Accumulation of Terrestrial Dissolved Organic Matter Potentially

Subscriber access provided by Queen Mary, University of London

Characterization of Natural and Affected Environments

Accumulation of terrestrial dissolved organic matter potentially enhances dissolved methane levels in eutrophic Lake Taihu, China Yongqiang Zhou, Qitao Xiao, Xiaolong Yao, Yunlin Zhang, Mi Zhang, Kun Shi, Xuhui Lee, David C Podgorski, Boqiang Qin, Robert G. M. Spencer, and Erik Jeppesen Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b02163 • Publication Date (Web): 24 Aug 2018 Downloaded from http://pubs.acs.org on August 27, 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 35

Environmental Science & Technology

1

Accumulation of terrestrial dissolved organic matter potentially enhances dissolved methane levels in

2

eutrophic Lake Taihu, China

3

Yongqiang Zhoua, b, c, 1, Qitao Xiaoa, 1, Xiaolong Yaoa, b, Yunlin Zhanga, b, *, Mi Zhangd, Kun Shia, b, Xuhui

4

Leed, David C. Podgorskie, Boqiang Qina, b, Robert G.M. Spencerf, Erik Jeppesenc, g

5

a

6

of Sciences, Nanjing 210008, China

7

b

8

c

9

d

State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy

University of Chinese Academy of Sciences, Beijing 100049, China

Sino-Danish Centre for Education and Research, Beijing 100190, China

Yale-NUIST Center on Atmospheric Environment, International Joint Laboratory on Climate and Environment Change

10

(ILCEC), Nanjing University of Information Science and Technology, Nanjing 210044, China

11

e

12

70148, Louisiana, USA

13

f

14

g

Pontchartrain Institute for Environmental Sciences, Department of Chemistry, University of New Orleans, New Orleans

Department of Earth, Ocean and Atmospheric Science, Florida State University, Tallahassee, FL 32306, USA

Department of Bioscience and Arctic Research Centre, Aarhus University, Vejlsøvej 25, DK-8600 Silkeborg, Denmark

15 16

*

17

of Sciences, 73 East Beijing Road, Nanjing 210008, P. R. China, Tel: +86-25-86882198, Fax:

18

+86-25-57714759. Email address: [email protected].

19

1

20

This manuscript has not been published or accepted elsewhere. We have not submitted it to any other

21

journals. The manuscript is organized according to the format and structure of your journal.

Corresponding author: Yunlin Zhang, Nanjing Institute of Geography and Limnology, Chinese Academy

These authors contributed equally to this manuscript.

1 / 35

ACS Paragon Plus Environment


22

Abstract: Inland waters play an important role for the storage of chromophoric dissolved organic matter

23

(CDOM) and outgassing of methane (CH4). However, to date, linkages between the optical dynamics of

24

CDOM and dissolved CH4 levels remain largely unknown. We used multi-year seasonal data series

25

(2012–2014) collected from Lake Taihu and 51 connecting channels to investigate how CDOM optical

26

dynamics may impact dissolved CH4 levels in the lake. High dissolved CH4 in the northwestern inflowing

27

river mouths coincided with high underwater UV-Vis light availability, dissolved organic carbon (DOC),

28

chemical oxygen demand (COD), DOM aromaticity, terrestrial humic-rich fluorescence, in situ measured

29

terrestrial CDOM, and depleted dissolved oxygen (DO), stable isotopic δ2H and δ18O compared with other

30

lake regions. Our results further revealed positive relationships between dissolved CH4 and CDOM

31

absorption at 350 nm, i.e. a(350), COD, DOC, terrestrial humic-rich fluorophores, and DOM

32

aromaticity, and negative relationships between dissolved CH4 and DO, δ2H, and δ18O. The central lake

33

samples showed a major contribution of terrestrial-sourced molecular formulas to the ultrahigh resolution

34

mass spectrometry data, suggesting presence of allochthonous DOM sources even here. We conclude that

35

an elevated terrestrial CDOM input likely enhances dissolved CH4 levels in Lake Taihu.

