University of Manchester - Fluvial organic carbon … · Web viewaAn average flow weighted % modern...

Fluvial OC variability within a peatland catchment.

Fluvial organic carbon composition and concentration variability

within a peatland catchment – implications for carbon cycling and

water treatment.

A.G. Stimsona#, T.E.H. Allotta, S Boultb, M. G. Evansa

aUpland Environments Research Unit, School of Environment, Education and

Development, The University of Manchester, Oxford Road, M13 9PL, United Kingdom.

bSchool of Earth and Environmental Science, University of Manchester, Oxford Road,

Manchester, M13 9PL, UK.

#Corresponding author

Email: [email protected]

Tel: 0044 (0) 161 275 5638

1

1

2

3

4

5

6

7

8

9

10

11

13

14

15

16

17

mailto:[email protected]


Abstract

Fluvial organic carbon (OC) transformations are an important component of carbon

cycling and greenhouse gas production in inland waters resulting in considerable recent

interest in the fate of fluvial OC exported from carbon rich soils such as peatlands.

Additionally, peatland catchments are important drinking water collection areas, where

high OC concentrations in runoff have water treatment implications. This analysis

presents the results from a year-round intensive study within a water treatment catchment

draining an area of peatland, considering carbon transformations along a continuum from

headwater river, through a storage reservoir and pipe, to a water treatment works. The

study uses a unique combination of methods (colourmetric, ultrafiltration and 14C

radiocarbon dating) to assess catchment wide changes in fluvial carbon composition

(colour, size and age) alongside concentration measures. The results indicate clear

patterns of carbon transformations in the river and reservoir, and dissolved Low Molecular

Weight (LMW) coloured carbon to be most subject to change, with both loss and

replacement within the catchment residence time. Whilst the evidence suggests Dissolved

OC (DOC) gains are from Particulate OC (POC) breakdown, the mechanisms of DOC loss

are less certain and may represent greenhouse gas losses or conversions to POC. The

transformations presented here appear to have minimal impact on the amount of harder to

treat (<10 kDa) dissolved carbon, although they do have implications for total DOC

loading to water treatment works. This paper shows that peatland fluvial systems are not

passive receptors of particulate and dissolved organic carbon but locations where carbon is

actively cycled, with implications for the understanding of carbon cycling and water

treatment in peatland catchments.

2

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41


Keywords: Fluvial OC composition, Fluvial OC transformations, Peatlands, Water

Treatment, POC, DOC

3

42

43


1. Introduction

Carbon composition and compositional change can be both a driver and consequence of

carbon cycling. Increased desire to constrain the role of inland waters in global carbon

budgets has shown that fluvial Organic Carbon (OC) levels do not remain static in the

downstream fluvial continuum (Cole et al., 2007; Battin et al., 2009; Ward et al., 2013;

Fasching et al., 2014). Additionally, the nature and mechanisms of downstream changes

are also of interest to water suppliers, as many water collection areas particularly at mid-

to-high northern latitudes (e.g. Lavonen et al., 2015; Ritson et al., 2014) contain substantial

areas of peatland, commonly resulting in high concentrations of dissolved OC (DOC), the

removal of which is a major water treatment cost (Worrall et al., 2004). Therefore,

understanding the within catchment relationships between fluvial OC composition,

concentration and transformations is important for effective management of these

environments.

Fluvial systems draining peatland dominated catchments, have potential for high OC

loading and rates of carbon cycling, as peatlands contain high densities of soil carbon (Yu,

2012; Scharlemann et al., 2014). High OC loads present several problems for drinking

water supply (Matilainen et al., 2011; Kastl et al., 2015), including i) effects on water taste

and colour, ii) increased need for chemicals and subsequent increase in unwanted by-

products such as trihalomethanes (THM), iii) bacterial growth in the distribution system,

and iv) OC binding with metals. Organic matter is commonly removed at treatment works

through the use of coagulants, with dosing rates linked to water colour (Edzwald et al.,

2009).

4

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68


Fluvial OC in peatland catchments takes both dissolved (DOC) and particulate (POC)

forms, and C cycling processes can result in alternations between the two, or transfers to

sedimentary or atmospheric C pools. DOC in the water column can be released as carbon

dioxide (CO2) through biological and photo-chemical degradation (bio and photo

degradation). Lower order streams are thought to be the sites of greatest gas release within

fluvial catchments (Butman and Raymond, 2011; Wallin et al., 2013) which is likely to be

linked to photodegradation. Photodegradation is suggested to occur faster than

biodegradation where DOC is terrestrially derived (Obernosterer and Benner, 2004;

Lapierre and Del Giorgio, 2014) and be linked to water colour (Lapierre et al., 2013). In

addition to breakdown through bio and photo degradation, DOC may be converted to POC

through coagulation or flocculation where different source waters mix (Sharp et al., 2006;

Palmer et al., 2015), a process likely to result in the loss of coloured higher molecular

weight humic materials (Asmala et al., 2014). Reductions in DOC concentrations

downstream may also be observed due to dilution by lower DOC waters (e.g. Worrall et

al., 2006; Tiwari et al., 2014). Processes of DOC conversion may improve drinking water

treatability through reducing overall load, however they are less likely to impact hard to

treat (lower molecular weight, less coloured) DOC (Weishaar et al., 2003; Worrall and

Burt, 2009; Gough et al., 2014). POC may enter long term sediment stores through

deposition or breakdown to DOC in the water column (Palmer et al., 2015; Stimson et al.,

2017a), CO2 or methane (CH4). Despite some exceptions (e.g. Dawson et al., 2012), most

studies on these processes and the mechanisms involved (Koelmans and Prevo, 2003;

Sobek et al., 2009; Kritzberg et al., 2014a) have not considered fluvial OC composition.

Further research into OC transformation mechanisms and composition are required, to

better understand the implications of fluvial carbon export from peatland catchments.

These catchments typically contain both lotic and lentic freshwater environments, so a

5

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93


study combining the two allows for a full appreciation of the range of likely processes.

Studies which combine river and lake locations are currently rare (Ilina et al., 2014), with

the majority focused on one or the other (Laudon et al., 2011; Kothawala et al., 2014), and

not accounting for POC.

The highly degraded peatlands of the UK uplands produce substantial fluvial DOC and

POC flux (Worrall et al., 2006; Pawson et al., 2008), resulting in the potential for

substantial rates of carbon transformation. Consequently these environments have been a

focus for research into fluvial OC fate (Evans et al., 2013; Moody et al., 2013; Worrall and

Moody, 2014; Palmer et al., 2015). UK peatlands contain large stores of soil carbon and

provide almost 70% of UK drinking water supply (Van der Wal et al., 2011), often via

steep (and consequently fast flowing) streams, which link peatland headwaters to water

supply reservoirs. Sediment loads in these streams can be very high (Evans et al., 2006)

and the age of fluvial DOC (assessed through 14C radiocarbon dating) has been found to

be older in comparison to other peatland streams at similar latitudes (Evans et al., 2014), a

phenomena which could be explained by the breakdown of old POC to DOC (see section

2.1). Streams draining UK peatlands along with those in mainland Europe and North

America have seen a rise in DOC concentrations. These increased DOC levels have been

linked to declining acidity (Monteith et al., 2007) and led to concerns over the impact on

water treatment (Ritson et al., 2014).

