted the importance of Hg dry

14, 19-22,

as well as for the

29, 30closely

tied to the

80

Recent advancement of stable Hg isotope techniques enables delineation of the origins

81

and accumulations processes of Hg in surface soil. Mercury undergoes both mass dependent

82

fractionation (MDF, reported as δ202Hg) and mass independent fractionation (MIF, reported as

83

Δ199Hg, Δ200Hg and Δ201Hg) in the environment 16. The MIF signatures of Hg isotopes are much

84

less likely to be altered by post-depositional processes than MDF is in the environment, making

85

the MIF signatures of Hg useful tracers to identify different sources 14, 16, 34. Atmospheric Hg2+

86

is characterized by positive Δ199Hg, Δ200Hg and Δ201Hg values while Hg0 in the atmosphere

87

shows negative values of Δ199Hg and Δ201Hg with negligible Δ200Hg14, 16, 34. Geological Hg pool

88

exhibits zero Δ199Hg, Δ200Hg and Δ201Hg14, 16, 34.

89

We hypothesize that the active processes driven by climate and vegetative development

90

influence soil Hg spatial distribution at continental and global scales. This study focuses on

91

atmospheric Hg0 deposition forced by climate and vegetative development, as well as its effect

92

on soil Hg spatial distribution at a global scale. Comprehensive data of Hg isotopic signatures

93

measured in global terrestrial ecosystems are analyzed, coupled with structural equation

94

modeling using geospatial datasets collected in China and the US, to verify the hypothesis.

95

Based on the analysis, global spatial distribution of Hg in the top 0-20 cm surface soil (consisted

96

of organic soil and upper mineral soil in terrestrial ecosystems) is constructed. The implications



97

Page 6 of 33

of the soil Hg storage on global Hg cycling under climate change scenarios are discussed.

98 99

2. Materials and Methods

100

2.1. Sites description, sampling and measurements. Hg concentration and isotopic

101

composition in vegetation and soil samples were measured at 13 agricultural land sites, 26 forest

102

sites, 11 grassland sites, and 2 Tibetan wetland sites across China (Figure S1). The samples,

103

collected from 2014 to 2018, include tree foliage, litterfall, grass, and straw, and soil at 0-20

104

cm and 100 cm depth. All samples were collected during 2014-2018. The site information is

105

provided in Table S1 of the Supported Information (SI). The sampling protocol has been

106

described in detail in our earlier studies

107

established for sample collection, and each ~500 g of samples for well mixed vegetation, 0-20

108

cm surface soils and 100 cm deep soils were collected. Yang, et al. 35 suggested that vegetation

109

samples freeze dried or oven-dried at < 65°C were suitable for determination of Hg. Vegetation

110

samples were dried in an oven at 60°C for 48 hours until 0.05, by Paired-T test).

14.

Briefly, five 5m×5m subplots at each site were

120

The procedure for Hg isotope measurement of vegetation and soil samples has been

121

described in our earlier work14(detailed in the SI). Mercury isotopic compositions are measured

122

using the Standard-Sample-Standard (SPS) protocol. The Hg-MDF is reported in δ notation

123

using the unit of permil (‰) referenced to the neighboring NIST-3133 solution:

124

δ202Hg (‰) =1000 × [(202Hg/198Hgsample)/(202Hg/198HgNISTSRM3133) - 1]

125

MIF is reported as ΔxxxHg following the convention suggested by Blum and Bergquist 40:

126

Δ199Hg (‰) = δ199Hg − 0.2520 × δ202Hg

(2)


(1)

Page 7 of 33


127

Δ200Hg (‰) = δ200Hg − 0.5024 × δ202Hg

(3)

128

Δ201Hg (‰) = δ201Hg − 0.7520 × δ202Hg

(4)

129

UM-Almadén secondary standard was analyzed for every 10 samples. The results of UM-

130

Almadén are consistent with reported values (δ202Hg=-0.53±0.04‰, Δ199Hg=-0.01±0.04‰,

131

Δ201Hg=-0.01±0.05‰, Δ200Hg=0.01±0.03‰,n=13) and BCR-482 (δ202Hg=-1.56±0.06‰,

132

Δ199Hg=-0.63±0.02‰, Δ201Hg=-0.63±0.02‰, Δ200Hg=0.03±0.03‰, n=4) and GSS-4 (δ202Hg=-

133

1.72±0.08‰,

134

Mean±1SD, n=4) 40, 41.

