ArticlePDF Available

Abstract

Dissolved organic matter (DOM) is an important constituent of freshwater that participates in a number of key ecological and biogeochemical processes but can be problematic during water treatment. Thus, the demand for rapid and reliable monitoring is growing, and spectroscopic methods are potentially useful. A model with 3 components—2 that absorb in the ultraviolet (UV) range and are present at variable concentrations and a third that does not absorb light and is present at a low constant concentration—was previously found to yield reliable predictions of dissolved organic carbon concentration [DOC]. The model underestimated [DOC] in shallow eutrophic lakes in the Yangtze Basin, China, however, raising the possibility that DOM derived from algae might be poorly estimated, an idea supported by new data reported here for eutrophic British lakes. We estimated the extinction coefficients in the UV range of algae-derived DOM from published data on algal cultures and from new data from outdoor mesocosm experiments in which high concentrations of DOC were generated under conditions comparable to those in eutrophic freshwaters. The results demonstrate the weak UV absorbance of DOM from algae compared to DOM from terrestrial sources. A modified model, in which the third component represents algae-derived DOM present at variable concentrations, allowed contributions of such DOM to be estimated by combining the spectroscopic data with [DOC] measured by laboratory combustion. Estimated concentrations of algae-derived DOC in 77 surface freshwater samples ranged from 0 to 8.6 mg L⁻¹, and the fraction of algae-derived DOM ranged from 0% to 100%.
Second revision submitted to Inland Waters September 2017
The contribution of algae to freshwater dissolved organic matter:
implications for UV spectroscopic analysis
Jessica L. Adams*, Edward Tipping, Heidrun Feuchtmayr, Heather T. Carter, Patrick Keenan
Centre for Ecology & Hydrology, Lancaster Environment Centre, Lancaster, LA1 4AP, UK
*Corresponding author: Jessica L. Adams jesams@ceh.ac.uk
1
1
2
3
4
5
6
7
8
9
10
ABSTRACT
Dissolved organic matter (DOM) is an important constituent of freshwater. It participates in a
number of key ecological and biogeochemical processes, and can be problematic during
water treatment. Thus, the demand for rapid and reliable monitoring is growing and
spectroscopic methods are potentially useful. A model with 3 components, 2 absorbing in the
ultraviolet (UV) range and present at variable concentrations, and a third that does not absorb
light and is present at a low constant concentration, was previously found to give good
predictions of dissolved organic carbon concentration; [DOC]. However, the model
underestimated [DOC] in shallow, eutrophic lakes in the Yangtze Basin, China, raising the
possibility that DOM derived from algae might be poorly estimated. This is supported by new
data reported here for eutrophic British lakes. We estimated the extinction coefficients, in the
UV range, of algae-derived DOM, from published data on algal cultures, and from new data
from outdoor mesocosm experiments in which high concentrations of DOC were generated
under conditions comparable to those in eutrophic freshwaters. The results demonstrate the
weak UV absorbance of DOM from algae compared to DOM from terrestrial sources. A
modified model, in which the third component represents algae-derived DOM present at
variable concentrations, allowed contributions of such DOM to be estimated by combining
the spectroscopic data with [DOC] measured by laboratory combustion. Estimated
concentrations of algae-derived DOC in 77 surface freshwater samples ranged from zero to
8.6 mg L-1, and the fraction of algae-derived DOM ranged from zero to 100%.
Key words: absorption spectroscopy, algal products, dissolved organic carbon,
eutrophication, modelling
2
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Introduction
Dissolved organic matter (DOM) is ubiquitous in surface, soil and ground waters, and chiefly
comprises partially decomposed plant and animal material (Thurman 1985). It provides a
source of energy for microbes, controls absorption of light and photochemical activity,
participates in nutrient cycling, buffers pH, sorbs metals and other organic pollutants, and
interacts with nanoparticles (Tipping 2002, Aiken et al. 2011, Tipping et al. 2016). Reactions
of DOM with chlorine during drinking water treatment produce by-products including
trihalomethanes and haloacetic acids, which are a risk to human health (Nguyen et al. 2005).
The need to monitor the quality and quantity of DOM has increased considerably in recent
years, partly because of the widespread observed increases in concentrations and fluxes of
dissolved organic carbon (DOC) in surface waters (Monteith et al. 2007), which have
implications for ecology and the costs of water treatment. The DOM produced by algae is
important in lake carbon cycling and storage (Heathcote et al. 2012) and is especially
problematic in water treatment (Nguyen et al. 2005, Henderson et al. 2008, Ly et al. 2017).
Dissolved organic matter is routinely quantified by the dissolved organic carbon
concentration [DOC], for example by infra-red detection of carbon dioxide (CO2) after
combustion. Significant correlations between optical absorbance and [DOC] mean that
approximate quantification can be achieved from UV-visible absorption spectroscopy at a
single wavelength (e.g., Grieve 1984, Moore 1987).
However, the spectroscopic properties of DOM vary temporally and spatially, a fact that is
exploited for example in the well-known use of specific ultra-violet absorbance (SUVA) as an
indicator of DOM quality (Chin et al. 1994, Weishaar et al. 2003). Such variability means that
the single wavelength approach cannot generally provide an accurate measure of [DOC].
3
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Therefore, Tipping et al. (2009) developed a 2-component model employing UV absorbance
data at 2 wavelengths, and showed that it could provide precise estimates of [DOC] in a
variety of surface water samples.
The 2-component model adopted the linear sum of the concentrations of component A
(DOCA) and component B (DOCB) representing strongly and weakly UV-absorbing material,
respectively. Further development of this modelling approach by Carter et al. (2012)
introduced a third component, ‘component C’, which represents non UV-absorbing DOC,
assumed to be present at the same concentration in all samples. The total [DOC] is then the
linear sum of [DOCA], [DOCB] and [DOCC]. Testing this 3-component model with data for
1700 river and lake samples (but few eutrophic waters) resulted in good, unbiased predictions
of [DOC] (r2 = 0.98) with fixed spectroscopic characteristics of the end members A and B,
combined with a small constant concentration of component C at 0.8 mg L-1. Because
[DOCC] was fixed, the model still only required absorbance data at 2 wavelengths. The dual
wavelength approach was therefore suggested as a means to estimate [DOC] accurately,
rapidly, and inexpensively, without the need for lengthier laboratory processing and
measurement and for in situ field monitoring.
However, for eutrophic shallow lakes of the Yangtze basin (Zhang et al. 2005), the
model underestimated [DOC] by an average factor of 2.1 (Carter et al. 2012). The average
extinction coefficient (absorbance/[DOC]) of 6.5 L g-1 cm-1 at 280 nm in these samples
suggested the presence of material that absorbs UV light more weakly than either component
A or B. Further, Zhang et al. (2005) found a positive relationship between DOM fluorescence
and the extent of eutrophication of the different Yangtze basin lakes, which indicated possible
influences from algal production. Therefore, it appears that the 3-component, dual
wavelength model may be effective only when the DOM under consideration is
predominantly terrestrial in origin. Consequently, further investigation of the optical
4
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
properties of algae-derived DOM, and how they affect the performance of the model, is
necessary.
UV spectroscopic data for DOM derived from different algal species grown in
laboratory cultures have been reported by Nguyen et al. (2005) who worked with axenic
(sterilised) cultures, and by Henderson et al. (2007) who worked with non-axenic cultures.
Nguyen et al. (2005) reported that the DOM produced comprised labile carbohydrates and
proteins with low SUVA values compared to those of terrestrially-sourced DOM. Henderson
et al. (2007) also found the DOM to absorb UV light weakly. De Haan and De Boer (1987)
concluded, from field observations of [DOC] and UV absorbance of the humic lake
Tjeukemeer, that water entering from the neighbouring eutrophic lake Ijsselmeer brought
weakly UV-absorbing DOM. Osburn et al. (2011) studied saline waters of the prairie lakes
region of the USA, which were rich in DOM of autochthonous (i.e., algal) origin, created by
bacterial processing of primary production, and reported optical absorption at 350 nm. Their
values were appreciably lower than those commonly observed for waters with comparable
[DOC] but with terrestrial sources of DOM (Carter et al. 2012). The results of these different
studies are consistent in suggesting that algae-derived DOM absorbs UV light weakly
compared to DOM from terrestrial sources.
Although these laboratory and field observations suggest that DOM derived from
algae has different absorption characteristics from terrestrially sourced material, they do not
permit a general quantitative assignment of spectroscopic parameters. We added to the data
from algal cultures reported by Nguyen et al. (2005) and Henderson et al. (2007) by making
new measurements on DOM generated by algae growing in outdoor mesocosms, under
conditions arguably more realistic than those in the cultures. Then we evaluated these
combined data to quantify UV absorption at different wavelengths, by deriving representative
extinction coefficients, for algae-derived DOM.
5
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
The new absorption parameters were then used to analyse the data for a new
freshwater sample set, biased towards eutrophic water bodies, to estimate concentrations of
algae-derived DOM and the fraction of total [DOC] that they account for. By this means, we
aimed to quantify the contribution of algae-derived DOM to freshwater [DOC], and to UV
absorbance, in order to (1) evaluate how the presence of such DOM in water samples would
affect estimation of [DOC] by UV spectroscopy, and (2) provide a means to quantify DOM
from different sources (the terrestrial system and algae) in rivers and lakes.
6
108
109
110
111
112
113
114
115
Study Site
Surface water samples representative of different states of eutrophication (defined by [Chl-a])
and DOM source were collected from catchments in the North of England during the summer
and autumn of 2014 and 2015 (Table 1, Tables S1a and S1b). The Shropshire – Cheshire
meres are situated in the North-West Midland outwash plains and drain predominantly small
agricultural, urban, and parkland catchments (Reynolds 1979, Moss et al. 2005). Fisher et al.