This research considers changes in fluvial OC composition and concentration

(transformations), between headwaters and treatment works in a UK water supply

catchment. The study seeks to locate, describe, and explain these changes and consider the

implications for the understanding of fluvial carbon cycling and water treatment. The

6

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118


study site is the Kinder Reservoir catchment in the South Pennines UK, which drains large

areas of degraded peat (e.g. Tallis, 1997), resulting in high fluvial OC fluxes. Three sites

of potential carbon transformation are considered; a) a 4km reach of the largest feeder

stream into the reservoir, b) the reservoir itself, and c) a 10km pipe transporting water to

the nearby treatment works.

2. Methodology

2.1. Methods selection

Three methods to determine carbon composition were selected in addition to

measurements to determine DOC concentration. These methods were ultraviolet-visible

(UV-VIS) spectrophotometry, tangential flow ultrafiltration (TFU) and 14C radiocarbon

dating.

Spectroscopic techniques are the most frequently used method to assess dissolved matter

composition (Filella, 2010), and offer easy comparison with many historic studies. The

predominant method is determination of absorbance across the ultraviolet-visible

spectrum, which is also commonly used as a proxy for DOC concentration (Peacock et al.,

2014). UV-VIS absorbance is also commonly used alongside DOC measurements to form

colour carbon ratios such as SUVA254 (Weishaar et al., 2003) and provide indirect

assessment of the likely molecular weight of DOC.

7

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142


Molecular weight can also be directly constrained using size fractionation, and for this

study the technique of TFU was employed. In the water industry ultrafiltration has been

employed both as a treatment and organic matter characterisation technique (Matilainen et

al., 2010, 2011). A substantial majority of organic matter has been found to be removable

with a 10 kDa filter (Schäfer et al., 2000), however it is suggested that material below this

size has substantial Trihalomethane Formation Potential (THMFP) (Chow et al., 2005).

For example a study in China (Wei et al., 2008) found that < 10 kDa material commonly

accounted for over 80% of THMFP.

Radiocarbon (14C) dating has also become more frequently used in recent years to assess

the nature of fluvial organic matter (Marwick et al., 2015). However as with

ultrafiltration it is a complex and costly technique and so cannot be routinely applied.

Determining degradation potential from 14C age can be difficult as, whilst plant derived

carbon such as lignin phenols is commonly very young in age (Martin et al., 2013), old

carbon has also been found to be highly bioavailable in some circumstances (Fellman et

al., 2014; Marín-Spiotta et al., 2014). One particular benefit of 14C dating however, is that

it can be used as an indication of transformations between POC and DOC. Terrestrial

POC is commonly older than DOC especially where high sediment loads are present

(Marwick et al., 2015) or human disturbance plays a role (Evans, 2014; Butman et al.,

2015). Therefore, a change in DOC age within catchment can be used as evidence for

POC breakdown to DOC.

Colour and DOC data (see 2.4), allowed for assessment of the magnitude and location of

carbon transformations and changes in DOC concentrations, whilst carbon size data

8

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167


allowed the impact of changes on a key aspect of water treatability to be determined, and

14C data allowed assessment of the likely mechanisms of carbon transformations.

2.2. Study area, sampling sites and calculation of transformations.

The reservoir and its surrounding catchment area along with sampling points (see also

Table A-I) are shown in Figure 1. For further information on the reservoir inlet

(KR,BC,WC) and outlet (VH,KRO) sampling points see Stimson et al. (2017a).

Additionally the study was conducted alongside work to understand nitrogen dynamics

(see Edokpa et al. 2015, 2016). To capture changes throughout the catchment, three

locations were considered: a) the major feeder stream to the reservoir draining a large area

of blanket peat, b) the reservoir, and c) the 10km pipe from the reservoir to the water

treatment works.

Changes within the river were calculated as the difference between headwater samples

taken at site B or KH and the reservoir inlet KR. The main purpose of site B was to assess

the impacts of restoration work taking place on the Kinder Plateau(Stimson et al., 2017b),

but it also proved useful in this study. However as site B drained a small catchment there

was not always a flow of water (making the total number of samples from here fewer than

from the other sites). As a consequence for TFU size fractionation and 14C dating, which

required larger sample volumes than those for determining concentration and colour, site

KH was used as a substitute for site B, as the flow at KH was greater and more reliable.

9

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191


Figure 1: Study area - Kinder Reservoir Catchment. Sampling site codes are explained further in the

text.

Comparison of the two sites showed some downstream declines in DOC concentration and

colour, however these were of a lower magnitude than changes observed between B and

KR (Table I). To take account of the effect of dilution within the river from other stream

inputs between B and KR, concentration data were adjusted to take account of the

upstream area of blanket peat. In the catchment area upstream of site KR the area of

blanket bog peat soils recorded equates to 52% of the total area. These blanket peat soils

are largely found on the headwater plateau so the upstream area of blanket peat at site B

was assumed to be 100%. For data from site KR dilution adjustment was applied by

dividing concentration data by 0.5. This approach is conservative as it assumes all fluvial

DOC inputs come from the blanket bog area, however other studies (e.g. Evans et al.,

10

192

193

194

195

196

197

198

199

200

201

202

203

204

205


2006a; Laudon et al., 2011) indicate all land cover types produce some fluvial DOC even

during low flow. Changes within the reservoir were calculated as the flow weighted

difference between the three inlet samples, and the two outlet samples, with the

contribution of each individual site weighted by discharge. For information on discharge

measurement see (Stimson et al., 2017a). Changes within the pipe to the treatment works

were calculated as the difference between samples VH and TW, that represents the water

as it enters the treatment works.

2.3. Water sampling

Samples to determine colour and OC concentration were collected fortnightly during 2013

from sites B and TW and the five reservoir sites. Fortnightly data collected from the

reservoir sites during 2012 is also presented in section 4.1.2 (see also Stimson et al.,

2017a). TFU size fractionation and 14C dating analysis was performed on samples from

six sites (sites as above except site KH was used instead of B, and site KRO was omitted).

Sampling for TFU took place on three separate dates in spring, summer and autumn 2013

and for 14C dating took place on one date in spring 2014. Additionally, some further

samples from four downstream surveys of the river reach between B and KR (sites K1-K7,

Figure 1) were taken on 14/02/2012, 19/05/2012, 10/12/2012 and 01/07/2013, and

analysed for colour and pH.

11

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230


2.4. DOC concentration and water colour measurements

Water samples were filtered on site shortly after collection using 0.45µm glass microfibre

syringe filters. Samples were then analysed for light absorbance using a Hach DR 5000

spectrophotometer and for DOC (2013 samples only), using a Shimadzu TOC analyser.

The majority of samples were analysed within two weeks for absorbance and three weeks

for DOC, with samples stored in the dark below 4°C in the intervening time.