135

2.2 Mapping of global surface soil Hg concentration. The data collection and availability are

136

described in the SI. Given the diverse spatial and temporal scales of these datasets, all data were

137

regridded into 1°×1° grid cell. The volume of dataset for 0-20 cm surface soil Hg concentration

138

in China (about 50,000 points) is much larger than the volume of dataset in the US (about 2,000

139

points) in this study. To ensure data comparability, the grid cell in the US with less than 3 data

140

points of surface soil Hg concentration was not considered. Similarly, the datasets of climate

141

and vegetation factors, population distribution, and anthropogenic Hg emission were regridded

142

into the same resolution of 1°×1°. Then, structural equation modeling and regression analyses

143

were processed by Amos 21 and SPSS 16. ArcGIS 10.4 was utilized to create the global map

144

of Hg concentration in surface soil. The global 0-20 cm surface soil Hg storage (HgP) is

145

estimated as:

146

Δ199Hg

=-0.34±0.03‰,

Δ201Hg

=-0.34±0.03‰,

HgP = HgC × BD × (1-f) × 20

Δ200Hg=-0.00±0.02‰,

(5)

147

where HgC is soil Hg concentration, BD is the soil bulk density, and f is the soil coarse

148

fragments (>2 mm) volume in 0-20 cm surface soil. Given the climate, vegetation and human

149

influences (details in the Section of Results and Discussion), we used the MLR (Multiple Linear

150

Regression) to obtain the empirical model to predict the global surface soil Hg concentration

151

as following:

152

log(HgC1) = -12.893 + 0.551log(P) + 7.651log(T) - 0.085log(NPP) – 1.366log(R) +

153

0.029log(Hg2+) + 0.357*log(SOC),

154

where HgC1 is the 0-20 cm surface soil Hg concentration (ng g-1, induced by climate, vegetation

155

and human influences), P is precipitation (mm), T is air temperature (K), NPP is net primary

156

production (g m-2 day-1), R is solar radiation (w m-2), Hg2+ is anthropogenic Hg2+emission (g

R2=0.57, P0.05). The Hg MIF signatures in foliage at rural sites exhibit consistent

205

values (Figures S1 and 1) across the globe.

16, 34, 36.

Combining our

206

Intriguingly, samples of submerged vegetation and surface sediment from Tibetan

207

wetlands exhibit positive Δ199Hg and Δ200Hg values (Figure 1), characteristic of the MIF

208

signatures of Hg2+ in precipitation water, and indicating a possibility that atmospheric Hg2+

209

deposition is the dominant sources. Estimates from the mixing model using Δ200Hg (detailed in

210

SI) yields Hg2+ (through precipitation) and Hg0 (through vegetative uptake) contributes to

211

68±21% and 32±19% of vegetation Hg, respectively. The triple endmember mixing model

212

using Δ199Hg and Δ200Hg (detailed in SI) suggests 26±19% of surface soil Hg in Tibetan

213

wetlands is from atmospheric Hg0 deposition, 40±17% from atmospheric Hg2+ deposition, and

214

34±18% inherited from parent soil material. This is different from earlier studies which

215

revealed that Hg in streams and runoffs from forest catchment is mainly derived from dry

216

deposition of atmospheric Hg0 45, 46. Mercury sources in Tibetan wetlands (mainly supplied by



217

precipitation and glacier meltwater47) clearly distinct from those for forest surface soil, showing

218

the characteristics of atmospheric Hg2+ deposition.

219

Surface soil in agricultural lands are mainly influenced by human activities, and therefore

220

has elevated Hg concentrations (>200 ng g-1) with isotopic signatures similar to Hg emission

221

from anthropogenic sources (~0 Δ199Hg and Δ201Hg signatures, p= 0.324 by T-test)16, 34, 48.

222

Except for Tibetan wetlands, the MIF signatures (Δ199Hg and Δ200Hg, Figure 1) of surface soil

223

Hg in forest, peatland, grassland and tundra ecosystems are more characteristic of vegetation

224

rather than those found in atmospheric water (positive Δ199Hg and Δ200Hg)14, 16, 34 and geological

225

inputs (~0 Δ199Hg and Δ200Hg)16, 49. This suggests wet deposition of atmospheric Hg cannot

226

explain the observed isotopic characteristics in soil samples. In addition, precipitation intensity

227

is the dominant factor (rather than Hg concentration in precipitation) that leads to Hg wet

228

deposition at remote sites globally6, 7. If Hg input from precipitation contributed significantly

229

to soil Hg, observed Δ199Hg shift in surface soil samples would exhibit positive correlation to

230

precipitation intensity. However, the data in this study show opposite trend, as indicated by the

231

negative correlation observed between precipitation and Δ199Hg in rural forest and grassland

232

ecosystems (r=-0.55, p < 0.001, n=48 sites, Figure S3). Moreover, the triple endmember mixing

233

model shows that 54±21% (mean±1SD) of Hg in the surface soil of global documented

234

ecosystems is derived from atmospheric Hg0 deposition, 12±9% from atmospheric Hg2+

235

deposition, and 34±18% inherited from parent soil material. This is consistent with earlier

236

findings that dry deposition of Hg0 is the primary source of soil Hg14, 19, 21, 22, 36, 37.

