14C Variation of Dissolved Lignin in Arctic River Systems - ACS Earth

Jun 28, 2017 - Marine Science Institute, University of Texas at Austin, Port Aransas, Texas ... of the Brown Alga Padina pavonica under Ocean Acidific...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Article 14

C variation of dissolved lignin in arctic river systems Xiaojuan Feng, Jorien E. Vonk, Claire Griffin, Nikita Zimov, Daniel B. Montluçon, Lukas Wacker, and Timothy I. Eglinton ACS Earth Space Chem., Just Accepted Manuscript • DOI: 10.1021/ acsearthspacechem.7b00055 • Publication Date (Web): 28 Jun 2017 Downloaded from http://pubs.acs.org on July 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.

ACS Earth and Space Chemistry 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 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

14

2

C variation of dissolved lignin in arctic river systems

3

Xiaojuan Feng,†,‡,§,* Jorien E. Vonk,ǁ Claire Griffin,

4

Lukas Wacker, and Timothy I. Eglinton§

5



6

Academy of Sciences, Beijing 100093, China

7



8

§

9

ǁ

1

10

Nikita Zimov,∇ Daniel B. Montluçon,§



State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese

University of Chinese Academy of Sciences, Beijing 100000, China

Geological Institute, ETH Zürich, Zürich, CH 8092, Switzerland

Department of Earth Sciences, VU University, Amsterdam, 1081 HV, The Netherlands



Department of Ecology, Evolution, and Behavior, University of Minnesota-Twin Cities, St. Paul,

11

Minnesota, MN 55108, USA

12

#

13

∇Pacific

14

Republic of Sakha (Yakutia), RU 690041, Russia

15

⊥, #



Marine Science Institute, University of Texas at Austin, Port Aransas, Texas, TX 78373, USA

Institute for Geography, Far-Eastern Branch of Russian Academy of Science, Cherskiy,

Laboratory of Ion Beam Physics (LIP), ETH Zürich, Zürich, CH 8093, Switzerland

16

1 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17

*Corresponding Author: Xiaojuan Feng, phone: +86 10 6283 6962; email:

18

[email protected].

Page 2 of 40

19

20

KEYWORDS: Compound-specific radiocarbon analysis, arctic rivers, dissolved organic

21

matter, lignin phenols, Keeling plot.

2 ACS Paragon Plus Environment

Page 3 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

22 23

ABSTRACT Assessing permafrost-release signals in arctic rivers is challenging due to mixing of 14

24

complex carbon components of contrasting ages. Compound-specific

25

terrestrially derived molecules may reduce the influence of mixed carbon sources and

26

potentially provide a closer examination on the dynamics of permafrost-derived carbon in

27

arctic rivers. Here we employed a recently modified method to determine radiocarbon

28

contents of lignin phenols, as a classic tracer for terrestrial carbon, isolated from the

29

dissolved organic matter (DOM) of two arctic river systems that showed contrasting seasonal

30

dynamics and age components in DOM. While dissolved lignin had relatively invariant 14C

31

contents in the Mackenzie, it was more concentrated and

32

but relatively diluted and 14C-depleted in the summer flow or permafrost thaw waters in the

33

Kolyma. Remarkably, the covariance between dissolved lignin concentrations and its

34

contents nicely followed the Keeling plot, indicating mixing of a young pool of dissolved

35

lignin with an aged pool of a constant concentration within the river. Using model

36

parameters, we showed that, while the young pool had similarly modern ages in both rivers,

37

Kolyma had a much higher concentration of aged dissolved lignin and/or with older ages.

38

With this approach, our study not only provided the first set of 14C data on dissolved lignin

39

phenols in rivers, but also demonstrated that the age and abundance of the old DOM pool

40

can be assessed by radiocarbon dating of dissolved lignin in arctic rivers related to

41

permafrost release.

14

C analysis of

C-enriched during spring thaw

14

C

3 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

42

Page 4 of 40

TOC Graphic

43

4 ACS Paragon Plus Environment

Page 5 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

44

1. INTRODUCTION

45

Arctic rivers transport ~5.8 Tg of particulate organic carbon (POC) and 34−38 Tg of

46

dissolved organic carbon (DOC) into the ocean annually.1,2 The latter outweighs annual

47

DOC export from Amazon, the largest fluvial system in the world (27 Tg),3,4 and hence

48

represents a major pathway and fate for terrestrial organic carbon (OC) released from arctic

49

watersheds. While mobilization and degradation of rapidly-cycling modern OC imparts no

50

change on atmospheric CO2 concentrations, release of ancient OC that was previously

51

“cryo-locked” in arctic permafrost may constitute a positive feedback to global warming.5,6

52

Hence, it is essential to investigate the age and source of OC carried by arctic rivers in order

53

to better understand global- and regional-scale carbon cycling and to constrain carbon

54

budgets in a changing climate.

55

Previous investigations have mainly focused on the bulk age of POC and DOC in

56

major arctic rivers. It is revealed that predominantly modern DOC is transferred into the

57

Arctic Ocean, especially during freshet when spring thaw causes a sharp rise in river

58

discharge, whereas POC is frequently much older in age.2,7-11 However, as riverine OC is

59

made up of complex mixtures of components, each potentially with varied sources and ages,

60

both riverine POC and DOC contain ancient permafrost-derived and modern OC

61

components with varied abundances.12-15 Alterations in the relative proportion of these

62

components may result in mixed bulk 14C signals that are difficult to directly inform on the

63

dynamics of permafrost-derived OC fluxes. Recently, we have used compound-specific

64

radiocarbon analysis of terrestrial plant biomarkers to show that different terrestrial carbon

65

pools exhibit contrasting mobilization pathways and ages in the sediments delivered by

66

arctic rivers.16,17 In particular, higher plant leaf wax lipids appear to trace aged permafrost

67

carbon mobilized by groundwater flows, whereas lignin-derived phenols in the same riverine

68

sediments are relatively young and more strongly influenced by surface runoff.16,17 Hence, 5 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

14

Page 6 of 40

69

although arctic riverine POC often shows old bulk

C ages, the release of permafrost OC

70

may be masked by the transport of modern POC pools with increasing river runoff.18

71

Similar to POC, riverine DOC is composed of heterogeneous components that likely

72

span a spectrum of ages, and is sensitive to environmental changes.19-22 Recent studies have

73

shown that permafrost-derived DOC is relatively easy to degrade upon release and its old

74

signature is subsequently concealed by modern DOC components that are much more

75

abundant in the river20,23. Such findings explain the predominantly modern age of DOC in

76

major artic rivers,8,9 and highlight the necessity to search for alternatives to trace ancient

77

dissolved components other than using bulk DOC. Compound-specific

78

terrestrially derived molecules may reduce the influence of mixed OC sources (such as

79

modern phytoplankton OC) and provide a closer examination on the changing sources and

80

dynamics of terrestrial DOC components in arctic rivers. However, the 14C composition and

81

variability of specific dissolved organic matter (DOM) constituents remain elusive. It is

82

unclear whether different terrestrial components of DOC have varied mobility and fate with

83

hydrographic changes or permafrost thaw, which may manifest itself in

84

certain DOC molecules. Different from POC, DOC consists of molecules with high

85

solubility and is generally devoid of hydrophobic components such as plant leaf wax lipids.4

86

However, as a well-established tracer of land plant-derived OC, lignin is prevalent in the

87

DOM as well as in sedimentary particles in arctic rivers,24-26 and is hence a good candidate

88

for assessing the release of ancient permafrost DOC to arctic fluvial networks.27 Elucidation

89

of the age of dissolved lignin components may also inform on the relationships between

90

lignin in the dissolved and particulate/sedimentary phases28.

91

14

C analysis of

14

C variations of

Here we employ solid phase extraction (SPE) coupled with a previously established

92

method16,29 to isolate lignin phenols from riverine DOM for

14

93

validated using authentic phenol standard solutions (both modern and

C dating. The method is 14

C-dead) and water 6

ACS Paragon Plus Environment

Page 7 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

14

94

extracts of fresh modern plant leaves. We then measure and compare

95

dissolved lignin phenols isolated from Kolyma and Mackenzie river system samples that

96

span ancient yedoma (ice complex) thaw water to fluvially-influenced coastal waters, and

97

include river mainstem samples collected during freshet and summer flow. To our

98

knowledge, this study represents the first attempt to radiocarbon date dissolved lignin

99

phenols in rivers and presents a potential approach of evaluating the release of ancient

100

C variability of

permafrost-derived DOC in the Arctic.