2.5. Colour measures

Two colour carbon ratios were used in the analysis. These were absorbance at 254 nm

(Abs254) and 400 nm (Abs400) against DOC concentration. Units for absorbance are in

absorbance units per metre (au-m-1), for DOC concentration in mg/l, and for ratios as L

mg-1 m-1. Abs254 divided by DOC concentration is also known as SUVA254 – standing for

specific ultra violet absorbance, and is commonly used as a measure of quality in water

treatment (Weishaar et al., 2003). Abs400 / DOC is also frequently used (Wallage et al.,

2006), as a carbon quality measure. The E4:E6 ratio (Abs465 / Abs665) is also used. This

has been used in a variety of studies (Kritzberg et al., 2014) and represents the degree to

which lower vs higher molecular weight coloured material, is present in a sample. The

ratio increases with declining molecular weight.

12

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253


2.6. 14C dating and TFU size fractionation

14C radiocarbon ages of DOC and POC were determined at the Scottish Universities

Environmental Research Centre (SUERC) accelerator mass spectrometry laboratory. For

further information on the method see (Stimson et al., 2017a). To perform TFU size

fractionation large (500ml) water samples were collected on 3 dates in May, July and

November 2013. The TFU size fractionation was performed within one week of sample

collection, using Polyether Sulfone (PES) membranes of 1000 kDa, 50 kDa, and 10 kDa,

with OC concentration measured in each size fraction. For further details on the methods

employed see (Gaffney et al., 2008). Results presented do not include all size fractions but

rather the relative proportion of <10 kDa material as a proportion of all material under 0.1

µm in size, as the smallest size fraction is likely to contain the least treatable material.

2.7. Statistical analysis

Average (mean) changes were calculated based on the difference between sample means,

calculated from sets of paired data (see Table I). To assess the statistical significance of

changes observed two-tailed paired students t-tests were performed to test for a significant

difference between each separate set of paired data. Changes for single tests were deemed

to be statistically significant where T-Test P values were below the 95% confidence

interval (α = 0.05). Where multiple tests were considered together (family wide

comparisons) P values had to be below a sidak corrected (Šidàk, 1967) reduced α value to

be considered statistically significant at the 95% confidence level. As the T-Test requires

normally distributed data, data were also assessed for normality using the Shapiro-Wilk

test (Shapiro and Wilk, 1965).

13

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278


3. Results

3.1. DOC and colour measurement

The river reach was the site of greatest change, with substantial declines in DOC

concentration and the E4:E6 ratio observed throughout the year, and yearly average

declines in the colour carbon ratios between the headwater and reservoir inlet sites (Table

I, Figure 2). DOC changes in the river adjusted for dilution effects (See section 2.2),

showed a monthly average decline (between headwaters and reservoir) in DOC

concentration commonly between 5 and 10 mg/l, with the greatest difference being 25

mg/l. The E4:E6 ratio declined by an annual average of 6.3, indicating that higher

molecular weight material formed a greater proportion of the coloured material in the

reservoir inlet sample, as declines in absorbance were proportionally greater at 465 nm

than 665 nm. The two colour carbon ratios also generally showed declines, but of lower

magnitude (or very slightly positive) in the summer, which indicates that carbon mostly

became less coloured downriver at the wavelengths of 254 nm and 400 nm. Average

drops in the ratios were 1 and 0.1 for colour compared to carbon at 254 nm and 400 nm

respectively. Comparison of sites B and KH (Table I) showed variation between the sites,

although of a lower magnitude than within the full river reach, however between B and

KH in this test only the average change in DOC concentration was shown to be statistically

significant. Average change calculated from downstream surveys (Table I), showed

statistically significant changes in Abs400 and pH after the Red Brook confluence (i.e. ∆

location K4-K5). Average change in Abs400 between K1 and K3 suggested the change

between these sites varied in direction and was greatest in the first 200m of the river;

however the changes here were not statistically significant.

14

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302


Table I: Average change and significance of change of water quality indices for the three main locations of change and changes within the river based on a comparison of change between the headwaters (B-KH) and full reach of the river (B-KR), and high resolution downstream river surveys (see sections 2.2 and 2.3)

∆ Location Determinand

DOC(mg/l) E4:E6 SUVA254

(L mg-1 m-1)Abs400:DOC(L mg-1 m-1)

River∆ -15.4/-11.1a -6.3 -1.0 -0.1

P Value <0.001 <0.001 <0.001 0.012

Reservoir∆ 1.7 1.6 1.1 0.2

P Value <0.001 0.007 <0.001 <0.001

Pipe∆ 0.4 0.4 0.0 0.0

P Value 0.009 0.41 0.921 0.299

B-KH∆ -2.5 1.9 0.8 0.1

P Value 0.012 0.795 0.05 0.07

B-KR∆ -10.5 -6.7 -1.7 -0.2

P Value 0.012 0.407 0.01 0.041

Abs400

(au m-1) E4:E6 pH

K1-K2∆ -4 -2.1 0.2

P Value 0.111 0.032 0.222

K2-K3∆ 1.5 1.2 0

P Value 0.166 0.281 0.562

K3-K4∆ -1.3 -0.3 0

P Value 0.16 0.852 0.205

K4-K5∆ -5.9 -2.8 1.9

P Value 0.006 0.281 0.002

K5-K6∆ -0.5 0.1 0.5

P Value 0.28 0.091 0.11

K6-K7∆ -0.3 -0.7 0.2

P Value 0.439 0.371 0.623

15

303

304305306307

308

309


Values for average (mean) changes (∆) are based on annual averages for the river, reservoir and pipe, three samples for B-KH and B-KR comparing the three samples taken at KH for TFU fractionation with the appropriate monthly average values from sites B and KR, and four samples for downstream survey sites K1-K7 (as detailed in section 2.3). P values represent results of a two tailed paired t-test to assess significance of changes. Sidak corrected α values for 95% confidence are 0.017 and 0.013 for 3 and 4 variables respectively. Shapiro-Wilk tests for normality (see Table A-III) showed that only the DOC changes in the river and changes in pH between K2 and K3 had potentially non-normal distributions adilution adjusted (see section 2.2).

16

310311312313314315316317318


As with the river reach, changes in the reservoir (between inflow and outflow) also showed

a consistent pattern, with year round increases in DOC concentration and the three colour

ratios (Figure 2). DOC in the reservoir showed a mean increase for the year of 1.7 mg/l.

The E4:E6 ratio rose by an average of 1.6, indicating the molecular weight of the coloured

material in the reservoir was declining. The in reservoir increases in DOC and the E4:E6

ratio were less than the declines seen in the river, however, the rise in the two colour

ratios through the reservoir was of a comparable magnitude, with average increases of 1.1

and 0.20 for colour compared to carbon at 254 nm and 400 nm respectively (Table I).

Unlike in the river and reservoir, carbon changes within the pipe did not show a consistent

pattern, and carbon processing here appeared to be of a smaller magnitude than in the river

and reservoir. Whilst annual average change in the river and reservoir were found to be

statistically significant at both the individual and family wide level, the changes in the pipe

were not except for the small rise in DOC concentration (Table I)

17

319

320

321

322

323

324

325

326

327

328

329

330

331

332


Figure 2: Monthly average differences in carbon and colour ratios in 2013 showing changes in the river, reservoir and pipe for a) DOC concentration, and carbon quality ratios b) E4:E6, c) SUVA254 and d) Abs400:DOC. In a) River Cadj represents river DOC change adjusted for in river DOC dilution effects (see text) and rainfall (mm) monthly rainfall totals recorded below the reservoir dam. Summary statistics for the data used to calculate these changes are shown in Table A-II.