(2009) reported a range of 2–68 μg L-1 for average chlorophyll a concentration, [Chl-a],
across the Shropshire – Cheshire meres region (Table S1a). Ten of the samples were from
small lakes in the Lake District National Park and 4 were from reservoirs in West Yorkshire,
all of which drain upland moorland. Ten further sites included small farm ponds in the Fylde
area of Lancashire and rivers and small streams draining lowland arable farmland and urban
areas in Yorkshire.
Methods
Application of the 3 component model of Carter et al. (2012)
The measure of optical properties used here is the extinction coefficient of the sample (E),
also known as specific absorbance, which is the ratio of the absorbance at a given wavelength
to [DOC] with units L g-1 cm-1 (Tipping et al. 2009). The basis of the model of Carter et al.
(2012) is that the DOM that absorbs UV light can be represented as a mixture of 2
components, A and B, each with a defined UV spectrum. The fraction of component A (fA) is
given by
f
A
=
E
B, 1λ
- R E
B, 2λ
R ( E
A, 2λ
- E
B, 2λ
) + ( E
B, 1λ
E
A, 1λ
)
(1)
7
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
where EA and EB are the extinction coefficients of components A and B at 2 given
wavelengths (λ1 and λ2) and R is the measured ratio of absorbance at the same 2
wavelengths. The value of fA can then be substituted into the following equation to obtain the
extinction coefficient for the sample being measured
EAB, λ = fA EA, λ + fB EB, λ = fA EA, λ + (1 – fA) EB, λ (2)
where EAB, λ is the extinction coefficient of the sample at either of the 2 chosen wavelengths
and fA and fB are the fractions of components A and B (fA + fB = 1).
To calculate the total UV-absorbing [DOC], the measured absorbance at either of the
wavelengths is divided by EAB, λ from equation (2), and the total (absorbing + non-absorbing)
[DOC] is obtained by adding a constant [DOCC] representing a small amount of non-
absorbing DOM present at the same concentration (0.8 mg L-1) in all water samples
(3)
Where the choice of wavelengths for the calculation is flexible, as long as they differ
sufficiently (by about 50 nm or more). Carter et al. (2012) reported extinction coefficients for
a number of wavelengths in the range 254 – 355 nm, and used various combinations to
analyse published data. The model is best-applied to filtered samples (as used in the present
work) and is assumed to apply to all freshwaters irrespective of pH or ionic composition.
Henceforth, we refer to the 3 component model with fixed [DOCC] as the Carter model.
Mesocosm experiments
The mesocosms are part of the CEH aquatic mesocosm facility (CAMF);
http://www.ceh.ac.uk/our-science/research-facility/aquatic-mesocosm-facility, accessed
January 2017. The facility contains 32 mesocosms, each of 2 metre diameter and 1 metre
8
[ D OC] = + [ D OC
C
]
E
AB,λ
A
λ
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
depth, simulating shallow lakes. Of the 32 mesocosms used for a multiple stressor
experiment, 4 were selected (mesocosms 4, 7, 15 and 20) to obtain a range of Chl-a
concentrations. In the stressor experiment, the mesocosms were subjected to different
treatments, including heating (40C above ambient) and the addition of nutrients free from
nitrogen or phosphorus. Mesocosms 4 and 20 were both unheated, with an average ambient
water temperature of 14.60C over the sampling period, and with intermittent nutrient addition.
Mesocosm 7 was heated with intermittent nutrient addition, and mesocosm 15 was heated
without intermittent nutrient addition. Sampling took place on 7 occasions between February
and August 2015. The dominant algal classes for each of the four mesocosms were
Chlorophyceae and Cyanophyceae, with a bloom of Euglena in mesocosm 7 in the early
summer. For our analyses, a 500 mL sample was collected from the four mesocosms in pre-
rinsed vessels.
We assumed that the DOM produced in the mesocosms during the observation period
resulted from the fixation of atmospheric CO2 by algae and its subsequent release in DOM.
Although some allochthonous sources could influence the mesocosm DOM, these can be
disregarded for the following reasons: (1) The simulation experiments commenced in 2013,
when sediment from a natural lake was added to the mesocosms, and therefore there has been
enough time for DOM in the water column to come to equilibrium with the sediment, (2) An
increase in pH could provide a mechanism for releasing DOM from sediment (Tipping 2002),
but during our observation period there were no systematic changes in pH, and thus it is
reasonable to assume that net DOM release did not occur, and (3) Addition of allochthonous
DOM to the mesocosms may have occurred through rainfall, but rainwater [DOC] is typically
low, around 0.6 mg L-1 for parts of the UK (Wilkinson et al. 1997) and < 2 mg L-1 globally
(Willey et al. 2000); quite insufficient to generate the large observed increases in [DOC].
9
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
Laboratory analyses
All samples were processed within 3 days of collection. Owing to the fact that the mesocosms
were primarily used for a separate study, there were minor methodological differences
between the analyses of the field and mesocosm samples. The determination of algal [Chl-a]
in field samples followed the method of Maberly et al. (2002). A known volume of the
sample was filtered through a Whatman GF/F (0.7 µm) filter paper, which was then
immediately submerged in 10 mL of industrial methylated spirit (IMS, 96% ethanol, 4%
methanol) and left overnight, in the dark at 4oC. The mesocosm samples were analysed
similarly for [Chl-a], but using a Whatman GF/C (1.2 µm) filter paper, which was submerged
in 96 % ethanol. The 2 different extraction solvents (IMS and 96% ethanol) are known to be
equally efficient (Jespersen and Christoffersen 1987). Following centrifugation at 4500 rpm,
optical absorbance readings at 665 and 750 nm were used to calculate [Chl-a], following
Marker et al. (1980). The mesocosm samples collected on 12 August 2015, were analysed for
[Chl-a] in situ using an AlgaeTorch (bbe Moldaenke, Germany), which had been calibrated
against [Chl-a] data obtained by ethanol extraction for all 32 mesocosms over the preceding
8-month period, yielding a regression with R2=0.67 (n=442, p<0.0001). Field samples were
analysed for pH and conductivity using a glass electrode with a Radiometer instrument and a
Jenway 4510 probe respectively, each instrument being calibrated for each set of samples.
For the mesocosm experiment, pH and conductivity were measured in situ, using a Hydrolab
DS5X multiparameter data sonde (OTT Hydromet), except that for samples collected on 12
August 2015 and 26 September 2015, pH and conductivity were measured using an EXO2
multiparameter data sonde (Exowater). Both multiparameter sondes were calibrated in the
laboratory before sampling the mesocosms.
All samples for absorbance spectroscopy and the determination of [DOC] were
analysed by the same procedure. A 125 mL sub-sample was filtered using a Whatman GF/F
10
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
(0.7 µm) filter. A 3 mL filtered sample was measured for absorbance in the UV-Vis range
(200 nm – 900 nm) using an Agilent 8453 diode array spectrophotometer with a 1 cm path
length quartz cuvette. Prior to each sample batch, measurements were made on a blank using
Milli-Q water, and used to correct the spectra of the samples. A 10 mg L-1 solution of
naphthoic acid was used as a quality control. Absorbance values at 270 nm, 350 nm and 700
nm were selected for [DOC] calculation with the model of Carter et al. (2012). Values of A270
and A350 for the calculations were obtained by subtracting A700 (near zero) from the raw values
to correct for instrument drift; it also corrects for suspended matter in unfiltered samples,
although these were not used in the present work. The remaining sample was acidified with 3
M hydrochloric acid and purged with zero grade air for 4 minutes to remove any inorganic
carbon. The sample was then combusted at 905oC with cobalt chromium and cerium oxide
catalysts, which converts all the remaining carbon to CO2. The CO2 was measured for [DOC]
through infra-red detection using a Skalar Formacs CA16 analyser.
Mathematical apportionment of DOM forms
The procedure to apportion 3 DOM forms (A, B and C, or A, B and C2) from measured
values of UV absorbance and [DOC] was as follows. Note that here we assume that
component C (no absorbance) or C2 (absorbance characteristics from Table 2) is present at a
variable concentration, and so the description differs from the Carter model which has fixed
[DOCC]. For simplicity, the following description is only in terms of A, B and C. The total
absorbance at a given wavelength is given by the linear sum of the absorbances of the 3
components
A = AA + AB + AC, (4)
and can be expressed in terms of the total DOC concentration, the fraction of each component
in the mixture (fA, fB, fC), and their extinction coefficients (EA, EB, EC)
11
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
A = [DOC] {fAEA + fBEB + fC2EC}. (5)
If A and [DOC] are known from measurement, then since fA, fB and fC2 must total unity,
equation (5) has 2 unknowns (e.g., fA and fB), and to calculate them it is necessary to have
measured values of A for 2 different, sufficiently separated, wavelengths (1 and 2). Since
the measurements cannot be error-free, the values of fA and fB cannot be calculated by
solution of simultaneous equations, and instead were estimated by minimisation of squared
residuals in observed and calculated A1 and A2. Calculated values (A1,calc and A2,calc) were
obtained from equation (5) for trial values of fA and fB, and fC by difference (1 - fA - fB). The
residuals are
r1 = A1,calc - A1,meas, (6)
r2 = A2,calc - A2,meas, (7)
where A1,meas and A2,meas are the measured absorbances at the 2 wavelengths. The sum of the
squared residuals (r12 + r22) was minimised by iterative improvement of the trial values of fA
and fB, to give the best fit of the data. Values of [DOCA], [DOCB] and [DOCC] were obtained
from the products of [DOC] with the derived fA, fB and fC respectively.