101 102

2. EXPERIMENTAL SECTION

103

2.1. Study area and sampling

104

Two important arctic river systems (the Kolyma and Mackenzie rivers) are selected

105

for this study, which exhibit contrasting seasonal dynamics and age components in the

106

DOM.8,9 The Kolyma River basin covers ~ 650,000 km2 of northeastern Siberia and

107

represents the largest watershed on Earth that is completely underlain by continuous

108

permafrost. By comparison, the Mackenzie River basin is the largest watershed in the North

109

American Arctic (1,750,000 km2) and has quite different DOC and dissolved lignin

110

dynamics relative to other major arctic rivers.25

111

Initially, a total of 16 DOM samples were collected from various locations on the

112

Kolyma and Mackenzie mainstem and tributaries in an attempt to depict spatial as well as

113

temporal variabilities in dissolved lignin

114

purified lignin phenols, we were only able to measure the 14C content of a limited number of

115

DOM samples for this study. These included four water samples collected from three sites

116

on the Kolyma River in 2011 (Figure 1a; Table 1): one representing a first-order mudstream

117

(permafrost thaw water) draining a Pleistocene yedoma exposure known as Duvanni Yar,

118

two from the Kolyma mainstem just upstream at Cherskiy during freshet (June 6th) and

14

C. However, constrained by the quantity of

7 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

119

summer flow (July 23rd), and one from the river mouth at the end of the freshet (salinity:

120

26‰). For the Mackenzie River, a total of three water samples were collected from two sites

121

in 2011 (Figure 1b): one under-ice sample from the East Channel at Inuvik, Canada just

122

before freshet (May 18th) and two from the mainstem at Tsiigehtchic, Canada during freshet

123

(May 27th) and summer flow (June 11th). In addition, a large-volume (662 L) sample was

124

collected at 9 m below water surface in the coastal ocean on a Canada Fisheries icebreaker

125

(CCGS Louis S. St-Laurent) on July 27th (water depth: 60 m; salinity: 31‰). Alkalinity

126

measurements on board showed that water from the Mackenzie River was pushed offshore

127

and upwelled at this site. Although only eight DOM samples were included in this study,

128

each sample represented different hydrological or geographic regimes and essentially

129

involved several

130

independent 14C data.

14

C measurements for different lignin phenols, resulting in a total of 32

131

All riverine samples were collected in acid-rinsed polyethylene plasticware and kept

132

in the fridge in the dark until being filtered (< 24 h) through pre-cleaned 0.45-µm filters.

133

Aliquots of the filtrates were frozen for subsequent DOC measurement on a

134

high-temperature Shimadzu TOC-V organic carbon analyzer at the Northeast Science Station

135

for the Kolyma samples or at Woods Hole Research Center for the Mackenzie samples.

136

DOC was estimated using chromophoric dissolved organic matter absorption at a

137

wavelength of 254 nm for the under-ice sample (Table 1; Figure S1),30 owing to the lack of

138

measured DOC data. The rest of the filtrate was acidified to pH 2 with hydrochloric acid,

139

and pure methanol was added to a final concentration of 0.5% (v) to improve extraction

140

efficiency.31 The samples were passed through Mega Bond Elut C18 solid phase extraction

141

(SPE) cartridges (10 g) by gravity to isolate DOM.32 All cartridges were dried and kept in a

142

fridge or freezer in the dark until DOM was eluted with 50–80 mL methanol, dried under N2

143

and stored at −20ºC until analysis. The Mackenzie coastal ocean water was directly filtered 8 ACS Paragon Plus Environment

Page 9 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

144

on board through pre-combusted 0.7-µm glass fiber filters (PAL-Gelman AF-1, 292 mm in

145

diameter) and immediately processed using SPE cartridges as described above. Due to the

146

large volume of this sample, a peristaltic pump was used to push water through the SPE

147

cartridges at a flow rate of ~60 mL min−1, and some of the filtrate was kept in the fridge in

148

the dark for up to 48 hours before SPE isolation. Unfortunately, DOC measurement was not

149

conducted for this sample.

150 151 152

2.2. Isolation, quantification and 14C measurement of lignin phenols

Lignin and hydroxy phenolic monomers were released from dried DOM using CuO

oxidation

performed

in

a

microwave-assisted

reaction

system.29

153

alkaline

154

Approximately 1 g of CuO, 0.4 g of ferrous ammonium sulfate, and 30 mL of N2-bubbled

155

NaOH solution (2 M) were loaded into teflon vessels containing DOM samples,

156

vacuum-purged with N2, and then oxidized at 150°C for 1.5 h. An aliquot of the oxidation

157

products was silylated and quantified as trimethylsilyl (TMS) derivatives on an Agilent 6890

158

Series gas chromatograph (GC) coupled to a mass spectrometer (Agilent Technologies,

159

USA).24 Total concentration of eight characteristic lignin phenols (vanillin, acetovanillone,

160

vanillic acid, syringaldehyde, acetosyringone, syringic acid, p-coumaric acid, and ferulic

161

acid) was quantified in the units of µg L−1 or mg g−1 DOC after being normalized to the DOC

162

content. Similarly, three hydroxy phenols (p-hydroxybenzaldehyde, p-hydroxyacetophenone,

163

and p-hydroxybenzoic acid) that may derive from proteins and “tannin-like” compounds or

164

from demethylation of lignin25,33,34 were quantified.

165

For 14C dating, the remaining oxidation products were first purified through two SPE

166

cartridges (Supelco Supelclean ENVI-18 and Supelclean LC-NH2) and then separated on an

167

Agilent 1200 high pressure liquid chromatography (HPLC) system coupled to a diode array

168

detector (DAD) and a fraction collector using a Phenomenex Synergi Polar-RP column (4 9 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 40

169

µm × 4.6 × 250 mm) and a ZORBAX Eclipse XDB-C18 column (5 µm × 4.6 × 150 mm).29

170

Approximately 2−39 µg C of individual phenols were collected after 8 injections on the

171

HPLC. Isolated compounds were eluted with ethyl acetate through a silica gel column (1%

172

deactivated) to further remove potential column bleed. A small aliquot of the isolated

173

compounds was used to check purity (as TMS derivatives) on an Agilent 7890A gas

174

chromatography coupled to flame ionization detection (GC-FID). Compounds with purities

175

> 99% were selected for subsequent

176

provide sufficient mass (> 7 µg C; average of 13.6 µg C; Table S1). To assess procedural

177

blanks, SPE isolation, chemical extraction and HPLC separation were carried out with only

178

solvents and reagents but no sample added. The HPLC effluent was collected at time

179

intervals corresponding to the retention time of targeted compounds and purified by the

180

aforementioned procedures. Purified compounds and procedural blanks were combusted

181

under vacuum at 850°C for 5 h. The resulting CO2 was cryogenically purified and quantified

182

in a calibrated volume. The CO2 of purified phenols was directly measured without

183

graphitization on the miniaturised radiocarbon dating system (MICADAS) at ETH Zürich

184

using a gas feeding system.35 Radiocarbon contents were reported as fraction modern carbon

185

(Fm).

186

14

C measurement. Some phenols were recombined to

To assess the accuracy of dissolved lignin

14

C measurements, we first isolated four

187

phenols from aqueous solutions prepared using authentic standards (dissolved in MilliQ

188

water; Table 2) and four phenols from the water extract of modern maple leaves employing

189

the same method involving SPE isolation, CuO oxidation and HPLC separation. We then

190

measured the radiocarbon content of the isolated phenols (11−37 µg C each) using the same

191

procedures and compared their measured Fm values with those of the original bulk materials

192

(i.e., authentic phenol standards or plant tissues). The offset between the measured

193

(uncorrected for procedural blanks) and original Fm values (∆Fm) averaged at –0.0628 ± 10 ACS Paragon Plus Environment

Page 11 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

194

0.0967 and was larger than those of previous studies12,29,36,37 mainly because the influence of

195

procedural blanks was relatively large for the small-sized samples measured in our study.

196

Consequently, ∆Fm decreased with an increasing

197

approached 20 µg C (Figure 2). The ∆Fm values and patterns suggest that small-sized

198

samples were strongly influenced by a constant background (blank) signal whose effect

199

became relatively small when the sample size was above ~20 µg C.