18

333334335336337


3.2. Carbon Size

The river showed a constant % gain in the proportion of <10 kDa OC, indicating the OC

was getting proportionally smaller. The reservoir by contrast showed a loss of <10KDa

OC indicating a proportional gain in larger carbon. The losses in the reservoir were half or

less the magnitude of the gains in the river. Changes in both the river and reservoir were

not statistically significant at the 95% confidence level (P>0.05), but were close to the

90% level (P>0.1). There was no clear trend in size changes within the pipe, which is

reflected by a much higher P value (~0.5) (Table II).

Table II: % change in truly dissolved OC(<10 kDa) in carbon below 1000 kDa in size.

Δ Location Date P valuea

13/05/13 01/07/13 04/11/13

River 8% 18% 19% 0.120Reservoirc -3% -3% -9% 0.112Pipe 13% -3% 1% 0.493

aP values represent results of a two tailed paired t-test to assess significance of changes for 3 dates combined. P values from Shapiro-Wilk tests for normality were 0.528, 0.062 and 0.705 for the river, reservoir and pipe respectively, where P >0.05 indicates a potentially non normal dataset. CVHout-Allin (see table A-IV for site data)

3.3. Carbon age

The 14C radiocarbon dating results are shown for samples taken on 12/05/2014 (Table III).

There was a rise in DOC age of 93 years in the Kinder River, where POC age also rose by

332 years. The age of POC entering the reservoir was between 1222 and 3093 years BP

and substantially greater than the age of DOC. POC age is likely to be influenced by

19

338

339

340

341

342

343

344

345

346

347348

349350

351352353354355356

357

358

359

360

361

362

363


material from deep erosional gullies on the Kinder Plateau, where peat formation began

circa 8 Ka BP (Conway, 1954). The reservoir showed the most dramatic change with

DOC at the VH outflow 323, 1147, and 1301 years older than the inflows of KR, WC and

BC respectively. There was a fall of 56 years in radiocarbon age through the pipe which is

19 years greater than the typical error on the measurement of 37 years.

Table III: Radiocarbon results for the study sites from 12/05/2014.

Site Type Code 14C Enrichment Conventional radiocarbon age

δ13CVPDB‰ ± 0.1

% Moderna +/-1σ Years

BP +/-1σ

KH DOC SUERC-54801 89 0.41 976 37 -28.2KR DOC SUERC-54802 88 0.4 1069 37 -27.7WC DOC SUERC-54803 97 0.45 245 37 -29.1BC DOC SUERC-54804 99 0.45 91 37 -28.7VH DOC SUERC-54806 84 0.39 1392 37 -29TW DOC SUERC-54809 85 0.39 1336 37 -28KH POC SUERC-54372 71 0.31 2761 35 -26.6KR POC SUERC-54373 68 0.30 3093 35 -26.9WC POC SUERC-54374 86 0.37 1222 35 -27.2BC POC SUERC-54375 85 0.37 1316 35 -28.4

aAn average flow weighted % modern 14C value for reservoir inputs (see section 4.1.2) was calculated based on weightings of 47%, 8% and 37% for sites KR, WC and BC respectively, plus a weighting of 9% for the ungauged catchment area where % modern 14Cvalues are assumed to be the mean of values for three inflow streams. The calculated averages are estimated at 93.06 % for DOC and 76.67% for POC.

20

364

365

366

367

368

369

370371

372373374375376377


4. Discussion

A summary of the nature of DOC transformations observed in this study (Table IV),

suggest that the river is the site of the greatest transformations in DOC concentration and

size, and dissolved matter colour, with the data pointing towards preferential loss of the

lower molecular weight coloured materials.

Table IV: Summary of downstream changes on DOC quantity and composition presented in section 3..

Δ Withina DOC charecteristicsb

Concentration and colour ratios Size Age

River Declinec Smaller Slightly older

Reservoir Increase Larger Substantially older

Pipe No trend(except

No trend Very slightly younger

small DOC increase)

aChange locations are defined in section 2.2.bConcentration and colour ratios refer to DOC (mg/l), E4:E6, SUVA254 and Abs400:DOC, size to the proportion of <10KDa OC as determined by TFU and age to the age of DOC in radiocarbon years BP (see sections 3.1, 3.2 and 3.3 respectively).cExcept for 2 months when colour carbon ratio rises very slightly.

The proportional increase in the smallest size fraction of OC, suggests the lightest and least

coloured material is relatively more resistant to such conversions. The reservoir shows the

opposite pattern to the river, albeit at a lower magnitude. Changes in DOC age show a

different pattern, with the river and reservoir presenting the same trend, although rates are

substantially lower in the river. There are no clear patterns of change observable in the

21

378

379

380

381

382

383

384

385386

387

388389390391392393

394

395

396

397

398


pipe in DOC size, age and dissolved matter colour. However there is a small but

statistically significant rise in DOC concentration.

4.1. Mechanisms of OC transformations

The different characteristics of the river and reservoir are likely to account for variations in

OC transformations between these two locations discussed above. Whilst the DOC losses

observed in the river are likely to be rapid due to short residence times, the reservoir

produces coloured DOC in an environment where reactions can take place over a longer

time period, and acts as a collection site for POC sediments which are periodically

oxidised during drawdown.

4.1.1. Within-river DOC loss

The declines in colour through the river are interpreted as evidence of UV

photodegradation of humic compounds. This corresponds with other studies that suggest

this is a rapid process (Evans et al., 2013; Moody et al., 2013), and is likely to exert

greater influence than biodegradation, given the short length of the river. Flow velocities

recorded at site KR generally ranged from a maximum of 1 m/s down to 0.25 m/s, which if

applied to the full river length (4km) equates to a residence time of approximately 1 - 4.5

hours. An alternative hypothesis could be preferential coagulation or flocculation of

coloured DOC (Palmer et al., 2015), after the Red Brook confluence. As the Red Brook

catchment drains a proportionally smaller area of the blanket bog plateau area than the

22

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422


main Kinder river at the point where the two meet (see Figure 1), then the other organo-

mineral soil types are likely to have a greater influence upon Red Brook water chemistry.

The rise in pH coupled with the fall in Abs400 at the confluence observed from the stream

surveys (Table I), provides some evidence for this, as increased pH can encourage

flocculation and be an indication of greater mineral soil influence. However the presence

of coagulants such as Fe or Al was not directly measured in this survey. In a recent study

in the Kolyma river basin, Siberia (Mann et al., 2012) it was found that that spring flush

samples had the highest SUVA254 values and suffered the most degradation. Average

SUVA254 values from these spring flush samples are comparable with those for headwater

samples in this study (around 4), suggesting similar degradation potential. At the Kinder

study site there was little deviation in SUVA254 values below this average, suggesting the

potential for carbon loss is similar year round.