237

The significantly negative correlation between precipitation and Δ199Hg can be explained

238

by vegetative development enhanced by precipitation, which increases biomass production and

239

therefore a greater atmospheric Hg0 uptake vegetation that leads to more negative Δ199Hg in

240

surface soil after litterfall. Such evidence has been documented in earlier Hg isotope studies

241

across North American forests and Eastern Tibetan Plateau forests that higher precipitation and

242

warmer temperature enhance plant productivity and subsequently increase Hg accumulation in

243

soil through atmospheric Hg0 deposition through plant uptake 8, 9. Since the spatial distribution

244

of vegetative development is closely tied to the climatic factors25-28, the changed vegetation

245

distribution will also be forced by climate change, thus shaping the soil surface Hg geospatial

246

distribution.


Page 10 of 33

Page 11 of 33


247

3.2 Influences of climate change and vegetation development on soil Hg. It has been

248

suggested that surface soil Hg concentration is correlated with soil organic carbon (SOC), leaf

249

area index (LAI), plant net primary production (NPP), and anthropogenic Hg emission14, 50-52.

250

We categorized these factors into three groups: climate factors (e.g., precipitation, temperature,

251

and solar irradiance), vegetation-related factors (e.g., LAI, NPP, and SOC), and human factors

252

(e.g., population and anthropogenic Hg emission). Few studies have reported the interplays

253

among surface soil Hg concentration and these factors at a continual or global scale. Cross

254

analysis of data collected (n=1541 1°×1° grid cells) in China and the US (Figure S2) shows that

255

Hg concentration in surface soil is highly correlated to climate factors, moderately correlated

256

to vegetation related factors, and weakly correlated to anthropogenic activities (population and

257

anthropogenic Hg emission). This suggests that climate and vegetation play an important role

258

in shaping the spatial distribution of Hg in surface soil.

259

Factors contributing to the spatial variation of soil Hg as explained by the structural

260

equation modeling pathway network are shown in Figure 2. The direct effect by vegetation is

261

0.66, compared to the much smaller effect by human activities (0.14). This suggests the direct

262

impact from human activities is largely invisible over a large spatial scale. This is plausible,

263

since anthropogenic emissions of Hg is primarily from point sources. In addition, the Hg

264

isotopic data suggest wet deposition (removing GOM in the atmosphere generally) is less

265

important than previously perceived as emissions are prone to long range atmospheric transport.

266

The variation of vegetation development is controlled by the climate factor (direct effect up to

267

0.93), leading to a confounded effect of climate-vegetation forcing at 0.61. This is much greater

268

than the total effect of 0.09 by climate-human activities as indicated by the isotopic evidence

269

that relates Hg accumulation in soil to vegetation productivity.

270

The high Hg0 concentrations caused by anthropogenic Hg release in industrialized areas

271

is greatly diluted during chemical transport processes over the long residence time of Hg0 (0.5-1

272

year)

273

accumulation in vegetated ecosystems. Since climate is a predominant driver for global

274

vegetation distribution, the impact caused by the changing climate and subsequently vegetation

275

distribution enriches soil Hg concentration in remote regions, such as Tibetan forests14, 54 and

276

arctic tundra37, 55. This is also supported by structural equation modeling results and Hg isotopic

53.

Dry deposition of Hg0 via vegetation uptake is the main source for soil Hg



277

signatures at large scale.

278

Analytical results from the structural equation modeling also infer that climate has a strong

279

impact on anthropogenic activities (direct effect=0.63, Figure 2). The significant effect of

280

climate on anthropogenic activities can be explained by population growth in the temperate

281

climate zone. For example, regions with >800 mm precipitation in China make up only 43% of

282

land area, but have 94% of the population and substantially elevated soil Hg concentration 11,

283

56.

284

hidden by climate changes. In addition, since human activities and vegetation development can

285

impose feedback to climate in Figure 2, the climate-vegetation-human interactions are likely to

286

occur under future land-use and climate change scenarios, which need to be considered when

287

assessing global Hg cycling.

288

3.3 Climate and vegetation influence on global distribution of Hg in surface soil. Using

289

multivariate curvilinear regression and geospatial mapping (Section 2.2), global distribution of

290

Hg concentration in top 0-20 cm surface soil is constructed using the WWF (World Wildlife

291

Fund) 14 terrestrial ecoregion classifications. The Hg concentration in soil varies from 10 ng

292

g-1 to 289 ng g-1 with a mean of 47±26 ng g-1 and a median of 45 ng g-1 (Figure 3, Table S3).

293

The model-predicted concentrations were verified using documented soil Hg spatial

294

concentration dataset in continental surface agricultural and grazing land soils in Europe, and

295

at 740 sites values across the globe. Figure 3 shows that the histogram of simulated Hg

296

concentration agrees with the observational data of continental surface soil in Europe. The

297

scatterplot of model-predicted Hg concentration versus observation yields a slope of 0.71 (R2

298

= 0.72, n=740, p

Climate and vegetation as primary drivers for global mercury storage

Recommend Documents