Carbon Emission from Cascade Reservoirs: Spatial Heterogeneity

Subscriber access provided by UNIV OF ESSEX

Article

Carbon Emission from Cascade Reservoirs: Spatial Heterogeneity and Mechanisms Wenqing Shi, Qiuwen Chen, Qitao Yi, Juhua Yu, Yuyu Ji, Liuming Hu, and Yuchen Chen Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b03590 • Publication Date (Web): 01 Oct 2017 Downloaded from http://pubs.acs.org on October 2, 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 26

Environmental Science & Technology

1

Carbon Emission from Cascade Reservoirs: Spatial Heterogeneity and

2

Mechanisms

3

Wenqing Shi†, Qiuwen Chen*,†, Qitao Yi‡, JuhuaYu†, YuyuJi†,§, Liuming Hu†, Yuchen

4

Chen†

5

†

6

Nanjing 210098, China

7

‡

8

Huainan 232001, China

9

§

Center for Eco-Environment Research, Nanjing Hydraulic Research Institute,

School of Earth and Environment, Anhui University of Science and Technology,

College of Water Conservancy and Hydropower Engineering, Hohai University,

10

Nanjing 210098, China

11

*

Corresponding author: Tel./Fax: +86 2585829765; E-mail: [email protected]

1

ACS Paragon Plus Environment


12

Abstract: Carbon emission from reservoirs is considered to tarnish the green

13

credentials of hydropower, and has been extensively studied in single reservoirs.

14

However, it remains unclear how carbon emission differs in cascade reservoirs and

15

the mechanism behind the differences. In this study, carbon dioxide (CO2) and

16

methane (CH4) emissions from cascade hydropower reservoirs were measured in the

17

Lancang River, the Chinese section of the Mekong River. Our results demonstrate that

18

carbon emissions from the river were increased by dam construction, but exhibited

19

spatial heterogeneity among cascade reservoirs. The first, most upstream reservoir

20

acted as the hotspot of CH4 and CO2 emissions, which were 13.1 and 1.7 times higher

21

than those in downstream reservoirs, respectively. Similarly, the CH4/CO2 ratio of

22

0.023 in the first reservoir was higher than the others and made a greater contribution

23

to the global warming effects of the cascade reservoirs. The sediment organic carbon

24

in downstream reservoirs was negatively correlated with reservoir age (r2 = 0.993)

25

and decreased at a rate of 0.389 mg g-1yr-1, suggesting a potential decrease of carbon

26

emission in the future. This study adds to our understanding of carbon emissions from

27

cascade reservoirs and helps to screen effective strategies for future mitigation of the

28

global warming effects from cascade hydropower systems.

2


Page 2 of 26

Page 3 of 26


29

Table of Contents

30

3



31

INTRODUCTION

32

Global warming has intensified extreme weather and climate events, such as

33

hurricanes, heat waves and floods.1-3 Carbon dioxide (CO2) and methane (CH4), the

34

two most important greenhouse gases (GHGs), contribute 64% and 19% to global

35

warming effects, respectively.4,5 Since the pre-industrial era, atmospheric CO2 has

36

increased by 35%, at an average rate of 1 ~ 3 ppmv yr-1 for the past several decades,

37

while CH4 has more than doubled from 0.7 to 1.7 ppmv over the same period.6 Inland

38

waters (rivers, reservoirs and lakes) have been identified as important sources of CO2

39

and CH4, and emit 2,100 TgC yr-1 and 700 TgC yr-1 (CO2-equivalents) to the

40

atmosphere, respectively.7,8

41

Rivers are active in the transport, mineralization and burial of organic matters,

42

playing a critical role in the global carbon cycle.7,9,10 It was estimated that 400 ~ 900

43

Tg organic carbon (OC) is transported to oceans every year through global river

44

networks.11 The mineralization of OC during river transport can lead to direct CO2

45

and CH4 emissions to the atmosphere. In many instances, carbon emissions from

46

rivers may meet or exceed export to the ocean, significant impacting the Earth’s GHG

47

budget.12,13 Carbon emission was shown to be significantly affected by river

48

hydrologic regimes.14,15 Major alterations to the physical structure in disturbed rivers

49

may change the hydrologic regime and thereby CO2 and CH4 emissions.