36 37

Keywords: dissolved methane (CH4); chromophoric dissolved organic matter (CDOM); Lake Taihu;

38

carbon cycling; parallel factor analysis (PARAFAC)

39 40 41 42

2 / 35


Page 2 of 35

Page 3 of 35


43

TOC

44 45 46 47 48 49 50 51 52 53 54

3 / 35



55 56

Page 4 of 35

Introduction Inland waters (rivers, lakes, ponds, etc.) play an important role for the storage and outgassing of 1, 2

57

greenhouse gases

58

organic carbon transformation and methane (CH4) emission 3. Globally, lakes cover 3.7% of the Earth’s

59

ice-free land surface area and contribute approximately 20% of the CH4 outgassing from all natural

60

ecosystems 4. Eutrophication and climate warming have together fueled the outgassing of CH4 from lake

61

ecosystems

62

methanogenesis and oxidation of CH4

63

(acetoclastic methanogenesis) in carbon-rich environments serving as the terminal step in the

64

decomposition of dissolved organic matter (DOM) has traditionally been considered as the primary

65

pathway of biological CH4 production 6, 7. However, recent studies have highlighted that a large fraction of

66

CH4 oversaturation in aquatic environments can be produced in oxygenated surface waters

67

paradoxical supersaturation of CH4 in oxygenated surface waters has been suggested to be a result of

68

microbial transformation of DOM

69

may be more important than microbial degradation in controlling the decomposition of newly exposed

70

terrestrial DOM, further stimulating the outgassing of greenhouse gases in Arctic waters

71

experiments revealed significant photochemical CH4 production in the anoxic water column 12, especially

72

in the inflowing river mouths in this case.

73

3, 5

. Lake ecosystems, especially the most common shallow lake types, are hotspots of

. The net outgassing of CH4 from lakes to the atmosphere is the difference between

8, 9

6, 7

. Anaerobic fermentation of acetate by certain archaea

8, 9

. The

. Recent studies have also suggested that photochemical degradation

10, 11

. Irradiation

DOM plays a vital role in biogeochemical processes in inland waters and the composition of lake 13, 14

74

DOM is highly variable, depending on its source, biological reactivity, and time of processing

75

Terrestrial humic-rich substances usually contribute primarily to the DOM pool in lakes in the monsoon

4 / 35


.

Page 5 of 35


76

region, while autochthonous inputs from algae and macrophytes can also be important substrates for

77

microbial loop 13-15. Both terrestrial and autochthonous DOM can be utilized and transformed by microbial

78

and photochemical degradation and are important sources of carbon in lake ecosystems

79

of terrestrial and autochthonous algal-derived CDOM in inflowing river mouths can further result in

80

enhanced degradation of carbon and nitrogen in lakes, known as the priming effect

81

humic-rich substances including polycyclic aromatics and vascular plant-derived polyphenols are

82

photoreactive and photochemical degradation augments the lability of CDOM to the subsequent biological

83

degradation 14, with CH4 being released as byproduct 8, 9. Urbanization, industrialization, and conversion of

84

land to agriculture have led to increased discharge of domestic sewage and agricultural and industrial

85

effluents to lakes, which likely enhances microbial- and photochemical-induced oxygen depletion of the

86

river mouth water column

87

aquatic ecosystems depends largely on its sources and chemical composition 19. Oxygen depletion in river

88

mouths resulting from an augmented terrestrial DOM input has, however, lowered bacterial and archaea

89

oxidation of CH4 (aerobic and anaerobic, respectively) to CO2, a major pathway of biological CH4

90

oxidation, consequently reducing the levels of dissolved CH4

91

composition of DOM for the levels of dissolved CH4 in lake ecosystems remains to be elucidated.

16

16, 17

. The mixing

18

. Terrestrial

. The susceptibility of DOM to microbial and photochemical degradation in

20

. To date, however, the role of the optical

92

Lake Taihu is well-suited for the undertaking of studies to reveal the linkage between CDOM optical

93

dynamics and dissolved CH4 levels as it has high inputs of both allochthonous and autochthonous DOM.