Statistics and minimisation
Calculations of standard deviations, t-tests, and regression analyses were carried out using
Microsoft Excel. The Solver function in Microsoft Excel was used to perform minimisations
in the apportionment calculations.
12
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
Results
Estimating extinction coefficients for DOM derived from freshwater algae
The 4 selected mesocosms represent enclosed systems where allochthonous inputs are
negligible. They therefore simulate conditions where the dominant DOM component is
derived from algae, but may be modified by subsequent microbial processing. Measured and
modelled [DOC], absorbance data, and [DOC] estimated with the Carter model, are shown in
Fig. 1 (see also Table S1c). Absorbance at 270 nm and 350 nm increased slightly through
time. The modelled [DOC] also increased slightly, but at a considerably lower rate than the
measured [DOC], which rose from 8.2 mg L-1 to 63.4 mg L-1 in mesocosm 4. The same
pattern was also seen in the mesocosms with lower [DOC] such as mesocosm 15, where
[DOC] increased from 4.5 mg L-1 to 14.1 mg L-1. Extinction coefficients derived from the
absorbance and [DOC] results of Fig. 1 decline with [DOC] for both wavelengths (Fig. 2).
There was a significant positive relationship (p < 0.001) between measured [DOC] and [Chl-
a] for the mesocosm samples (Table S2). The average pH for the mesocosms was 9.7 and
there was no significant relationship observed between measured [DOC] and pH.
The extinction coefficients of the additional DOM produced were estimated by
considering the changes in [DOC] and optical absorbance in the mesocosms during the
sampling period. First, the increase in [DOC] was calculated for each of the mesocosms by
finding the differences between the first data point and each of the last 4. Then, the same was
done for the absorbance values at 270 nm and 350 nm, and also for 254 nm, 280 nm and 355
nm to permit comparison with results from other studies. Extinction coefficients were
calculated as the averages of the ratios of the absorbance and [DOC] increases during algal
growth. Similar results were obtained for the different mesocosms, yielding reasonably well-
defined extinction coefficients, which are considerably lower than those estimated by Carter
et al. (2012) for terrestrially-derived freshwater DOM (Table 2). We also calculated extinction
13
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
coefficients at 254 nm of DOM produced in laboratory cultures from the results of Nguyen et
al. (2005) and Henderson et al. (2008). The results of these 2 studies showed only minor
differences in the E254 values of DOM from different algal species.
The average E254 for DOM produced in the mesocosms does not differ significantly (t-
test, p>0.05) from the value for DOM in the non-axenic cultures (Henderson et al. 2008).
Although it is significantly (t-test, p<0.05) greater than the value for DOM in the axenic
cultures (Nguyen et al 2005), the difference is modest.
Therefore, the results suggest that the UV absorption properties of DOM derived from
freshwater algae can reasonably be represented by a single set of extinction coefficients; there
is no evidence that different algal species, or collections of species, produce greatly different
types of DOM, at least with respect to their UV spectra. For further modelling analysis (see
below), we used the average extinction coefficients derived from the mesocosm data.
Natural water samples
Samples collected from the field sites had a wide range of [DOC], from 1.7 mg L-1 in a soft
water lake to 63.5 mg L-1 in a peat dominated lake. Overall, the Carter model predicted
[DOC] reasonably well (Fig. 3), with an average modelled:measured ratio of 0.96. However,
model predictions for seven sites were too low (average modelled:measured ratio = 0.70) and
these were all situated in the Shropshire-Cheshire meres region, which features eutrophic
lakes. In our judgement, the results from these 7 sites cannot be satisfactorily explained by
the Carter model. Combining the data from all of the Shropshire - Cheshire meres sites with
the Yangtze Basin samples (Zhang et al. 2005) shows that the Carter model fails with
eutrophic lakes, especially for samples with relatively low [DOC] (Fig. 4).
Spectroscopic modelling with 3 variable components
14
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
The underestimation of [DOC] in samples from eutrophic lakes suggests the presence of
DOM that absorbs weakly in comparison to the terrestrially-derived components A and B,
and is present at concentrations greater than the fixed value of 0.8 mg L-1 for component C
assumed in the Carter model. Clearly, DOM derived from algae is a likely explanation for
this DOM, and so we analysed the data for the natural water samples by assuming the DOM
to comprise variable amounts of components A, B and algae-derived DOM, which we refer to
as component C2 and which has the extinction coefficients (Table 2) derived as described
above. In this application, the model was not used to estimate [DOC]; instead, we combined
the measured [DOC] value with spectroscopic data to estimate the fractions of components A,
B and C2 in each sample (see Methods). For the new data reported here, we used
wavelengths of 270 nm and 350 nm, while for the Yangtze basin samples (Zhang et al. 2005)
the wavelengths were 280 nm and 355 nm (Table 2). Errors in the modelled values of fA, fB
and fC2 were estimated (Table S3) using representative errors in the input values (measured
UV absorbance and [DOC]) and errors in the extinction coefficients for algae-derived DOM
(Table 2). The errors in fA, fB and fC2 were modest, the largest (average 0.03) being due to
uncertainty in [DOC], the next largest (average 0.009) to extinction coefficient errors, and the
smallest (average 0.003) to errors in measured absorbance.
The results indicate that algae-derived DOM is most prevalent in the eutrophic
Yangtze basin (YB) lakes with a mean [DOCC2] of 4.9 mg L-1, and all fC2 values greater than
0.66 (mean = 0.87; Table 3, Fig. 5, Table S4). Of the UK sites, the Shropshire-Cheshire meres
(SCM) have the highest amounts of algae-derived DOM; the mean concentration of 3.6 mg L-
1 for [DOCC2] was appreciably greater than the Carter model fixed [DOCC] value of 0.8 mg L-
1, and this explains why the Carter model predicts [DOC] poorly in some of the samples.
However, it remains the case that in only 4 of the 21 SCM samples did fC2 exceed 0.5,
indicating that the majority of the DOM was from algae. Therefore in most instances the
15
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
catchment was the main supplier of DOM to the SCM lakes. For the remaining UK site
categories of Table 1 (LD, PR, YR) the mean values of [DOCC2] were in the range 0 to 1.0 mg
L-1, with an overall mean of 0.7 mg L-1. This is very similar to the fixed value of [DOCC] of
0.8 mg L-1 (equation 4), which implies that if these samples contain algae-derived DOM then
it is present at sufficiently low concentrations to be accounted for by the fixed component C
of the Carter model.
The possible dependence of the derived [DOCC2] values on measured [Chl-a] was
examined by regression analysis for the samples collected and analysed in the present study
(Table S5). There was no relationship when all data were analysed together. However, if data
for the 5 site categories of Table 1 were analysed separately, there was a positive relationship
in each case, although only for LD (n = 10, r2 = 0.46) and FP (n = 5, r2 = 0.73) were the
relationships significant (p<0.05). Zhang et al. (2005) did not report [Chl-a], and so we
compared our estimated [DOCC2] values for the YR sites with total phosphorus
concentrations; again there was a positive but not significant (p>0.05) relationship.
For comparison, we also performed the apportionment calculations with the non-
absorbing component C as the third variable, that is, we found fA, fB and fC, together with
[DOCA], [DOCB] and [DOCC]. Note that this is different from the Carter model, where
[DOCC] is a constant. The results did not differ greatly from those obtained with C2 (Table
S4) and in linear regression there was a strong correlation between the estimates of [DOCC]
and [DOCC2] (R2 = 0.99, p < 0.001, n = 77); on average, the calculated values of [DOCC]
were 80% of those of [DOCC2].
For completeness, we examined whether the assumption of a fixed concentration of
DOCC2, instead of DOCC, affected application of the Carter model to data from 426 UK
surface water samples previously used by Carter et al. (2012) to derive model parameters.
This was done by re-optimisation of the parameters, assuming the weakly UV absorbing
16
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
component C2, rather than the non-absorbing C, to be present at a fixed concentration; in
other words we attributed all DOM not accounted for by components A and B to algae-
derived DOM. The derived parameters using component C2 were almost the same as the
original values; the new fitted extinction coefficients for components A and B differed by less
than 0.5% from the original ones, and the fixed concentration of C2 was greater by only 0.06
mg L-1 than the original fixed concentration of component C.
17
361
362
363
364
365
366
367
Discussion
The mesocosm experiments provided a valuable simulation of a eutrophic shallow lake
system, and as explained in Methods it could reasonably be assumed that the DOM produced
during the observation period resulted from the fixation of atmospheric CO2 by algae and its
subsequent release in DOM. The assumption is further supported by the highly significant
relationship (P<0.001) between [DOC] and [Chl-a] obtained for the mesocosms (Table S2).
In the mesocosms, the relationship is likely strengthened by both the high [Chl-a] and the
lack of flushing, so that the production of DOM (Fig. 1) follows the change in algal biomass
fairly closely. This is less likely in the field sites, where the relationship may be confounded
by the time gap between the formation of Chl-a by primary production and the subsequent
conversion of algal biomass to DOM, together with variations in flushing rates within and
between the natural waters. Therefore, although we found that modelled [DOCC2] showed
positive relationships with [Chl-a] or total [P] (Table S5), the relationships were not strong,
and only significant (P<0.05) in 2 cases with rather few numbers of samples. Nonetheless, the
results overall show that modelled [DOC] generally deviates from the measured value in field
waters classified as eutrophic, as judged by their generally relatively high [Chl-a] values.
This supports the assumption that [DOC] not explained by the Carter model is associated
with algae.