14

C sample size until the sample size 14

C

200

Procedural blanks yielded 2.5 ± 0.5 µg C for eight HPLC injections and the size was

201

not dependent on the elution or trapping time of compound. We did not directly measure the

202

radiocarbon content of our procedural blanks as sample sizes were too small. Instead, we

203

indirectly estimated their Fm value using a mass balance approach,38 assuming that isolated

204

phenols (from “real” samples and standards) were diluted with a constant amount of blank

205

(2.5 ± 0.5 µg C) with a constant radiocarbon content which caused an offset between the

206

measured and original Fm values of phenol standards. A range of Fm values (from 0.000 to

207

1.000) were tested to correct the measured Fm values of all phenol standards (Table 2; Figure

208

S2). An Fm value of 0.25 and a relatively high uncertainty (± 0.07) were assigned for

209

subsequent corrections, which decreased the Fm offset to an average of 0.0000 ± 0.0367

210

(Table 2). This Fm value is consistent with that measured for procedural blanks collected for

211

particulate lignin phenols in the same laboratory during the same period (mean value of 0.23

212

± 0.07; n = 6; Table S1). Potential sources for the procedural blank include

213

carbon-containing materials from the HPLC column bleed, glassware and/or background

214

CO2 on the vacuum line and MICADAS system. Although every effort was made to

215

minimize extraneous carbon from such sources, it is still challenging (if not impossible) to

216

completely remove all background influences for such small-sized samples. There was no

217

relationship between the (corrected) Fm offset and sample size (Figure 2), suggesting

218

effective removal of the background

14

C influence. After being corrected for procedural 11

ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 40

219

blanks, the Fm values of dissolved phenols isolated by our method are identical to those

220

measured for the authentic standard compounds or bulk leaf tissues within measurement

221

uncertainty. The Fm offset between individual phenols (< ± 0.0578) is comparable to those

222

reported previously.12,29,36 All radiocarbon values are subsequently corrected for procedural

223

blanks with the errors propagated.36 The efficacy of our blank subtraction method was

224

further confirmed through comparing the measured (uncorrected) and blank-corrected Fm

225

values of dissolved phenols in arctic river waters against their 14C sample sizes (see Section

226

3.2 and Figure 3).

227 228 229

2.3. Data analysis

As bulk

14

C contents were not measured for our DOM samples, we used DOC

14

C

230

data measured in the same river and on the most approximate sampling time in the

231

literature8,9,39

232

(www.arcticgreatrivers.org) to compare with dissolved lignin 14C contents. A t test was used

233

to compare the 14C content of different phenols and between different samples. Relationships

234

between the Fm values of dissolved phenols and phenol concentrations or sample sizes were

235

tested by Pearson correlation tests. Differences or correlations are considered to be

236

significant at a level of p < 0.05 or marginally significant at a level of p < 0.10.

or

from

the

Arctic

Great

Rivers

Observatory

database

237 238

3. RESULTS AND DISCUSSION

239

3.1. Variation of dissolved phenol concentrations in the arctic rivers

240

Both Kolyma and Mackenzie exhibited DOC variations typical for waters heavily

241

influenced by snow melting,1,40 with < 6 mg L−1 during summer flow or near river mouth

242

and > 11 mg L−1 during freshet (Table 1). The yedoma mudstream carrying high amounts of

243

particulate matter had the highest DOC content (155 mg L−1). Accordingly, lignin and 12 ACS Paragon Plus Environment

Page 13 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

244

hydroxy phenols showed the highest concentration in the yedoma mudstream, followed by

245

the Kolyma mainstem water at freshet. The Mackenzie coastal ocean water had the lowest

246

concentration of lignin and hydroxy phenols. When normalized to the DOC content, lignin

247

phenols were highest in the Kolyma mainstem and river mouth during freshet (34.3 and 33.3

248

mg g−1 DOC) and decreased sharply during summer flow (4.7 mg g−1 DOC). Hydroxy

249

phenols had the highest DOC-normalized concentrations at the Kolyma river mouth, which

250

may be attributed to inputs of ancient peat leachates near the coast (see discussions below

251

based on

252

concentrations of lignin and hydroxy phenols in the Kolyma River, suggesting that

253

non-lignin components, such as aliphatic constituents23,41 or carboxyl-rich alicyclic

254

molecules (CRAM) abundant in arctic river DOM in the winter,42 constitute a higher fraction

255

of DOC leaching from permafrost deposits.43 By comparison, dissolved lignin and hydroxy

256

phenols had less variable DOC-normalized concentrations (1.4−5.2 and 0.7−1.4 mg g−1 DOC,

257

respectively) in the Mackenzie River, likely buffered by wetlands and lakes that are

258

prevalent in the watershed.25

14

C). Notably, the DOC-rich yedoma mudstream had the lowest DOC-normalized

259 260

3.2. 14C characteristics of dissolved phenols

261

The measured (blank-uncorrected) Fm value of individual lignin and hydroxy phenols

262

varied linearly with their isolated mass in the Kolyma and Mackenzie DOM due to the

263

strong influence of procedural blanks for these small-sized

264

blank correction, individual lignin phenols exhibited relatively uniform

265

DOM samples with the sample size effect largely removed (except the yedoma mudstream;

266

Figure 3). This result re-confirms the efficacy of our blank subtraction method. The Fm

267

variability (∆Fm < 0.1494) of lignin phenols within the same DOM was similar to those in

268

arctic river sediments16,17 or in the particulate organic matter (POM) of Mekong River.12

14

C samples (Figure 3). After 14

C contents in all

13 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 40

269

Furthermore, if lignin phenols isolated from the same DOM are treated as replicates with a

270

similar 14C content, the standard deviation of all Fm values is similar to the propagated errors

271

of the corrected Fm for individual phenols, corroborating the precision of our

272

Nonetheless, it is worth mentioning that while blank correction reduced sample size effect

273

for dissolved phenols in the yedoma mudstream, their corrected Fm values were still

274

marginally correlated with their sample size (p = 0.051; Figure 3a). This indicates that, while

275

our analytical and blank assessment methods seem to be adequate for radiocarbon dating

276

small-sized (7−20 µg C) phenols with relatively high Fm values, larger sample sizes (> 20 µg

277

C) are needed for

278

atmospheric CO2 during sample processing. Hence, for the yedoma sample, we used the 14C

279

data of only one phenol (syringic acid) in the following discussion, which had 20 µg C

280

isolated for radiocarbon dating and yielded the lowest Fm value among all phenols analyzed.

281

Our approach is justified given the overall proximity of

282

mudstream-derived phenols (∆Fm < 0.0845).

283

14

14

C data.

C-dead samples that are very sensitive to modern contamination by

Similar to previous lignin phenol

14

C contents for all the yedoma

14

C studies,12,16,17,29 there was no significant Fm

284

offset between vanillyl and syringyl phenols in the same DOM (t test; p > 0.05). Hence,

285

abundance-weighted average Fm values were calculated for all lignin phenols analyzed of

286

each sample using the following equation with the errors propagated:36

287

Average Fm = (Fm-1 × c1 + Fm-2 × c2 +… + Fm-n × cn ) / (c1 + c2 +… + cn)

(1)

288

where Fm-n and cn represent the 14C content and concentration of individual lignin phenols in

289

a given DOM sample, respectively. Overall, dissolved lignin phenols exhibited

290

predominantly modern

291

waters (except in the yedoma mudstream and Kolyma mainstem during summer flow). This

292

agrees with an overall modern age of bulk DOC carried by arctic rivers.7-9 Also in line with

293

the temporal

14

14

C contents in the Kolyma and Mackenzie riverine and coastal

C shifts in bulk DOC,8,9 dissolved lignin had the highest Fm values during 14 ACS Paragon Plus Environment

Page 15 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

294

freshet in the Kolyma River (Fm of 1.1656 ± 0.0185; Figure 4a), showing progressively

295

lower values during the summer flow (Fm of 0.8753). The most 14C-depleted (oldest) lignin

296

phenols (Fm of 0.0426 ± 0.0124) were derived from the mudstream of an exposed

297

Pleistocene yedoma deposit along the Kolyma riverbank. As ice complex deposits (yedoma)

298

undergo increasing thaw during the summer, more aged DOC components (including

299

dissolved lignin) may be mobilized into the Kolyma mainstem and contribute to the

300

variabilities. By comparison, the Mackenzie River had relatively invariant Fm values for

301

dissolved lignin in the mainstem and coastal waters from freshet to the summer (Figure 4b),

302

consistent with the invariant bulk DOC

303

river.9,25 In the East Channel of the Mackenzie delta, however, DOM sampled (under ice)

304

before the spring freshet carried somewhat older dissolved lignin (Fm of 1.0561 ± 0.0472; p

305

< 0.05), possibly mobilized by groundwater flow from deeper soil layers or from further

306

south in the catchment.