94

The Lake Taihu watershed is located in the Yangtze River Delta, the most urbanized area in China with a

95

high population density and extensive economic development. Excessive input of nutrients has led to the

96

occurrence and persistence of nuisance blooms of cyanobacteria and high heterogeneity of DOM

5 / 35



Page 6 of 35

97

molecular weight and dissolved CH4 concentrations in the lake 21. The direction of the water flow of the

98

channels connecting Lake Taihu with the fluvial plain is primarily controlled by meteorological and

99

anthropogenic disturbance (e.g. floodgates) 22. However, to date, the relative importance of allochthonous

100

and autochthonous CDOM, regulated by the inflow and backflow discharge (Qnet, net inflow discharge),

101

for the dissolved CH4 levels has not been assessed.

102

We studied how the optical compositional dynamics of CDOM mediated by net inflow discharge

103

quantitatively drives the levels of dissolved CH4. A total of 700 CDOM samples and 552 headspace CH4

104

samples were collected in Lake Taihu and the connecting channels in the watershed from February 2012 to

105

November 2014. We expected to find large variations in dissolved CH4 levels among lake regions and

106

hydrological seasons due to different net inflow of terrestrial CDOM.

107 108

Materials and methods

109

Study sites

110

Lake Taihu has an area of 2,338.1 km2 and a water retention time of ~ 309 d

23

, and the lake

111

watershed covers 36,500 km2. A total of 172 channels are connected with the lake and the elevation of the

112

water surface of the connecting channels is mostly lower than 5 meters 24. The water exchange between the

113

lake and the connecting channels is complex and similar to that of lagoons. This leads to large variations in

114

the amount and composition of DOM in different areas of the lake. Information about hydraulic

115

sub-watersheds and hydrological data collection, multi-year variations of daily δ2H and δ18O (Fig. S2), and

116

monthly Qnet (inflow – backflow runoff) in each hydraulic sub-watershed (Fig. S3; Table S1) can be found

117

in the Supporting Information.

6 / 35


Page 7 of 35


118 119

Light availability, Secchi disk depth (SDD), dissolved oxygen (DO), and CDOM optical sampling

120

collection

121

Water depth (WD) was measured in situ using a scaled straight stick, while Secchi disk depth (SDD)

122

was determined in situ using a 30 cm diameter black and white Secchi disk. The dissolved oxygen (DO)

123

concentration was determined in situ at 0.5 m depth using a YSI 6600 V2 multi-sensor sonde. WD, SDD,

124

and DO data were only available from the seasonal campaign conducted in Lake Taihu from February

125

2012 to November 2014 (Table S2).

126

CDOM absorption and fluorescence measurements were conducted to trace the sources and optical

127

properties of CDOM and to investigate the impacts of CDOM optical compositional dynamics on the

128

levels of dissolved CH4. Detailed information about the seasonal CDOM sampling collection in the lake

129

and the connecting channels from 2012 to 2014 can be found in the Supporting Information (Table S2).

130

Previous studies have revealed that relatively high concentrations of dissolved organic carbon (DOC) 25, 26

131

are exported to Lake Taihu from the northwestern sub-watershed via Zhushan Bay

132

sampling campaign (n = 61) was therefore carried out in Zhushan Bay, located in northwestern Lake Taihu,

133

in April 2014 with the aim to trace the sources of CDOM and assess light availability (Fig. 1). Detailed

134

information about the measurement of CDOM fluorescence using an in situ CDOM fluorescence sensor,

135

light availability indices including the diffuse attenuation coefficient, Kd, of light in the UV region and the

136

photosynthetically active radiation (PAR, 400-700 nm), as well as measurements of total suspended matter

137

(TSM) and inorganic suspended matter (ISM), can be found in the Supporting Information (Table S2).

138

7 / 35


. An extensive


139

Page 8 of 35

Stable isotopic δ2H and δ18O sampling collection and measurements

140

Stable isotopic composition of surface water provides information on hydrological and meteorological

141

processes as evaporation fuels the accumulation of the heavier isotopic water molecules (H2HO and H218O)

142

in the liquid water

143

evaporation in Lake Taihu. We assumed that the source of riverine CDOM in Lake Taihu is consistent with

144

the depletion of δ2H and δ18O. Detailed information about the stable isotopic δ2H and δ18O sampling

145

collection in the lake and the connecting channels can be found in the Supporting Information (Table S2).