The extinction coefficient at 254 nm for DOM derived from algae in the mesocosm
experiments (Table 2) is similar in magnitude to the averages of the values for a range of
algal species that can be calculated from data reported by Nguyen et al. (2005) and
Henderson et al. (2008). We therefore can assume that the UV absorption properties of the
mesocosm material are generally representative of algae-derived DOM. The similarity holds
for both axenic (Nguyen et al. 2005) and non-axenic (Henderson et al. 2008; our mesocosms)
conditions, implying that although bacterial processing of the DOM may affect its
18
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
composition (Rochelle-Newall et al. 2004) this does not significantly alter its UV spectrum.
The UV absorption characteristics of DOM derived from freshwater algae can be compared
to those of open ocean DOM, which is largely algal-derived (Biddanda and Benner 1997, Jiao
et al. 2010). We estimated UV extinction coefficients for marine DOM from the Mid-Atlantic
Bight region by combining absorbance data (Helms et al. 2008) with a measured [DOC] of
0.9 mg L-1 (Guo et al. 1995). We obtained values at 270 nm and 350 nm of 6.4 L g cm-1 and
1.0 L g cm-1 respectively, which are similar to the freshwater values of Table 2. The much
lower extinction coefficients of DOM derived from algae, compared to those for terrestrially-
sourced DOM (components A and B; Table 2) must reflect the paucity of conjugated or
aromatic moieties in algal biomass; in particular algae lack the lignin phenols that account for
the spectra of terrestrial DOM (Del Vecchio and Blough 2004).
We focused here on eutrophic waters in which algae-derived DOM was expected to be
present. In this context it was justified to replace component C in the Carter model by
component C2, which has the UV absorption characteristics of algae-derived material; this is
equivalent to assuming that all the DOM not attributable to components A and B was algal in
origin. Then the contributions of algae-derived DOM in the different waters could be
estimated by optimising the values of fA, fB and fC2 (Table 3, Fig. 5). This approach provides
the best estimates of [DOCA], [DOCB] and [DOCC2] for the present samples, and demonstrates
that C2 can be the dominant component, particularly in the Yangtze basin lakes (Zhang et al.
2005), total [DOC] values of which were poorly predicted by the Carter model. More extreme
examples of freshwaters in which autochthonous sources dominate the DOM are the 27
saline, generally eutrophic, prairie lakes of the U.S.A. Great Plains, studied by Osburn et al.
(2011). These had [DOC] in the range 13 to 330 mg L-1 (median 28 mg L-1), and the mean
whole-sample extinction coefficient at 350 nm was 1.5 (SD 1.1) L gDOC-1 cm-1, in fair
agreement with our value for algae-derived DOM (Table 2).
19
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
Another circumstance in which significant amounts of weakly-absorbing DOM occur
was reported by Pereira et al. (2014), who found that headwater streams of tropical
rainforests in Guyana contained between 4.1% and 89% optically “invisible” DOM following
rainfall events, the likely sources of the material being fresh leaf litter and/or topsoil. The
“invisible” DOM was taken to be the difference between DOM measured by combustion and
that estimated with the Carter model. It may be that the material identified by Pereira et al.
(2014) was not truly invisible, that is, completely lacking in chromophores; rather it may
have been weakly-absorbing, as for algae-derived DOM. It is unlikely that the DOM from
these terrestrial sources is the same as the algae-derived DOM of Table 2, and so it would
have different extinction coefficients. However, because the algae-derived and tropical
headwater DOM both have low UV extinction coefficients, then should they occur together
there would be little prospect of distinguishing them, especially against a “background” of A
and B. For the same reason, when we assumed algae-derived DOM to be the same as
component C (i.e., non-UV-absorbing), the estimates of [DOCC] were quite similar to (on
average 80% of) the estimates of [DOCC2] (Table S4).
Implications for UV spectroscopic analysis
Apportionment of DOM forms using measured [DOC]: The approach used in the present
work allowed the contribution of algae-derived DOM to the total to be estimated, using
combustion-measured [DOC] as an input to the calculation, and with the extinction
coefficients estimated from the mesocosm results. This type of analysis could be useful in
biogeochemical and ecosystem studies of eutrophic freshwaters. It could also benefit the
characterisation of DOM in water undergoing treatment for supply, bearing in mind the
difficulty of treating algae-derived DOM (see Introduction). If the absorption characteristics
of the non-A, non-B material could be determined or assumed, the analysis method could be
20
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
used in other circumstances. For example, it might be applied to the tropical headwaters
studied by Pereira et al. (2014); as noted above, Pereira et al. (2014) assumed it to be non-
absorbing.
Continued use of Carter model: The samples used by Carter et al. (2012) to obtain [DOC]
and absorbance data to construct their model were representative only of temperate
freshwaters with mainly allochthonous DOM, formed in terrestrial ecosystems and leached
into water courses. It remains the case that for such waters the Carter model is likely to be an
accurate and rapid means of both estimating total [DOC] and obtaining information about the
division of the DOM between components A and B. For such waters, the assumption of a
small amount of component C works satisfactorily, and we showed here that even if a fixed
concentration of component C2 were substituted for component C the results would hardly
differ. Periodic checking against [DOC] measured by combustion would of course be
necessary. The Carter model has considerable potential for use in continuous monitoring,
although it would not reveal unexpected excursions from ambient conditions.
Derivation of a “universal” model: The outstanding question is whether the present
findings can be exploited to make a “universal” model that would permit [DOC] to be
estimated in most or many freshwaters. The logical extension of the Carter model would be to
replace the fixed invisible component C by a variable component with a defined UV
absorbance spectrum, representative of different contributors, including algae-derived DOM
and the DOM in tropical headwaters. As discussed above, incorporation of more than one
weakly-absorbing component is unlikely to be feasible. To extract concentrations of 3
components would require data for 3 wavelengths at least. As well as fitting the data to 3
components in a mixing model, information might also be obtained from the spectral slope,
following Fichot and Benner (2011); these workers showed, for estuarine water samples, a
monotonic relationship between specific absorption (equivalent to extinction coefficient) and
21
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
the spectral slope in the range 275 to 295 nm. The use of derivative spectra may also prove
helpful (Causse et al. 2017). To explore the feasibility of a truly generally-applicable model,
absorption and [DOC] data from as wide as possible a range of contrasting waters need to be
gathered and analysed. Experience with the Carter model suggests that a model of this type
would probably be most effective for water samples with moderate proportions of weakly-
absorbing DOM; if weakly-absorbing DOM dominates, calculated total [DOC] would likely
prove sensitive to spectral variations among its different types.
Conclusions
We have defined, for the first time to our knowledge, generally-applicable average extinction
coefficients for algae-derived DOM. The values are based on data from outdoor mesocosm
experiments in which high concentrations of algae-derived DOM were generated, supported
by literature data from axenic and non-axenic culture experiments with freshwater algae.
Combining the extinction coefficients of algae-derived DOM with extinction coefficients for
terrestrially-sourced material, and with measured [DOC], permits the apportionment of DOM
among the three components. The results show that the algae-derived DOM can account for
nearly all the DOM in some eutrophic lakes. The presence of algal DOM and of other forms
of weakly-absorbing DOM in tropical headwaters, mean that a previously developed dual
wavelength spectroscopic model, assuming 2 variable UV-absorbing components and a fixed
concentration of non-UV-absorbing DOM, cannot be applied to all waters. However, that
model remains applicable to temperate waters in which terrestrial sources account for most or
all of the DOM. A more widely-applicable spectroscopic model for freshwater DOM will
require the use of absorbance data for at least 3 wavelengths.
Acknowledgements
22
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
This research was funded by the UK Science and Technology Facilities Council CLASP
project (Grant No: ST/K00672X/1). The mesocosm experiment was part of the MARS
project, funded under the 7th EU Framework Programme (contract no. 603378). We are
grateful to J. Richardson, M. De Ville and other team members of the CEH aquatic mesocosm
facility (CAMF) for collecting samples and providing data, P. Tipping for help with field
sampling, S.C. Maberly for practical guidance, and D.T Monteith and P. Scholefield for
helpful comments on the draft manuscript. We thank C. Williams, an anonymous referee, and
the Associate Editor R.L. North for their constructive review comments and suggestions.
23
493
494
495
496
497
498
499
500
References
Aiken GR, Hsu-Kim H, Ryan JN. 2011. Influence of dissolved organic matter on the
environmental fate of metals, nanoparticles, and colloids. Environ Sci Technol.
45:3196–3201.
Biddanda B, Benner R. 1997. Carbon, nitrogen and carbohydrate fluxes during the production
of particulate and dissolved organic matter by marine phytoplankton. Limnol
Oceanogr. 42:506-518.
Carter HT, Tipping E, Koprivnjak J-F, Miller MP, Cookson B, Hamilton-Taylor J. 2012.
Freshwater DOM quantity and quality from a two-component model of UV
absorbance. Water Res. 46:4532-4542.
Causse J, Thomas O, Jung A-V, Thomas M-F. 2017. Direct DOC and nitrate determination in
water using dual pathlength and second derivative UV spectrophotometry. Water Res.
108:312-319.
Chin YP, Aiken GR, O’Loughlin E. 1994. Molecular-weight, polydispersity, and
spectroscopic properties of aquatic humic substances. Environ Sci Technol. 28:1853
1858.
De Haan H, De Boer T. 1987. Applicability of light absorbance and fluorescence as measures
of concentration and molecular size of dissolved organic carbon in humic Lake
Tjeukemeer. Water Res. 21:731-734.
Del Vecchio R, Blough NV. 2004. On the origin of the optical properties of humic substances.
Environ Sci Technol. 38:3885-3891.