50

In an effort to ameliorate the growing water shortages and energy demands, over

51

70,000 large dams have been built worldwide, with more dams under construction,

52

planned or proposed.16 Many of these reservoirs are built in a cascade configuration, 4


Page 4 of 26

Page 5 of 26


53

especially in large rivers. The hydropower generated by these dams has been

54

considered to be green energy; however, there has been an ongoing scientific debate

55

over the role of hydropower in GHG emissions to the atmosphere. River segmentation

56

and disruption by damming can block the transport of suspended and bedloaded

57

particles, leading to sediment accumulation in reservoirs.15,17 It has been estimated

58

that river segmentation and damming have caused the transport of terrestrial OC to

59

the ocean to be attenuated by 26%.18 Sediment and associated carbon trapped in

60

reservoirs can be mineralized and released to the atmosphere as CO2 and CH4, turning

61

reservoirs into hotspots of GHG emissions. Until now, previous research has mainly

62

focused on carbon emissions from single reservoirs, while the spatial variations of

63

carbon emissions from cascade reservoirs and the underlying mechanisms have been

64

less studied.15,19,20

65

In this study, the fluxes of CO2 and CH4 from cascade hydropower reservoirs were

66

investigated in the Lancang River, the upstream portion of the Mekong River in China.

67

The OC in water and sediment was analyzed to identify the pathways through which

68

OC was mineralized to CO2 and CH4. Bacteria are considered as main drivers of OC

69

mineralization, and the quantitative polymerase chain reaction (qPCR) method was

70

employed to analyze sediment bacteria abundance to illustrate the associated

71

molecular mechanism. The objective of this study was to investigate spatial patterns

72

of carbon emissions from cascade reservoirs, with a goal of informing future efforts to

73

mitigate the global warming effects of cascade hydropower exploitation.

74 5



Page 6 of 26

75

MATERIALS AND METHODS

76

Study Area. The Lancang-Mekong River (Lancang River is used hereafter), one of

77

the largest rivers in the world, originates from the Tibetan Plateau and discharges into

78

the South China Sea. It has a length of 4.9 × 103 km, a watershed area of 7.6 × 105

79

km2, and a mean annual discharge of 4.6 × 102 km3 at a rate of 1.5 × 104 m3 s-1.21,22

80

Because of its rich hydropower resources, the Lancang River has already been heavily

81

dammed, with many more dams under construction or in the planning stage. In China,

82

six hydropower dams had been built by 2016 (Figure 1), including Gongguoqiao

83

(GGQ), Xiaowan (XW), Manwan (MW), Dachaoshan (DCS), Nuozhadu (NZD) and

84

Jinghong (JH). These dams create a series of cascade reservoirs, whose main features

85

are described in table 1.

86

Table 1. The main features of cascade hydroelectric reservoirs in the Lancang River.

GGQ

XW

MW

DCS

NZD

JH

Age a (yr)

4

7

23

13

2

7

Dam height (m)

105

292

132

115

261.5

108

Water level (m)

1307

1240

994

899

812

602

Storage capacity (108 m3)

3.5

149.1

5.0

9.4

237.0

11.4

Discharge volume (108 m3)

318.5

381.6

388.0

419.0

545.6

574.0

HRTb (yr)

0.01

2.36

0.78

0.30

1.87

0.40

Installed capacity (106 kW)

0.90

4.20

1.50

1.35

5.85

1.75

87

a

88

b

In 2016 when sampling campaign was carried out. HRT is hydrological residence time. 6


Page 7 of 26


89

Sampling and Chemical Analysis. The sampling campaign was carried out at 23

90

sites (Figure 1) in September 2016, including 5, 3, 3, 3, 3, 3 and 3 sites in the

91

upstream river, GGQ, XW, MW, DCS, NZD and JH, respectively. Reservoirs can be

92

segmented into riverine, transitional and lacustrine zones, with sediment accumulation

93

maximized in the lacustrine zone, potentially intensifying carbon emissions to the

94

atmosphere15. Thus, samplings in each reservoir were mainly conducted in the

95

lacustrine zone in this study. At each site, surface water (2 L) and sediment (100 g)

96

samples were collected in triplicate using a stainless-steel bucket and an Ekman grab

97

sampler, respectively. The collected water and sediment samples were kept frozen in

98

the dark and transported to the laboratory for analysis within three days. Additional

99

information detailing the field campaign (sampling time, water level, air temperature

100

and wind speed) can be found in Table S1 in the supporting information.

101

At each sampling site, water temperature, pH, conductivity and turbidity were

102

measured in situ using a Yellow Springs Instruments multi-parameter meter (Yellow

103

Springs, Ohio, USA). Total nitrogen (TN), total phosphorus (TP) and chemical

104

oxygen demand (COD) in water were determined according to the Monitoring

105

Analysis Method of Water and Wastewater.23 The analysis of dissolved organic

106

matters in water was performed by measuring DOC, ultraviolet absorbance at 254 nm

107

(UV254) and excitation-emission matrix (EEM) fluorescence spectra. Water samples

108

were filtered through 0.45-µm membrane filters before analysis. DOC was measured

109

using a TOC analyzer (Liqui TOC II, Elementar, Germany). UV254 was determined

110

using a UV-Vis spectrophotometer (756PC, Shanghai Sunny Hengping Scientific 7



111

Instrument Co. Ltd., China) at 254 nm with a quartz cell path of 1 cm. The EEM

112

fluorescence spectra were measured using a fluorescence spectrophotometer (F-7000,