27

and these can therefore be used to trace the sources of riverine CDOM and

146 147

Chlorophyll-a (Chl-a), chemical oxygen demand (COD), DOC and CDOM optical measurements

148

Previous studies have indicated that the degradation of algal mats can impact the levels of dissolved 1, 3, 21

149

CH4

150

and dissolved CH4 concentrations in Lake Taihu was investigated. DOC, COD, and optical measurements

151

of CDOM were used to trace the sources and optical composition of CDOM (for information about Chl-a,

152

DOC, COD, and CDOM optical measurements see Supporting Information, Table S2).

. In this study, Chl-a was used as a surrogate for algal biomass, and the linkage between Chl-a

153

CDOM absorption indices included the absorption coefficient at 350 nm, a(350), the ratio of

154

a(250)/a(365), the spectral slope (S275–295), and the slope ratio (SR), which can be used to trace the amount

155

of CDOM

156

about the measurement and calculation of CDOM absorption indices is given in the Supporting

157

Information (Table S2).

17, 19, 28

and the relative molecular size and aromaticity of CDOM

29, 30

. Detailed information

158

CDOM fluorescence (excitation-emission matrices, EEMs) measurements and the calibration

159

processes including corrections of Raman and Rayleigh peaks, inner-filter effects, and normalization of

8 / 35


Page 9 of 35


160

Raman units (R.U.) can be found in the Supporting Information. PARAFAC is a three-way multi-variable

161

data analysis technique that statistically decomposes the complex mixture of CDOM fluorophores into

162

trilinear components with a single emission maximum and corresponding to a group of highly covarying

163

analytes 31, 32. Using PARAFAC modeling, a six-component model was well validated (Fig. 2; Fig. S4; Fig.

164

S5) (see details about the modeling and validation in the Supporting Information, Table S2).

165 166

Dissolved CH4 sampling collection and measurements

167

In total, dissolved CH4 concentrations were determined in 552 headspace samples (29 lake sites × 4

168

seasons × 3 years and 51 river sites × 4 seasons) (Fig. 1; Table S2). Direct measurement of dissolved CH4

169

was performed by headspace equilibration following the method detailed in Xiao et al.

170

Information, Table S2). We determined only the levels of dissolved CH4 but not CH4 emission as wind

171

speed and flow velocity data for the river samples were not available. Although the relationship between

172

the levels of dissolved CH4 and CH4 emissions is not 1:1, in this study we used dissolved CH4

173

concentrations as a proxy for the potential CH4 emission flux in different areas of Lake Taihu since the

174

levels of dissolved CH4 largely control the emission from the lake 21 (Fig. S1).

21

(see Supporting

175 176

Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) analyses

177

Details on the pretreatment of the FT-ICR MS samples can be found in the Supporting Information.

178

Briefly, in November 2016, FT-ICR MS samples collected from central Lake Taihu and headwater streams

179

of the lake basin as well as algal-derived samples to determine degradation were solid-phase extracted with

180

PPL Bond Elut (Agilent) resins before analysis by negative-ion electrospray FT-ICR MS

9 / 35


33, 34

to trace the


181

sources of DOM in the lake. Different sources of CDOM exhibit distinct molecular signatures

182

FT-ICR MS can be a useful tool in tracing the sources of CDOM in Lake Taihu.

Page 10 of 35

35

and

183

The assigned molecular formulas were categorized according to Ohno et al. 36. Briefly, the chemical

184

classes classified by the van Krevelen diagrams include: (i) lipids (O/C = 0 – 0.3, H/C = 1.5 – 2.0), (ii)

185

proteins and amino sugars (O/C = 0.3 – 0.67, H/C = 1.5 – 2.2), (iii) lignins (O/C = 0.1 – 0.67, H/C = 0.7 –

186

1.5), (iv) carbohydrates (O/C = 0.67 – 1.2, H/C = 1.5 – 2.2), (v) unsaturated hydrocarbons (O/C = 0 – 0.1,

187

H/C = 0.7 – 1.5), (vi) condensed aromatics (O/C =0 – 0.67, H/C = 0.2 – 0.7), and (vii) tannins (O/C = 0.67

188

– 1.2, H/C = 0.5 – 1.5) 36.