Fichot CG, Benner R. 2011. A novel method to estimate DOC concentrations from CDOM
absorption coefficients in coastal waters. Geophys Res Lett. 38:L03610.
Fisher J, Barker T, James C, Clarke S. 2009. Water quality in chronically nutrient-rich lakes:
the example of the Shropshire – Cheshire meres. Freshwat Rev. 2:79-99.
24
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
Grieve IC. 1984. Determination of dissolved organic matter in streamwater using visible
spectrometry. Earth Surf Process Landf 10:75-78.
Guo L, Santschi PH, Warnken W. 1995. Dynamics of dissolved organic carbon (DOC) in
oceanic environments. Limnol Oceanogr. 40:1392-1403.
Heathcote AJ, Downing JA. 2012. Impacts of eutrophication on carbon burial in freshwater
lakes in an intensively agricultural landscape. Ecosystems. 15:60–70.
Helms JR, Stubbins A, Ritchie JD, Minor EC. 2008. Absorption spectral slopes and slope
ratios as indicators of molecular weight, source and photobleaching of chromophoric
dissolved organic matter. Limnol Oceanogr 53:955-969.
Henderson RK, Baker A, Parsons SA, Jefferson B. 2008. Characterisation of algogenic
organic matter extracted from cyanobacteria, green algae and diatoms. Water Res.
42:3435-3445.
Jespersen AM, Christoffersen K. 1987. Measurements of chlorophyll – a from phytoplankton
using ethanol as an extraction solvent. Arch Hydrobiol. 109:445-454.
Jiao N, Herndl GJ, Hansell DA, Benner R, Kattner G, Wilhelm SW, Kirchman DL,
Weinbauer MG, Luo T, Chen F, Azam F. 2010. Microbial production of recalcitrant
dissolved organic matter: long-term carbon storage in the global ocean. Nature Rev
Microbiol. 8:593-599.
Ly QV, Maqbool T, Jin Hur J. 2017. Unique characteristics of algal dissolved organic matter
and their association with membrane fouling behavior: a review. Environ Sci Poll Res.
24: 11192–11205.
Maberly SC, King L, Dent MM, Jones RI, Gibson CE. 2002. Nutrient limitation of
phytoplankton and periphyton growth in upland lakes. Freshwater Biol. 47:2136-
2152.
25
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
Marker AFH, Nusch EA, Rai H, Riemann B. 1980. The measurement of photosynthetic
pigments in freshwaters and standardization of methods: conclusions and
recommendations. Archiv fur hydrobiology beihefte ergebnisse der limnologie 14:97-
106.
Monteith DT, Stoddard JL, Evans CD, De Wit HA, Forsius M, Hohasen T, Wilander A,
Skjelvale BL, Jeffries DS, Vourenmaa J, Keller B, Kopacek J, Vesely J. 2007.
Dissolved organic carbon trends resulting from changes in atmospheric deposition
chemistry. Nature 450:537-540.
Moore TR. 1987. An assessment of a simple spectrophotometric method for the
determination of dissolved organic carbon in freshwaters. N.Z. J Mar Freshwat Res.
21:585-589.
Moss B, Barker T, Stephen D, Williams AE, Balayla DJ, Beklioglu M, Carvalho L. 2005.
Consequences of reduced nutrient loading on a lake system in a lowland catchment:
deviations from the norm? Freshwater Biol. 50:1687-1705.
Nguyen M-L, Westerhoff P, Baker L, Hu Q, Esparza-Soto M, Sommerfield M. 2005.
Characteristics and reactivity of algae produced dissolved organic carbon. J. Environ.
Eng. 131:1574-1582.
Osburn CL, Wigdahl CR, Fritz SC, Saros JE. 2011. Dissolved organic matter composition
and photoreactivity in prairie lakes of the U.S. Limnol Oceanogr. 56:2371-2390.
Pereira R, Isabella Bovolo C, Spencer RGM, Hernes PJ, Tipping E, Vieth-Hillerbrand A,
Pedentchouk N, Chappell NA, Parkin G, Wagner T. 2014. Mobilization of optically
invisible dissolved organic matter in response to rainstorm events in a tropical forest
headwater river. Geophys Res Lett. 41:1202-1208.
Reynolds CS. 1979. The limnology of the eutrophic meres of the Shropshire-Cheshire plain.
Field studies. 5:93-173
26
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
Rochelle-Newall E, Delille B, Frankignoulle M, Gattuso J-P, Jacquet S, Riebesell U,
Terbruggen A, Zondervan I. 2004. Chromophoric dissolved organic matter in
experimental mesocosms maintained under different pCO2 levels. Mar Ecol Prog Ser.
272:25-31.
Thurman EM. 1985. Organic geochemistry of natural waters. Dordrecht: Kluwer.
Tipping E. 2002. Cation binding by humic substances. Cambridge University Press:
Cambridge.
Tipping E, Corbishley HT, Koprivnjak J-F, Lapworth DJ, Miller MP, Vincent CD, Hamilton-
Taylor J. 2009. Quantification of natural DOM from UV absorption at two
wavelengths. Environ Chem. 6:472-476.
Tipping E, Boyle JF, Schillereff DN, Spears BM, Phillips G. 2016. Macronutrient processing
by temperate lakes: a dynamic model for long-term, large-scale application. Sci Tot
Environ. 572:1573–1585.
Weishaar JL, Aiken GR, Bergamaschi BR, Fram MS, Fuji R, Mopper K. 2003. Evaluation of
specific ultra violet absorbance as an indicator of the chemical composition and
reactivity of dissolved organic carbon. Environ Sci Technol. 37:4702-4708
Wilkinson J, Reynolds B, Neal C, Hill S, Neal M, Harrow M. 1997. Major, minor and trace
element composition of cloudwater and rainwater at Plynlimon. Hydrol. Earth Syst
Sc. 1:537-569.
Willey JD, Kieber RJ, Eyman MS, Brooks Avery Jr G. 2000. Rainwater dissolved organic
carbon: concentrations and global flux. Global Biogeochem Cycles. 14:139-148.
Zhang Y, Qin B, Zhang L, Zhu G, Chen W. 2005. Spectral absorption and fluorescence of
chromophoric dissolved organic matter in shallow lakes in the middle and lower reaches of
the Yangtze River. J Freshwater Ecol. 20:451-459.
27
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
Tables
Table 1. Mean values of dissolved organic carbon concentration [DOC], pH, conductivity (cond) and chlorophyll concentration [Chl- a] for the
field sites. Numbers of samples are denoted by n. Modelled refers to application of the Carter model.
Code Site category n[DOC] mg L-1 pH cond µs cm-1 [Chl-a] µg L-1
Measured modelled
SCM Shropshire-Cheshire meres 21 14.1 11.7 8.2 358 39.2
LD Lake District lakes 10 2.9 2.9 7.6 86 14.1
PR Pennine reservoirs 4 8.9 10.4 7.2 96 16.7
FP Fylde farmyard ponds 5 21.7 22.6 8.0 311 91.5
YR Lowland Yorkshire rivers 15 3.9 4.0 7.9 627 14.7
28
596
597
598
599
Table 2. Extinction coefficients (Eʎ L g DOC-1 cm-1) for dissolved organic matter (DOM)
derived from algae, and parameters from the Carter model (components A and B). Mesocosm
values were derived from data in Fig. 3, with 16 measurements at each wavelength. The value
for axenic cultures is averaged from 12 values of Nguyen et al. (2005), and that for non-
axenic cultures is from 4 values of Henderson et al. (2008). Error terms are 95% confidence
margins. All Eʎ values are significantly greater than zero (P<0.001 for E254, E270, E280; P<0.01
for E350, E355).
Source E254 E270 E280 E350 E355
Mesocosms
5.7
(±1.7)
4.9
(±1.4)
4.4
(±1.3)
1.1
(±0.5)
1.0
(±0.5)
Axenic cultures1
3.7
(±0.7)
----
Non-axenic cultures2
5.4
(±0.4)
- - - -
Model component A 77.1 69.3 63.9 30.0 27.9
Model component B 21.3 15.4 12.0 0.0 0.0
1 Average of results for Scenedesmus quadricauda, Chaetoceros mulleri, Oscillatoria
Prolifera (Nguyen et al. 2005).
2 Average of results for Chlorella vulgaris, Microcystis aeruginosa, Asterionella formosa,
Melosira sp., at stationary phase growth (Henderson et al. 2008).
29
600
601
602
603
604
605
606
607
608
609
610
611
612
613
Table 3. Measured dissolved organic carbon concentration [DOC], calculated fractions of A,
B and C2, and calculated [DOCA], [DOCB] and [DOCC2], ordered by [DOCC2]. See Table 1 for
key to the UK sites; sample details are given in Supplemental material 1. YB = Yangtze basin
SCM = Shropshire Cheshire meres, LD = Lake District lakes PR = Pennine reservoirs, FP =
Fylde farm ponds, YR = lowland Yorkshire rivers.