4.1.2. Reservoir DOC-POC interactions

Changes within the reservoir show a year round increase in DOC and coloured material,

accompanied by an increase in DOC 14C age. The most likely explanation for the increase

in carbon is release from the breakdown of POC in the water body or bed sediments

(Stimson et al., 2017a), a hypothesis also supported by laboratory experiments (Koelmans

and Prevo, 2003; Goulsbra et al., 2016). A simple mixing model calculation based on this

hypothesis can be used to estimate the proportion of DOC which must be derived from

POC in order to produce the observed decline in % modern 14C values at the catchment

outlet. This calculation compared a flow weighted average of % modern 14C values from

the inflow steams with the outflow value at site VH and indicated that up to 54.7% of

outlet DOC may be derived from POC. Given that reservoir residence times are likely to

23

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447


be of the order of months this process could take place over a much longer timescale than

the river DOC losses.

The data presented in section 3.1 suggest that transformations within the reservoir result in

an overall net increase in DOC concentration. However, this effect is of a lower

magnitude than the reduction of DOC concentration in the river reach, and other data from

this study suggest that DOC concentrations can also fall within the reservoir (Stimson et

al., 2017a). These alternate trends are apparent in Figure 3, which includes additional

colour data for 2012 and considers changes in the raw (approximately fortnightly) data,

rather than the monthly average data presented earlier.

Figure 3a and Figure 3b show raw (fortnightly) data for 2013, with changes in DOC

plotted against those in SUVA254 and the E4:E6 ratio respectively. Changes in both

SUVA254 and the E4:E6 ratio are positive in the vast majority of cases when DOC

concentrations increase through the reservoir (see also Figure 2a). In the opposite situation

when DOC concentrations decline within the reservoir, the relationship to colour appears

weaker. This pattern can also be seen in combined E4:E6 and colour data from 2012 and

2013 (Figure 3c), and is likely to account for the low r2 values and some P values which

are greater than 0.05 and so not statistically significant at the 95% confidence level.

Consideration of the reservoir input DOC concentrations may provide an explanation for

this. Figure 3d which again shows combined data for 2012 and 2013 shows that decreases

in DOC concentrations through the reservoir are associated with higher input rates, with

the relationship between change in colour and input colour having both a significant

relationship, and a much better r2 value than the other three models.

24

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471


Figure 3: Patterns of DOC and colour change in the Kinder reservoir. The plots show within reservoir change in DOC concentration plotted against change in a) SUVA254 and b) E4:E6 carbon quality ratios for 2013 and c) against the E4:E6 carbon quality ratio using data from 2012 and 2013; d) plots change in absorbance through the reservoir against the Abs400 value of the combined input waters for 2012 and 2013. Model parameters for best fit lines on plots a, b, c and d respectively are as follows. Equation: y = 0.1487x + 0.8988, y = 0.8246x + 0.2936, y = 0.4098x + 0.7225, y = -0.8543x + 3.0036; R2 0.09, 0.41, 0.20, 0.83; P value (intercept): 0.002, 0.587, 0.067, <0.001; P value (slope) 0.132, <0.001, 0.001, <0.001.

25

472473474475476477


It is therefore hypothesised that high carbon waters entering the reservoir are much more

subject to the competing processes of carbon breakdown and addition, resulting in a mixed

impact on colour. By contrast low carbon waters may be closer to their limits of degradation

(especially through UV photo breakdown of humic compounds), so the addition of carbon

from reservoir POC exerts a more dominant influence. Laboratory experiments (Evans et al.,

2013; Goulsbra et al., 2016) simulating POC-DOC interactions within stream systems, by

adding peat-derived Particulate Organic Matter (POM) to river water, also provide support

for this hypothesis. In these experiments reductions in lower molecular weight coloured were

observed when the water was untreated but increases occurred when sterilised river water

(said to be similar to photodegraded waters) was used.

The breakdown of POC observed in the reservoir is also likely to result in greenhouse gas

release (e.g. Ferland et al., 2014). For example the addition of POM to river water is argued

to result in the production of both CO2 and DOC, with CO2 release possibly taking place

through a DOC intermediary (Evans et al., 2013). The same study also found that POM

added to river water resulted in lower CO2 and DOC release than with sterilised water. The

implications of this are that coloured waters potentially act to reduce CO2 release from bed

sediments, however further studies which record greenhouse gas release are needed to verify

this mechanism.

26

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496


4.2. Implications for water treatment

These results indicate that the magnitude of OC removal required at water treatment works is

reduced by in catchment processes, although the increases in coloured DOC in the reservoir

are an exception to this trend. This study provides evidence that DOC increases in the

reservoir are linked to reservoir POC loads and the presence of organic bed sediments (see

section 4.1.2). Catchment processes appear not to increase the total amount of the less

treatable >10 kDa OC size fraction and although there is a proportional increase in this size

fraction in the river it is likely to be balanced by comparatively larger declines in the total

amount of DOC and the opposite trend in the reservoir.

27

497

498

499

500

501

502

503

504

505

506

507

508


5. Conclusions

The fluvial system is shown to be an important site of carbon transformation in waters

draining from degraded peatlands with headwater streams, reservoirs and water

supply pipes playing distinct roles in carbon cycling.

Rapid removal of coloured DOC is observed in the fluvial system. In the reservoir

there is evidence of the addition of coloured DOC. DOC along the pipe to the

treatment works shows comparatively less change with no statistically significant

changes in colour ratios.

Dissolved Low Molecular Weight (LMW) coloured carbon is shown to be most

subject to change, and subject to both loss and replacement within the catchment

residence time.

The evidence suggests that the mechanism for addition of DOC in the reservoir

system is POC breakdown, most likely associated with large quantities of organic bed

sediments in the reservoir.

The carbon transformations presented here appear to have minimal effect on the

amount of material (<10 kDa) that is hard to remove during water treatment, however

the potential role of reservoir sediments in POC-DOC transformations may mean that

this process is important in total DOC loading to the treatment works.

28

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530


Acknowledgements

This work was part funded by United Utilities, the Moors for the Future Partnership and the

National Trust. Additionally, this work was supported by the NERC Radiocarbon Facility

NRCF010001 (allocation number 1657.1012). From these partners we would particularly

like to thank Dr Mark Garnett at NRCF East Kilbride and reservoir manager Matthew Ethell

for their support of the project.

Thank you to all the people who helped with fieldwork including Alan Heath, Adrienne King,

Andrew Harding and Roger Braithwaite, Michael Pilkington and Tom Spencer and to Donald

Edokpa for both field and laboratory assistance.

Thank you to Jon Yarwood and John Moore at Manchester University Geography

Laboratories for providing technical support.

We also thank the anonymous reviewers for helpful comments which improved the

manuscript.