14

14

C

C and dissolved lignin concentration in the

307

Hydroxy phenols showed similar Fm values to lignin phenols in all DOM analyzed

308

except at Kolyma river mouth, where p-hydroxy benzoic acid exhibited a relatively old 14C

309

age. As phytoplankton, a potential source for hydroxy phenols,34 usually carries a modern

310

14

311

for hydroxy phenols. Alternatively, as shown previously,17 hydroxy phenols in East Siberian

312

arctic rivers contain significant inputs from pre-aged peat deposits that are prevalent in the

313

Kolyma basin and along its coast.45 Coastal erosion and leaching processes may thus have

314

transferred aged hydroxy phenols to the estuarine waters.

C signature in the surface ocean,44 marine input is unlikely to be the cause of the old age

315

The young age of dissolved lignin in the Kolyma and Mackenzie mainstem and river

316

mouth stands in contrast with the much older age of lignin phenols in the surface sediments

317

of great arctic rivers, i.e., Fm of 0.6256 and 0.7261 in the Kolyma estuary and Mackenzie

318

shelf, respectively.16,17 The contrasting

14

C ages may be attributed, among other 15

ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 40

319

complicating factors, to (i) the age difference between the sources of dissolved and

320

particulate lignin within terrestrial systems (i.e., litter and soil, respectively) and/or to (ii) the

321

short residence time of dissolved components in contrast to repeated deposition and

322

resuspension cycles for sedimentary particles within the river.46,47 Currently, there is no

323

information on the

324

previous studies on the Amazon and Ganges Rivers, surface POM was considerably younger

325

than riverbed sediments due to a higher contribution of modern biospheric carbon to the

326

former.48,49 Lignin phenols in the POM of the Mekong River also displayed consistently

327

modern 14C contents throughout the year, suggesting rapid export of particulate lignin in the

328

tropical watershed.12 Hence, particulate lignin in the surface water of arctic rivers may

329

exhibit younger ages compared to the sediments. On the other hand, arctic river basins have

330

considerably different hydrological and geochemical properties from the tropics, including

331

longer residence time of terrestrial carbon,50 thicker organic horizons, large reservoirs of

332

frozen, aged organics in the deep soil51 and, last but not least, complex permafrost export

333

dynamics coupled with changing hydrological pathways.17 Hence, particulate lignin is

334

expected to be older in the arctic than the tropic rivers and to display differential sources as

335

well as 14C characteristics. Overall, it remains to be determined whether POM carries lignin

336

of varied 14C contents compared to DOM and which of the aforementioned factors contribute

337

more to the age offset between dissolved and sedimentary lignin in the arctic rivers.

338

Nonetheless, the age contrast between dissolved and sedimentary lignin implies that

339

exchange between the two pools through dissolution, sorption or co-precipitation processes

340

must be relatively limited in these rivers, at least in terms of 14C composition.

341

14

C characteristics of lignin phenols in arctic riverine POM. Based on

Dissolved lignin had a similar

14

C content as bulk DOC in the Kolyma yedoma

342

mudstream and was younger than bulk DOC in both rivers during freshet and in the

343

Mackenzie mainstem summer flow (Figure 5). These observations suggest that lignin traces 16 ACS Paragon Plus Environment

Page 17 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

344

the relatively young component of terrestrial DOC transported in these arctic rivers, which is

345

consistent with the younger age of lignin compared with bulk OC in sediments16,17 or POM12

346

in other riverine systems. Dissolved lignin was however older than bulk DOC in the Kolyma

347

mainstem during summer flow (p < 0.05), suggesting an increased input of younger DOC

348

components relative to lignin in the summer. The nature and source of this young DOC

349

(such as in-situ primary productivity products) remain to be determined. Nonetheless, the Fm

350

offset between dissolved lignin phenols and bulk DOC (up to 0.18) emphasizes that the age

351

variation of specific components and bulk DOC may be different and needs to be assessed

352

separately.

353 354

3.3. Covariance between the 14C and concentration of dissolved lignin phenols

It has been previously demonstrated that the covariance between oceanic DOC

355

14

356

concentrations (c) and its

357

inter-annual scales and at different depths of a given location.52,53 Under this model, DOC

358

consists of an aged, background (bg) component of a constant concentration and a

359

14

360

the measured DOC 14C content is a linear function of c−1:

361

C contents can be explained by Keeling plot models on

C-enriched, excess (xs) component of varied concentrations as binary mixtures.53,54 Hence,

Fm = (Fm-bg − Fm-xs) × cbg × c−1 + Fm-xs

(2)

362

where Fm, Fm-bg, Fm-xs are the 14C composition (Fm values) of the measured, background and

363

excess DOC components, respectively, while cbg and c are the concentration of background

364

DOC and total DOC of a given water, respectively. This model is consistent with the

365

uniform distribution of an aged DOC fraction with depth in the ocean55 and may be used to

366

estimate the isotopic composition of the excess (modern) DOC component.

367 368

Similarly, arctic riverine DOC undergoes obvious temporal

14

C variations due to

supply shifts between modern DOC delivered by surface runoff during freshet and aged 17 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 40

369

components mobilized from deeper soils by base (or summer) flow.8,9 Assuming that

370

dissolved lignin in the arctic rivers follows the binary mixing model, we applied the Keeling

371

plot (by replacing DOC with lignin phenols in Equation 2) to our data in the mainstem and

372

tributary of Kolyma (excluding the yedoma mudstream) and Mackenzie to compare source

373

variations of dissolved lignin in the two rivers. Remarkably, the Fm value of dissolved lignin

374

was nicely correlated with the inverse of dissolved lignin concentration (p = 0.07 for Kolyma

375

and < 0.01 for Mackenzie; Figure 6), even though our dataset included samples from varied

376

locations and sampling time for both rivers. Given that the majority of the analyzed lignin

377

phenols yielded similar amount of carbon for

378

correction method removed the sample size effect on Fm, this relationship is unlikely to

379

result from greater blank contributions at lower lignin concentrations. Instead, it suggests

380

that dissolved lignin

381

that temporal variations of dissolved lignin age and abundance are closely coupled. This

382

observation stands in contrasts with the consistent modern age of lignin phenols in the POM

383

of Mekong River throughout the year,12 suggesting dynamic changes in the sources of

384

dissolved lignin in the arctic rivers. Spring freshet is characterized by concentrated and

385

modern dissolved lignin (the excess component) that is presumably leached from litter and

386

surface soils.25 By contrast, aged lignin (the background component) carried by groundwater

387

or base flow from deeper soils is often diluted with non-lignin DOC.8,9,19,41 This explanation

388

agrees with the low concentration of lignin phenols in the yedoma mudstream DOM per unit

389

of DOC. The age-concentration relationship of dissolved lignin falls in line with the general

390

decreasing age of riverine DOC with increasing DOC concentrations in global rivers,46 also

391

reflecting the input of concentrated, modern DOC during high-flow conditions. However, the

392

global compilation of riverine DOC concentration and 14C content do not follow the Keeling

393

plot due to large variations in the background and excess components for different rivers. In

14

14

C analysis (Table S1) and our blank

C is a relatively conservative property during riverine transport and

18 ACS Paragon Plus Environment

Page 19 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

394

addition, the Fm of dissolved lignin is not correlated with bulk DOC content in Kolyma or

395

Mackenzie (p > 0.05), implying heterogeneous compositions and mixed sources of DOM

396

across seasons.