Sample ID [DOC]meas
mg L-1 fAfBfC2
[DOCA]
mg L-1
[DOCB]
mg L-1
[DOCC2]
mg L-1
YR3b 2.7 0.31 0.69 0.00 0.8 1.9 0.0
YR3a 2.4 0.65 0.35 0.00 1.6 0.9 0.0
PR2 8.9 0.83 0.17 0.00 7.4 1.5 0.0
PR4 8.9 0.42 0.58 0.00 3.7 5.2 0.0
FP3 28.7 0.33 0.67 0.00 9.6 19.1 0.0
PR3 9.6 0.84 0.16 0.00 8.1 1.5 0.0
PR1 8.3 0.70 0.30 0.00 5.9 2.5 0.0
SCM7a 13.8 1.00 0.00 0.00 13.8 0.0 0.0
FP5 32.4 0.67 0.33 0.00 21.8 10.6 0.0
YR2a 2.8 0.40 0.57 0.02 1.1 1.6 0.1
SCM9a 9.9 0.17 0.81 0.01 1.7 8.1 0.1
LD4 2.2 0.26 0.68 0.07 0.6 1.5 0.1
YR5a 3.5 0.36 0.54 0.10 1.2 1.9 0.3
YR4a 3.7 0.22 0.65 0.13 0.8 2.4 0.5
LD10 2.1 0.24 0.54 0.23 0.5 1.1 0.5
LD9 3.9 0.38 0.44 0.19 1.5 1.7 0.7
YR2b 2.3 0.25 0.42 0.32 0.6 1.0 0.8
LD1 3.6 0.24 0.53 0.22 0.9 1.9 0.8
LD2 1.7 0.32 0.20 0.48 0.5 0.3 0.8
LD7 1.9 0.18 0.40 0.42 0.4 0.8 0.8
LD3 2.9 0.19 0.51 0.31 0.6 1.5 0.9
YR2c 3.7 0.29 0.46 0.26 1.1 1.7 1.0
YR3c 3.7 0.28 0.43 0.29 1.0 1.6 1.0
YR4c 3.7 0.24 0.47 0.28 0.9 1.8 1.1
LD6 2.2 0.33 0.16 0.51 0.7 0.4 1.1
YR4b 3.7 0.25 0.43 0.32 0.9 1.6 1.2
YR5c 3.6 0.31 0.32 0.37 1.1 1.2 1.3
FP1 15.3 0.30 0.60 0.09 4.6 9.2 1.4
SCM6a 10.0 0.19 0.66 0.15 1.9 6.6 1.5
YR1c 6.6 0.26 0.50 0.24 1.7 3.3 1.6
YR1a 5.1 0.57 0.12 0.31 2.9 0.6 1.6
YR5b 4.7 0.23 0.43 0.34 1.1 2.0 1.6
LD8 2.8 0.19 0.22 0.59 0.5 0.6 1.6
SCM13 7.4 0.16 0.58 0.26 1.2 4.3 2.0
FP4 14.5 0.23 0.64 0.14 3.3 9.2 2.0
LD5 5.2 0.28 0.31 0.40 1.5 1.6 2.1
30
614
615
616
617
618
Table 3 (continued)
YB8 2.7 0.03 0.15 0.82 0.1 0.4 2.2
SCM8 20.1 0.22 0.64 0.14 4.4 13.0 2.8
SCM14 7.5 0.18 0.45 0.37 1.3 3.4 2.8
SCM3 7.7 0.13 0.50 0.37 1.0 3.8 2.8
SCM16a 11.3 0.11 0.64 0.25 1.2 7.2 2.9
YR1b 7.3 0.19 0.41 0.40 1.4 3.0 2.9
SCM5 10.7 0.12 0.58 0.31 1.3 6.2 3.3
YB22 4.1 0.03 0.13 0.84 0.1 0.5 3.4
YB2 4.9 0.04 0.21 0.76 0.2 1.0 3.7
YB5 4.1 0.03 0.06 0.91 0.1 0.3 3.7
SCM9b 10.9 0.13 0.54 0.34 14 5.9 3.7
FP2 17.8 0.29 0.50 0.21 5.1 9.0 3.7
SCM15 11.5 0.13 0.53 0.33 1.5 6.2 3.8
YB16 4.4 0.03 0.08 0.89 0.1 0.4 3.9
YB19 4.7 0.00 0.15 0.85 0.0 0.7 4.0
SCM12 16.7 0.13 0.63 0.24 2.1 10.5 4.1
YB7 5.6 0.03 0.21 0.76 0.2 1.2 4.3
YB3 4.3 0.00 0.00 1.00 0.0 0.0 4.3
SCM10 7.4 0.08 0.34 0.58 0.6 2.5 4.3
YB13 6.7 0.06 0.28 0.66 0.4 1.9 4.4
SCM4 8.42 0.07 0.40 0.53 0.6 3.4 4.5
SCM16b 11.5 0.09 0.52 0.39 1.1 5.9 4.5
SCM11 7.8 0.07 0.35 0.58 0.5 2.7 4.5
YB4 4.9 0.00 0.04 0.96 0.0 0.2 4.7
YB9 5.6 0.04 0.10 0.86 0.2 0.6 4.8
YB6 5.5 0.01 0.09 0.90 0.0 0.5 4.9
YB18 5.0 0.00 0.02 0.98 0.0 0.1 4.9
SCM6b 11.3 0.09 0.48 0.43 1.0 5.4 4.9
YB1 6.4 0.01 0.18 0.81 0.1 1.1 5.2
YB14 5.8 0.01 0.09 0.90 0.1 0.5 5.2
SCM1 27.7 0.26 0.56 0.19 7.2 15.4 5.2
SCM7b 15.1 0.12 0.53 0.35 1.8 8.0 5.3
YB15 5.6 0.00 0.00 1.00 0.0 0.0 5.6
YB20 6.5 0.00 0.04 0.96 0.0 0.3 6.2
YB21 7.7 0.00 0.13 0.87 0.0 1.0 6.7
YB17 7.5 0.01 0.08 0.91 0.1 0.6 6.8
YB11 8.4 0.06 0.09 0.85 0.5 0.8 7.1
SCM2 63.5 0.47 0.40 0.13 30.1 25.1 8.3
YB12 10.1 0.09 0.06 0.85 0.9 0.6 8.6
31
619
620
Figure captions
Fig. 1 Monthly time-dependence of DOC concentration [DOC] and absorbance for
experimental mesocosms; see Methods for experimental treatments. Measured and modelled
[DOC] are shown on the primary (left) axis, represented by hollow and filled squares,
respectively. Absorbance values at 270 nm and 350 nm are on the secondary (right) axis,
represented by filled and hollow triangles, respectively. Mesocosm 4 = panel A, mesocosm 7
= panel B, mesocosm 15 = panel C and mesocosm 20 = panel D.
Fig. 2 Extinction coefficients (E) at 270 nm (A) and 350 nm (B) plotted against measured
[DOC] for the mesocosms. Samples were collected between February and August 2015.
Fig. 3 Comparison of DOC concentration [DOC] estimated using the Carter model with
measured [DOC] for all samples collected in this study. Hollow circles represent the
mesocosm samples and triangles the field sites. Filled triangles show 7 Shropshire – Cheshire
meres sites that were not satisfactorily explained by the Carter model. The 1:1 line is shown.
Fig. 4 Comparison of DOC concentration [DOC] estimated using the Carter model with
measured [DOC] for the Shropshire Cheshire mere water samples (triangles) and Chinese
lakes (Zhang et al. 2005; hollow squares).The filled triangles show the 7 mere sites that were
unsatisfactorily predicted by the Carter model. The 1:1 line is shown.
Fig. 5 The fraction of the variable component C2 (fC2) vs the [DOCC2] for UK field sites
(Table 1) and the Yangtze basin (YB) samples. Category PR (Pennine Reservoirs) values are
not plotted because all fC2 values were close to zero (Table 3).
32
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
Figure 1
33
645
646
647
648
649
650
651
652
653
654
655
656
657
658
Figure 2
34
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
Figure 3
35
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
Figure 4
36
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
Figure 5
37
717
718
719
38
720
39
721
... This unexpected decrease in TOC removal efficiency during sedimentation could be attributed to the activities of microorganisms involved in the formation and transformation of organic matter, which are more prominent during the high temperatures of the dry season (Abdelrady et al. 2019). The presence of algal blooms and sludge formation during the sedimentation stage can also contribute to increased TOC levels due to nutrient loading and sufficient residence time (Volk 2001;Adams et al. 2018;Wurtsbaugh et al. 2019). Additionally, the resuspension of settled organic matter and the desorption of organic compounds from suspended solids in the sedimentation stage can lead to a net increase in the TOC concentration in the effluent from this treatment stage. ...
Article
Full-text available
This study evaluated the seasonal performance of the Koka water treatment plant in removing natural organic matter (NOM) and the implications for disinfection by-product (DBP) formation potential. Raw and treated water samples were collected during the dry and wet seasons and analyzed using physicochemical parameters and fluorescence spectroscopy. The results revealed significant seasonal variations in raw water (RW) quality, with higher turbidity, pH, temperature, conductivity, total organic carbon (TOC), dissolved organic carbon (DOC), and UV254 absorbance during both seasons and across treatment processes. The overall NOM removal efficiency of the treatment plant was poor, with mean TOC removal of 46 and 43% and DOC removal of 15.8 and 15.2% during dry and wet seasons, respectively. The sedimentation unit demonstrated negative TOC removal, indicating NOM accumulation likely due to biochemical reactions in the unit. Fluorescence analysis and the correlation between specific ultraviolet absorbance (SUVA) and DBP formation potential suggest a higher risk of DBP formation in chlorinated drinking water. These findings highlight the influence of seasonal variations, RW quality, and the treatment process dynamics on the plant's performance in removing NOM. There is a need to implement adaptable strategies to enhance NOM removal, accounting for seasonal fluctuations in RW quality.
... When eutrophication results from the rapid growth of phytoplankton, the content of proteins and humic substances rises dramatically (Fukuzaki et al., 2014;Romera-Castillo et al., 2011). According to a different study, the synthesis and importance of components that resemble humus were intimately linked to the growth and degradation processes of phytoplankton (Adams et al., 2018). Natural organic matter makes up the majority of DOM in pristine lakes, with humic substances being the most prevalent type (Henderson et al., 2009;Hudson et al., 2007). ...