189 190

Statistical analyses

191

Statistical analyses, including mean values, standard deviations (SD), and t-tests, were performed

192

applying R-studio 0.97.551 (R i386 2.15.2) software. Linear regressions and van Krevelen diagrams were

193

made using MATLAB R2016b. Results with p < 0.05 for t-tests and linear regressions were reported as

194

significant. Means were shown ± SD.

195 196

Results

197

PARAFAC modeling results

198

Fluorescent DOM (FDOM) is composed of humic substances and amino-acids, both free and bound

199

in proteins, and alterations in PARAFAC-derived FDOM components are sometimes used as a surrogate

200

for changes in the composition of the wider DOM pool

201

explains > 99.9% of the variables of the whole EEMs dataset. The spectral characteristics of the six

32

. The six-component PARAFAC model

10 / 35


Page 11 of 35


202

components (C1-C6, Fig. 2; Fig. S4) were compared with PARAFAC results for samples collected from

203

other aquatic ecosystems using an online spectral library called Openfluor 37. C1 had two Ex maxima at ≤

204

230 and 285 nm and an Em maximum (340 nm) as have amino acid-associated tryptophan-like

205

components 38, 39. C2 displayed two Ex maxima at 240 and 350 nm and an Em maximum (468 nm) and is

206

categorized as a typical terrestrial humic-like substance

207

230/420 nm, similar to those of agricultural-soil-derived humic-like or fulvic-like materials

208

peaked at 380 nm and is categorized as a microbial humic-like substance

209

(275)/316 nm) and C6 (Ex/Em: ≤ 230 (270)/ ≤ 300 nm) had spectral characteristics similar to those of

210

red-shifted tyrosine and typical tyrosine substances, respectively

211

Fmax ranged from 0.24 ± 0.09 R.U. (microbial humic-like C4) to 5.04 ± 2.04 R.U. (tryptophan-like C1) for

212

the PARAFAC-derived C1-C6 in Lake Taihu with the relative contribution percentages for the six

213

components ranging from 2.2 ± 0.8% (C4) to 46.9 ± 19.0% (C1) (Table S3).

13, 38, 40-42

. C3 exhibited Ex/Em maxima at ≤

38, 46

38, 44, 45

40, 42-44

. C4

. C5 (Ex/Em: ≤ 230

. The multi-year (2012-2014) mean

214 215

Spatial variability of light availability, stable isotopic δ2H and δ18O, CDOM-related indices, COD, DO,

216

dissolved CH4, and Chl-a

217

SDD in the lake ranged from 0.0 to 1.7 m with a mean of 0.4 ± 0.3 m, high values being observed in

218

the northern and northwestern bays for the campaigns conducted seasonally from February 2012 to

219

November 2014 (Fig. S6). The ratio of SDD to WD ranged from 0.06 to 0.61 with a mean of 0.19 ± 0.12

220

and, similarly, high values of the ratio of SDD to WD were recorded in the northern and northwestern bays

221

of the lake (Fig. S6).

222

As evaporation enhances the accumulation of heavier water isotopic molecules of δ2H and δ18O in the

11 / 35



Page 12 of 35

223

liquid water, more depleted δ2H and δ18O imply a strong riverine signal, while enriched δ2H and δ18O

224

indicate evaporation induced by prolonged water retention time in the lake

225

from 2012 to 2014, δ2H and δ18O ranged from -63.4‰ to -16.5‰ (-33.9‰ ± 8.2‰) and from -8.9‰ to

226

0.4‰ (-4.6‰ ± 1.3‰), respectively. Multi-year mean δ2H and δ18O increased notably from the

227

northwestern bays to the southeastern bays in all seasons (Fig. 3; Fig. S7). δ2H and δ18O in the samples

228

collected from the inflowing rivers, i.e. in the northwestern sub-watersheds (e.g. Huxi and Wuchengxiyu)

229

were generally lower than in the corresponding bays and coastal lake regions.

27

. For all samples collected

230

In all samples collected from 2012 to 2014, DOC and a(350) ranged from 1.0 to 17.2 mg L-1 (4.4 ±

231

1.3 mg L-1) and from 0.4 to 16.0 m-1 (4.0 ± 1.6 m -1), respectively. Multi-year (February 2012 to November

232

2014) mean concentrations of DOC and a(350) and the Fmax of C1-C4 and C6 decreased markedly from

233

the northwestern part of the lake, especially Zhushan Bay, to the southeastern lake regions (Fig. 3; Fig. S7).

