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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
Keywords: Fluvial OC composition, Fluvial OC transformations, Peatlands, Water
Treatment, POC, DOC
3
42
43
Fluvial OC variability within a peatland catchment.
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 variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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
Fluvial OC variability within a peatland catchment.
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