... The BIX value was the highest in March, while the chlorophyll concentration was the lowest, indicating the biodegradability of DOM and the proportion of newly produced DOM increase. Algae, as one of the sources of DOM, released DOM from cells during growth and death (Adams et al., 2018). When the explosive growth of phytoplankton leads to eutrophication, the content of proteins and humic substances in DOM increased significantly. ...
Article
Dissolved organic matter (DOM), as the primary form of organic carbon in lakes, is crucial for the lake carbon cycling. As an important component of lake ecosystems, phytoplankton plays a crucial role in carbon fixation. However, we are not yet clear about the contribution of phytoplankton composition to DOM. To clarify this issue, we conducted we conducted continuous monitoring for 6 months in Lake Erhai, a plateau lake in China. Through quantitative detection of phytoplankton and spectral characteristic analysis, four fluorescence components were identified: tryptophan-like (C1), tyrosine-like (C2), fulvic-like (C3), and humic-like (C4). The Fluorescence Index (FI) ranged from 1.70 to 2.54, and the Biogenic Index (BIX) was above 0.8, indicating the DOM in Lake Erhai’s overlying water was predominantly autochthonous. Cyanophyta prevailed in September and November, while Chlorophyta dominated in other months. Phytoplankton diversity was significantly related to DOM concentration that may be related to the difference in the DOM contribution of different phytoplankton. The effect of phytoplankton biomass on C3 and C4 was prominent. Whereas, phytoplankton did not stimulate C1 and C2 significantly, indicating that protein-like may originate from terrestrial input and microbial activity. The composition of phytoplankton significantly affects the DOM composition in Lake Erhai. Therefore, the autochthonous DOM production by phytoplankton warrants further consideration in plateau lake systems.
... Thus, the CDOM will not always explain the DOC variation sufficiently. Adams et al. (2018) observed that the fraction of non-colored DOC derived from algae is variable and can assume high values in eutrophic lakes. This study showed that the estimated algal-derived DOC concentrations in 77 surface freshwater samples ranged from 0 to 8.6 mg L-1, and the algal-derived DOM fraction ranged from 0 to 100%. ...
Article
Full-text available
Inland waters are important components of the global carbon cycle as they regulate the flow of terrestrial carbon to the oceans. In this context, remote monitoring of Colored Dissolved Organic Matter (CDOM) allows for analyzing the carbon content in aquatic systems. In this study, we develop semi-empirical models for remote estimation of the CDOM absorption coefficient at 400 nm (aCDOM) in a tropical estuarine-lagunar productive system using spectral reflectance data. Two-band ratio models usually work well for this task, but studies have added more bands to the models to reduce interfering signals, so in addition to the two-band ratio models, we tested three- and four-band ratios. We used a genetic algorithm (GA) to search for the best combination of bands, and found that adding more bands did not provide performance gains, showing that the proper choice of bands is more important. NIR-Green models outperformed Red-Blue models. A two-band NIR-Green model showed the best results (R² = 0.82, RMSE = 0.22 m⁻¹, and MAPE = 5.85%) using field hyperspectral data. Furthermore, we evaluated the potential application for Sentinel-2 bands, especially using the B5/B3, Log(B5/B3) and Log(B6/B2) band ratios. However, it is still necessary to further explore the influence of atmospheric correction (AC) to estimate the aCDOM using satellite data.
Article
Full-text available
Advanced oxidation processes (AOPs) are one of the highly effective alternatives for treatment of algal toxins in drinking water. Water that contains algal toxins commonly has organic matter of algal...
Article
Dissolved organic carbon (DOC) is a powerful regulator of the ecotoxicity of ciprofloxacin (CIP), a widely applied fluoroquinolone antibiotic. In this study, we investigated the impact of DOC from a variety of sources on CIP bioavailability, using a cyanobacteria growth inhibition test with Microcystis aeruginosa. We analyzed the impact from two perspectives: (1) DOC concentration, and (2) water brownness, defined in this work as the light absorbance of DOC solutions. The toxicity tests were conducted with (1) unprocessed freshwater DOC in the naturally occurring state, (2) DOC extracted from a freshwater stream (Schwarzbach stream, Küchelscheid, Belgium), and (3) the commercial DOC product Suwannee River organic matter. Across all DOC sources investigated, a strong negative correlation was observed between CIP ecotoxicity and light absorbance at four wavelengths across the ultraviolet-visible range (e.g., A350), whereas CIP ecotoxicity correlated poorly with the DOC concentration. In addition, the interactions between CIP and DOC were modelled as a CIP-DOC binding process to allow the quantification of the inhibitory effects of DOC on CIP toxicity via binding constants (Kd,CIPx, with x being the ionic charge + or +/-, L g-1). Processed DOC sources showed higher binding potency than most of the unprocessed DOC sources, suggesting that toxicity tests employing only processed DOC potentially overestimates the impact of DOC in natural environments. Nonetheless, the light absorption coefficient (i.e., ε350) appeared a reliable predictor of the Kd,CIP+/- (and thus of the potential of the DOC source to reduce ecotoxicity of CIP) of both processed and unprocessed DOC. The relationship can be further incorporated into model simulations to estimate CIP bioavailability in dynamic environments. It is concluded that the brownness of water is a better predictor of the impact of DOC on CIP bioavailability than the DOC concentration itself.
Article
The land‐ocean dissolved organic carbon (DOC) flux represents a significant term within the global carbon budget, with peatland‐dominated regions representing the most intense sources of terrestrial DOC export. As the interface between freshwater and marine systems, estuaries have the potential to act as a filter of the land‐ocean carbon flux, removing terrestrially derived DOC, which is present at low concentrations in the oceans, via a combination of physicochemical and biological processes. However, the fate of peat‐derived DOC within estuaries remains poorly quantified, partly due to the complicating influences of heterogeneous soils, land‐use, point sources, and upstream modification of organic matter. To minimize these modifying factors, we studied DOC and inorganic nutrients in four small, peat‐dominated, minimally disturbed, and oligotrophic Falkland Island estuaries. Contrary to expectations, we found limited evidence of physicochemical estuarine DOC removal, and instead observed apparent “hot zones” of biogeochemical activity, where terrestrially‐derived silicate mixed with inorganic nitrogen and phosphorus entering the estuaries from the nutrient‐rich marine ecosystem. In two estuaries, this coincided with apparent in situ DOC production. We suggest that the observed phenomena of marine nutrient subsidy of estuarine productivity, and flexible utilization of multiple nutrients within the oligotrophic system, may once have been widespread in temperate estuaries. However, this function has been lost in many ecosystems due to catchment eutrophication by agricultural and urban development. We conclude that the estuaries of the Falkland Islands provide a valuable pre‐disturbance analogue for natural biogeochemical functioning in temperate estuaries receiving high organic matter inputs.
Article
Full-text available
Over the last several decades, the frequent occurrence of algal bloom in drinking water supplies, driven by increasing anthropogenic input and climate change, has posed serious problems for membrane filtration processes, resulting in reduced membrane permeability and increased energy consumption. It is essential to comprehensively understand the characteristics of algal dissolved organic matter (DOM) and the subsequent effects on the filtration processes for better insight into membrane fouling mitigation. Many studies have revealed that algal DOM has displayed unique characteristics distinguished from other sources of DOM with respect to the chemical composition, the structures, and the molecular weight distributions. Algal DOM is considered to be a major obstacle in understanding membrane fouling due to its complicated interactions among dissimilar algal DOM constituents as well as between algal DOM and membrane material matrices. The present review article summarizes (1) recent characterizing methods for algal DOM, (2) environmental factors affecting the characteristics of algal DOM, (3) the discrepancies between algal DOM and other sources of aquatic DOM, particularly terrestrial sources, and (4) potential fouling effects of algal DOM on membrane filtration processes and their associations with algal DOM characteristics. A broad understanding of algal DOM-driven membrane fouling can lead to breakthroughs in efficient membrane filtration processes to treat algal bloom water sources.
Article
Full-text available
Dissolved organic matter (DOM) of 27 prairie saline lake ecosystems was investigated in the Northern and Central Great Plains of the United States using absorbance, fluorescence, lignin concentration, and stable C isotope values. The majority of variation in DOM fluorescence among lakes was due to humic (peak C) and microbially formed (peak M) fluorescent components, which appear to be derived from autochthonous primary production. Strong correlations between peak M and nutrients allow us to model total phosphorus (TP) concentration using peak M fluorescence and chromophoric dissolved organic matter (CDOM) absorption. The rate of primary production (PP) was positively correlated with peak M fluorescence and negatively with lignin concentration. Lignin phenol yields in the DOM were generally smaller than those of most freshwater systems. δ13C values of dissolved organic carbon (DOC) ranged from −25.0‰ to −20.1‰ and were generally enriched relative to typical freshwaters (ca. −27‰). Terrestrial DOM is degraded in prairie lakes, spanning a gradient from mixotrophic to eutrophic, as determined by a color-nutrient model. The photodegradation of autochthonous DOM was significant: CO2 fluxes from these prairie lakes, modeled from peak M fluorescence, ranged from 5 to 228 mmol C m−2 d−1 (median, 37 mmol C m−2 d−1) and was similar to community respiration estimated from protein fluorescence (median, 50 mmol C m−2 d−1). The combined estimates were about 50% of the global mean total C release previously reported for saline lake ecosystems. The implication of these new results is that the global C release from saline lake ecosystems is likely underestimated.