234

In comparison, a(250)/a(365), S275-295, and SR increased notably from the northwestern part of the lake to

235

the southeastern bays (Fig. S7). COD concentrations in the samples collected from the lake from 2012 to

236

2014 ranged from 2.43 to 15.56 mg L-1 with a mean of 4.27 ± 1.28 mg L-1, and COD decreased gradually

237

from the northwestern regions to the southeastern bays in all seasons (Fig. S6). DO concentrations in the

238

samples collected from the lake ranged from 2.47 to 12.28 mg L-1 with a mean of 8.86 ± 1.69 mg L-1, and

239

depleted DO was observed in the northwestern inflowing river mouths (Fig. S6).

240

Zhushan Bay located in northwestern Lake Taihu has relatively high concentrations of terrestrial,

241

agricultural, and microbial humic-like fluorophores as well as depleted DO (Fig. 3; Fig. S6), and the

242

sources of CDOM in the bay were investigated based on more comprehensive samplings

243

exclusively terrestrial humic-like C2 decreased from the northwestern Zhushan Bay and the northeastern

12 / 35


25

. The

Page 13 of 35


244

coastal lines to the southeastern bays in all seasons from 2012 to 2014 (Fig. 3; Fig. 4), coinciding with the

245

higher inflow runoff from the tributaries in the Huxi and Wuchengxiyu sub-watersheds than in the

246

remaining sub-watersheds. During the Zhushan Bay sampling campaign in April 2014, DOC ranged from

247

3.7 to 7.1 mg L-1 with a mean of 5.3 ± 0.8 mg L-1 (Fig. 4). In comparison, Fmax of terrestrial humic-like C2

248

and in situ measured fluorescence intensity (370/460 nm) ranged from 0.1 to 2.7 R.U. (0.8 ± 0.6 R.U.) and

249

from 21.5 to 200.0 µg L-1 (90.8 ± 54.2 µg L-1), respectively (Fig. 4). DOC concentrations, Fmax of terrestrial

250

humic-like C2, and measured in situ fluorescence intensity (370/460 nm) all decreased notably from the

251

northwestern inner Zhushan Bay towards the south entrance of the bay (Fig. 4). The sampling in Zhushan

252

Bay therefore indicated that high concentrations of terrestrial humic-like and anthropogenic

253

tryptophan-like fluorophores were discharged into the lake via the bay, potentially enhancing the levels of

254

dissolved CH4. Kd of light in the UV region and the PAR regions was fairly low, corresponding to a

255

relatively low TSM and high SDD at the inflowing river mouth compared with the open water region (Fig.

256

4; Fig. S6; Fig. S8). Correspondingly, a significant negative relationship was found between the Kd of light

257

in the UV region and the PAR regions and SDD (r2 = 0.89, p < 0.001) and with in situ measured

258

fluorescence intensity (370/460 nm) (r2 = 0.32, p < 0.001) (Fig. 4; Fig. S8).

259

Dissolved CH4 samples collected from the macrophyte-dominated lake regions (Fig. 1) were excluded

260

in our analyses since macrophytes influence the dynamics of the greenhouse gas exchange 1, 3. Submerged

261

macrophytes affect biological and chemical processes in shallow lakes, including the biomass of organisms

262

across trophic levels, and can therefore affect the dynamics of greenhouse gas 1. Submerged macrophytes

263

fuel the transport of dissolved and ebullitive CH4 from the sediment to the water column and can thus

264

increase dissolved CH4 concentrations in the water 3. This is supported by a recent study showing that

13 / 35



Page 14 of 35

265

organic sediments in macrophyte-dominated lake areas strongly enhanced the levels of dissolved CH4 21. In

266

this study, the boundaries of macrophyte-dominated lake regions were delineated following the results

267

detailed in Dong et al.

268

showed that East Lake Taihu Bay, Xukou Bay, and Guangfu Bay in the southeastern part of the lake were

269

macrophyte-dominated, which was well validated by the remote sensing results obtained by Liu et al.