113

Hitachi High-Tech. Corp., Japan). The spectra were collected with subsequent

114

scanning emission spectra from 250 to 550 nm at 5 nm increments by varying the

115

excitation wavelength from 200 to 400 nm at 5 nm increments. The excitation and

116

emission slits were maintained at 5 nm and the scanning speed was set at 1000 nm

117

min-1. The spectrum of pure water was used as the blank. Sediment OC was analyzed

118

using a vario MACRO cube elemental analyzer (Elementar Inc., Germany). Fresh

119

sediment was freeze-dried and ground before analysis. Approximately 30 mg of

120

sample was weighed in tin cups and acidified with 2 drops of 8% H3PO4 to remove

121

inorganic carbonates before OC analysis.

122 123

Figure 1. Location of cascade dams and sampling sites in this study.

124

CH4 and CO2 Fluxes. The CH4 and CO2 fluxes across the water-air interface were

125

quantified using the Thin-Boundary Layer method described by Liss and Slater 8


Page 8 of 26

Page 9 of 26


126

(1974)24 in equation (1),

= − ∙

127

(1)

128

where Fgas is the gas flux from water to air (mg m-2h-1), Cw is the gas concentration in

129

surface water (mg L-1), and Ceq is the gas concentration in surface water that is in

130

equilibrium with the atmospheric concentration (mg L-1), Kgas is the gas transfer

131

velocity (m h-1).

132

The dissolved CH4 and CO2 concentrations in surface water were analyzed using

133

the headspace equilibration method.25 Briefly, a 20-ml water sample was collected

134

from 5 ~ 10 cm below the surface using a 60-ml polypropylene syringe equipped with

135

a three-way stopcock. 20 ml ambient air was added to the syringe to create a

136

headspace. Then, the sample syringe was shaken vigorously for 2 min and left to

137

stand for 30 min. The equilibrated headspace gas was injected into a pre-evacuated

138

Exetainer® vial (839 W, Labco, UK) for storage until analysis using a gas

139

chromatograph (7890B, Agilent Technologies, USA). The ambient air at each

140

sampling site was also analyzed. The dissolved CH4 and CO2 concentrations in

141

surface water were calculated according to Henry’s law26,27.

142 143 144

The gas transfer velocity was calculated using the following equation (2) from Liss and Merlivat (1986),28

. K = 2.07 + 0.215 ∙ U ∙ !"

#

(2)

145

where U10 is the wind speed at a height of 10 m above the water surface (m s-1), Sc is

146

the gas Schmidt number at the water surface temperature according to Wanninkhof

147

(1992).29 n is -2/3 for U10 ≤ 3.7 m s-1 and -1/2 for U10 > 3.7 m s-1.30 9



Page 10 of 26

148

Since CH4 has a 25 times higher global warming potential than CO2 on a mass

149

basis over a hundred-year horizon, the CH4/CO2 ratio significantly affects the global

150

warming effects of carbon emission31. In this study, the CH4/CO2 ratios were also

151

calculated and compared among different sampling sites.

152

Microbial Abundance Analysis. The total bacteria and methanogens in sediment

153

were quantified using the qPCR method. DNA extraction was undertaken with

154

Powersoil DNA Isolation Kits according to the manufacturer’s instructions. The

155

qPCR assay was performed using primers targeting total bacterial 16S rDNA (primer

156

set,

157

1106F/1378R).32,33 Gene copies were amplified and quantified in a Bio-Rad cycler

158

equipped with the iQ5 real-time fluorescence detection system and software (version

159

2.0; Bio-Rad). All reactions were completed in a total volume of 20 µL containing 10

160

µL SYBR® Premix Ex TaqTM (TOYOBO, Japan), 0.5 mM of each primer, 0.8 µL

161

BSA (3 mg mL-1, Sigma), double-distilled water and template DNA. The qPCR

162

program for total bacterial 16S rDNA commenced with 95°C for 60 s, followed by 40

163

cycles of 95°C for 30 s and 60°C for 15 s and 72°C for 30 s. The qPCR program for

164

archaeal 16S rDNA was as follows: 95°C for 60 s, followed by 40 cycles of 95°C for

165

25 s and 57°C for 30 s and 72°C for 60 s. The standard curve was established by serial

166

dilution (10-2 ~ 10-8) of known concentration plasmid DNA with the target fragment.

167

All PCRs were run in triplicate on 96-well plates (Bio-Rad) sealed with

168

optical-quality sealing tape (Bio-Rad). Three negative controls without DNA template

169

were included for each PCR run.