Article
Full-text available
The influence of inland water bodies on the global carbon cycle and the great potential for long-term carbon burial in them is an important component of global limnology. We used paleolimnological methods to estimate changes in carbon burial rates through time in a suite of natural lakes in the US state of Iowa which has watersheds that have been heavily modified over the last 150 years. Our results show increasing carbon burial for all lakes in our study as agriculture intensified. Our estimates of carbon burial rates, before land clearance, are similar to the published worldwide averages for nutrient-poor lakes. In nearly all the cases, burial rates increased to very high levels (up to 200 g C m−2 y−1) following agricultural development. These results support the idea that the increased autochthonous and allochthonous carbon flux, related to anthropogenic change, leads to higher rates of carbon burial. Further, these results imply that the fraction of global carbon buried by lakes will be increasingly important in the future if worldwide trends in anthropogenic eutrophication continue.
Article
Full-text available
1. Lake restoration from eutrophication often rests on a simple paradigm that restriction of phosphorus sources will result in recovery of former relatively clear-water states. This view has apparently arisen from early successful restorations of deep lakes in catchments of poorly weathered rocks. Lakes in the lowlands, however, particularly shallow ones, have proved less tractable to restoration. This study of three lowland lakes provides insights that illuminate a more complex picture. 2. The lakes lie in a sequence along a single stream in a mixed urban and rural landscape. Severely deoxygenating effluent from an overloaded sewage treatment works was diverted from the catchment in 1991. Effects on two lakes, Little Mere (zmax <2 m) and Rostherne Mere (zmax 31 m) were followed until 2002. Mere Mere (zmax = 8 m), upstream of the former works, acted as a comparison for changes in water chemistry. Mere Mere showed no change in total phosphorus (TP), total inorganic nitrogen, or planktonic chlorophyll a concentrations. Increased winter rainfall was associated with higher winter soluble reactive phosphorus (SRP) and ammonium concentrations in its water. 3. Little Mere changed from a deoxygenated, highly enriched, fishless system, with large populations of Daphnia magna Straus, clear water and about 40% aquatic plant cover, to a slightly less clear system following diversion. Daphnia magna was replaced by D. hyalina Leydig as fish recolonised. Spring peaks of chlorophyll a declined but summer concentrations increased significantly. Annual mean chlorophyll a concentrations thus showed no change. Submerged plants became more abundant (up to 100% cover), with fluctuating community composition from year to year. Summer release of SRP from the sediment was substantial and has not decreased since 1993. The summer phytoplankton was apparently controlled by nitrogen availability perhaps with some influence of zooplankton grazing. SRP was always very abundant. The lake appeared to have reached a quasi-stable state by 2002. 5. Rostherne Mere showed a steady decline in TP and SRP concentrations following effluent diversion apparently as a result of steady dilution by water with lower phosphorus concentration. Decline in phosphorus concentrations was much less rapid than expected because of internal remobilisation from the hypolimnion and sediments. There have been no changes in chlorophyll a concentration or of nitrogen availability and by 2002 the phytoplankton probably remained limited by a combination of mixing, grazing and nitrogen. 6. A seeming paradox is, thus, that immense changes in phosphorus budgets have shown no consequences for phytoplankton chlorophyll concentrations in either of the lakes, although the seasonal distribution has altered in Little Mere. Although these case studies deviate from others, for both shallow and deep lakes, they represent distinctive situations rather than undermining conventional models.
Article
Full-text available
Environmental context. Dissolved organic matter (DOM) is part of the global carbon cycle, ecologically and geochemically active, and costly to remove in water treatment. Spectroscopic monitoring at a single wavelength provides some indication of DOM concentration, but variations in optical properties mean that accurate determinations currently rely on slow and costly laboratory methods. We show that for water samples containing non-anthropogenic DOM, ultraviolet absorbance at two wavelengths can quantify DOM rapidly, cheaply and accurately, and also indicate its quality. Abstract. The precise simulation of ultraviolet absorption by 23 contrasting surface-water DOM samples was achieved with a model based on two components, one absorbing light strongly (A) and the other weakly (B). The parameterised model can be used to predict [DOC] in water samples simply from absorbance values at two wavelengths, while information on DOM quality is provided by the calculated fractionation into A and B. The model was tested by predicting [DOC] for a separate dataset obtained by combining results for 12 samples each from surface waters in the UK, Canada and the USA, and from UK groundwaters. A close correlation (R² = 0.997) was obtained, with only slight underestimation of the true [DOC]. The proportions of components A and B varied considerably among the sites, which explains why precise prediction of [DOC] from absorbance data at a single wavelength was not possible. When the model was applied to samples collected from river locations in a heterogeneous UK catchment with areas of industry and high human population, [DOC] was underestimated in many cases, which may indicate the presence of non-absorbing pollutant DOM.
Article
UV spectrophotometry is largely used for water and wastewater quality monitoring. The measurement/ estimation of specific and aggregate parameters such as nitrate and dissolved organic carbon (DOC) is possible with UV spectra exploitation, from 2 to multi wavelengths calibration. However, if nitrate determination from UV absorbance is known, major optical interferences linked to the presence of suspended solids, colloids or dissolved organic matter limit the relevance of UV measurement for DOC assessment. A new method based on UV spectrophotometric measurement of raw samples (without filtration) coupling a dual pathlength for spectra acquisition and the second derivative exploitation of the signal is proposed in this work. The determination of nitrate concentration is carried out from the second derivative of the absorbance at 226 nm corresponding at the inflexion point of nitrate signal decrease. A short optical pathlength can be used considering the strong absorption of nitrate ion around 210 nm. For DOC concentration determination the second derivative absorbance at 295 nm is proposed after nitrate correction. Organic matter absorbing slightly in the 270e330 nm window, a long optical pathlength must be selected in order to increase the sensitivity. The method was tested on several hundred of samples from small rivers of two agricultural watersheds located in Brittany, France, taken during dry and wet periods. The comparison between the proposed method and the standardised procedures for nitrate and DOC measurement gave a good adjustment for both parameters for ranges of 2e100 mg/L NO3 and 1 e30 mg/L DOC.
Article
We developed a model of the biogeochemical and sedimentation behaviour of carbon (C), nitrogen (N) and phosphorus (P) in lakes, designed to be used in long-term (decades to centuries) and large-scale (10(4)-10(5)km(2)) macronutrient modelling, with a focus on human-induced changes. The model represents settling of inflow suspended particulate matter, production and settling of phytoplankton, decomposition of organic matter in surface sediment, denitrification, and DOM flocculation and decomposition. The model uses 19 parameters, 13 of which are fixed a priori. The remaining 6 were obtained by fitting data from 109 temperate lakes, together with other information from the literature, which between them characterised the stoichiometric incorporation of N and P into phytoplankton via photosynthesis, whole-lake retention of N and P, N removal by denitrification, and the sediment burial of C, N and P. To run the model over the long periods of time necessary to simulate sediment accumulation and properties, simple assumptions were made about increases in inflow concentrations and loads of dissolved N and P and of catchment-derived particulate matter (CPM) during the 20th century. Agreement between observations and calculations is only approximate, but the model is able to capture wide trends in the lakewater and sediment variables, while also making reasonable predictions of net primary production. Modelled results suggest that allochthonous sources of carbon (CPM and dissolved organic matter) contribute more to sediment carbon than the production and settling of algal biomass, but the relative contribution due to algal biomass has increased over time. Simulations for 8 UK lakes with sediment records suggest that during the 20th century average carbon fixation increased 6-fold and carbon burial in sediments by 70%, while the delivery of suspended sediment from the catchments increased by 40% and sediment burial rates of N and P by 131% and 185% respectively.
Article
This study emphasizes the importance of rainstorm events in mobilizing carbon at the soil-stream interface from tropical rainforests. Half-hourly geochemical/isotopic records over a 13.5 hour period from a 20 km2 tropical rainforest headwater in Guyana show an order of magnitude increase in dissolved organic carbon (DOC) concentration in less than 30 minutes (10.6-114 mg/L). The composition of DOC varies significantly and includes optically invisible dissolved organic matter (iDOM) that accounts for a large proportion (4-89%) of the total DOC, quantified using size exclusion chromatography (SEC). SEC suggests that iDOM is comprised of low molecular weight organic moieties, which are likely sourced from fresh leaf litter and/or topsoil, as shown in soils from the surrounding environment. Although poorly constrained at present, the presence of iDOM further downstream during the wet season suggests that this organic matter fraction may represent an un-quantified source of riverine CO2 outgassing in tropical headwaters, requiring further consideration.
Article
This book focuses on the important binding properties of these compounds which regulate the chemical reactivity and bioavailability of hydrogen and metal ions in the natural environment. Topics covered include the physico-chemical properties of humic matter and interactions of protons and metal cations with weak acids and macromolecules. Experimental laboratory methods are also discussed, together with mathematical modelling. Finally the author looks at how the results of this research can be used to interpret environmental phenomena in soils, waters and sediments.
Article
Dissolved organic carbon (DOC) is a major component of both marine (23 muM) and continental (161 muM) rain, present in concentrations greater than nitric and sulfuric acids combined. Rain is a significant source of DOC to surface seawater (90×1012gCyr-1), equivalent to the magnitude of river input of DOC to the open ocean and half the magnitude of carbon buried in marine sediments per year on a global scale. Current models of global carbon cycling focus primarily on inorganic forms of carbon and are unable to account for approximately 20% of the global carbon dioxide, suggesting a significant missing carbon sink. Quantification of the average DOC concentration in marine rain allows calculation of the global rainwater flux of DOC of 430+/-150×1012gCyr-1. When inorganic carbon is included, this rainwater carbon flux becomes 510+/-170×1012gCyr-1, which, although not the same carbon, is equivalent in magnitude to over one third of the missing carbon sink.