29

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548


Bibliography

Asmala E, Bowers DG, Autio R, Kaartokallio H, Thomas DN. 2014. Qualitative changes of

riverine dissolved organic matter at low salinities due to flocculation. Journal of

Geophysical Research: Biogeosciences 119 (10) 1919-1933–566. DOI:

10.1002/2014JG002722

Battin TJ, Luyssaert S, Kaplan LA, Aufdenkampe AK, Richter A, Tranvik LJ. 2009. The

boundless carbon cycle. Nature Geoscience 2 (9): 598–600 DOI: 10.1038/ngeo618

Butman D, Raymond PA. 2011. Significant efflux of carbon dioxide from streams and rivers

in the United States. Nature Geoscience 4 (12): 839–842 DOI: 10.1038/ngeo1294

Butman DE, Wilson HF, Barnes RT, Xenopoulos MA, Raymond PA. 2015. Increased

mobilization of aged carbon to rivers by human disturbance. Nature Geoscience 8

(February): 112–116 DOI: 10.1038/NGEO2322

Chow A, Dahlgren R, Gao S. 2005. Physical and chemical fractionation of dissolved organic

matter and trihalomethane precursors: A review. Journal of water supply research and

technology 54: 475–507

Cole JJ, Prairie YT, Caraco NF, McDowell WH, Tranvik LJ, Striegl RG, Duarte CM,

Kortelainen P, Downing JA, Middelburg JJ, et al. 2007. Plumbing the global carbon

cycle: Integrating inland waters into the terrestrial carbon budget. Ecosystems 10 (1):

171–184 DOI: 10.1007/s10021-006-9013-8

Conway VM. 1954. Stratigraphy and Pollen Analysis of Southern Pennine Blanket Peats.

Journal of Ecology 42 (1): 117–147. www.jstor.org/stable/2256982

Dawson JJC, Adhikari YR, Soulsby C, Stutter MI. 2012. The biogeochemical reactivity of

suspended particulate matter at nested sites in the Dee basin, NE Scotland. Science of the

30

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572


Total Environment 434: 159–170 DOI: 10.1016/j.scitotenv.2011.08.048

Edokpa DA, Evans MG, Rothwell JJ. 2015. High fluvial export of dissolved organic nitrogen

from a peatland catchment with elevated inorganic nitrogen deposition. Science of the

Total Environment 532: 711–722 DOI: 10.1016/j.scitotenv.2015.06.072

Edokpa DA, Evans MG, Rothwell JJ. 2016. Reservoirs are hotspots of nitrogen cycling in

peatland catchments. Hydrological Processes 30: 3666–3681. DOI: 10.1002/hyp.10892

Edzwald JK, Dee PE, Kaminiski GS. 2009. A practical method for water plants to select

cooagulant dosing. Journal NEWWA March: 11–27

Evans C. 2014. Biogeochemistry: Old carbon mobilized. Nature Geoscience 8 (2): 85–86

DOI: 10.1038/ngeo2334

Evans C, Allott T, Billett M, Burden A, Dinsmore K, Evans M, Freeman C, Holden J, Jones

D, Jones T, et al. 2013. Greenhouse gas emissions asscoaited with non gaseous losses of

carbon from peatlands - Fate of Particulate and dissolved carbon - Defra Project SP1205

Final report.

Evans CD, Cooper DM, Juggins S, Jenkins A, Norris D. 2006a. A linked spatial and temporal

model of the chemical and biological status of a large, acid-sensitive river network.

Science of the Total Environment 365 (1–3): 167–185 DOI:

10.1016/j.scitotenv.2006.02.037

Evans CD, Page SE, Jones T, Moore S, Gauci V, Laiho R, Hruška J, Allott TEH, Billett MF,

Tipping E, et al. 2014. Contrasting vulnerability of drained tropical and high-latitude

peatlands to fluvial loss of stored carbon. Global Biogeochemical Cycles 28 (11): 1215–

1234 DOI: 10.1002/2013GB004782

Evans M, Warburton J, Yang J. 2006b. Eroding blanket peat catchments: Global and local

implications of upland organic sediment budgets. Geomorphology 79 (1–2): 45–57 DOI:

10.1016/j.geomorph.2005.09.015

31

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597


Fasching C, Behounek B, Singer GA, Battin TJ. 2014. Microbial degradation of terrigenous

dissolved organic matter and potential consequences for carbon cycling in brown-water

streams. Scientific reports 4: 4981 DOI: 10.1038/srep04981

Fellman JB, Spencer RGM, Raymond PA, Pettit NE, Skrzypek G, Hernes PJ, Grierson PF.

2014. Dissolved organic carbon biolability decreases along with its modernization in

fluvial networks in an ancient landscape. Ecology 95 (9): 2622–2632 DOI: 10.1890/13-

1360.1

Ferland ME, Prairie YT, Teodoru C, Del Giorgio PA. 2014. Linking organic carbon

sedimentation, burial efficiency, and long-term accumulation in boreal lakes. Journal of

Geophysical Research: Biogeosciences 119 (5): 836–847 DOI: 10.1002/2013JG002345

Filella M. 2010. Quantifying ‘humics’ in freshwaters: purpose and methods. Chemistry and

Ecology 26 (sup2): 177–186 DOI: 10.1080/02757540.2010.494159

Gaffney JW, White KN, Boult S. 2008. Oxidation state and size of Fe controlled by organic

matter in natural waters. Environmental Science and Technology 42 (10): 3575–3581

DOI: 10.1021/es702880a

Gough R, Holliman PJ, Willis N, Freeman C. 2014. Dissolved organic carbon and

trihalomethane precursor removal at a UK upland water treatment works. Science of the

Total Environment 468–469: 228–239 DOI: 10.1016/j.scitotenv.2013.08.048

Goulsbra CS, Evans MG, Allott TEH. 2016. Rates of CO2 efflux and changes in DOC

concentration resulting from the addition of POC to the fluvial system in peatlands.

Aquatic Sciences 78 (3): 477–489 DOI: 10.1007/s00027-016-0471-6

Ilina SM, Drozdova OY, Lapitskiy SA, Alekhin Y V., Demin V V., Zavgorodnyaya YA,

Shirokova LS, Viers J, Pokrovsky OS. 2014. Size fractionation and optical properties of

dissolved organic matter in the continuum soil solution-bog-river and terminal lake of a

boreal watershed. Organic Geochemistry 66: 14–24 DOI:

32

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622


10.1016/j.orggeochem.2013.10.008

Kastl G, Sathasivan A, Fisher I. 2016. A selection framework for NOM removal process for

drinking water treatment. Desalination and Water Treatment 57 (17): 7679-7689. DOI:

10.1080/19443994.2015.1044476

Koelmans AA, Prevo L. 2003. Production of dissolved organic carbon in aquatic sediment

suspensions. Water Research 37 (9): 2217–2222 DOI: 10.1016/S0043-1354(02)00581-X

Kothawala DN, Stedmon CA, Müller RA, Weyhenmeyer GA, Köhler SJ, Tranvik LJ. 2014.

Controls of dissolved organic matter quality: Evidence from a large-scale boreal lake

survey. Global Change Biology 20 (4): 1101–1114 DOI: 10.1111/gcb.12488

Kritzberg ES, Granéli W, Björk J, Brönmark C, Hallgren P, Nicolle A, Persson A, Hansson

LA. 2014a. Warming and browning of lakes: Consequences for pelagic carbon

metabolism and sediment delivery. Freshwater Biology 59 (2): 325–336 DOI:

10.1111/fwb.12267

Kritzberg ES, Villanueva AB, Jung M, Reader HE. 2014b. Importance of boreal rivers in

providing iron to marine waters. PLoS ONE 9 (9), 1-10. DOI:

10.1371/journal.pone.0107500

Lapierre J-F, Guillemette F, Berggren M, Del Giorgio PA. 2013. Increases in terrestrially

derived carbon stimulate organic carbon processing and CO2 emissions in boreal aquatic

ecosystems. Nature Communications 4: 2972 DOI: 10.1038/ncomms3972

Lapierre JF, Del Giorgio PA. 2014. Partial coupling and differential regulation of biologically

and photochemically labile dissolved organic carbon across boreal aquatic networks.