397

The Fm values for the excess component, i.e., the intercepts on the y axis of the

398

Keeling plot, were 1.1928 for Kolyma and 1.2258 for Mackenzie. This suggests that the

399

excess (young) pool of dissolved lignin had relatively similar ages in the two rivers and was

400

influenced by the bomb-derived carbon. More importantly, the Keeling plot showed discrete

401

slopes for the Kolyma and Mackenzie mainstem samples. As the slope is a function of the

402

concentration of background dissolved lignin (cbg) as well as the Fm values of background

403

and excess lignin (Fm-bg and Fm-xs in Equation 2, respectively), cbg can be estimated using the

404

following equation given a range of Fm-bg values:

405

cbg = Slope / (Fm-bg − Fm-xs)

(3)

406

As shown in Figure 7, cbg is much lower in the Mackenzie than the Kolyma River for any

407

given Fm-bg values. For instance, when the background lignin is assumed to be 14C-dead (i.e.,

408

Fm-bg = 0.0000), its concentration is 5.5 and 0.5 µg C L−1 in the Kolyma and Mackenzie

409

Rivers, respectively. Hence, the Kolyma River, with a much steeper slope for the Keeling

410

plot, may have a higher concentration of aged dissolved lignin and/or with older ages. This

411

explanation is consistent with the widespread occurrence of yedoma in the Kolyma

412

watershed,8,9,20,45 which supplies ancient DOC and lignin into the river.

413

Our constructed Keeling plot is more of a statistical relationship for the Mackenzie

414

River (p < 0.01) than for the Kolyma (p = 0.07), although we are limited by the number of

415

data points included in our study. Alternatively, while the binary mixing model describes the

416

behavior of dissolved lignin phenols in the Mackenzie River, it may be inadequate for

417

modeling DOC components in the Kolyma, which is more complicated with seasonal inputs

418

of yedoma thaw water as a third DOC pool. It also remains to be determined whether the 19 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 40

419

age-concentration correlation varies inter annually and spatially for dissolved lignin within a

420

river. However, as shown by Beaupré and Druffel,52 the isotopic composition of the

421

background component can be better constrained based on the slopes of multiple Keeling

422

plots stemmed from different sampling sites or time series (assuming a common

423

background). Given that the age-concentration relationship of bulk DOC is complicated by

424

mixed sources of varied age components, monitoring dissolved lignin

425

and/or seasonally may provide a key clue to the age of the background DOC component

426

using multiple Keeling plots. Such information is critical for assessing permafrost carbon

427

response to climatic and human disturbances but is difficult to measure directly.

14

C across the Arctic

428 429

4. Conclusions and scopes for future work

430

Overall, our study provides the first set of 14C data on dissolved lignin and hydroxy

431

phenols in arctic rivers and presents analytical advancements as well as challenges

432

associated with such measurements. Through comparing the measured

433

individual phenols isolated from the same DOM sample but with varied masses, we show

434

that our analytical and blank correction method is adequate for radiocarbon dating

435

small-sized (7−20 µg C) dissolved phenols with relatively high Fm values, while larger

436

sample sizes (> 20 µg C) are ideal for 14C-depleted DOM samples that are more sensitive to

437

modern contamination. Our study hence pushes the lower sample size limit for

438

compound-specific 14C analysis. Moreover, by measuring the 14C content of dissolved lignin

439

collected from various hydrological regimes and locations, we demonstrate distinct age

440

characteristics of dissolved lignin in two arctic rivers with different watershed properties.

441

While dissolved lignin had relatively invariant

442

stronger seasonal variation in the Kolyma River. Applying Keeling plot to the dissolved

443

lignin data, we find that the Kolyma River has an older background component of dissolved

14

C content of

14

C contents in the Mackenzie, it had a

20 ACS Paragon Plus Environment

Page 21 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

444

lignin than the Mackenzie, which is likely associated with the widespread occurrence of

445

yedoma in its watershed. Such information is not discernable using bulk DOC

446

measurements although dissolved lignin phenols exhibited an overall similar

447

bulk DOC in our investigated rivers. More importantly, as this study represents the first

448

attempt to radiocarbon date dissolved lignin phenols in arctic rivers, it remains to be

449

investigated whether dissolved lignin displays similar residence time to bulk DOC in other

450

aquatic systems (such as temperate and tropical rivers, lakes and ocean). For instance, using

451

lignin phenols as a tracer for terrestrial organic matter, it will be very interesting to examine

452

the residence time of land-derived DOC components in surface ocean and lakes, which have

453

high modern DOC inputs from phytoplankton. Hence, the meticulous molecular-level

454

analysis we present should have a wider application in marine and lacustrine carbon cycles

455

as well.

14

C

14

C signal to

456

Admittedly, our study is constrained by the limited number of DOM samples due to

457

challenges associated with collecting ample DOM for radiocarbon dating dissolved lignin

458

phenols. Therefore, future endeavor is needed for more extensive seasonal sampling to

459

ascertain the Keeling plot patterns as well as contrasts between the two rivers. Nonetheless,

460

this study puts forward an analytical and modelling approach to utilize the

461

preserved in dissolved lignin to evaluate the age of old DOC components associated with

462

permafrost release. Such information is not directly available using bulk DOC

463

measurement as complicated by mixed OC sources. Given the large and dynamic flux of

464

terrestrially-derived DOC in arctic rivers,1,2 dissolved lignin 14C seems worthy of monitoring

465

in the context of a warming Arctic. Meanwhile, as mentioned previously, lignin phenols are

466

relatively low in abundance in the permafrost-derived DOM. Therefore, in the attempt to

467

search for “permafrost-release sentinels”, aliphatic constituents that constitute a higher

468

fraction of DOC leaching from permafrost deposits43 may provide complementary

14

C signal

14

C

21 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 40

469

information in their radiocarbon characteristics. Although the specific molecular structures

470

of such aliphatic constituents remain elusive currently,41 neutral sugars that are abundant in

471

riverine DOM42,56 seem to be a promising candidate. We therefore urge future studies to

472

target these molecules for 14C dating to build a more comprehensive understanding of arctic

473

DOC dynamics in relation with permafrost release.

474 475

ACKNOWLEDGMENTS

All molecular

476

14

C data are available in Supporting Information Table S1. Members of

477

LIP (ETH) are acknowledged for support with the radiocarbon measurements. X.F.

478

acknowledges support from the National Natural Science Foundation of China (41422304,

479

31370491). T.I.E. acknowledges support from the Swiss National Science foundation (SNF)

480

(#200021_140850), and grants OCE-9907129, OCE-0137005, and OCE-0526268 from the

481

US National Science Foundation (NSF), and ETH Zürich. J.E.V. thanks support from NWO

482

Rubicon (#825.10.022) and Veni (#863.12.004). X.F. thanks ETH Zürich for postdoctoral

483

support. Support for the Arctic Great Rivers Observatory comes from NSF #ARC-0732821

484

and #ARC-1107774. The “Young Thousand Talent” recruiting plan of China is

485

acknowledged for start-up support to X.F.

486

487

Supporting Information Available:

488



489

absorbance at 254 nm (Figure S1), changes in the average Fm offset between isolated phenols

490

and the original bulk materials in response to Fm change of procedural blanks (Figure S2),

491

original Fm values and concentrations of individual phenols isolated from DOM of the

Comparison of DOC concentration to chromophoric dissolved organic matter

22 ACS Paragon Plus Environment

Page 23 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

492

Kolyma and Mackenzie river systems and the Fm values of procedural blanks collected for

493

particulate lignin phenols (Table S1).

494

495 496

Notes

The authors declare no competing financial interest.

497 498

REFERENCES

499

(1) Holmes, R. M.; McClelland, J. W.; Peterson, B. J.; Tank, S. E.; Bulygina, E.; Eglinton,

500

T. I.; Gordeev, V. V.; Gurtovaya, T. Y.; Raymond, P. A.; Repeta, D. J.; Staples, R.;

501

Striegl, R. G.; Zhulidov, A. V.; Zimov, S. A. Seasonal and annual fluxes of nutrients and

502

organic matter from large rivers to the Arctic Ocean and surrounding seas. Estuar. Coast.

503

2012, 35, 369-382.

504

(2) McClelland, J. W.; Holmes, R. M.; Peterson, B. J.; Raymond, P. A.; Striegl, R. G.;

505

Zhulidov, A. V.; Zimov, S. A.; Zimov, N.; Tank, S. E.; Spencer, R. G. M.; Staples, R.;

506

Gurtovaya, T. Y.; Griffin, C. G. Particulate organic carbon and nitrogen export from

507

major Arctic rivers. Global Biogeochem. Cy. 2016, 30, 629-643.