270

Dissolved CH4 in all the remaining lake samples collected from 2012 to 2014 ranged from 4.2 to 618.3

271

nmol L-1, with a mean of 64.9 ± 92.3 nmol L-1 (Fig. 3). In the non-macrophyte-dominated lake regions,

272

dissolved CH4 decreased notably from the northwestern bays or the western coastal regions to the

273

southeastern lake regions in all hydrological seasons from 2012 to 2014 (Fig. 3). Dissolved CH4 in the

274

river samples ranged from 7.9 to 1597 nmol L-1, with a mean of 307.8 ± 347.5 nmol L-1 (Fig. 3), and was

275

significantly higher than in the lake samples from the Huxi, Wuchengxiyu, Zhexi, and Hangjiahu

276

sub-watersheds in all seasons (t-test, p < 0.05; Fig. 3; Table S4).

47

and Liu et al.

48

(Fig. 1). The field investigation conducted by Dong et al.

47

48

.

277

As the temporal resolution of the sampling campaign was seasonal, we found highly depleted δ2H and

278

δ18O, and enhanced input of terrestrial C2, and thereby high concentrations of dissolved CH4, in the

279

northwestern lake regions in November compared with August due to drought on hot August days in the

280

Lake Taihu watershed (Fig. 3). An increase in the temporal resolution of the sampling campaign can

281

probably capture the rainy season enhancement of dissolved CH4, while in the present results the seasonal

282

enrichment signal was surpassed by the spatial signal.

283

Significantly higher Chl-a concentrations were observed in August (37.8 ± 27.6 µg L-1) and

284

November (21.0 ± 18.1 µg L-1) than in February (9.4 ± 3.8 µg L-1; t-test, p < 0.005) and May (12.4 ± 7.1 µg

285

L-1; t-test, p < 0.05). Chl-a decreased from the northern regions of the lake to the southeastern bays in all

14 / 35


Page 15 of 35


286

seasons except in February from 2012 to 2014 (Fig. 3), coinciding with higher external loading from the

287

northwestern sub-watersheds, for instance Huxi and Wuchengxiyu. Chl-a in the river samples ranged from

288

0.5 to 770 µg L-1, with a mean of 42.9 ± 74.3 µg L-1, and showed no significant difference from the mean

289

Chl-a in the lake samples.

290 291

Relationships between dissolved CH4 and stable isotopic, CDOM-related indices, Chl-a and hydrological

292

indices

293

Significant positive relationships were found between mean dissolved CH4 and mean a(350) (r2 =

294

0.11, p < 0.001), DOC (r2 = 0.36, p < 0.001), Fmax of PARAFAC-derived C1-C3 (r2 = 0.07~0.50, p < 0.01),

295

COD (r2 = 0.43, p < 0.001), and monthly net inflow runoff (Qnet) (p < 0.001) (Table S5) for all the samples

296

collected in the lake watershed. It should be noted that the coefficients of determination between mean

297

dissolved CH4 and mean terrestrial PARAFAC-derived C2 and the products of C2 and Qnet, i.e. Qnet × C2,

298

can be as strong as 0.50 and 0.27, respectively (Table S5). In comparison, significant negative

299

relationships were found between dissolved CH4 and a(250)/a(365) (r2 = 0.14, p < 0.001), S275-295 (r2 =

300

0.31, p < 0.001), SR (r2 = 0.28, p < 0.001), stable isotopic δ2H (r2 = 0.25, p < 0.001) and δ18O (r2 = 0.10, p

301

< 0.001), and DO (r2 = 0.11, p < 0.001) for all the samples collected (Table S5; Fig. S9). No significant

302

relationship emerged between dissolved CH4 and Chl-a or with C4-C6 if all data were used during linear

303

fitting, but exclusion of river samples resulted in significant positive relationships between dissolved CH4

304

and Chl-a (r2 = 0.18, p < 0.001; Fig. S9), C4 (r2 = 0.10, p < 0.001), and C6 (r2 = 0.08, p < 0.01). A close

305

relationship was found between dissolved CH4 and Chl-a if we include only August samples (r2 = 0.47, p

Accumulation of Terrestrial Dissolved Organic Matter Potentially

Recommend Documents