Biogeosciences 11 (20): 5969–5985 DOI: 10.5194/bg-11-5969-2014

Laudon H, Berggren M, Agren A, Buffam I, Bishop K, Grabs T, Jansson M, Kohler S. 2011.

Patterns and Dynamics of Dissolved Organic Carbon (DOC) in Boreal Streams: The

Role of Processes, Connectivity, and Scaling. Ecosystems 14 (6): 880–893 DOI:

33

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647


10.1007/s10021-011-9452-8

Lavonen EE, Kothawala DN, Tranvik LJ, Gonsior M, Schmitt-Kopplin P, Köhler SJ. 2015.

Tracking changes in the optical properties and molecular composition of dissolved

organic matter during drinking water production. Water Research 85: 286–294 DOI:

10.1016/j.watres.2015.08.024

Mann PJ, Davydova A, Zimov N, Spencer RGM, Davydov S, Bulygina E, Zimov S, Holmes

RM. 2012. Controls on the composition and lability of dissolved organic matter in

Siberia’s Kolyma River basin. Journal of Geophysical Research: Biogeosciences 117

(1): 1–15 DOI: 10.1029/2011JG001798

Marín-Spiotta E, Gruley KE, Crawford J, Atkinson EE, Miesel JR, Greene S, Cardona-Correa

C, Spencer RGM. 2014. Paradigm shifts in soil organic matter research affect

interpretations of aquatic carbon cycling: Transcending disciplinary and ecosystem

boundaries. Biogeochemistry 117 (2–3): 279–297 DOI: 10.1007/s10533-013-9949-7

Martin EE, Ingalls AE, Richey JE, Keil RG, Santos GM, Truxal LT, Alin SR, Druffel ERM.

2013. Age of riverine carbon suggests rapid export of terrestrial primary production in

tropics. Geophysical Research Letters 40 (21): 5687–5691 DOI:

10.1002/2013GL057450

Marwick TR, Tamooh F, Teodoru CR, Borges A V., Darchambeau F, Bouillon S. 2015. The

age of river-transported carbon: A global perspective. Global Biogeochemical Cycles 29

(2): 122–137 DOI: 10.1002/2014GB004911

Matilainen A, Gjessing ET, Lahtinen T, Hed L, Bhatnagar A, Sillanpää M. 2011. An

overview of the methods used in the characterisation of natural organic matter (NOM) in

relation to drinking water treatment. Chemosphere 83 (11): 1431–1442 DOI:

10.1016/j.chemosphere.2011.01.018

Matilainen A, Vepsäläinen M, Sillanpää M. 2010. Natural organic matter removal by

34

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672


coagulation during drinking water treatment: A review. Advances in Colloid and

Interface Science 159 (2): 189–197 DOI: 10.1016/j.cis.2010.06.007

Monteith DT, Stoddard JL, Evans CD, de Wit HA, Forsius M, Høgåsen T, Wilander A,

Skjelkvåle BL, Jeffries DS, Vuorenmaa J, et al. 2007. Dissolved organic carbon trends

resulting from changes in atmospheric deposition chemistry. Nature 450 : 537–540 DOI:

10.1038/nature06316

Moody CS, Worrall F, Evans CD, Jones TG. 2013. The rate of loss of dissolved organic

carbon (DOC) through a catchment. Journal of Hydrology 492: 139–150 DOI:

10.1016/j.jhydrol.2013.03.016

Obernosterer I, Benner R. 2004. Competition between biological and photochemical

processes in the mineralization of dissolved organic carbon. Limnol. Oceanogr 49 (1):

117-124.

Palmer S, Evans C, Chapman P, Burden A, Jones T, Allott T, Evans M, Moody C, Worrall F,

Holden J. 2015. Sporadic hotspots for physico-chemical retention of aquatic organic

carbon: from peatland headwater source to sea. Journal of Aquatic Sciences 78: 491–504

DOI: 10.1007/s00027-015-0448-x

Pawson RR, Lord DR, Evans MG, Allott TEH. 2008. Fluvial organic carbon flux from an

eroding peatland catchment, southern Pennines, UK. Hydrology and Earth System

Sciences Discussions 12 (2): 625–634 DOI: 10.5194/hessd-4-719-2007

Peacock M, Evans CD, Fenner N, Freeman C, Gough R, Jones TG, Lebron I. 2014. UV-

visible absorbance spectroscopy as a proxy for peatland dissolved organic carbon (DOC)

quantity and quality: considerations on wavelength and absorbance degradation.

Environmental science. Processes & impacts 16 : 1445–1461 DOI:

10.1039/c4em00108g

Ritson JP, Graham NJD, Templeton MR, Clark JM, Gough R, Freeman C. 2014. The impact

35

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

691

692

693

694

695

696

697


of climate change on the treatability of dissolved organic matter (DOM) in upland water

supplies: A UK perspective. Science of the Total Environment 473–474: 714–730 DOI:

10.1016/j.scitotenv.2013.12.095

Schäfer AJ, Fane AG, Waite TD. 2000. Fouling effects on rejection in the membrane

filtration of natural waters. Desalination 131 (1–3): 215–224 DOI: 10.1016/S0011-

9164(00)90020-1

Scharlemann JP, Tanner EVJ, Hiederer R, Kapos V. 2014. Global soil carbon: understanding

and managing the largest terrestrial carbon pool. Carbon Management 5 (1): 81–91

DOI: 10.4155/cmt.13.77

Shapiro SS, Wilk MB. 1965. An Analysis of Variance Test for Normality (Complete

Samples). Biometrika 52 (3/4): 591–611 DOI: 10.1093/biomet/52.3-4.591

Sharp EL, Parsons SA, Jefferson B. 2006. The impact of seasonal variations in DOC arising

from a moorland peat catchment on coagulation with iron and aluminium salts.

Environmental Pollution 140 (3): 436–443 DOI: 10.1016/j.envpol.2005.08.001

Šidàk Z. 1967. Rectangular confidence region for the means of multivariate normal

distributions. Journal of the American Sta- tistical Association 62: 626–633

Sobek S, Durisch-Kaiser E, Zurbrügg R, Wongfun N, Wessels M, Pasche N, Wehrli B. 2009.

Organic carbon burial efficiency in lake sediments controlled by oxygen exposure time

and sediment source. Limnology and Oceanography 54 (6): 2243–2254 DOI:

10.4319/lo.2009.54.6.2243

Stimson AG, Allott TEH, Boult S, Evans MG. 2017a. Reservoirs as hotspots of fluvial

carbon cycling in peatland catchments. Science of The Total Environment 580: 398–411

DOI: 10.1016/j.scitotenv.2016.11.193

Stimson AG, Allott TEH, Boult S, Evans MG, Pilkington M, Holland N. 2017b. Water

quality impacts of bare peat revegetation with lime and fertiliser application. Applied

36

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722


Geochemistry 85: 97–105 DOI: 10.1016/j.apgeochem.2017.09.003

Tallis JH. 1997. The southern Pennine experience: an overview of blanket mire degradation.