508 509

(3) Moreira-Turcq, P.; Seyler, P.; Guyot, J. L.; Etcheber, H. Exportation of organic carbon from the Amazon River and its main tributaries. Hydrol. Process. 2003, 17, 1329-1344.

510

(4) Raymond, P. A.; Spencer, R. G. M. Riverine DOM. In: Biogeochemistry of Marine

511

Dissolved Organic Matter; Hansell, D. A., Carlson, C. A., Eds.; Academic Press:

512

Burlington, 2015; pp 509-533.

23 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 40

513

(5) DeConto, R. M.; Galeotti, S.; Pagani, M.; Tracy, D.; Schaefer, K.; Zhang, T.; Pollard, D.;

514

Beerling, D. J. Past extreme warming events linked to massive carbon release from

515

thawing permafrost. Nature 2012, 87, 484-492.

516

(6) Schuur, E. A. G.; Bockheim, J.; Canadell, J. G.; Euskirchen, E.; Field, C. B.;

517

Goryachkin, S. V.; Hagemann, S.; Kuhry, P.; Lafleur, P. M.; Lee, H.; Mazhitova, G.;

518

Nelson, F. E.; Rinke, A.; Romanovsky, V. E.; Shiklomanov, N.; Tarnocai, C.; Venevsky,

519

S.; Vogel, J. G.; Zimov, S. A. Vulnerability of permafrost carbon to climate change:

520

Implications for the global carbon cycle. Bioscience 2008, 58 (8), 701-714.

521

(7) Benner, R.; Benitez-Nelson, B.; Kaiser, K.; Amon, R. M. W. Export of young

522

terrigenous dissolved organic carbon from rivers to the Arctic Ocean. Geophys. Res. Lett.

523

2004, 31, L05305, doi:10.1029/2003GL019251.

524

(8) Neff, J. C.; Finlay, J. C.; Zimov, S. A.; Davydov, S. P.; Carrasco, J. J.; Schuur, E. A. G.;

525

Davydova, A. I. Seasonal changes in the age and structure of dissolved organic carbon

526

in

527

doi:10.1029/2006GL028222.

Siberian

rivers

and

streams.

Geophys.

Res.

Lett.

2006,

33,

L23401,

528

(9) Raymond, P. A.; McClelland, J. W.; Holmes, R. M.; Zhulidov, A. V.; Mull, K.; Peterson,

529

B. J.; Striegl, R. G.; Aiken, G. R.; Gurtovaya, T. Y. Flux and age of dissolved organic

530

carbon exported to the Arctic Ocean: A carbon isotopic study of the five largest arctic

531

rivers. Global Biogeochem. Cy. 2007, 21, GB4011, doi:10.1029/2007GB002934.

532

(10) Guo, L.; Macdonald, R. W. Source and transport of terrigenous organic matter in the

533

upper Yukon River: Evidence from isotope (δ13C, ∆14C, and δ15N) composition of

534

dissolved, colloidal, and particulate phases. Global Biogeochem. Cy. 2006, 20, GB2011,

535

doi:10.1029/2005GB002593.

24 ACS Paragon Plus Environment

Page 25 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

536

(11) Guo, L.; Ping, C.-L.; Macdonald, R. W. Mobilization pathways of organic carbon from

537

permafrost to Arctic rivers in a changing climate. Geophys. Res. Lett. 2007, 34, L13603,

538

doi:10.1029/2007GL030689.

539

(12) Martin, E. E.; Ingalls, A. E.; Richey, J. E.; Keil, R. G.; Santos, G. M.; Carlson, L. T.;

540

Alin, S. R.; Druffel, E. R. M. Age of riverine carbon suggests rapid export of terrestrial

541

primary production in tropics. Geophys. Res. Lett. 2013, 40, 5687-5691.

542

(13) Tao, S.; Eglinton, T. I.; Montluçon, D. B.; McIntyre, C.; Zhao, M. Pre-aged soil organic

543

carbon as a major component of the Yellow River suspended load: Regional

544

significance and global relevance. Earth Planet. Sci. Lett. 2015, 414, 77-86.

545

(14) Coppola, A. I.; Ziolkowski, L. A.; Masiello, C. A.; Druffel, E. R. M. Aged black carbon

546

in marine sediments and sinking particles. Geophys. Res. Lett. 2014, 41, 2427-2433.

547

(15) Repeta, D. J.; Aluwihare, L. I. Radiocarbon analysis of neutral sugars in

548

high-molecular-weight dissolved organic carbon: Implications for organic carbon

549

cycling Limnol. Oceanogr. 2006, 51 (2), 1045-1053.

550

(16) Feng, X.; Gustafsson, Ö.; Holmes, R. M.; Vonk, J. E.; van Dongen, B. E.; Semiletov, I.

551

P.; Dudarev, O. V.; Yunker, M. B.; Macdonald, R. W.; Wacker, L.; Montluçon, D. B.;

552

Eglinton, T. I. Multimolecular tracers of terrestrial carbon transfer across the pan-Arctic:

553

14

554

Global Biogeochem. Cy. 2015, 29 (11), 1855-1873.

C characteristics of sedimentary carbon components and their environmental controls.

555

(17) Feng, X.; Vonk, J. E.; van Dongen, B. E.; Gustafsson, Ö.; Semiletov, I. P.; Dudarev, O.

556

V.; Wang, Z.; Montluçon, D. B.; Wacker, L.; Eglinton, T. I. Differential mobilization of

557

terrestrial carbon pools in Eurasian Arctic river basins. Proc. Natl. Acad. Sci. 2013, 110

558

(35), 14168-14173.

25 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 40

559

(18) Peterson, B. J.; Holmes, R. M.; McClelland, J. W.; Vörösmarty, C. J.; Lammers, R. B.;

560

Shiklomanov, A. I.; Shiklomanov, I. A.; Rahmstorf, S. Increasing river discharge to the

561

Arctic Ocean. Science 2002, 298, 2171-2173.

562

(19) Mann, P. J.; Davydova, A.; Zimov, N.; Spencer, R. G. M.; Davydov, S.; Bulygina, E.;

563

Zimov, S.; Holmes, R. M. Controls on the composition and lability of dissolved organic

564

matter in Siberia's Kolyma River basin. J. Geophys. Res. 2012, 117, G01028,

565

doi:10.1029/2011JG001798.

566

(20) Mann, P. J.; Eglinton, T. I.; McIntyre, C. P.; Zimov, N.; Davydova, A.; Vonk, J. E.;

567

Holmes, R. M.; Spencer, R. G. M. Utilization of ancient permafrost carbon in

568

headwaters

569

doi:10.1038/ncomms8856.

of

Arctic

fluvial

networks.

Nat.

Commun.

2015,

6,

7856,

570

(21) Vonk, J. E.; Mann, P. J.; Dowdy, K. L.; Davydova, A.; Davydov, S. P.; Zimov, N.;

571

Spencer, R. G. M.; Bulygina, E. B.; Eglinton, T. I.; Holmes, R. M. Dissolved organic

572

carbon loss from Yedoma permafrost amplified by ice wedge thaw. Environ. Res. Lett.

573

2013, 8, 035023, doi:10.1088/1748-9326/8/3/035023.

574

(22) Butman, D.; Raymond, P. A.; Butler, K.; Aiken, G. Relationships between D14C and the

575

molecular quality of dissolved organic carbon in rivers draining to the coast from the

576

conterminous

577

doi:10.1029/2012GB004361.

United

States.

Global

Biogeochem.

Cy.

2012,

26,

GB4014,

578

(23) Spencer, R. G. M.; Mann, P. J.; Dittmar, T.; Eglinton, T. I.; McIntyre, C.; Holmes, R.

579

M.; Zimov, N.; Stubbins, A. Detecting the signature of permafrost thaw in Arctic rivers.

580

Geophys. Res. Lett. 2015, 42, doi:10.1002/2015GL063498.

581

(24) Feng, X.; Gustafsson, Ö.; Holmes, R. M.; Vonk, J. E.; van Dongen, B. E.; Semiletov, I.

582

P.; Dudarev, O. V.; Yunker, M. B.; Macdonald, R. W.; Montluçon, D. B.; Eglinton, T. I.

583

Multi-molecular tracers of terrestrial carbon transfer across the pan-Arctic: Comparison 26 ACS Paragon Plus Environment

Page 27 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

584

of hydrolysable components with plant wax lipids and lignin phenols. Biogeosciences

585

2015, 12, 4841-4860.