F341GC/R518)

and

methanogenic

archaeal

16S

10


rDNA

(primer

set,

Page 11 of 26


170

Data Analysis. Significant differences in carbon emission, sediment OC, and

171

bacterial abundance between sites were determined by one-way analysis of variance

172

followed by a post-hoc Duncan’s multiple range test using SAS (SAS 9.2, SAS

173

Institute Inc.). The level of significance was P < 0.05 for all tests.34

174

RESULTS

175

Environmental Characteristics. The characteristics of water and sediment in the

176

upstream river and downstream cascade reservoirs are presented in Table 2. Water

177

temperature and pH maintained steady levels at 21.2 ~ 25.7oC and 8.3 ~ 9.6,

178

respectively, while water conductivity and turbidity showed a general decrease along

179

the flow direction. The water conductivity and turbidity decreased from 367.9 us cm-1

180

and 87.9 FNU in the upstream river to 260.7 us cm-1 and 12.3 FNU in JH, respectively.

181

From the upstream river to the cascade reservoirs, sediment TP showed no significant

182

difference (P < 0.05) and remained steady at 0.44 ~ 0.60 mg g-1, while sediment TN

183

increased from 0.27 mg g-1 in the upstream river to 0.87 ~ 1.60 mg g-1 in the

184

downstream reservoirs.

11



Page 12 of 26

Table 2. The characteristics of water and sediment in upstream river and cascade reservoirs

185

Upstream

GGQ

XW

MW

DCS

NZD

JH

Temperature (oC)

21.2 ± 0.9

23.1 ± 1.0

25.7 ± 2.8

21.0 ± 1.0

25.7 ± 0.3

27.8 ± 3.2

22.9 ± 0.5

pH

8.3 ± 0.7

8.6 ± 0.2

9.0 ± 0.4

8.6 ± 0.3

8.3 ± 0.1

9.6 ± 0.1

8.6 ± 0.3

Conductivity (us cm-1)

367.9 ± 30.4

371.8 ± 12.8

296.8 ± 66.2

289.7 ± 5.8

286.7 ± 15.1

241.6 ± 29.4

260.7 ± 2.3

Turbidity (FNU)

87.9 ± 29.1

35.8 ± 10.1

5.0 ± 1.0

6.6 ± 1.9

30.4 ± 16.8

2.7 ± 0.7

12.3 ± 1.5

CODMn (mg L-1)

1.82 ± 0.26

1.13 ± 0.47

2.10 ± 0.10

1.20 ± 0.10

1.70 ± 0.44

1.53 ± 0.31

1.10 ± 0.10

Sediment TN (mg g-1)

0.27 ± 0.03

1.14 ± 0.08

1.34 ± 0.07

0.87 ± 0.14

1.34 ± 0.10

1.60 ± 0.43

1.53 ± 0.06

Sediment TP (mg g-1)

0.54 ± 0.02

0.60 ± 0.02

0.44 ± 0.04

0.62 ± 0.03

0.57 ± 0.04

0.49 ± 0.08

0.45 ± 0.00

186

12


Page 13 of 26


187

Water-Air Carbon Flux. The CH4 flux in the upstream river kept at a low level of

188

0.03 mg d-1m-2; however, high CH4 fluxes were observed in the cascade reservoirs,

189

especially in GGQ. The CH4 flux was 3.02, 0.14, 0.42, 0.32 0.10 and 0.10 mg d-1m-2 in

190

GGQ, XW, MW, DCS, NZD and JH, respectively (Figure 2A). The CO2 flux varied

191

with sampling sites, but was the highest in GGQ as well, where it reached 4.07 × 102

192

mg d-1m-2. Other sampling sites yielded lower levels of CO2 fluxes, at 1.23 × 102, 0.79 ×

193

102, 1.58 × 102, 2.63 × 102, 0.58 × 102 and 2.03 × 102 mg d-1m-2 in the upstream river,

194

XW, MW, DCS, NZD and JH, respectively (Figure 2A). Compared to the upstream

195

river, carbon emission had a higher CH4/CO2 flux ratio in the cascade reservoirs, which

196

was maximized in GGQ. The CH4/CO2 flux ratio in GGQ, XW, MW, DCS, NZD and

197

JH was 0.023, 0.005, 0.007, 0.005, 0.004 and 0.001, respectively (Figure 2B).

198 199

Figure 2. Carbon flux across the water-air interface in the upstream river and cascade

200

reservoirs. (A) CH4 and CO2 fluxes, (B) CH4/CO2 flux ratio. Error bars indicate

201

standard deviations.

202

Sediment and Water Organic Carbon. The sediment OC in the upstream river was

203

about 1.91 mg g-1, which was significantly lower than in the cascade reservoirs (P

Carbon Emission from Cascade Reservoirs: Spatial Heterogeneity

Recommend Documents