In Blanket Mire Degradation: Causes, Concequ Ences and Challenges, Tallis, J.H.,

Meade, R., Hulme PD (ed.).Macaulay Land Use Research Institute, Aberdeen; 7–16.

Tiwari T, Laudon H, Beven K, Ågren AM. 2014. Downstream changes in DOC: Inferring

contributions in the face of model uncertainties. Water Resources Research 50 (1): 514–

525 DOI: 10.1002/2013WR014275

Van der Wal R, Bonn A, Monteith D, Reed M, Blackstock K, Hanley N, Thompson D,

Alonso I, Allott T, Armitage H, et al. 2011. Mountains , Moorlands and Heaths. In UK

National Ecosystem Assessment. The UK National Ecosystem Assessment Technical

Report. UNEP-WCMNCambridge; 105–159.

Wallage ZE, Holden J, Mcdonald AT. 2006. Drain blocking : An effective treatment for

reducing dissolved organic carbon loss and water discolouration in a drained peatland.

367: 811–821 DOI: 10.1016/j.scitotenv.2006.02.010

Wallin MB, Grabs T, Buffam I, Laudon H, Agren Å, Öquist MG, Bishop K. 2013. Evasion of

CO2 from streams - the dominant component of the carbon export through the aquatic

conduit in a boreal landscape. Global change biology 19 (3): 785–97 DOI:

10.1111/gcb.12083

Ward ND, Keil RG, Medeiros PM, Brito DC, Cunha AC, Dittmar T, Yager PL, Krusche A V,

Richey JE. 2013. Degradation of terrestrially derived macromolecules in the Amazon

River. Nature Geoscience 6 (7): 530–533 DOI: 10.1038/ngeo1817

Wei Q, Wang D, Wei Q, Qiao C, Shi B, Tang H. 2008. Size and resin fractionations of

dissolved organic matter and trihalomethane precursors from four typical source waters

in China. Environmental Monitoring and Assessment 141: 347–357 DOI:

10.1007/s10661-007-9901-1

37

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747


Weishaar JL, Fram MS, Fujii R, Mopper K. 2003. Evaluation of Specific Ultraviolet

Absorbance as an Indicator of the Chemical Composition and Reactivity of Dissolved

Organic Carbon. Environmental Science & Technology 37 (20): 4702–4708.

Worrall F, Burt TP. 2009. Changes in DOC treatability : Indications of compositional

changes in DOC trends. Journal of Hydrology 366 (1–4): 1–8 DOI:

10.1016/j.jhydrol.2008.11.044

Worrall F, Moody CS. 2014. Modeling the rate of turnover of DOC and particulate organic

carbon in a UK, peat-hosted stream: Including diurnal cycling in short-residence time

systems. Journal of Geophysical Research: Biogeosciences: 1–13 DOI:

10.1002/2014JG002671. Worrall F, Burt TP, Adamson J. 2006. The rate of and controls

upon DOC loss in a peat catchment. Journal of Hydrology 321: 311–325 DOI:

10.1016/j.jhydrol.2005.08.019

Worrall F, Harriman R, Evans CD, Watts CD, Adamson J, Neal C, Tipping E, Burt T, Grieve

I, Monteith D, et al. 2004. Trends in dissolved organic carbon in UK rivers and lakes.

Biogeochemistry 70(3), 369-402 DOI: 10.1007/s10533-004-8131-7

Yu ZC. 2012. Northern peatland carbon stocks and dynamics : a review. Biogeosciences 9:

4071–4085 DOI: 10.5194/bg-9-4071-2012

38

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766


Appendices

Table A-I: Location details for sampling sites used in this study (note some have two names).

Site ID Description UK National Grid Reference

K1/B Weir SK 08706 88571K2 Junction with River Kinder SK 08837 88619K3/KH Above Kinder Downfall waterfall SK 08341 88967K4 Below Kinder Downfall waterfall SK 08152 88838

K5 Just below Kinder River / Red Brook confluence SK 07512 88398

K6 Half way between K5 and K7 SK 07001 88538K7/KR Reservoir inlet SK 06442 88326WC Reservoir inlet SK 05953 88684BC Reservoir inlet SK 06121 87832KRO Reservoir outlet SK 05389 88005VH Reservoir outlet SK 05474 88089TW Treatment Works SK 96053 85579

39

767

768769770

771772


Table A-II: Summary statistics for carbon and colour ratio data.

Determinand Value

Site

River start(B)

River end(KR)

Lake ina

(WC, KR, BC)

Lake out(VH, KRO)

Pipe start(VH)

Pipe end(TW)

Number 10 10 12 12 12 12

DOC(mg/l)

Max 30 8 7 8 7 7Min 14 2 2 4 4 3Mean 20 4 4 6 6 6St dev 6 2 2 1 1 1

E4:E6

Number 9 9 12 12 12 12Max 15 7 7 8 8 11Min 9 2 1 3 2 3Mean 11 4 4 5 5 6St dev 2 2 2 1 1 2

SUVA254

(L mg-1 m-1)

Number 10 10 12 12 12 12Max 5 4 3 5 5 8Min 3 1 2 2 2 2Mean 4 3 2 4 4 4St dev 1 1 1 1 1 1

Abs400:DOC(L mg-1 m-1)

Number 10 10 12 12 12 12Max 0.9 0.6 0.5 0.8 0.8 1.3Min 0.3 0.2 0.2 0.4 0.4 0.3Mean 0.6 0.5 0.4 0.6 0.6 0.6St dev 0.2 0.1 0.1 0.1 0.1 0.2

aflow weighted

40

773

774775776777


Table A-III: Results of Shapiro-Wilk test to assess normality of the in catchment changes in water quality indices summarised in Table II. P >0.05 indicates a potentially non normal dataset.

∆ Location P values for determinand (Shapiro Wilk normality test)

DOC E4:E6 SUVA254 Abs400:DOC

River <0.010a 0.07 0.569 0.641Reservoir 0.423 0.315 0.147 0.498Pipe 0.828 0.563 0.796 0.472

B-KH 0.848 0.863 0.944 0.747B-KR 0.431 0.578 0.064 0.709

Abs400 E4:E6 pH

K1-K2 0.567 0.785 0.146K2-K3 0.396 0.438 0.015K3-K4 0.348 0.558 0.453K4-K5 0.726 0.356 0.332K5-K6 0.482 0.741 0.49K6-K7 0.497 0.99aValue for concentration adjusted DOC is 0.044

Table A-IV: Percentage of truly dissolved OC(<10 kDa) in carbon below 1000 kDa in size.

Site Date13/05/13 01/07/13 04/11/13

KH 11% 13% 11%KRin 13% 24% 24%WCin 22% 22% 24%BCin 27% 23% 25%Allin

a 21% 24% 24%VHout 17% 20% 15%TW 30% 18% 17%aFlow weighted average of reservoir inputs

41

778779780

781

782

783

784785786

787788

789

Date post:	21-Jun-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

University of Manchester - Fluvial organic carbon … · Web viewaAn average flow weighted % modern...

Documents