586

(25) Amon, R. M. W.; Rinehart, A. J.; Duan, S.; Louchouarn, P.; Raymond, P.; Holmes, R.

587

M.; McClelland, J. W.; Peterson, B. J.; Guggenberger, G.; Prokushkin, A.; Stedmon, C.;

588

Walker, S. A. Dissolved organic matter sources in large Arctic rivers. Geochim.

589

Cosmochim. Ac. 2012, 94, 217-237.

590

(26) Jex, C. N.; Pate, G. H.; Blyth, A. J.; Spencer, R. G. M.; Hernes, P. J.; Khan, S. J.; Baker,

591

A. Lignin biogeochemistry: from modern processes to Quaternary archives. Quat. Sci.

592

Rev. 2014, 87, 46-59

593

(27) Vonk, J. E.; Sánchez-García, L.; van Dongen, B. E.; Alling, V.; Kosmach, D.; Charkin,

594

A.; Semiletov, I. P.; Dudarev, O. V.; Shakhova, N.; Roos, P.; Eglinton, T. I.; Andersson,

595

A.; Gustafsson, Ö. Activation of old carbon by erosion of coastal and subsea permafrost

596

in Arctic Siberia. Nature 2012, 489, 137-140.

597

(28) Feng, X.; Feakins, S. J.; Liu, Z.; Ponton, C.; Wang, R. Z.; Karkabi, E.; Galy, V.;

598

Berelson, W. M.; Nottingham, A. T.; Meir, P.; West, A. J. Source to sink: Evolution of

599

lignin composition in the Madre de Dios River system with connection to the Amazon

600

basin and offshore. J. Geophys. Res.-Biogeos. 2016, 121, 1316-1338.

601

(29) Feng, X.; Benitez-Nelson, B. C.; Montluçon, D. B.; Prahl, F. G.; McNichol, A. P.; Xu,

602

L.; Repeta, D. J.; Eglinton, T. I. 14C and 13C characteristics of higher plant biomarkers in

603

Washington margin surface sediments. Geochim. Cosmochim. Acta 2013, 105, 14-30.

604

(30) Green, S.; Blough, N. Optical absorption and fluorescence properties of chromophoric

605

dissolved organic matter in natural waters. Limnol. Oceanogr. 1994, 39 (8), 1903-1916.

606

(31) Spencer, R. G. M.; Aiken, G. R.; Dyda, R. Y.; Butler, K. D.; Bergamaschi, B. A.;

607

Hernes, P. J. Comparison of XAD with other dissolved lignin isolation techniques and a

27 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 40

608

compilation of analytical improvements for the analysis of lignin in aquatic settings.

609

Org. Geochem. 2010, 41, 445-453.

610

(32) Louchouarn, P.; Opsahl, S.; Benner, R. Isolation and quantification of dissolved lignin

611

from natural waters using solid-phase extraction and GC/MS. Anal. Chem. 2000, 72 (13),

612

2780-2787.

613 614

(33) Hedges, J. I.; Mann, D. The characterization of plant tissues by their lignin oxidation products. Geochim. Cosmochim. Acta 1979, 43 (11), 1803-1807.

615

(34) Goñi, M. A.; Hedges, J. I. Sources and reactivities of marine-derived organic matter in

616

coastal sediments as determined by alkaline CuO oxidation. Geochim. Cosmochim. Acta

617

1995, 59, 2965-2981.

618

(35) Wacker, L.; Fahrni, S. M.; Hajdas, I.; Molnar, M.; Synal, H.-A.; Szidat, S.; Zhang, Y. L.

619

A versatile gas interface for routine radiocarbon analysis with a gas ion source. Nucl.

620

Instrum. Methods Phys. Res. B 2013, 294, 315-319.

621

(36) Hou, J.; Huang, Y.; Brodsky, C.; Alexandre, M. R.; McNichol, A. P.; King, J. W.; Hu,

622

F.; Shen, J. Radiocarbon dating of individual lignin phenols: A new approach for

623

establishing chronology of late Quaternary lake sediments. Anal. Chem. 2010, 82 (17),

624

7119-7126.

625

(37) Ingalls, A. E.; Ellis, E. E.; Santos, G. M.; McDuffee, K. E.; Truxal, L.; Keil, R. G.;

626

Druffel, E. R. M. HPLC purification of higher plant-dervied lignin phenols for

627

compound specific radiocarbon analysis. Anal. Chem. 2010, 82 (21), 8931-8938.

628

(38) Ziolkowski, L. A.; Druffel, E. R. M. Quantification of extraneous carbon during

629

compound specific radiocarbon analysis of black carbon. Anal. Chem. 2009, 81 (24),

630

10156-10161.

631

(39) Vonk, J. E.; Mann, P. J.; Davydov, S.; Davydova, A.; Spencer, R. G. M.; Schade, J.;

632

Sobczak, W. V.; Zimov, N.; Zimov, S.; Bulygina, E.; Eglinton, T. I.; Holmes, R. M. 28 ACS Paragon Plus Environment

Page 29 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

633

High biolability of ancient permafrost carbon upon thaw. Geophys. Res. Lett. 2013, 40

634

(1-5), doi:10.1002/grl.50348.

635

(40) Finlay, J.; Neff, J.; Zimov, S.; Davydova, A.; Davydov, S. Snowmelt dominance of

636

dissolved organic carbon in high-latitude watersheds: implications for characterization

637

and

638

doi:10.1029/2006GL025754.

639 640

flux

of

river

DOC.

Geophys.

Res.

Lett.

2006,

33,

L10401,

(41) Ward, C. P.; Cory, R. M. Chemical composition of dissolved organic matter draining permafrost soils. Geochim. Cosmochim. Ac. 2015, 167, 63-79.

641

(42) Cao, X.; Aiken, G. R.; Spencer, R. G. M.; Butler, K.; Mao, J.; Schmidt-Rohr, K. Novel

642

insights from NMR spectroscopy into seasonal changes in the composition of dissolved

643

organic matter exported to the Bering Sea by the Yukon River. Geochim. Cosmochim.

644

Ac. 2016, 181, 72-88.

645

(43) Cory, R. M.; Crump, B. C.; Dobkowski, J. A.; Kling, G. W. Surface exposure to

646

sunlight stimulates CO2 release from permafrost soil carbon in the Arctic. Proc. Natl.

647

Acad. Sci. 2013, 10 (9), 3429-3434.

648

(44) Pearson, A.; McNichol, A. P.; Benitez-Nelson, B. C.; Hayes, J. M.; Eglinton, T. I.

649

Origins of lipid biomarkers in Santa Monica Basin surface sediment: A case study using

650

compound-specific ∆14C analysis. Geochim. Cosmochim. Acta 2001, 65 (18),

651

3123-3137..

652

(45) Kremenetski, K. V.; Velichko, A. A.; Borisova, O. K.; MacDonald, G. M.; Smith, L. C.;

653

Frey, K. E.; Orlova, L. A. Peatlands of the western Siberian lowlands: Current

654

knowledge on zonation, carbon content and Late Quaternary history. Quat. Sci. Rev.

655

2003, 22, 703-723.

29 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 40

656

(46) Marwick, T. R.; Tamooh, F.; Teodoru, C. R.; Borges, A. V.; Darchambeau, F.; Bouillon,

657

S. The age of river-transported carbon: A global perspective. Global Biogeochem. Cy.

658

2015, 29, 122-137.

659 660

(47) Newbold, J. D.; Mulholland, P. J.; Elwood, J. W.; O'Neill, R. V. Organic carbon spiralling in stream ecosystems. Oikos 1982, 38, 266-272.

661

(48) Bouchez, J.; Galy, V.; Hilton, R. G.; Gaillardet, J.; Moreira-Turcq, P.; Perez, M. A.;

662

France-Lanord, C.; Maurice, L. Source, transport and fluxes of Amazon River

663

particulate organic carbon: Insights from river sediment depth-profiles. Geochim.

664

Cosmochim. Acta 2014, 133, 280-298.

665

(49) Galy, V.; France-Lanord, C.; Beyssac, O.; Faure, P.; Kudrass, H.; Palhol, F. Efficient

666

organic carbon burial in the Bengal fan sustained by the Himalayan erosional system.

667

Nature 2007, 450, 407-410.

668

(50) Carvalhais, N.; Forkel, M.; Khomik, M.; Bellarby, J.; Jung, M.; Migliavacca, M.; Mu,

669

M.; Saatchi, S.; Santoro, M.; Thurner, M.; Weber, U.; Ahrens, B.; Beer, C.; Cescatti, A.;

670

Randerson, J. T.; Reichstein, M. Global covariation of carbon turnover times with

671

climate in terrestrial ecosystems. Nature 2014, 514, 213-217.

672

(51) Tarnocai, C.; Canadell, J. G.; Schuur, E. A. G.; Kuhry, P.; Mazhitova, G.; Zimov, S.

673

Soil organic carbon pools in the northern circumpolar permafrost region. Global

674

Biogeochem. Cy. 2009, 23, GB2023, doi:2010.1029/2008GB003327.

675

(52) Beaupré, S. R.; Druffel, E. R. M. Constraining the propagation of bomb-radiocarbon

676

through the dissolved organic carbon (DOC) pool in the northeast Pacific Ocean.

677

Deep-Sea Res. I 2009, 56, 1717-1726.

678

(53) Mortazavi, B.; Chanton, J. P. Use of Keeling plots to determine sources of dissolved

679

organic carbon in nearshore and open ocean systems. Limnol. Oceanogr. 2004, 49 (1),

680

102-108. 30 ACS Paragon Plus Environment

Page 31 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

681 682 683 684

ACS Earth and Space Chemistry

(54) Keeling, C. D. The concentration and isotopic abundances of atmospheric carbon dioxide in rural areas. Geochim. Cosmochim. Acta 1958, 13, 322-334. (55) Williams, P. M.; Druffel, E. R. M. Radiocarbon in dissolved organic matter in the central North Pacific Ocean. Nature 1987, 330 (6145), 246-248.

685

(56) Shen, Y.; Fichot, C. G.; Benner, R. Floodplain influence on dissolved organic matter

686

composition and export from the Mississippi-Atchafalaya River system to the Gulf of

687

Mexico. Limnol. Oceanogr. 2012, 57 (4), 1149-1160.

31 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 32 of 40

Table 1: Sample information, dissolved organic carbon (DOC), phenol concentration and composition of the dissolved organic matter in Kolyma and Mackenzie Rivers. River

Kolyma

Mackenzie

Sample name

Yedoma mudstream

Mainstem freshet

River mouth

Mainstem summer flow

East Channel

Mainstem freshet

Mainstem summer flow

Coastal ocean

Latitude (ºN)

68.631

68.741

69.660

68.742

68.357

67.468

67.468

69.997

Longitude (ºE)

159.150

161.290

162.451

161.282

−133.742

−133.685

−133.685

−139.998

Sampling time

2011/07/13

2011/06/06

2011/07/08

2011/07/23

2011/05/18

2011/05/27

2011/06/11

2011/07/27

1.4

1.4

7.0

10.0

28.6

Volume (L) −1

7.8

13.8

662

c

DOC (mg L )

155

13.6

1.8

4.5

2.5

11.8

5.4

n.a.

Lignin phenols (µg L−1)a

605

466

61.0

21.1

3.6

61.4

15.8

0.1

Lignin phenols (mg g−1 DOC)b

3.9

34.3

33.3

4.7

1.4

5.2

2.9

Hydroxy phenols (µg L−1)a

119

97.7

23.4

9.1

1.7

16.8

6.6

Hydroxy phenols (mg g−1 DOC)b

0.8

7.2

12.8

2.0

0.7

1.4

1.2

n.a. 0.1 n.a.

n.a.: not available. a

Concentrations of eight lignin-derived phenols or three hydroxy phenols.

b

DOC-normalized concentrations.

c

Estimated from chromophoric dissolved organic matter absorption at 254 nm (a254). See Figure S1 for details.

32 ACS Paragon Plus Environment

Page 33 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

ACS Earth and Space Chemistry

Table 2:

14

C content of lignin phenols isolated from dissolved standards or leaf water extracts relative to that of the original bulk

materials. Isolated phenol (µg C)

Measured values on isolated phenols

Original

Source

Phenol

AMS #

Standard

Syringic acid

45128.1.1

24

AMS-corrected only Fm error 0.0281 0.0029

from Sigma

Vanillin

45129.1.1

20

1.0324

0.0112

1.1456

0.0307

1.1343

Standard

Vanillic acid

45131.1.1

23

0.0579

0.0033

0.0348

0.0106

0.0053

from Acros

Acetovanillone

47488.1.1

37

0.0390

0.0017

0.0237

0.0063

0.0241

p-Hydroxyacetophenone

45127.1.1

17

0.9031

0.0100

1.0122

0.0304

1.0573

Acetovanillone

45132.1.1

11

0.8271

0.0129

1.0020

0.0531

1.0573

p-Coumaric acid

46413.1.2

31

0.9946

0.0090

1.0595

0.0183

1.0573

Ferulic acid

46414.1.1

21

1.0098

0.0113

1.1151

0.0289

1.0573

Water extract of maple leaves

Procedural blank-correcteda Fm error 0.0019 0.0107

a

Procedural blank contains 2.5 ± 0.5 µg C with Fm of 0.25 ± 0.07.

b

Original values were measured on the same authentic standards or bulk leaf material (blank-corrected).

Fmb 0.0017

33 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 34 of 40

Figures

Figure 1: Map of sampling locations (yellow stars) in the Kolyma (a) and Mackenzie (b) watersheds and coasts. The satellite images are obtained from Google earth (Pro 7.1.8.3036; http://www.earth.google.com).

34 ACS Paragon Plus Environment

Page 35 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

Figure 2: Changes in the Fm offset (∆Fm) between isolated phenols and the original bulk material of phenol standards in response to the sample size of the isolated phenols. Note that ∆Fm (uncorrected for procedural blanks) decreased with sample size (p < 0.05) until the mass of isolated phenols reached > 20 µg C. Blank correction significantly reduced ∆Fm.

35 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 40

Figure 3: Changes in the uncorrected (blue) and blank-corrected (orange) Fm values of isolated phenols against the 14C sample size of individual phenols in the yedoma mudstream (a), Kolyma mainstem (b), Kolyma river mouth (c), Mackenzie mainstem (d) and Mackenzie coastal ocean (e). Note that the uncorrected Fm values were correlated with sample sizes while this effect was largely removed after blank correction.

36 ACS Paragon Plus Environment

Page 37 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

ACS Earth and Space Chemistry

Figure 4: The Fm values of individual dissolved lignin and hydroxy phenols (colored diamonds and squares) compared with the abundance-weighted average Fm of dissolved lignin phenols (black lines with grey shades representing propagated standard errors) in the Kolyma (a) and Mackenzie (b) Rivers collected in 2011. All values are corrected for procedural blanks with the standard errors of analytical measurement propagated. Some phenols were pooled for 14C measurement and represented by the most abundant monomer here (refer to Table S1 for detailed Fm values). *Note that only the Fm value of syringic acid (> 20 µg C) was used to represent the yedoma sample.

37 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 40

Figure 5: Comparison of Fm values between dissolved lignin phenols (abundance-weighted average) and bulk dissolved organic carbon (DOC) in the Kolyma and Mackenzie Rivers. *DOC Fm values are obtained from Vonk et al.39 (Duvannyi Yar yedoma mudstream; sampled on 2010/7/22), Raymond et al.9 (Mackenzie freshet at Tsiigehtchic on 2004/6/17), Neff et al.8 (Kolyma summer flow at Cherskiy on 2003/7/7) and Arctic Great Rivers Observatory database (www.arcticgreatrivers.org; Kolyma freshet at Cherskiy on 2011/6/10 and Mackenzie summer flow at Tsiigehtchic on 2011/06/14).

38 ACS Paragon Plus Environment

Page 39 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

Figure 6: Keeling plot model for the abundance-weighted average Fm values of dissolved lignin phenols against the inverse of their concentrations (µg L−1) in the Kolyma and Mackenzie mainstems. Note that the yedoma mudstream (a first-order stream not on the mainstem) is excluded for the Kolyma.

39 ACS Paragon Plus Environment

ACS Earth and Space Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 40

Figure 7: Changes in the concentration of background dissolved lignin phenols with their changing Fm values based on the Keeling plot parameters (Equation 3).

40 ACS Paragon Plus Environment